95 LT-1 Idle Cell Comparison - Humidity? [Archive]

spfautsch

08-03-2019, 09:40 PM

Been struggling with this problem for roughly 2 months. After a bunch of experimentation I started looking for compression loss because it was struggling to start and stay running, and most of the time died when it switched to closed loop.

While I'm not leaning toward the ignition controller being the cause, as most of you know I'm running my experimental coil per cylinder controller (link in sig). It seems to run great once up to temperature, but it is certainly possible the ignition controller is a contributing factor.

At this point I can barely get the thing to start - it seems like it's not getting enough air. I've drilled the IAC bypass hole from 11/32 to 1/4 inch with little change. Cold starts barely stay running with the IAC pegged at 160 until the engine warms up. Once warm IAC counts are around 80 at 950 rpm.

So I'm looking for a simple explanation because the problem only appeared as the humidity started to climb right after the 43kpa frame was captured. Note the huge difference in metered airflow. I know the barometric pressure is higher in the 43kpa frame, but so is air temp. Aside from this the biggest difference between the conditions the data was captured in is that the 43kpa frame was from late spring with relatively low humidity (and 11/32 IAC bypass hole). On the other hand the 48kpa frame was captured a week ago in 65-70% relative humidity and 17/32 bypass hole. Can humidity and air density account for this much difference? All the MAF logs I have to compare to are 2+ years old and I couldn't possibly guess what the relative humidity was.

14445 14446

Edit: I failed to mention that this is the first summer I've run this after adding the stroker crank (384ci - 4.040" x 3.750")

kur4o

08-03-2019, 10:56 PM

I vote for compression loss due to valves not fully closed.
I guess adding more fuel didn`t cure the problem. Iac counts seems normal.
On the second log engine is not making good vaccuum, possibly the result of lost compression.

Humid air is heavier and harder to move. It takes more effort to move the same amount of air. It doesn`t seem to have that much impact as you expect. If you suspect engine chocking, enlarging the 8 ports on the intake manifold is better than drilling the bypass hole.

I hope your individual cylinder trims are not off chart. They do contribute to engine surging and stalling.

spfautsch

08-04-2019, 12:14 AM

I ruled out compression loss. Engine was torn down to replace head gaskets and installed solid lifters.

More fuel isn't generally the direction I felt it needed to go when it smells so obscenely rich as it is. This is why I was suspecting loss of compression and / or very weak spark. But I just can't imagine the spark is weak enough to cause incomplete combustion.

I've been dabbling with the startup fueling and the injector flow constant since the initial post and have got it starting and idling relatively well from warm (approx 60c coolant temp). I really need to buy some known good / unmodified injectors because I've added an entire lb/hr to the setting I was running in the 43kpa frame and it seems to be getting better. I don't know if it's a meaningful measure of anything, but now when it switches to closed loop you can't even feel a difference in idle. A few days ago it would either die or stumble and then switch back to OL for another 90 seconds depending on how warm it was before starting. I guess I'll have to wait until it cools completely to test if I've made any progress in this regard.

The water vapor in humid air also displaces oxygen and contributes to cooling in the cylinder, which is why I'm trying to get a grip on how much this should be impacting things. Bear in mind I'm speaking in terms of "oppressive humidity". It was 82F/28C in the garage this morning when I began working on it, and my shirt was soaked in sweat in 15 minutes.

Individual trims were set using the wideband well before the problem began to present itself. It didn't take much - cyl 6 is at 0.98 and cyl 5 is at 0.99, all others are at 1.0. I haven't re-checked lately but don't suspect this is a major problem.

kur4o

08-04-2019, 12:49 AM

Rechecking valve lash with solid lifters might help with cold hard starting conditions. It takes a little for a valve to hang down.
If it is a start up fueling problem, adjust the cranking VE table. It is reponsible for fueling at startup.

At cold only hesitation, fueling will be the main suspect. The hotter the engine, the less fuel and air it needs to keep the idle.
Bump the fuel all over to see if it cures it, and start looking for good known injectors. These knock off, no brand reworked always make more troubles than help.

spfautsch

08-04-2019, 01:15 AM

I haven't checked hot lash yet, but I doubt I'm losing compression here. Primarily because it acted the same way with hyrdraulic lifters, but also because lash has probably not shrunk during break-in, rather increased.

Startup fueling is getting my complete attention now, as is injector flow constant. I'll report back tomorrow once it cools completely.

Trust me - I've read "the gospel" on modified injectors and couldn't agree more, but I ran these injectors enough when the engine was a 5.7L to know they aren't the problem. If anything increasing displacement should have made them less problematic for startup and idle scenarios.

Thanks for your thoughts nonetheless.

steveo

08-04-2019, 05:05 AM

just for fun lets force your IAC into operating range. crack the throttle plates open until cold start yeilds iac counts <160 and see if things improve. the air will be less even than a properly sized idle passage but that shouldn't affect startup too much.

More fuel isn't generally the direction I felt it needed to go when it smells so obscenely rich as it is.

it takes a lot for it to 'smell' rich. often a 'fuel' smell is a misfire due... lean will hit a misfire state more easily than rich

spfautsch

08-04-2019, 04:48 PM

I'll try that if things aren't improved today. I actually tapped my throttle body to take a 1/4 pipe plug that I'm drilling to make different sized bypass orifices. 1/4" is the largest orifice I can create with the hex socket drive plugs I currently have. Yesterday evening I had a fairly successful start from 38C. IAC was pegged at 160 for the first few minutes but it was at least able to make the target rpm. I was really hoping to hear from someone who's done a 383 LT-1 in a climate that sees swamp-a$$ type humidity like St Louis. Houston, TX and the Carolinas are the only other areas I've been to with comparable humidty.

I didn't want to hear the word "misfire". That points me back in the direction of the ignition controller.

I would completely drop the humidity idea if I didn't also have a problem with my daily driver - a 1.9L turbo diesel at this same time. The vane actuator pot was worn causing the ecu to lose control at higher rpm and go into limp mode where it would command minimum boost all the time. The difference between a 90hp engine trying to climb a moderate incline with little or no boost in May versus July humidity was immense. In 95F air I could very plainly feel the "boost" from drafting semis.

steveo

08-04-2019, 05:49 PM

so the only difference between the two snapshots above was humidity? no other engine modifications?

spfautsch

08-04-2019, 10:00 PM

no other engine modifications?

There were - solid lifters and larger IAC bypass holes. But this problem started as soon as it started to get hot & humid, well before any modifications. Nothing really changed with the addition of the lifters.

Cracking the throttle a bit made a huge difference. The MAF was actually registering slightly less airflow at the same engine speed. I'm working on checking hot valve lash so can't post too many details. Will try a cold start in the morning and post an update.

Rocko350

08-04-2019, 11:21 PM

You're at the point where you need to enlarge the idle passages in the intake manifold.

spfautsch

08-04-2019, 11:33 PM

Here's a short log of a drive this afternoon. In general it's not running very impressively, but as I've mentioned the transition to closed loop was unnoticeable if that means anything. I noticed some popping in the exhaust when decelerating from a few PE "events".

I suspect that the humidity is causing the MAF to register inflated airflow #s. I say this because I've always logged less spark knock in humid conditions, and I'm presuming that's caused by the water vapor increasing the cooling effect on hot / heated parts. I haven't tried swapping to SD mode yet but will if I can get it to start and stay running reliably.

Wideband is a LC-2 on D27 using default settings. I'm not sure how meaningful any of the data is except to use for cylinder balance testing. I pay it almost no attention unless I drive the car a few minutes to get the exhaust hot. Even then, Missouri's laws on ethanol are stupid - it's considered required in some concentration but they don't have to tell you how much or that it's even present. I'm guessing the 93 octane I'm running has a stoich about 14.5:1.

By the way, with the help of a definition file I found in kur4o's build folder I found I could display the wideband units in Lambda with this in the definition.csv:

54,WIDEBAND_D27,Wideband O2 @ Pin D27,^,EXTENDED,8,2,0.007874016,,0x01,0xFF,,,,,

Rocko350

08-04-2019, 11:56 PM

Rocko350

08-05-2019, 12:04 AM

kur4o

08-05-2019, 12:52 AM

That 02s and BLM reading looks weird enough. Did you mess with the closed loop settings. Does the AIR system present and functioning. 02s looks frozen at stratup and really slow moving after that. They should be pig rich at startup and I noticed blms were 120 at open loop meaning they are pulling fuel.

You can set mode4 controls before starting the engine. Like setting target afr and spark for startup.

Humidity revealed some hidden problem you already had. Is it a tune or mechanical.
Humid air takes alot more energy to ignite, You lose about 10-20% power and there is water vapor condensation inside the cylinder untill it gets warmed up, which can foul the spark plugs with water.
But that is likely to happen at much lower temperatures.

The engine gets less efficient and draws more volume for the same amount of oxygen. That can explain the maxed out IAC counts at stratup.

To elimimnate CL tune problem I will start by running Open loop and than CL SD and OL SD.

It could be also 2 different problems mixed up.
1 is hard startup, which can be the result of spark syncronisation, Crank ve table, or slower fuel pressure build up.
2 is crap runninng untill warmed up, the result of weak spark, CL setting off chart, MAf table off chart.

spfautsch

08-05-2019, 12:52 AM

You're at the point where you need to enlarge the idle passages in the intake manifold.

Anyone have a second on this opinion? I'm thinking the same but damned if I didn't just put the manifold back on 3 weeks ago.

What would baro tell me about? I thought $EE only uses SD if the MAF stops working.

kur4o

08-05-2019, 01:17 AM

At startup the IAC are set at 144, I think that setting can be increased somewhere in the bin. The intake passage will affect only idle quality and you should eliminate every other possible reason before doing it.
Too much and you will never have a good idle again, The individual cylinder trims will get all around and will need a fresh tune from zero. Even GM made different size holes to compensate for the unequal air distribution.
Once I made the mistake to make the holes the same size and the idle was crap with constant surging. Only fine tuning ind cyl trims make it run stable.

Humidity hits the ignition system to the limit, so it will be good to take a look at that first.

spfautsch

08-05-2019, 01:18 AM

steveo

08-05-2019, 01:36 AM

the maf isn't really good with air density, im sure it uses barometric pressure to tweak fueling in some way in MAF mode (although i haven't looked at the code). either way ~1 KPA of baro difference shouldnt' do this to you. something else is going on.

spfautsch

08-05-2019, 02:16 AM

the maf isn't really good with air density

I'll have to dig up where I read this, but the word "mass" is used in the name because it's designed to measure air mass through the sensor. This implies it compensates for temperature and pressure by design. I'm going to have to dig but I believe this was the same source that mentioned humidity being the only thing a maf can't compensate for, only the O2s could do that. It might have been a Jeff Hartmann book or it might have been something I found online.

I'm going to work with the throttle blades to give it more air at startup and try to tune around that b/c it responded very well there. I suspect most of my other problems stem from moving the injector constant too far.

I'm also going to bump up the dwell in my controller to see if anything improves because I'm not sure where I came up with the dwell figure I used for these coils. I'm 99% confident there are no other issues with the ignition system. When the air was cooler and drier it was firing immediately every time and idling better than it ever did with the mechanical distributor. When I first saw 42-43kpa at idle I had to double-take on it because the last I recall seeing that good of idle vacuum was when it was stock.

Thanks for all the ideas guys.

I'd really like to know about the BLMs question two posts back. If no-one is sure I'll have to dig into the disassembly.

Terminal_Crazy

08-05-2019, 12:46 PM

I’ve lowered my IAC values at startup around 40 i think.
Thereseems to be startup values in tables but it usualy starts as 160 ish.

I also reduced the air bleed size in the throttle body after reducing the idle timing as the iac valve couldn’t go lower.

spfautsch

08-05-2019, 06:57 PM

Mitch am I mis-remembering, or did I read somewhere that you'd also drilled out your port orifices (the holes pointing at the injector tips)?

It sounds like there might be some misunderstanding and confusion on the subject - the how-to I've seen about setting a base bypass hole size and adjusting the throttle blades fails to mention that it should be done at maximum air density for the climate you drive in. For example if you're in a cold weather season and consequently high air density, you want lower IAC counts - say in the 20s at hot idle. This is the point where the engine needs the least amount of idle airflow.

I'm at the other extreme of the spectrum. In minimum air density situations you want enough bypass air to reach your idle target speed during a cold start without maxing out the IAC motor (160). This is when the engine needs the maximum amount of idle airflow.

Thereseems to be startup values in tables but it usualy starts as 160 ish.

If you're logging 160 IAC counts during cold starts it's an indication that the engine speed is below your idle target and the ecu has opened the IAC as far as it can trying to correct the situation with more airflow.

All this of course assumes your IAC position has been learned by the ecu. There's no feedback mechanism for the pintle position. The ecu has a relearn procedure it uses to extrapolate when the pintle is fully closed and then stores that position (not sure where or how) and tracks it whenever it commands a change in IAC position. But if you replace your IAC motor or the position is otherwise lost by the ecu, the counts aren't going to be very meaningful until the relearn routine runs. I other words, if your ecu thinks the IAC motor is at 50 and it's actually fully open, there's no way for it to know and it will continue to command it open in steps until the ecu's tracking variable is also at 160.

Terminal_Crazy

08-05-2019, 09:10 PM

Hiya Scott,
Yes I drilled the manifold bleeds, no 2 i think is smaller of the top of my head. I drilled them as the drilling was crap, flash left in so needed deburring and they come out all over the sides of port rather than consistantly at the bottom like the chevy manifold.
Throttle body was drilled originally to get iac counts down. As timing was increased vac would improve and counts went down. As the tune and idle has improved (timing lowered) counts started dropping very low and ive fitted a restrictor back in. That’s why i suggested not drilling until your idle settles.

I’ll check logs tomorrow but i’m sure this always starts with counts around 160 and drops. Idle starts about 1100 rpm when cold. I’ve replaced iac and done relearns etc, iac vs temp tables all lowered and remembered values don’t seem to work.

Mitch

spfautsch

08-05-2019, 10:43 PM

What are your ambient temps like this time of year?

spfautsch

08-06-2019, 05:42 AM

Had to play used car salesman tonight so no progress on any front of interest. Thought I'd share this tidbit I found - not the source I recall finding years ago, but makes me feel better knowing I'm not hallucinating memories / facts / whatever.

From this page talking about some newer humidity sensor equipped MAFs being used on the newest Duramax engines.

https://gearsmagazine.com/magazine/a-breath-of-fresh-air-8-pin-mass-airflow-sensors/

Humidity has a major impact on the density of the air and the amount of oxygen contained in the air column traveling into the intake manifold. Humidity takes up space, leaving less room for oxygen molecules.

Standard 5-pin MAF sensors are unable to compensate for the water vapor in the air, which leads to rich air fuel ratios when high water vapor conditions are present.

Terminal_Crazy

08-06-2019, 05:19 PM

What are your ambient temps like this time of year?
Anywhere from 5C to 35C. Generally it’s very cloudy so in 20’s. 30 ish when the sun comes out.

spfautsch

08-06-2019, 10:15 PM

We generally don't see lows near 5C from about mid May to mid October. I'm guessing your average monthly temp (i.e. several hundred samples per day averaged over 30 days) is a solid 3-6C lower. That doesn't sound like much but in terms of air density it is.

Whatever the case, I think a case of mistaken identity has been a major contributing factor. I won't be able to test until later tonight, possibly even tomorrow but I think the combination of all of the aforementioned factors have stacked up to cause me the severe cold start problems.

This, boys and girls is the part of the show where Scott eats his hat in front of you all. :-) Anyone have any ketchup?

After a bit of research it seems like this coil, which is what I have:

http://www.pfautsch.com/wp-content/uploads/GM12658183-150x150.jpg (http://www.pfautsch.com/wp-content/uploads/GM12658183.jpg)

Is not one of the supposed "hot" or "truck" LS coils. Seems it's electrically similar to the common LS coils and needs a target dwell of 6ms.

Whereas this coil which looks quite similar:

http://www.pfautsch.com/wp-content/uploads/GM12573190-150x150.jpg (http://www.pfautsch.com/wp-content/uploads/GM12573190.jpg)

Is a "hot" or "truck" LS coil and needs a target dwell of 4.5ms.

Anyone want to guess what target dwell I was using?

Terminal_Crazy

08-07-2019, 12:50 AM

Not the one you thought! ��

Would anyone consider opening the manifold idle air passages up any bigger?....

Went to the track on Saturday. (1/8) Had a fantastic day but traction was diabolical.
Spun all through first even feathering the throttle and clutch. It’s never that bad on the road.
Best 9.18 @81. 60’ 2.399 so more mods over the winter.
However the rear seal on the manifold needs resealing after I blew it out arsing about with pcv so I’ve just ordered manifold gaskets and need to pull it off again.

Mitch

spfautsch

08-07-2019, 03:07 AM

Oh no, it was exactly what I'd intended it to be. I'd just labored through the whole diy-ltcc project under the assumption the coils I have were the 4ms "hot" ones. I wish there were datasheets available for these items like injectors.

Unfortunately bumping dwell up didn't miraculously solve any problems. Cold start (albeit from 33C and ~50% RH) was pretty much the same. IAC was pegged until about 4 minutes in. O2s hovered around 460-560mv and the ecu dropped out of CL for a frame or two numerous times during the 5-1/2 minute log. So I guess I'm back to the drawing board.

I think I'll try a SD flash and see what happens.

When / if I ever get this figured out I intend to experiment with different sized bypass orifices to answer the question of whether enlarging the port feed holes is in any way necessary / beneficial. In fact at this point I have nothing to lose by removing my 1/4" npt orifice plug (leaving a 1/2" bypass hole) and see if it doesn't let the engine run away when started.

vilefly

08-07-2019, 08:22 AM

Well, I was going to say, "unplug the MAF, tune for speed density", but you beat me to it.

All this talk about humidity being a major factor reminds me of when I determined the exact nature of the MAF sensor. A typical sensor heats an element 100 deg C above ambient temp to determine the specific heat of the airflow. This crudely compensates for humidity as well, but not in an exacting way. So technically, the output of the sensor is less sensitive when it is humid out, since the cooling effect of humid air is less (and also less o2 by way of water vapor displacing the usable oxygen). Crude, but effective for the emissions at the time of manufacture.
I just hope the MAF is accurate at low RPM with a choppy idle, but I suspect not. Tinkering with the MAF calibration might be viable, only if the sensor is still good. I say clean the MAF sensor, just in case. I suspect the MAF sensor won't be used to a 383 sucking on it as opposed to a 350. Best to eliminate it for now, then come back to it.

spfautsch

08-07-2019, 05:48 PM

Hey vilefly - have any plans this Saturday? Care to make a trip to central misery? I'll pay for your fuel. I could use some help measuring dwell time and led and coil igniter current draws with my circa 1984 CRT oscilloscope.

I'm nearly tempted to ask the mods to delete this thread. There's enough inaccurate information in it regarding the nature of mass air flow sensor operation that it's quite rudely reminded me why I hate asking for help.

All I was hoping for is someone who perhaps

a) also has a MAF enabled 383 gen 2 LT-1

and

b) lives in a similarly hot and humid climate

to chime in with any thoughts about the idle airflow numbers I'm seeing and IAC bypass size so I can assess how much effort I should exert looking for a consistent misfire / weak spark situation.

MAFs aren't evil. I can't believe there aren't thousands of cammed up 6.0, 6.2, and 7.0L LS based cars running around still using the airflow sensor for it's intended purpose. However, those guys all have the benefit of ETC to control idle air, so there's no need for an IAC motor, or a bypass. I can however believe no-one is doing the same with a 6.3L gen 2 LT-1, because most of the tuning "experts" encourage disabling the MAF and using speed density only. Things like this are what drive me to drink.

kur4o

08-07-2019, 08:16 PM

I should say you are running really lean on startup.
Don`t forget you are spraying cold fuel straight in the cylinder, There is not enough time to evaporate and it should be extra rich to run smooth.
The engine doesn`t make enough power and wants to stall, drawing more unused air, making things worse, that`s why you are seeing 160 iac and high map and low rpm. The engine is fighting to run.
It gets a liitle better when the 02s start to climb.

Compare that with high humidity and possible weak spark not having enough energy to ignite the lean humid mixture.

Engine is not making enough power to stay running.

Maf doesn`t play at engine starting at all it is the crank VE table that matters.

And that 02s are like frozen they should be jumping like mad from rich to lean. There is something very bad happening there. Load the stock CL settings and see if there is some improvements.
If you still can`t dial the idle closed loop, I have a remedy called Open loop idle patch.

You should also investigate the controversial readings between narrowband 02s and wideband readings.

spfautsch

08-07-2019, 09:54 PM

I should say you are running really lean on startup.
...
The engine doesn`t make enough power and wants to stall, drawing more unused air, making things worse, that`s why you are seeing 160 iac and high map and low rpm. The engine is fighting to run.
It gets a liitle better when the 02s start to climb.

Compare that with high humidity and possible weak spark not having enough energy to ignite the lean humid mixture.

Engine is not making enough power to stay running.

Now you're making sense. I'll give it a try.

Maf doesn`t play at engine starting at all it is the crank VE table that matters.

I understand that - the startup / cranking fueling tables are well out of the picture after the first 8-12 logged frames of startup. But immediately after those initial frames the MAF and Injector constant pretty much run the whole show. Agreed?

And that 02s are like frozen they should be jumping like mad from rich to lean. There is something very bad happening there.

As I mentioned previously, I think I've raised the injector constant enough to cause this. In the first post, both frames (43kpa and 48kpa) were taken with 42.036 for the injector constant. It's currently at 43.547. I suppose it's possible the injectors flow less than 42, or my dwell problem was wholly to blame and I've moved it from rich (or ideal) to lean fighting what I believed to be a problem of too much fuel and not enough air.

Load the stock CL settings and see if there is some improvements.
If you still can`t dial the idle closed loop, I have a remedy called Open loop idle patch.

If you would, take a look at these logs (http://www.pfautsch.com/wp-content/uploads/drivelog-05-31.zip) and critique the closed loop behavior. The 43kpa idle frame was taken from one of these logs, and the only closed loop setting I've changed since is the Stoich AFR Target.

You should also investigate the controversial readings between narrowband 02s and wideband readings.

I'll worry about that another time. As discussed in other threads it's using the D27 analog input with the default ADC output settings. It's also mounted about 18" downstream from the narrowbands in the X pipe. It doesn't get hot enough from idling alone for me to use for calibration purposes.

Edit: Acutally the x pipe is more like 24" downstream of the header collectors.

Terminal_Crazy

08-07-2019, 11:04 PM

Hi Scott, I know you know what you're doing.

Have you played with injector startup PW and crank VE tables.

Mine cranks right away now cold but doesn't always fire right away when warm... but is ok hot.
I can't tell whether it needs more or less fuel, Probably more.

Just played back a log starting. My IAC starts at 130 & drops to 108 as soon as the motor fires, then drops as motor warms up.
I've lowered most of the tables re IAC and nothing is set that high.
IAC keep alive :20
present motor position loaded at reset : 40
Park position crank offset vs coolant
-4 20
8 21
20 21
32 21
IAC Initial position
-4 77
8 70
20 63
32 56

So no idea where the 130 comes from

IAC relearn done with new iac several times. No change anywhere

I'm going to mix it up with a new Throttle body - yay.

How does AFR & WB compare in EEHack?

Mine sits initially at 3.30:1 at 20C

Jumps to 5.70 as the motor starts turning
Jumps to 12.1 as the motor flares up 13/1400 rpm (iac at 108 so starting to control revs)
at about 30 secs, the wideband starts recording & matching AFR at 12.6

As Rocko350 also mentioned increasing the manifold idle ports, Who would agree it's a good idea?
I've ordered 2 pair of gaskets for the manifold.

Hmm the IAC does seem to control the idle though.
It also depends on idle timing where it sits.
As I lowered idle from mid 30's to 20 I needed to re restrict the bleed hole in the throttle body.
When idle was in 30's the iac was hitting zero and my idle is set at 800.

HTH
Mitch

spfautsch

08-08-2019, 12:19 AM

Have you played with injector startup PW and crank VE tables.

Extensively. That's really a subject for a completely different discussion - the point of this discussion is to make it stay running once it fires. But since all my threads eventually turn into into free-for-alls here's what I've learned.

Mine cranks right away now cold but doesn't always fire right away when warm... but is ok hot.

I had the same problem after my injector swap, before the stroker kit. It would flood when restarting warm. I spent a lot of time fiddling back then and got it "pretty good" restarting warm. Then when I looked at these tables again a couple weeks ago I realized I'd scaled the prime pulse tables in the wrong direction (made them bigger) and then compensated by removing fuel from the cranking VE tables. Totally wrong way to go about it, but it worked.

Look at the prime pulse width tables - this is a raw pulsewidth that does not reference the injector constant. The ecu fires the injectors for this period in batch mode (I believe all 8 at once) when it sees the first low res pulse from the opti. This is to re-establish wall film in the intake ports. So you'll want to scale this table by the relative difference between your stock injector constant and your current injector size. I started with about 25% (42 / 24 = 1.75) or multiplying by 0.75, but I also factored in the additional displacement so ended up around 0.82 as the multiplier. The warm and hot cells of this table will need to be massaged by hand in hex because the stock values are like 0x02, 0x03 and 0x04 and there is very limited resolution for removing an accurate amount of fueling. You'll have to settle for "close".

Then move on to the cranking VE table and use this to add / subtract fuel based on how it behaves in those first few firing events. The fueling calculations behind this table do reference the injector constant so you shouldn't need to factor that in. Just changes in pumping efficiency caused by cam & displacement.

Lastly, the Crank AFR vs Low Res Pulse vs Coolant Temp can be tweaked. This is the target AFR the calculations are using with the cranking VE table for. Notice the first two cranking pulses are much leaner than the rest. I assume this is because the first and second prime pulses aren't factored into the fueling and wall film modeling. I've no idea however if this then adds to the prime pulse or what.

Mine is far from perfect, but I'm not trying to get it perfect yet. This can be a slow & tedious process b/c you have to let it cool down to test.

So no idea where the 130 comes from

That's a question kur4o might be able to answer - he mentioned it a few posts back as being set to 144 but not the address for the table / constant. But long story short the ecu uses IAC for coarse idle speed control, and spark timing for fine control. So it will move the IAC as the ecu sees the need (there's a RPM Variation pid you can monitor / graph in eehack).

How does AFR & WB compare in EEHack?

My WB controller pulls the analog pin to ground until the warmup timer has expired (30 or 60 seconds). So it reads 0 (in units of lambda).

Once it starts reporting AFR it's all over the place. Sometimes showing rich of the narrowbands in OL and then usually 20% lean in closed loop. I'm really not paying it much attention because at cold start, or in any other instances where you have unburned fuel getting in the exhaust the effect is lean output from the O2. For example one of the Banish tuning guides I talked to you about says at cold start you should see the widebands showing stoich because the unburned fuel isn't seen by the sensor. Only the concentration of oxygen in comparison to other gases. Bear in mind his wideband probably costs more than my car, so I doubt we'll be able to use this method to favorable results with a $200 instrument.

As Rocko350 also mentioned increasing the manifold idle ports, Who would agree it's a good idea?

I didn't get in a hurry to pull my intake off to try that, and I don't intend to unless nothing else gives results. I'd like to see some quasi-scientific data that compels me towards that before I jump out of that airplane. I already had a plan to fix my IAC bypass before I decided to drill it. Doing the same with the port feeds is a whole different animal.

kur4o

08-08-2019, 01:01 AM

It looks like the 02s you have are dead shot. Not swinging enough in either direction. They can crack from moisture but yours are completely dead, if something else is not going on.

Get a good afs75 delco ones.

Here are some logs i made showing good voltage swing. The car is perfectly untuned but you can get the idea. There is a cold start and some idle showing the IAC in stock form can easily support 15 AFGS. It has a midlength SLP headers and some tweeks in the delay times. The wideband is set at 0v-11.4 5v-16.5 at d27.

The way you modified the injector scalar can lean the mixture alot. For the sake of experiments set it at 41 and see how it goes.

I will have to dig out the iac tables. Sure there is one for startup counts vs iat or coolant.

spfautsch

08-08-2019, 03:37 AM

It looks like the 02s you have are dead shot. Not swinging enough in either direction.

Are you saying that based solely on analyzing the two big logs from 5/31?

... but yours are completely dead, if something else is not going on.

Trust me, there's the possibility of a great many other "somethings" going on. Not least of which is that this (http://www.pfautsch.com/wp-content/uploads/0912182003.jpg) is the entirety of the calibration data I received with my injectors. Be sure you zoom in because it's hard to see. At the very least I'd like to determine the main injector slope within reason through extrapolation. I know this isn't the way to do this, but hear me out.

In a couple weeks I'll be in a state where I can buy ethanol free gas. I intend to bring 10 gallons back. I'm already braced for my wife's response when she finds out why I'm bringing two empty gas cans along for our 25th anniversary trip.

Assuming I can tweak the ADC output and get my wideband to a spot where I feel it can be trusted to report stoich, I believe I can extrapolate injector flow in a manner more objective than guessing. Obviously I'll want to be forcing OL and AFR to 14.7:1 during this experiment, and running at a load that yields a pulsewidth above the slope knee (aka low pulse adder).

This theory is based on the (relatively huge) assumption that my MAF transfer curve is accurate. But assuming it's in the ballpark, you should now be able to more completely understand my fascination with the effect of humidity on the MAF function. Adjusting for humidity, and assuming I have healthy enough spark for complete combustion, it should be realistic to be able to scale my injector constant based on however rich or lean of stoich the results are.

Get a good afs75 delco ones.

When I get to fine tuning closed loop I'll likely create a new discussion. I might even break down and buy some new ACD O2s. I would be honored and grateful if you'd contribute to it. But that's not my goal for today.

The way you modified the injector scalar can lean the mixture alot. For the sake of experiments set it at 41 and see how it goes.

Based on my faulty assumptions, that was my intention. I've discovered through experience that trying to improve on the factory MAF transfer function isn't a terribly productive exercise.

I'm getting ready to test a cold start with 41.92 for the injector constant. Fingers (and plug wires) crossed...

spfautsch

08-08-2019, 04:45 AM

No miracles were witnessed but I think I learned something. Only change to tune from yesterday was injector constant from 43.xxx to 41.92.

Take a look at this log pls.

Sluggish and pegged IAC from cold again. I did notice however that the BLMs ramped down from 128 to 120 in frames 14-18. If this is affecting OL fueling I'm baffled as to why. I'd just flashed the e-side with the injector constant. I always assumed that flashing a new bin (programming mode / running code from ram) reset BLM data. But even if it didn't, it hasn't run in 24 hours. If the ecu is using stale BLM data to trim startup (edit: meant to reference OL, not cranking) fueling that would probably explain the lame cold start. And it makes me ponder the question "why in the hell did they do that"?

Edit: I've always labored under the assumption that BLM data was NOT being used to trim fueling if the BLM box in eehack wasn't green.

Closed loop acted pretty much the same as before.

At frame 2243 I forced open loop at 14.5:1 and immediately it jumped 50 rpms and the IAC count started falling. O2s started to read rich and exhaust smell changed from eye watering to what I imagine Steve refers to as "chemical fire". I fooled around with forced AFR a bit to see where the wideband reported stoich, but not placing much faith in accuracy when the O2s aren't being heated by moderate driving.

I'm going to reset the BLMs in eehack and run it some tomorrow in forced OL.

vilefly

08-08-2019, 05:23 AM

Wish I could take that trip, spfautsch, but I will be installing a fuel pump in the wagon and getting it ready to pass inspection. Just got the exhaust done, now this. It's doing the vapor lock stumble of death. So much for joyriding in the vette. (sigh)

But in the meantime I will dig up some binaries from a '39 Business Coupe with a 383 lt-1 engine. I had the problem of running rich when cold, and lean at cruise. Can't remember what I changed, so I will compare bins. It had come in with a bad MAF sensor, and an untuned speed density map. It did not run lean like yours, and had a choppy cam. It looked like whoever worked on it didn't really tune it at all. Yours should have behaved in a similar fashion, but instead it is leaning out.
When diagnostics fail, I start the testing back at the beginning again. It hurts, but it gets you there.
Some basic questions perturb me.
1) Timing test - make a timing mark for #1 TDC on the balancer/timing cover, and see if a timing light agrees with your scan readings. Requires a timing light with advance/retard feature or knob (or timing tape on the balancer). It will at least prove the ignition system has no logic errors, and no delays due to dwell mismatch.
2) The bucket test - With the engine running, dump a large cup of water on the intake manifold areas, and watch for misfires or sucking noises. Injector o-rings can be dastardly vacuum leak. So can EGR gaskets.
3) Intake manifold gasket test - remove the pcv valve from the valve cover, but leave it connected to the hose. Plug off the pcv vent and valve ports in the valve covers, and
see if the engine pulls a vacuum in the crankcase when you open one side up. It should show slight positive pressure from blowby. If there is a vacuum, you have a intake gasket problem that will also suck oil into the intake also.
4) O2 test - supply an alternate fuel source into the engine while watching o2 readings. (carb cleaner spray) If you can make them twitch, they work.
5) running compression test - Install compression gauge, start engine, watch for equal readings that do not vary more than 7psi. Will tell you how well the engine is breathing, and whether or not the valves are sealing.
6) Vacuum Reading - should be steady. If you have a bouncing needle, look for a broken valve spring.
7) Map sensor - check with vacuum pump and scan data. Make sure it reads barometric pressure correctly, also. It calculates air density from this, so it can really mess things up if it is wrong. Yours should be making it run rich in speed density mode with a oem tune.

With these basics out of the way, only then can we go after the software. Not that you didn't try all this, but I had to go back to the beginning, since I forget a test or two when I think I know something, and sabotage myself.

vilefly

08-08-2019, 06:08 AM

Ack. It was a 1949 truck with a 383 lt-1 in it.
Here's the original bin and the VE modified @idle speeds.
The pcm/engine combo is from a 1996 firebird.

kur4o

08-08-2019, 11:55 PM

Take a look at this stock log from vette calibration. The 02s are constantly running from 100-150 to 850-900 mv. Yours are barely moving from 300-650

http://www.gearhead-efi.com/Fuel-Injection/attachment.php?attachmentid=14475&d=1565148272

It really doesn`t matter the fuel, the stoich and the tune you have. Unless the blms are maxed out you weill always get the correction close to the swing voltage settings.

spfautsch

08-09-2019, 01:58 AM

Take a look at this stock log from vette calibration. The 02s are constantly running from 100-150 to 850-900 mv. Yours are barely moving from 300-650

That's great, thanks for taking the time to post that.

What I'm really curious about now that I've gone open loop until I figure this out is...

Why does the BLM # drop from 128 to 120 at every start?

I assumed this was stored trims being applied, but I've reset my BLMs numerous times and it still happens every time it's started.

I didn't think it was trimming open loop fueling, but when it's running and I clear BLMs and see them jump from 120 to 128 immediately the IAC drops a few counts.

I'm working on measuring the dwell time of my ignition controller tonight. I'm hoping to find a really dumb math error causing far less dwell than intended, which would explain a weak spark scenario.

Wish I could take that trip, spfautsch, but I will be installing a fuel pump in the wagon and getting it ready to pass inspection.

No worries, it was just a ploy to get you to come out and then once you'd driven 3 hours I'd change my mind and want to re-lash the valves and then change the diff oil in my wife's daily driver. You know, work that's dirty and sucky. ;-) I was going to grill porksteaks and make you drink a couple cold adult beverages though.

Valve lash has been re-set and even made a few laps around town. The good news is after spending $1000 on complete kit, then putting my old pressure plate back in my clutch is working better than it has in 2 years. Two different cheap chinese pressure plates exhibited the same problems - chunky engagement.

I feel ya. Knock on wood, most of my daily driver dramas are sorted for the moment.

spfautsch

08-15-2019, 08:35 AM

Mods feel free to change the thread title to something more appropriate like "Idiot attempts tuing LT-1".

I spent a good bit of time over the weekend testing the ignition system. I haven't yet ruled out weak spark, or some mechanical issue. When I have time I'll outline all the things that have been attempted to those ends such as swapping the original injectors back in, leak testing valve seats, etc. But here's where I'm at right now.

Item #1 - starts and then immediately dies.

I think this might have been VATS. Since disabling it two days ago the cold start behavior has been fairly consistent.

Unfortunately that consistency hasn't been great.. "Cold" starts have still required a bit of initial throttle to keep it from dying and expedite the warmup process.

Item #2 - unknown injector slope

As mentioned, I don't have an overwhelming amount of faith in the actual slope of my injectors, other than it's somewhere between 41 lb/hr and 44 lb/hr @ 43.5 psi.

In an effort to extrapolate the injector slope based on what sensors I have at my disposal, I adjusted my wideband controller to give similar analog outputs to what kur4o mentioned a few posts back. This is what my wideband config looks like currently. Note the -0.200 afr adjustment, due to the fact that I know there is some amount of ethanol in all the pump fuel I'm able to purchase (because water dissolves in it).

http://www.pfautsch.com/wp-content/uploads/wideband-settings-150x150.png (http://www.pfautsch.com/wp-content/uploads/wideband-settings.png)

Using these settings I went for a short drive tonight to get the wideband heated by some load greater than idling, and afterwards ran a cylinder balance test. Log attached.

However, my primary concern is determining injector slope. If there's something I'm overlooking that invalidates my theory here please point it out - I'm very much an amateur at this stuff. But...

Assuming the wideband is within the "ball park", looking at the data between frame # 2580 and 2794 there's a block of data where TPS and AFR target were absolutely steady for nearly 20 seconds. Running at about 2150 RPM, this range of operation should be just beyond the range where cam overlap / reversion would cause unsteady airflow numbers, and subsequently wideband readings.

http://www.pfautsch.com/wp-content/uploads/steady-cruise-graph-150x150.png (http://www.pfautsch.com/wp-content/uploads/steady-cruise-graph.png)

What I can't seem to discount here is that the wideband is averaging 13.0:1 afr while commanded AFR is 14.3:1 through this entire stretch. Even if my assumption is correct about airmass sensors causing rich operation in humid conditions, I would think this is a bit excessively rich.

Just looking for some honest opinions. My plan for tomorrow is to increase my injector constant from 42.114 to somewhere around 42.5 lb/hr, repeat the above test and see if the wideband comes closer to agreeing with commanded AFR.

Edit: Running OL MAF currently

Terminal_Crazy

08-15-2019, 04:01 PM

Hiya
You running OLSD?
Is it generally rich all over? Then increase injector constant
What is your VE table like?

Mitch
Recently I had to reduce the injector flow value as my VE table was maxing out

spfautsch

08-16-2019, 09:03 PM

First attempt with 42.5 ifc looked near perfect, until I realized BLMs had reverted to 120 after initial startup. I wish I knew why this was happening.

Third attempt looks pretty good if I remove the ethanol stoich adjustment - 42.62lb/hr injector slope w/ BLMs reset immediately after start.

http://www.pfautsch.com/wp-content/uploads/steady-cruise-42.62ifc-150x150.png (http://www.pfautsch.com/wp-content/uploads/steady-cruise-42.62ifc.png)

I'm noticing that my narrowbands almost never cross below 450mv except during dfco. Lean spikes hitting the limit of the wideband's analog range (16.5:1) are still really close to stoich according to the narrowbands. If they're not completely shot then my wideband must be. Though I still need to confirm that what eehack is seeing on D27 agrees with readings from the wideband controller's serial data.

I was in a rush to get back home so there's not much in this log, but open to any thoughts if anyone cares to analyze.

I did further tinkering and things got much worse after this log was captured - possibly because I removed too much fuel from the one cylinder that shows the largest lean delta during cylinder balance tests (#6). By the end of the day it almost wouldn't idle at all. I assumed this was because of heat soak and / or my ignition controller. I'm not completely convinced I'm not tuning around ignition problems, but driveability was very good until the idle started getting screwy.

NomakeWan

08-17-2019, 10:12 AM

Kind of a shot in the dark here, but if both your O2 sensors are in agreement with one another--that is, both narrowbands--then I have to ask. How long has it been since you fresh-air-calibrated your wideband? Could it simply be outside of its calibration?

In my experience with multiple cars that have had narrowband O2 sensors fail, they tend to fail either completely by dropping out (in my case, due to physical damage from afterfire in the exhaust melting the catalytic converter onto it) or by getting incredibly lazy compared to other O2s on the vehicle. It should be fairly obvious in the log as one will flick back and forth rapidly as normal while the other sits relatively static. I've never seen an O2 sensor fail by operating "normally."

spfautsch

08-18-2019, 04:04 PM

The wideband was calibrated when it was installed - thinking that was around 4 months ago. It's due for it's initial 3 month re-calibration, though it hasn't seen much actual use in the time it's been installed. It's on my todo list. The only other thing that could be causing it to read less accurately is it's location in my x-pipe. The recommendations are to avoid places where the exhaust system pressure might change. Unfortunately that's the only point in the exhaust where both sides converge.

I suppose the same phenomenon (pressure change) could be affecting the binaries which are located in the header collectors.

However, in general I would agree and go on to state I don't think I've ever seen a binary O2 sensor actually fail. Speaking in the context of tuning there are a lot of other factors that can cause them to appear to be causing a problem. Poor individual cylinder trims, poor closed loop settings to accommodate for header and / or cam, being relocated from factory placement, etc. All these are possibilities here. But I'm probably going to replace them simply because when I installed new ones back in 2017 I saved myself $20 by getting the Bosch units instead of the AC Delco 75s.

Before I pulled the engine apart looking for a mechanical problem I swapped back to my stock injectors. I should have reverted back to the factory tune at that point just to be sure, but for whatever reason just copied the injector and startup tables. I may repeat that test but revert back to the factory tune this time to eliminate my modded mystery injectors. I don't think they're leaking but I could have completely wrong offsets for them.

When I had the heads off I leak tested all the valve seats and found nothing wrong other than goopy fouling on the piston tops and in the chambers. This could have been from oil getting sucked into the intake ports from the rocker stud holes. I used better sealant on these bolts when it was re-assembled. The plugs didn't show signs of fouling and in fact looked relatively good - a light tan color on the porcelains.

Assuming it's not running miraculously better when I get back to it, and the injector swap proves fruitless I'm thinking I better start looking at the ignition system.

I'm traveling this week so will likely have no progress to report other than the acquisition of some ethanol free fuel.

Rocko350

08-18-2019, 09:05 PM

MAF sensors work on a heated wire principal. The wire is heated and then the resistance is checked, then the wire is pulsed to return i to a "known" resistance. The amount of pulses it takes to return that wire to the known or base resistance is reported to the ecm as a Hz signal. Hence the lookup table is referenced in Hz. MAF sensor performance is based on is internal processor, the supply voltage and ground.

As for humidity, if BARO is not reading within 3 of reported actual baro, the air density calc can be off. The humidity flowing over the MAF wire cools the wire, misreporting actual airflow. We are trying to calibrate a sensor table to report a mass of air that happens to include oxygen. Temperature (iat), mass flow(MAF) BARO (key on MAP) reading and engine vacuum (VE) determine the air density being delivered to the cylinder. So the actual humidity is not that big a problem. Its impact the actual combustion process in the chamber not so much the calibration of the engine. The stfts are fast enough to handle the humidity.

The enlarging the idle passages is for allowing the air passages to do their jobs. There is not a "closed loop idle" like there is in a GEN3 engine. If opening the idle bypass does nothing, but opening the throttle blades does, the passages need to be larger. Air in the main intake chamber (non idle passages of intake manifold) can be robbed from cylinder to cylinder if your just gonna open the throttle blades. Cylinder to cylinder fueling will need to be redone either way you choose. If you haven't adjusted the eoit i would visit that as well. Maybe stick a set of injectors in that you have excellent data for, then swap back the the njectors you need for power and adjust the injector data around the idle data from the known injectors.

Chris

vilefly

08-20-2019, 02:45 AM

I just remembered something from my obd-I days that really screwed with me at the time. Narrowband O2 sensors sometimes go nuts and put out NEGATIVE voltage on the output wire. This really made things crazy, since each computer brand reacted differently to this. The OBD-1 ecms could only see voltage differential, did not ignore the O2 sensor like they should have, skewing everything backwards. Lean became leaner and rich became richer.
Not that this might help, but one should always check for positive voltage output from the O2 sensors on OBD-1 setups if suspicions are raised.

I have an old injector flow tester that tests the injectors on the vehicle (with some temporary fuel line rerouting). Let me see if I can find it and check to see if the display works still. It has a faded lcd display that may or may not work.

spfautsch

08-20-2019, 06:04 PM

I have an old injector flow tester that tests the injectors on the vehicle (with some temporary fuel line rerouting). Let me see if I can find it and check to see if the display works still. It has a faded lcd display that may or may not work.

Don't waste a bunch of time messing with that - if I go to the trouble to try characterizing the injectors I think I'll find a 4 cylinder fuel rail at a junk yard and build my own test bench. I'm just not sure it's worth the trouble. A new set of 42lb Ford Racing injectors with good data is only $400. The only reason I'm going to this trouble with the ones I have is because I know they were working pretty darn well before summer.

At the point I had to stop working on it the drivability wasn't that bad save for slightly lazy cold start (at hot ambient) and falling on it's face at hot idle. If my hot idle is still crap when I revert the last couple changes I'll probably switch back to the factory 25lb injectors and see if I have better luck.

vilefly

08-21-2019, 05:40 PM

Don't waste a bunch of time messing with that - if I go to the trouble to try characterizing the injectors I think I'll find a 4 cylinder fuel rail at a junk yard and build my own test bench. I'm just not sure it's worth the trouble. A new set of 42lb Ford Racing injectors with good data is only $400. The only reason I'm going to this trouble with the ones I have is because I know they were working pretty darn well before summer.

At the point I had to stop working on it the drivability wasn't that bad save for slightly lazy cold start (at hot ambient) and falling on it's face at hot idle. If my hot idle is still crap when I revert the last couple changes I'll probably switch back to the factory 25lb injectors and see if I have better luck.

A quick way to test for equal injector flow w/o removing it involves a fixed injector pulsewidth with about 10 cycles or so, and a fuel pressure gauge. Pressurize the fuel system and turn off the fuel pump, send 10 x 4mS pulses to an injector, note the fuel pressure drop, write it down, and move to the next injector - repeat. This way, you can get a rough idea whether or not your injectors flow equally. The injector pulsewidth can be 4mS - 500mS. Your choice, just make sure it is enough to move the gauge far enough to get a good reading. Don't forget to WOT start the engine afterwards to clear the fuel. I bought the injector pulser tool a long time ago, before the arduino or Basic Stamp was born.

kur4o

08-22-2019, 11:27 AM

bosch sensors explain the odd readings you have, even new they are garbage and can`t switch fast enough for the superior GM pcms.
Get a delco ones and for the sake of money savings, don`t buy anything bosch related for GM cars.

There is a minimum keep alive blm settings, that explain the 120 blms at startup. For vettes it is set at 120, changing it to 128 must fix the lower blms at startup.

Look at the wideband reading not as extremely accurate tool, but as a fixed constant to dial the ve and maf curves and get a match between commanded and actual afr. After that offsetting the whole tables up or down will get you the sweet spot you are looking for.

SInce I also have elongated start times at clod startup around 25-35 C, and great start time less than 0.3 seconds at warmer engine, went through the code to see what is happening.

The prime pulse is complex as hell, in your case you will have to adjust for new injectors along with the extra cube. The good way is to convert the stock BPW readings to grams of fuels. than add some fuel for the extra cubes and convert from grams to bpw of the new fuel flow constant. I still haven`t found the switch point between prime pulse and the crank ve calculations. There are 2 different afr crank tables that switch around 0.4 seconds of cranking, which must be the norm of cranking time around 0.3 seconds.

The good news is that afr and delta ve mode 4 control all works in cranking mode. Changing these before cranking with igntion on will cut the guess time. Only crank spark is unchangeable. I have noticed that you set the crank spark at 12 degrees, stock is 7. Any explanation on that why it is needed.

spfautsch

08-26-2019, 11:05 PM

bosch sensors explain the odd readings you have, even new they are garbage and can`t switch fast enough for the superior GM pcms.
Get a delco ones and for the sake of money savings, don`t buy anything bosch related for GM cars.

I wish I'd known this a couple years ago. Bosch is generally OE on my daily driver (a diesel VW) so I thought they were of decent quality. The cost differential wasn't even $20 - more like $12. AFS75s arrived couple days ago.

I still haven`t found the switch point between prime pulse and the crank ve calculations.

I had always assumed it switched to the crank VE table after the first two low res pulses simply because the ecu should have determined sequence by that point. But your understanding of machine language trumps my theory.

I have noticed that you set the crank spark at 12 degrees, stock is 7. Any explanation on that why it is needed.

I haven't modified the cranking spark table. This must be another Y-body specific thing.

12075 Crank Spark Advance vs Coolant Temp
-40 14
-28 14
-16 14
-04 14
08 13
20 13
32 12
44 11
56 10
58 9
80 8
92 7
104 6
116 5
128 4
140 4
152 4

After more research I'm finding that the location of my O2s (both binary and UEGO) shouldn't be causing an outright skew of AFRs - the source I'm referencing says they should still report stoichiometry accurately. Linearity above and below stoich is where exhaust pressure can affect the output. So I'm hoping they will agree once I get the new binary sensors in and some pure dino gasoline running through it.

kur4o

09-01-2019, 01:13 AM

I went through the code and there is 1 pulse on first opti event, and there is 1 more on the second opti event. All pulses are corrected with baro reading. Than it starts to calculate fuel based on crank afr and ve tables.

First at reset a value from table $2682 is loaded for initial BPW, not sure if injectors fires with this value.

AT first opti event a value from table $2691 is added to the initial BPW, injectors for sure fires with this BPW.
At second opti event a value from table $26A0 is added on top of the already calculated BPW, injectors also fires.
On second opti event injectors also fires with values from crank ve and afr tables.

There are some other tables that seems to do nothing and are labeled with crank names.

For the first 16 opti events[2 full engine revolutions] the $2504 crank afr table is used, AFter that the pcm switch to $25f7 crank afr tables.
The tables are vs coolant and opti event number.

UNtill the engine running flag is set it could be a batch fire mode or it could be sequential. It is decided from the TPU processor, which we don`t have access to the built in rom.

Bigger cubes might need less spark advance at startup. Trying the f-body crank table could help abit.

A healthy starting time is around 0.3 seconds. The prime pulse plays the biggest role here. Usually the crank ve and afr tables comes into play after that.

spfautsch

09-01-2019, 03:03 AM

That's awesome, thanks for taking the time to research that.

I've finally been able to spend some "quality time" methodically troubleshooting. Step one was to put my stock multec injectors in and revert to stock cranking tables and injector characterization. That made zero difference.

Continuing on, the first thing I found is that my fuel pressure was hovering around 50-52psi while the pump was running (engine not running). After the pump switched off pressure dropped immediately to 42-44 and held. Leakdown was about 2 psi every 10 minutes. I'm kind of embarrassed that I didn't check fuel pressure sooner - I thought I had when this problem first started to materialize but I guess I didn't. Anyway, I replaced it with a 3 bar regulator I happened to have in my parts bin that fit and now the pressure maxes out at 42-44 and holds at 42 after the pump stops running.

The other simple thing I hadn't looked at very recently was plugs. They were a little wet and fouled, but I wanted to try and salvage them since they're almost new AC double platinum 41-906s. Cleaning the first one with carb cleaner I then hooked it up with a grounding clip and tested the spark by hooking up my spare opti and spinning it by hand (with the injector and fuel pump fuses out). All I can say is I'm confused. The spark kernel was seemingly avoiding the platinum discs and jumping from the perimeter of the center electrode to the outsides of the ground electrode in an orange-ish ball. Very odd... I took some 600 grit wet or dry paper and cleaned up the platinum discs but if anything that made it worse. So a dirt cheap set of NGK TR55s were bought, gapped to 0.045 and a hot blue spark was witnessed.

Planning to finish installing the plugs and the new O2s tomorrow, will report back soon.

kur4o

09-01-2019, 09:36 AM

That might be the culprit of the weak spark you had recently. The energy of the ls1 coils is too much for the platinum disks to handle on these plugs and the spark scatters taking the least resistance path around the platinum disk. I should say they are either defective or the ls1 coils are overkill for 41-906s. Tr55s are always better for non stock application, but they must be replaced quite often for optimal performance.

If your fuel pump can support 60psi and you have a regulator for 60psi, go for it. Much better spay pattern and atomization with the newer ls1 style injectors.

spfautsch

09-08-2019, 11:40 PM

That might be the culprit of the weak spark you had recently.

I think I might have possibly eliminated weak spark as the problem. Read on...

The energy of the ls1 coils is too much for the platinum disks to handle on these plugs and the spark scatters taking the least resistance path around the platinum disk. I should say they are either defective or the ls1 coils are overkill for 41-906s.

Since this is the first time I've ever used a double-platinum plug to check spark strength (normally I use a new + clean champion that I keep in my toolbox for this specific purpose) I 'm going to go out on a limb and say I think these type of spot-welded disk platinum plugs are just junk. I also dug out my old OE plugs that have been run an unknown amount (possibly original to the car) exclusively with the optispark / hei2 setup. This is the plug on top, the plug below is essentially new. Notice how the center electrode is eroded away around the perimeter of the platinum disc. Also, the ground electrode has long since expelled it's platinum.

http://www.pfautsch.com/wp-content/uploads/0908191452-150x150.jpg (http://www.pfautsch.com/wp-content/uploads/0908191452.jpg)

Tr55s are always better for non stock application, but they must be replaced quite often for optimal performance.

I just wanted something I could get cheap at the local parts house. Since I'll need to order some water pump gaskets from my favorite online parts house I'll probably buy several different exotic plugs (double iridium, ruthenium, ???) to compare. The prices there are almost $4 per plug cheaper than the local wallet sucker parts house.

... and you have a regulator for 60psi, go for it. Much better spay pattern and atomization with the newer ls1 style injectors.

This makes me wonder if there is a OE style 4 bar regulator that's a direct fit for this application. But unfortunately this pressure problem was with the original non-adjustable 3 bar regulator. I'm going to cut it open when all this is over and see what's going on internally, because it seems like the problem could be caused by clogging in the return section. The lines don't seem to be the problem because the new regulator doesn't exhibit this behavior.

Anyway, as I parked it three weeks ago it wouldn't even start without giving it pedal. I wasn't sure what had happened but in my troubleshooting process I noticed I'd left the radiator cap sitting loose so I likely boiled the coolant during my last few test drives. This made me wonder if I'd burned some exhaust valves. So after the plugs and factory injectors made no difference I started out by doing a complete compression check. Every cylinder was within 2 psi, so I breathed a huge sigh of relief.

I then installed my new binary O2s and recalibrated the UEGO, thinking all the while it may be the sixth Tuesday of the month of Never before I get to enjoy the fruits of that labor.

After that I decided to check off a few other remedial things vilefly suggested. Hats off to you sir.

Since it wouldn't run, I flashed a bin with 0 degrees cranking spark advance and had the wife crank her while I observed with a timing light. To my surprise I was never able to see the painted-on timing mark. After confirming by checking the opti outputs with my ignition controller's timing check mode, I found the low res pulse was retarded a massive amount. Seems like when it got hot the loctite "high temperature" sleeve and bearing retainer I used to secure the two parts of the opti hub together got soft enough to allow the hub to slip.

http://www.pfautsch.com/wp-content/uploads/0908191431b-150x150.jpg (http://www.pfautsch.com/wp-content/uploads/0908191431b.jpg)

I had assembled this around December of 2018 when I had bought a spare opti base and decided to do some disassembly to discover what rebuild options there were. This is the rebuilt hub that came with the car. I removed the 1/8" roll pin that originally filled the hole because according to my degree wheel it was off by about 6 degrees - 2+ of which was due to my cam needing to be installed advanced to make target ICL.

Anyway, I'm debating putting this back together and re-drilling the hub for another roll pin. That, or looking for an adjustable MSD base I can get cheap off eBay so I won't have to trial-and-error 20+ times to get the junk opti hub installed back at zero like I did back in December 2018 (with the engine comfortably mounted on a stand).

Thinking back along the progression of this problem, it seems highly possible this was the entire cause, and that the opti hub slipped more than once. An initial slip of a few (8-16) degrees of retard would more than explain the difficult cold start, lean readings and rich smell I was describing at the beginning of this thread. Where it's at currently is at least 30-40 crankshaft degrees retarded. I'm surprised it would start at all.

vilefly

09-09-2019, 02:29 PM

Timing lights are still not obsolete, folks. It still urks me that they are fading away from the public's reach, due to the illusion that computers don't make timing errors. One of these days I am going to build a timing pointer with its own LED strobe built in. Stroboscopes used to rule the days of old due to their usefulness. Still used to balance driveshafts. Mystery solved.

spfautsch

09-09-2019, 05:52 PM

I'll be sure to pass your outrage along to the dealership the next time I buy a new car. I need a damper that's keyed and marked for TDC and a dipstick on that automatic transmission or there's no deal. Pfffft!

vilefly

09-10-2019, 02:43 AM

Don't forget to make sure it has more than just one keyhole in it. The new camaros are so wrong when it comes to unlocking it with a dead battery in the back. I'd just have to drill holes in it to install my own tumblers. The door handles are electrical membrane switches, no mechanicals.

Fire that engineer.....now. And call off christmas!

spfautsch

09-10-2019, 03:08 AM

Well if you knew me the only thing funnier than my demands was the part where I said "buy a new car". Never gonna happen (unless I win powerball).

A mechanic buddy of mine owned a '18 Camaro. Notice the past tense. Turns out the key fobs don't like cold much and he found it was damned hard to get the car to start at 5F when there's no manual ignition and the bcm isn't seeing the key fob.

NomakeWan

09-10-2019, 04:34 AM

Agreed with you there. Guy wanted to sell my friends and I a Mercedes Benz V12 Biturbo. For an unbelievably low price, because he couldn't get it to run. So we went over thinking, sure, why not, let's see what we're in for. Well it turns out there's no physical key for the ignition, so you can't start the car without the fob. But even more hilariously, even if you have the fob, you can't start the car if the car's been left with a dead battery for too long, because then the backup battery for the security system fails and the only solution is to have the vehicle towed to a Mercedes dealer along with you and all the paperwork proving you own the car so they can install a brand new security computer and give you new keys.

Needless to say we walked away and wished him luck.

Here's me wishing you luck that the slipped rotor has been the only issue this whole time. :)

vilefly

09-11-2019, 12:53 AM

I was thinking that you could add 20 degrees advance across the board to your timing tables to confirm your suspicions about the distributor without taking stuff apart. Turn the key and see if it liked it.

spfautsch

09-11-2019, 05:52 PM

As far retarded as it was I think the best I could've done was lock timing at max (46 btdc) with eehack. That might have given me 16 degrees advance best case, and it would have still ran like crap that way. With the egr the cam generates it wants at least 28 degrees spark advance at 925 rpm idle.

It's kind of immaterial - I'll have it back together this weekend if the "baby" degree wheel shows up when it's supposed to. My good wheel interferes with the k-member by about 2 inches.

After moving the hub yesterday I feel like this had to have happened gradually over numerous driving sessions. As hot as I was comfortable getting the hub without burning the lube out of the bearing, it still took quite a bit of force (with a channel lock pliers no less) to turn the hub back.

vilefly

09-12-2019, 02:41 AM

https://www.blocklayer.com/degree-wheel.aspx

Here. Print out one of these babies on card stock paper. Works for me in a pinch. Lots of options, too.

spfautsch

09-20-2019, 01:17 AM

Print out one of these babies on card stock paper. Works for me in a pinch.

I'd have to be in a pretty big "pinch" to use card stock, but I guess it could work. The nice thing about the neat little 7" wheel I got off slamazon for $15 is that even though it's aluminum it's a nice high-density extrusion and when I'm not using it for timing an engine I can hang it on my drum kit as a gong for when I want to attempt Rush's YYZ.

Once again I got called into service as a used car salesman over the weekend, otherwise this would have been done last sunday. If I could have found an adjustable MSD dizzy for a reasonable price I'd have been all over that - the time it would have saved would have almost made buying one new worthwhile. After getting the hub back to zero it took almost 6 hours and numerous 3/32" cobalt drill bits to get a hole drilled in the hub to lock it in place. It seems like it's some kind of powdered metal and damn is it hard!

Anyway, I've got a couple odd things going on now - first log was with -2 degrees spark advance by way of setting the 0x012044 "Spark Reference Angle" to 2 because I thought my hub was a couple degrees more advanced than I wanted. Stock injectors with original characterization and individual cylinder trims. Judging by the idle quality I think the timing was actually spot on and my determining of geometric TDC with the engine in the car (and heads on) was off by +2 degrees.

Right bank is quite a bit lean especially before bumping idle timing back +2 degrees and resetting blms to 128. I'm going to pull my plugs from that side to see if one of the iridium electrodes "blew" off.

Spark plug discussion detour: I compared my ACD double platinums to a Bosch 9655 double iridium and a Denso 4713 TT iridium using the highly scientific "eyeball" method, and the Denso TTs gapped at 0.040" produced a far superior spark to all the above (even the copper NGKs I bought in desperation). Detour over, back to the main subject...

For the second log I pulled the 2 degrees of retard back out of 0x012044 and reverted to the individual cylinder trims I arrived at months ago. Right bank still pretty lean. I forced closed loop for a short time and other than having +10 blms on the right bank it wasn't idling too shabbily. I saw a few data frames with 41.5 kpa at ~9.8 g/s on the MAF. The IAC counts hit 24 which tells me I probably need to go to the next smaller IAC bypass orifice because today was another 93F scorcher with swamp-ass humidity levels. This seems to invalidate the usefulness of enlarging the idle air feeds in the manifold.

At this point I think I only have one other problem to locate. But...

In both of these logs it didn't start right away like it was before this problem manifested itself. So I have more troubleshooting to do, because before this started I never needed to crank more than 1-1/2 engine revolutions before it took off.

Hopefully it won't take me another two weeks to track down these problems. I'm about exhausted...

kur4o

09-20-2019, 11:39 AM

There is a lean condition on both sides, that evolve to one side later, that can be caused by numerous things.
The mechanical issues must be ruled off first. I would start at tapping the hole in the intake blade and load the last good ind cyl trims you have on record. The extra air from the hole can screw up alot of the fueling so as a first measure just make it temporaly tapped.
At this point make an idle log. If there is no improvement on lean bank to bank condition, swap injectors per bank. Now you can rule out a clogged or pissy injector. If the lean bank moves with the injectors the injectors are toast.
The last mechanical issue could be small exhaust leaks. It can affect 02s reading even if the leak is after them. Just for the record did you swap the 02s with delco ones.

If nothing of this help it is time to do some tuning. The maf table needs to be dialed first, than fine tune the individual cyl trims. Than tune the spark and again fueling since spark and fuel at idle are closely related and couple of degrees less or more advance can shoot your afr off chart.

I am having the same hard cold starting issue that developed out of nowhere. It will start right away after it has been running for a while cold or hot. The issue shows if it have been sat for a couple of hours no matter cold or hot is the engine. NO tuning helped at all so far, so i am thinking a mechanical issue could be the problem.

Spark reference angle offset all timimg tables by the amount specified. It doesn`t correct injector timing and any other opti related calculations.

spfautsch

09-20-2019, 05:46 PM

There is a lean condition on both sides

If you look at the first log the extreme lean condition on both sides improves on the odd bank when I manually increase idle timing from 27 to 29.

Another factor is the BLMs - I keep forgetting to increase the keepalive value to 128. I don't necessarily intend to continue running open loop, but that would be something anyone who is (running open loop full time) should do. Anyway, when I reset them to 128 the lean condition improves markedly.

I would start at tapping the hole in the intake blade

There are no holes in the throttle blades and I've re-adjusted them back to fully closed so there should be very little air getting past the throttle at idle.

... If there is no improvement on lean bank to bank condition, swap injectors per bank.

I believe my plan of attack will be putting my big injectors back in. I have the feeling the 150k mile stock Multecs have had the varnish in them solidify from sitting dry in my parts bin for 2+ years. I'm crossing my fingers this is my starting problem also.

Just for the record did you swap the 02s with delco ones.

Yes, these logs were taken with the AFS75s. If you scan through the second log there's a ~30 period starting at 142.1 where I forced closed loop. When I switched back to open loop the odd bank BLM was still 128 but the even bank had climbed all the way to 142. This was with the factory closed loop parameters restored - not a single item has been touched, even 0x26fb Integrator / Transport Delay.

If nothing of this help it is time to do some tuning. The maf table needs to be dialed first

I think my methodology is going to be leaving the MAF calibration alone until all other avenues have been explored. From what I've been able to find on the subject I don't think I'm going to be able to considerably improve on what GM did there.

I am having the same hard cold starting issue that developed out of nowhere. It will start right away after it has been running for a while cold or hot. The issue shows if it have been sat for a couple of hours no matter cold or hot is the engine. NO tuning helped at all so far, so i am thinking a mechanical issue could be the problem.

I had the same warm restart issue before increasing displacement, and I was able to find a sweet spot between prime pulse and cranking VE. Unfortunately the prime pulse table lacks the resolution necessary for much fine tuning in the warm / hot regions, especially with larger than stock injectors.

spfautsch

09-23-2019, 11:14 PM

Switched back to the modded 42lb/hr LS injectors. Still lean on the right bank, but if I force closed loop and let the BLMs adjust a bit it's much better and I'm seeing ~40kpa at idle once it gets up to temp. I think my individual cylinder trims need a lot of work here because looking at idle BLMs on some older closed loop logs the right bank has been 6-12 counts lean of left.

Initial start is still not great so I'm going to re-work the cranking fueling beginning with the base pulsewidth tables.

The prime pulse is complex as hell, in your case you will have to adjust for new injectors along with the extra cube. The good way is to convert the stock BPW readings to grams of fuels. than add some fuel for the extra cubes and convert from grams to bpw of the new fuel flow constant.

If you have a chance please look at the attached and check my math, particularly the lb/hr to g/s conversion factor (<lb/hr>*0.175070002). If this is anywhere close, my previous method of scaling arrived at numbers that were way off. There is scaling in this worksheet for cylinder volume changes and I also added another scalar to account for a delta in pumping efficiency due to cam swap, etc. Note that the cylinder volume scaling doesn't convert volume to air mass because I didn't see a difference when I calculated it outside of the spreadsheet.

kur4o

09-24-2019, 03:25 PM

per google conversions are

lb/hour to g/s x*0.125997881

g/s to lb/hour x*7.93664144

24lb/hr*0.125997881=3.02394 grams/s
42lb/hr*0.125997881=5.29190 grams/s

Lets take or example 50ms BPW and convert it for 383ci @42lb/hr
50ms[x/1000=0.05second]*3.02394=0.151197 grams of fuel

so 50ms BPW delivers 0.151197 grams of fuel at 24 lb/hour injectors.

346 ci to 383 ci is 10.7% difference.
Hope the 383 ci is correctly calculated and not marketing cubes like 350 is.

0.151197 grams * 10.7% = 0.1673751 grams needed or the new displacement.

SO for 383ci you need 0.1673751 grams /3.02394 =0.05535second BPw[55.35ms] at 24 lb/hr

or 0.1673751 grams /5.29190 =0.03162seconds BPW[31.62ms] at 42 lb/hr.

I doubt the cam will be a factor here since cammed only lt1 fires straight up with stock tables.

spfautsch

09-24-2019, 07:21 PM

lb/hour to g/s x*0.125997881

I'd used that initially but wasn't completely confident in the accuracy.

346 ci to 383 ci is 10.7% difference.
Hope the 383 ci is correctly calculated and not marketing cubes like 350 is.

I was dividing the "new" ml/cyl by the ml/cyl from the original bin and it gives me 1.0999 so essentially 10%. I believe I used one of the online calculators to come up with 788.97ml/cyl, but the new setup is 3.750" x 4.040". I'll double-check this. The stock bin has 717.25 ml/cyl.

Adjusting the lb/hr to g/s conversion factor, a 50ms pw gives 32.87ms adjusted for injector flow and your displacement figures. This should be close enough in that region (8C / 46F). The real trouble spots are the warm and hot restart cells where I can only adjust between 0x02 and 0x03 (3.1 and 4.7ms respectively).

Edit: I noticed in your example you rounded all the precision out of the stock injector flow, which is closer to 25lb/hr than 24. Using 24 and 42 gives 31.4 to your 31.6 so more than close enough...

I doubt the cam will be a factor here since cammed only lt1 fires straight up with stock tables.

Just trying to be thorough. I haven't dialed my VE in yet but the old setup had roughly 10% lower numbers in the low RPM regions compared to stock bin.

Either way it still doesn't want to start worth a damn. I drove it to work today and thought I was going to spend a half hour tuning next to the gas pump after filling it up. I'll repopulate the prime pulse tables with the current results but I doubt it will make a difference.

Attached are the logs from this morning and one from last night. At idle the right bank BLM is about %9 lean of the left bank. Once the throttle is opened the lean condition diminishes. In fact there's a short bit in the log from last night where I held the throttle at 4.7% tps for 10 seconds or so and the right bank O2 came right up and stabilized around 910mv.

The logs from this morning are mostly closed loop because I didn't want to chance stalling it at a stop light, and mostly interstate cruise in 6th gear. Here's the BLM analysis from the longest log.

14660

Last night I ruled out a lower intake manifold vacuum leak - I got a noticeable puff of blowby after roughly 7 seconds. I also tried adding some fuel to cyls 2, 4 and 6 in the individual trims, but all that seemed to accomplish was making my eyes water from the unburned fuel.

Tonight I plan to try vilefly's bucket test to check for leaks on top but I highly doubt there's a problem there. The only other thing I haven't ruled out is an exhaust leak on the right side, but I would think I'd hear something if the header was leaking.

I'm starting to think one side of the idle air plenum might be clogged with something. If nothing else shows improvement I might try opening the throttle blades some to give more idle air through the main plenum.

Things are falling into place, but it sure doesn't want to start cleanly and I can't help but think the lean right bank O2 readings aren't related to this.

kur4o

09-24-2019, 11:35 PM

Some more info I need. What are the headers your using. How long are the primary tubes, and are they equal lenght accross all cylinders. If not it is good idea to measure each tube`s lenght. The distance the 02s are mounted from exhaust ports. It can help dialing the initial ind cyl trims. If the 02s are mounted too far from engine than stock, there is some delay for the exhaust gas to reach the sensor and it must be fixed in the bin CL settings.

The new 02s perform much better now, still the cross counts are little too low.

The engnie runs too cold to my linking at 78 degrees. The thermostat might need attention or the coolant sensor is not good. At 85-90 degrees it runs so much smoother and better. Get some cardboard in front of the radiator and increase the fan points by 5 degrees. The fuel evaporates best even at 95-105 degrees coolant, you get a higher fuel milage and the engine is more efficient with all the shit from oil turning into vapors, but is too much for built lt1. The best compromise from performance and drivability is around 90 degrees.

I see you got higher blms everywhere on the right side meaning it is adding fuel off idle too. It does seems like exhaust leak to me. Try swapping the 02s and look very carefully for black sooth and for exhaust smell in the engine bayat cold stratup. Sometimes even the tiniest hole can bring much trouble and split blms.

The wideband also shows the lean condition so it is real.
If you suspect a cylinder s cyl cut test can point which ones is giving trouble. Set the afr to 12.5 and start cutting cylinders with 5-10 second delay between them. Record how much the wideband readings change when a cylinder is cut.

Did you tried to lower the rpm and the advance for more stable idle. Stock cubes cammed lt1 idles happy at 650 at Park and 550rpm at drive. I doubt the 383 can`t keep stable idle at 750-850rpm with 20-25 degrees advance. At lower rpms the surge from bad tune is much more pronounced.
How is the idle ovespeed/underspeed spark retard/advance tables set. Having 1-3 degrees there offload the iac motor jumping all the time.

spfautsch

09-25-2019, 05:18 AM

I wasn't really intending for this thread to turn into a tuning workshop, but I appreciate the value of having your attention so what the heck. I want the tune to be perfect (eventually) but I want to know how to do it just a bit more.

These are the headers I have (link (https://stainlessworks.net/chevy-corvette-1992-96-headers)). Cats aren't currently installed, although I doubt they would do much of anything.

... If the 02s are mounted too far from engine than stock, there is some delay for the exhaust gas to reach the sensor and it must be fixed in the bin CL settings.

I'm aware of the 26FB integrator delay / transport delay table, and until just recently I had the values almost doubled from stock. They're back to stock now.

I did a bit of "visual" measuring against the stock exhaust system before taking it to the scrapper for this purpose. I can't find any pictures at the moment but if I'm remembering correctly the left bank O2 bung wasn't considerably closer / further away from the exhaust valves than they are through the header primaries. On the right bank the O2 bung was several inches closer because they tucked the cat much closer to the exhaust manifold.

In the grand scheme of things I'm not sure there's a lot of difference in transport delay though, when you consider the velocity of gasses traveling through a 1-5/8 tube versus the huge log manifold and then into the 2-1/2" exhaust pipe. But I'm open to suggestions.

The new 02s perform much better now, still the cross counts are little too low.

As mentioned the right bank has always been slightly lazy, but I'm starting to think this is a result of poor cylinder balance and improper injector characterization. Not to mention the old regulator producing higher than intended rail pressure.

To update the current state of affairs - since flashing the most recent prime pulsewidth tables using values from the correct conversion formulas, starts have been good so far. I suspect my old fuel pressure regulator has had me chasing my tail trying to find the injector slope "sweet spot" when in fact it was a moving target. I've been decrementing the 12B4C injector constant by 0x0001 with every flash and will continue until I see slightly rich OL numbers on the wideband. I suspect this will make the cross counts increase on both banks.

The engnie runs too cold to my linking at 78 degrees. The thermostat might need attention or the coolant sensor is not good.

I've had a 160F thermostat installed for several years. When I first bought the car it had numerous cooling system issues from all the dexcool sludge that had accumulated, so I put this in out of an abundance of caution. I see less need for that extra insurance now that I know the cooling system is 100% healthy. I understand the pros and cons of the lower temps, and I think I'll leave it alone until I'm happy with the state of the tune so a before / after comparison can be made when I install a OE spec 180F t-stat.

I see you got higher blms everywhere on the right side

Sorry, I'm not sure I'm seeing that (everywhere). Looking at the big log from earlier today stacked with the one attached from this afternoon I'm only seeing averaged BLMs on the right that are higher than 129 in cells 2 and 16-18.

Edit: I see what you're saying though - the right bank is on average lean of the right.

If you suspect a cylinder s cyl cut test can point which ones is giving trouble. Set the afr to 12.5 and start cutting cylinders with 5-10 second delay between them. Record how much the wideband readings change when a cylinder is cut.

That's what I've been doing minus dropping the AFR to 12.5. Nothing has really "jumped out" so far. I'll try to get a fresh balance test tomorrow.

Did you tried to lower the rpm and the advance for more stable idle. Stock cubes cammed lt1 idles happy at 650 at Park and 550rpm at drive. I doubt the 383 can`t keep stable idle at 750-850rpm with 20-25 degrees advance. At lower rpms the surge from bad tune is much more pronounced.
How is the idle ovespeed/underspeed spark retard/advance tables set. Having 1-3 degrees there offload the iac motor jumping all the time.

I've never been happy with the idle quality below 29 degrees spark or 900 rpm - I like to hear the cam, but I don' t like to feel it. I'm sure some of this is a function of cylinder balance. So hopefully this is something that will improve as I find the right individual cylinder trims. Unfortunately MBT is happening around 29 degrees so there's not much room for the underspeed idle spark control to do it's intended job.

Thanks to everyone for all your help and suggestions. I feel pretty confident now that all the problems encountered were not a direct result of the experimental ignition system, and that's what I was most concerned with.

kur4o

09-25-2019, 11:58 PM

Here is a starter pic of the intake manifold for reference for individual cyl trims. You can clearly see why in the stock applications #7 is always set at 0.9.
WIth long tubes with different lenght primaries there is considerable difference in cylinder efficiency. The longer the tube the more air it draws. SO the difference will be how much readily available air is left for cylinder to draw from the intake passage and how much vacuum is made in the exhaust tube.

If you haven`t equalize the stock intake passage holes, you can start with the stock cyl trims settings and work from there. The idea is to find out which cylinders draw more air and which ones` less, Than adjust the fueling.

I always wondered why the CL worked for you at idle even at that cam and didn`t experience the dreaded enleanment at idle CL. When you mentioned that the 02s were down the pipe, I checked the bin settings for your calibration and it is seems like it is preset for some headers already. I will test them as soon as possible and will report how it does with midlenght slp headers.
You can still make some tweeks to get more crosscounts at off idle cruising.

MBT is ideal at wide-open throttle (WOT), but not desirable when the engine is at idle.
per wikipedia.
Too much heat in exchange for really lean mixture to utilize the extra spark advance. Maybe that`s why you are getting that unburned fuel smell. The afr is too fat for the advance and the fuel can`t burn and it is dumped in the exhaust.

I agree and terminal_crazy also reported similar results.
I have also zeroed the underspeed tables and the result was a disaster in idle quality. So I increased it to a max of +-2 degrees. It really smooths and offload the IAC valve from constantly adjusting.

The only reason I see for idle rpm increase is if the engine stalls at take off or it gets unstable at near idle conditions and is trying to die. Also the lower map readings you get are because the higher rpms. I am sure that the map will jump around 45-50kp area if you lower the rpms by 100. It takes alot of tuning to get it right with no surging but the results is clean and efficient combustion.

I also have 160* thermostat but never have seen under 80-81 C degrees cruising even at colder weather. I can suggest to leave the thermostat and increase the fan points. At colder weather block part of the radiator. Just for sake of experiments try the idle quality at 90*C and see how it calms and quiets down. I have never run it hotter but can imagine stock at 100*C. There is one very nice settings for fan temp with mph threshold. You can set higher temps with low speed cruising and colder settings with high speed performance trips.

Sometimes it is better to tune by feel and sound. The preffered method is also seat in the pants tuning.

spfautsch

09-26-2019, 12:41 AM

When you mentioned that the 02s were down the pipe, I checked the bin settings for your calibration and it is seems like it is preset for some headers already.

Am I correct to interpret that to mean the O2s are in the manifolds on other LT1s? That would explain a lot of the difficulties f-body guys have with headers.

per wikipedia.
Too much heat in exchange for really lean mixture to utilize the extra spark advance. Maybe that`s why you are getting that unburned fuel smell. The afr is too fat for the advance and the fuel can`t burn and it is dumped in the exhaust.

I would suggest you search "spark hook test". Maximum Brake Torque is the spark advance where, with all other variables removed (speed and AFR) the most torque is generated, and consequently the lowest resultant manifold pressure. Adding more spark advance beyond MBT for a given mixture yields no additional torque. This is why you want spark at MBT everywhere except at idle, because the underspeed adder can't increase engine speed if you're already at MBT.

The only reason I see for idle rpm increase is if the engine stalls at take off or it gets unstable at near idle conditions and is trying to die.

I've never attempted to get it to idle below 875. The cam lope shakes the entire car and I'm not a fan of it. If I can find any improvements with cylinder balance I may attempt to work on lowering idle speed and spark.

Also the lower map readings you get are because the higher rpms. I am sure that the map will jump around 45-50kp area if you lower the rpms by 100.

I was under the impression lower map at idle was the goal, not higher. ???

It takes alot of tuning to get it right with no surging but the results is clean and efficient combustion.

When you write a book I'll be the first in line to purchase it.

spfautsch

09-26-2019, 03:24 AM

cut test can point which ones is giving trouble. Set the afr to 12.5 and start cutting cylinders

Is this something that's only possible with your v4 patches? The fork of eehack I'm using has the cylinder disable button group greyed out when AFR is overridden, but when I commented out the line that greys them out and tested it seems like the AFR reverts back to the tune default when a cylinder is disabled.

As a workaround I changed the OL AFR target to 13.2 in all the cells I thought would be affected. I should have gone one temperature band higher, the result was 13.4-13.6 during the balance test. In the attached log it starts at timestamp 4410 and I disabled the cylinders in numerical order.

14667

If you watch the narrowbands, it looks like the right bank is robbing fuel from #3 pretty efficiently as both sides go flat when it's injector is disabled. I'm not sure what this means (if anything).

Other than that I don't see any really obvious red flags except that MAP took a nosedive after #6 had been disabled for a few seconds.

At the end I played with idle speed and spark and managed to get a reasonable idle at 800 with 25-26 degrees advance (albeit at 13.4:1 OL afr). But then I let it revert to closed loop and couldn't tell much difference. I guess I need to try these settings out in a tune and see how it feels as it's warming up.

The attached log is about 80 minutes of mixed driving in closed loop until I got in the driveway and played with cylinder balance and then idle. If you look at the CL Performance tab on analysis in eehack note that my two primary cruising cells (2 and 6) seem to trade places - in cell 2 the right bank is lean, and in cell 6 the left bank turns lean. The BLMs almost trade places.

On an unrelated subject, I think I came up with a semi-reasonable method for working out the cranking prime pulse tables. I need to clean up my worksheet and write up a how-to for you to look at and see what you think. What I did is took the actual pulsewidths after the floats have been converted to hex and calculated the sum of the actual fuel charge for the first two pulses. Then I adjust one or the other until the summed fuel charge most closely matches the desired sum. It seems to be producing decent results. I've seen folks mention running 80lb Ford Racing injectors in this application, and I have to wonder how they've managed to get crank fueling dialed in with no more resolution than is available here.

spfautsch

09-26-2019, 06:25 PM

Hot target idle speed set to 800rpm with 26 degrees. Idle over and underspeed tables restored to original. Also added a bit more fuel to 2, 4 and 6 trims. Bin included with log this time.

Cold start this morning was great. No problems whatsoever - sitting idle was great but I might have felt a slight amount of surge at ~20mph before pulling in the parking lot at work. This is the first 50+ mile log I can recall with zero knock events.

The idle BLM split didn't improve much but the cell 2 split is slightly better. Still significantly fewer cross counts on the right. Planning on pulling the throttle body tonight to see if anything might be clogging up the 2 and 4 idle ports.

If nothing breaks and warm / hot restarts are ok I'm going to drive this a while and see how it performs. If I change anything it will be continuing to reduce the injector constant gradually as PE seems lean.

Hoping to work on SD VE tables this weekend.

kur4o

09-27-2019, 02:19 PM

Great you are having serious progress.
The afr target and cyl cut use the same byte in mode4 message and are exclusive.
v4 has the same limitation. In v4 you can change the cyl trims on the fly.

The cyl balance test wasn`t very reliable. The fans skew the map readings and afr is moving a bit and is not too fat. The wideband is almost topped at 16.5.
A new test with constant variables will give better picture, since you also change the rpm target.

The cut test produce some good results showing some lean cylinder on the right side. Could be #6 since it is the one that fires after #3. More air left for #6 to draw. Or it could be #2 the companion firing cylinder of #3.Generally all of the right side calibrations have more fuel added. I checked the bin and the trims looks really conservative. It is not uncommon to have 5-10% difference even with stock form.
I ended with this settings in firing order.

1.070
0.984
1.031
0.938
1.039
1.008
0.977
1.047

The result of drilling the passage holes even. #1 had much smaller hole since it draws much more air and now I got it at 1.07.
I tried to get them better with wideband test but failed. The test looked better and even but the feel and sound of the engine sucked so I revert to the best feel.
I always have 1-3 problems cylinders that didn`t react to changes as expected. I am planning to play more with them if I have the time. I suspect you need much more air on the right side for some reason. It could be 1 cylinder leaning or it could be mutiple.

As I remember the passage goes to mid of the intake plug and than goes to both sides center and than goes to intake ports. 2 90* turns till it reach the ports.
#2 and #4 have the shortest path. #6 and #8 the longest. #7 fires after #5 and for #7 there is almost no air left. Also there are 2 different designed intakes. the one have a center plug the other doesn`t and there are some other visible changes in the bottom shape. Do you remember which one you have.

I noticed that the rpm and map blm boundaries are also higher in the bin. The crosscounts could be lower because of the different settings in the vette bin. I will try the settings and compare the results with some older logs.

kur4o

09-27-2019, 05:45 PM

I couldn`t resist and made a quick log with the vette CL settings. To my surprise it worked perfect with SLP midlenght headers and GM 846 cam. The right one is a little shorter and the 02 bung is the header flange, the left is longer and the 02s is down the pipe, like 8-10 inches down the flange. SO there is considerable difference in the 02s placement.

So GM had perfect CL idle settings for radical cams and headers all the time and nobody noticed. Now I will have to ditch my Open loop idle patch, but first I will do some driving and see how it goes on the road.

Later I see what settings are changed and make analysis how it works in the bin.

spfautsch

09-27-2019, 06:19 PM

The afr target and cyl cut use the same byte in mode4 message and are exclusive.
v4 has the same limitation.

So changing the Target AFR table is the only way to lock AFR for this test?

In v4 you can change the cyl trims on the fly.

I may break out a mouse and make a serious effort to utilize it this weekend.

I have a few interesting items to report from last night's changes. First I tamed down some of my individual trims on the right bank because I was smelling fuel at closed loop idle. I also took a good bit more out of the injector constant which is now at 41.88.

The first thing that I noticed is that startup went to hell when it was starting almost perfectly yesterday afternoon prior to the individual trims change. So I think I was on the right path and removing fuel from the right was a mistake. It seems to me (and this is just a theory) that the prime pulses have to be spot-on to get the usual immediate fire and run we all know so well with this setup. I had just never considered that the individual trims were applied to the prime pulses.

Also, this log of my morning commute compared to the other three days has the lowest difference between cross counts bank to bank - 7218L, 6010R of 31811 records. I don't know if this is coincidence or a direct result of one of the things I changed - it is pretty humid and there are spot showers popping up today. I guess I need a weather station in my car.

Lastly, AFR at PE is closer to commanded than ever, so I think I'm pretty close to the right injector slope.

The cyl balance ... The fans skew the map readings

I've been forcing the fans on low because I don't want them kicking on during the test and changing accessory load. I assume this means your method is to force them off?

The cut test produce some good results showing some lean cylinder on the right side. Could be #6 since it is the one that fires after #3. More air left for #6 to draw. Or it could be #2 the companion firing cylinder of #3.

I'm not sure it matters but my assumption is that #4 is the one robbing fuel from #3 because it's adjacent in firing order and closer to #3 injector than #6. It will be on the intake stroke while the #3 injector is firing, whereas when #6 is on intake #3 would have just finished depleting any fuel left in it's intake port. I think #6 is what's robbing fuel from #5 because it also has the same relationship as 4 to 3.

Generally all of the right side calibrations have more fuel added. I checked the bin and the trims looks really conservative.

I arrived at this after setting them all to 1.0. I tried using the stock trims at one point and it wasn't good, but I could have had other things way out of whack at that point so I'll revisit that this weekend. I feel like I'm going to spend 90% of my tuning time nailing down the cylinder trims.

there are 2 different designed intakes ... Do you remember which one you have.

I think I have a picture with the oil shield off somewhere. Will update if I can find it.

I noticed that the rpm and map blm boundaries are also higher in the bin.

I did customize the BLM boundaries when I first got it running as a 350 with this cam. I may revert this back and see if the BLMs are better distributed. It seems like 95% of normal driving falls into two or three cells. But I suspect this is just the nature of running a built engine on the street.

spfautsch

09-27-2019, 06:24 PM

So GM had perfect CL idle settings for radical cams and headers all the time and nobody noticed.

Knowing that the CL logic is controlled by a PID type feedback loop, I'd always wondered why no-one having these problems was discussing changing the gain and proportional tables.

spfautsch

09-27-2019, 09:41 PM

Reverting to trims from previous tune and then adding a slight bit to 1, 2 and 4 caused noticeable improvement to starting and idle sound. Also it seems to have much more torque when I let the clutch out too fast and bog it down towards 400 RPM.

Edit: Also, surge feels almost non-existent. When I pulled in the parking lot I left it in 2nd and let it slow down until it started bucking. I didn't log it but it was perfectly smooth down to what seemed to be idle target (800).

Sorry to quote myself but I had a revelation while out getting some air.

... my assumption is that #4 is the one robbing fuel from #3 because it's adjacent in firing order ... I think #6 is what's robbing fuel from #5 because it also has the same relationship as 4 to 3.

It may seem counter intuitive, but what if the way to get more fuel to #4 and #6 is to also increase fuel to #3 and #5?

Here's a back of the napkin sketch of how I envision parasitic fuel flow in a gen 2 LT-1.

1 -> 2

3 -> 4

5 -> 6
^
7 ⅃ 8

Obviously all these dynamics change when the throttle blades open, but maybe the way to get perfect balance has less to do with the size of the idle feed holes and more to do with parasitic flow.

LeMarky Dissod

09-28-2019, 04:44 AM

I did customize the BLM boundaries when I first got it running as a 350 with this cam. I may revert this back and see if the BLMs are better distributed. It seems like 95% of normal driving falls into two or three cells. But I suspect this is just the nature of running a built engine on the street.You should [continue to] customise the BLM RpM and MAP boundaries, based on how you'd drive if you were trying to achieve better MpGs.
The OEM boundaries are silly even for a bonestock car doing the EPA's CAFE MpG test, nevermind driven normally, nevermind modded, etc

spfautsch

09-28-2019, 05:08 AM

Thanks, I'll give it some thought once my SD VE table is dialed in.

Funny you should mention MPGs and CAFE what-n-such. I filled up tonight to find I've averaged (http://www.gearhead-efi.com/Fuel-Injection/showthread.php?6767-DIY-LTCC-or-similar-system-for-LT1s&p=77974&viewfull=1#post77974) better than 27 mpg over the last four days. I've nearly worn a hole in my left forearm pinching myself. Sure enough, I'm not dreaming!

LeMarky Dissod

09-28-2019, 06:35 AM

One mechanical issue that is nearly universally overlooked, is the idlespider.

The throttlebody butterflies feed the two larger main holes on top which empty into their respective cylinder runners.
The Idle Air Controller feeds the smaller hole underneath the two larger main holes, and follow paths parallel to the main runners, emptying into each respective cylinder's runner near the very end of their respective paths.

So far, I have not seen a single LT1 inlet manifold that has had its idlespider properly cleaned during a socalled cleaning. This will definitely have an impact on your idle, as it does for us all.

Hope this helps.

spfautsch

09-28-2019, 03:44 PM

Clogged ports certainly crossed my mind (http://www.gearhead-efi.com/Fuel-Injection/showthread.php?8105-95-LT-1-Idle-Cell-Comparison-Humidity&p=77936&viewfull=1#post77936). I did pull the throttlebody with the intention of getting an inspection camera in there, but forgot that the idle passage makes an immediate 90 degree turn so that was a bust. The best I could do was verify the holes weren't plugged by viewing them from the primary ports. I wonder if a bath in sea foam might be possible without yanking the intake.

With that said, I'm not seeing a massive BLM split at idle, and it switches sides as the throttle blades open so I'm somewhat of the mind this might be a function of airflow dynamics caused by the cam and headers.

kur4o

09-28-2019, 11:29 PM

Untill the engine run flag is set there are no cyl trims adjustments.
The run flag settings comes from 2 tables 22b4 and 22bd.

#1 inj - #7,#2 cyl
#8 inj - #2,#1 cyl
#4 inj - #1,#8 cyl
#3 inj - #8,#4 cyl
#6 inj - #4,#3 cyl
#5 inj - #3,#6 cyl
#7 inj - #6,#5 cyl
#2 inj - #5,#7 cyl

The fuel robbing theory is really interesting and explain some of the issues I have had before. Running great when cold and turns to crap slowly with heat build up.
Since the end of injection target is set to inject fuel straight into the cylinder at colder temp, there isn`t anytime left for fuel to move around ports. When the engine gets warmer the target slowly move to 180* for y and f-body and there is 180* duration that fuel is sprayed and stays in the port before the intake valve opens and it get sucked in the cylinder.

Another addon to the theory that is not mentioned is the valve overlap, under some circumstances it provides natural egr effect and can reduce the oxygen concentration in the mix.

So there is 180 degrees before the fuel get sucked and can move to other cylinders. I made a quick chart at possible fuel robbing combos the first is the injector firing and than cyl#s intake open at 0* and *90 after end of injection occured for the given injector #.

If #3 don`t fire #4 and #8 can get really lean.

I will have to do another chart for least resistance path for fuel robbing and link it to the previous chart to find the most possible combos. The diagonal path seems with the least resistance.
6,8 gets fuel from 1,3
5,7 gets fuel from 2,4
And between adjacent ports could be the other least resistance path
1<->3
5<->7
2<->4
6<->8

spfautsch

09-29-2019, 01:04 AM

I'm shocked that the individual trims aren't applied during cranking because I've had it starting great and then reset the trims to 1.0 and it wants to crank forever.

I tried your eehack fork on the trims and the engine died when I clicked 'enable' in the trims group. If I clicked enable before starting it would die when I changed a trim cell. It also cut out for no apparent reason while driving it around the block to warm it up. The source I have is from a zip file named eehack_4.9_source06_nov_18.rar.zip that extracts a root folder named eehack4.7_darkmagic_06_nov_2018.

Whatever the problem is I'm pretty happy with how it idles with these trims. When it goes CL there's about a 10 point BLM split to the right, and then it moves to the left as the throttle opens more. It's running incredibly well when I get on the pedal and I'm still in utter disbelief that it made 29.4 mpg over all the driving I did between Tuesday and Friday.

closed / open trims:
1-82 86
8-7d 7f
4-84 81
3-80 80
6-83 81
5-80 80
7-7e 80
2-83 84

I tried adding a lot to the right side and it didn't seem to make any difference other than idle got choppy. I also tried the above trims but added 3 points to #3 and the split seemed to get bigger.

I looked for an exhaust leak, but the only place the header could be leaking is between the primaries in the collector and I'm not finding any soot there.

Here's a balance test at 12.8:1. I forgot to switch the fan off but IAC and spark were locked even though I noticed advance was dropping to 26 once in a while.

I'm working on VE now. The only major driveability problem left is hot restarts.

I'm probably going to take the intake off and check the idle passages out for blockage later this fall / winter. It seems like a diode is gone in the alternator too because it's having tough time keeping both fans running. It's always something...

kur4o

09-29-2019, 09:26 PM

There was some minor bug of v4 patch regarding cyl trim and maf pump shot controls. The tables didnt copy to ram properly. Actually they were copied and cleared afterwards.

Here is the corrected bin_file.cpp to compile the fork. Don`t forget to rename the file to bin_file.cpp

Now it is tested and there is no stalling of the engine. There is a workaround to manually populate the tables. When connected press the cyl trims number 1 to 8 with 1 second interval and it will copy the values to the ram table.

Another thing that I had missed in the vette calibration is the off idle cyl trims. All set to 1 in f-bodies. It looks like Gm put alot of effort to tune the vette properly and the strategy is the front cylinders get more air and the rear less.
No clue how to tune that but it will be miss or hit I guess.
Some cylinders runnig rich and other lean but the mixture equalize in the end. A real nightmare for tuners.

A play with the closed loop controls might pinpoint the blm split at idle. It is very likely to be in the tune or tiny exhaust leak. The smallest pinhole in the solder joints can lead to major problems.

spfautsch

09-29-2019, 11:37 PM

Thanks for the update / patches. I'll try to test them out soon. I think I'm going to park it until the alternator comes in though. I have an electric water pump and if I lose another diode in this heat it could end badly.

Another thing that I had missed in the vette calibration is the off idle cyl trims. All set to 1 in f-bodies. It looks like Gm put alot of effort to tune the vette properly and the strategy is the front cylinders get more air and the rear less.

Strange they'd only do that on the y body. That's where 90 percent of driving happens.

Also worth noting are the constants at 126da and 126db that apparently dictate the transition points between using the closed, open and no individual trims.

Some cylinders runnig rich and other lean but the mixture equalize in the end. A real nightmare for tuners.

If a tunnel ram has any redeeming qualities it's airflow at WOT. The end.

I'm sure they learned a lot from the mini-ram and it's no accident just about every intake design afterwards used long runners with the inlet funnels as far apart as possible. I would imagine it also had something to do with the LS engines' different firing order. Airflow dynamics are everything...

A play with the closed loop controls might pinpoint the blm split at idle. It is very likely to be in the tune or tiny exhaust leak. The smallest pinhole in the solder joints can lead to major problems.

I'm sort of thinking it's an actual lean condition on the right side because I'm still fighting with hot restart.

kur4o

09-30-2019, 12:29 AM

Do you suspect less fuel than needed. You mentioned that the injectors you have are reworked version of some stock gms with p/n on them. If you can read the part number it might be possible to rip some data for stock application. At least the low pulse slope. I bet it is more related to the coil than the actual hole of the injector.

I noticed that ls1 type injectors are very sensitive on lower voltage. At startup it is not uncommon for the voltage to drop below 8-10volts during cranking. So the injectors prime pulses will be at that lower voltage. It could be that the shorter open time at high coolant temp can lead to not opening the injector.

You can test cranking for half second than stop, wait 5 seconds, if it starts right on the second time it could be less fuel than needed on the first attempt.

Can you describe the exact condition when hot restart fails. I am only having problems on the first start if it had been sitting for more than 4-5 hours no matter cold or hot.

spfautsch

09-30-2019, 01:34 AM

I'm honestly baffled by what it's wanting. I started suspecting offsets because my alternator isn't maintaining the battery charge.

But, get this: I've tried changing the cranking AFR from 10:1 to 16:1 for the 80c row of the cranking AFR table and it wants to crank until it switches to using the MAF and then it fires immediately.

Once the coolant temp starts to dip below 75c it starts firing sooner and when below 72c it fires immediately pretty much every time. I think I'm going to look at the end of injection table and see if it may be involved, but I'm sweaty and tired.

The injectors I have are GM 12561462. I need to look at the datasheet I used to interpolate the offsets from. It's been 2+ years and I honestly don't remember.

Incidentally, what would I be able to do with the low slope on these? I was under the impression $EE could only handle low pulsewidth adders for injectors that flow less below the knee point. It was my understanding that all the Gen 3 bosch injectors flow more on the low slope.

I've hung up my hat for the day before I brick my ecu or burn my starter up.

Edit: The more I think about it, the more I'm considering zeroing all the cranking tables above 50c and see if it starts reliably. I'm a firm believer that having oil pressure before fire is a good thing, and this may be an idiot's way to make that happen. Obviously I may have problems if the MAF dies - guess I better test with and without it.

kur4o

09-30-2019, 02:34 AM

I think Gm made it start hot from maf reading, and it starts hard hotter in stock applications too. I too think that it starts longer when it gets hotter, above 80 degrees.

There are 2 tables that set engine run flag. When it is set it calculates fuel from maf.
$122b4 are the rpm above which the flag is set. The scalar for the table is x*12.5, on the xdf is set at x*25, so fix that first to get accurate readings.
At 80* you can lower the rpms to 160-180.

Than the other table is $122bd. This is a counter of how many cycles the routine will run with the rpm target reached, before the engine run flag is set.
Set it to 0 above 80 degrees.

The injectors are from 01-02 ls6 vette. I will dig a bin and see what are the stock offsets.
You can even try the ls1 patch if we score some good data. The fuel flow is linear with vacuum manifold so it won`t be hard to interpolate.

The low pulse slope is very important, when you have that much low pulse driving, below 4ms the flow is not linear and needs adjusting, usually adding time.

The end of injection target is pretty much the same above 44*.

spfautsch

09-30-2019, 03:08 AM

I think Gm made it start hot from maf reading, and it starts hard hotter in stock applications too. I too think that it starts longer when it gets hotter, above 80 degrees.

I don't recall anything like this before the stroker build. Whatever the case, as long as it starts reliably I'm happy.

I guess I need to put a test light on one of the injectors and see if they're being pulsed at all before the engine run flag is set.

The injectors are from 01-02 ls6 vette. I will dig a bin and see what are the stock offsets.
You can even try the ls1 patch if we score some good data. The fuel flow is linear with vacuum manifold so it won`t be hard to interpolate.

Wasn't that a 4 bar application? If so the offsets would be different for 3 bar.

spfautsch

09-30-2019, 04:22 AM

Here's the datasheet I found (somewhere????) for the modified LS6 injectors. As with all Bosch datasheets it specifies 39.15 psi as the target pressure.

I'll work over this tomorrow to see if I adjusted the offsets for 3 bar or just copied + pasted.

spfautsch

09-30-2019, 06:38 PM

I think my prime pulsewidth foolishness might revolve around the $12680 multiplier that I'd completely forgotten about. It looks like we talked about this a couple years ago here (http://www.gearhead-efi.com/Fuel-Injection/showthread.php?6616-EE-LT1-Injector-Swap-Running-Rich/page3&highlight=12680), and I'd just blocked it out of memory.

I'd completely forgotten changing this multiplier because I didn't put it in the Startup AFR category in my .xdf, but it would explain why it wouldn't start with the stock injectors. This multiplier would give a lot more resolution for these prime pulsewidth tables so I suppose I need to figure out how it works so I can account for it in my spreadsheet.

Edit: take a look - the table on the left is what the default parameter for the 12691 table gives (1.5625*X). The one on the right is using the 12680 multiplier X*($12680*0.00015625). The conversion formula is suspect but I believe the prime pulsewidths I was using from the stock bin were way off so fuel mass conversion was equally way off.

14693

kur4o

09-30-2019, 10:11 PM

That scalar is used to convert 8bit value to 16 bit BPW. AT stock it is $6666.
It will definitely offset the pw to hell if you had it set at that value all the time.

SUb_7781 is where the conversion takes place. It is something like $6666 [66{=a}66{=b}]
c= 8bit value from table. It is something like (c*a) + (c*b)

Revert to stock and start from there. I just run a hot restart and it started immediately.
With that lower scalar it could be sub 1 ms pw you are getting.

EDIT: the calculation formula is a little different
You take from C*A only the first byte if the result is 00CC you take 00 and add it to c*b

$05 8 bit value will be converted like this 05*66{b}= $01FE
05*66{a}=1FE +01 =1FF * 0.0152587891= 7.7972412301ms

If you change the scalar follow the rule to change it like $5555 or $4444 , than set your own scalar calculations in tuner pro for PW to see the corrected results.

edit2:
(c*a) + (c*b) should be c*b + c*a

spfautsch

10-01-2019, 12:38 AM

Looks like by scaling the table at 12691 I had 2s in the 80c and hotter rows. With the multiplier I had in 12680 that resulted in 2.7ms prime pulses. Not too short to fire the injector, but an awful lot less fuel mass.

I guess it never occurred to me back in 2017 when it was being discussed that the 12691 and 12682 tables were presented in the wrong units in the xdf. I've created new definitions for them showing the decimal value and with a big "DO NOT EDIT" note.

kur4o

10-01-2019, 12:33 PM

So you got this injectors redrilled.
here (https://www.bosch-automotive-catalog.com/en/product-detail/-/product/0280155931)

The datasheet you have are for some ford injectors which some belief are the same one. I will rip some bins to get you the offsets. Hope they don`t change with the redrill and are coil dependant. Some testing will prove it.

spfautsch

10-01-2019, 05:27 PM

I just noticed your math on the base pulsewidth scalar - very interesting. Confounding actually. No idea how to pull the MSB and LSB out of a table value in a xdf definition.

Quick question on a related subject, does anyone know if the injector offsets are applied to the prime pulses? I'm guessing that's handled in the processor we don't have the ROM for because I'm not finding any references to the table address in the public disassembly?

Edit: So if I'm following you correctly, you're saying the pulsewidths in the left table of the screencap in post #95 were correct for 0x6666 scalar? I'm struggling to understand why this scalar is used like this and why the limitation of having 0x3333, 0x4444, 0x5555.

kur4o

10-01-2019, 08:25 PM

It is 8bit processor calculating 16bits value. I too haven`t found way to get a usable decimal scalar out of it.

It is just easier to calulate it that way, you can throw any number in there but the first byte is the most important. You can rescale the PWs with new scalar but have to find how to convert it to decimal. I guess it is compromise between low resolution and max usable value. If you lower the scalar you can get better resolution but will loose the max possible value.
01*6666 = $66
02*6666 = $CC
03*6666 = 1+ 132= $133
ff*6666= $65+$659a=$65ff

01*5555= $55
02*5555=$AA
03*5555= $FF
FF*5555=54+54ab=$54FF

01*4444=$44
02*4444=$88
03*4444=$CC
ff*44444=43+43bc=$43ff

to convert the results to bpw ms the scalar is x*0.0152587891
SO with different scalar the minimum step will be
$66[102]*0.0152587891 =1.556316ms resolution
$55[85]*0.0152587891 = 1.29693 ms
$44[68]*0.0152587891 = 1.007028ms

and the max usable value will be
$65ff[26111]*0.0152587891 = 398ms
$54ff[21759]*0.0152587891 =331 ms
$43ff[17407]*0.0152587891 =265ms

I just noticed that the 1 resolution is really close to the scalar in the xdf (1.562500 * X) + 0.000000 vs 1.556316ms
Actually the numbers missed the rounding so it it is a close match.

I guess you need to find the 1 resolution for the scalar and use it as a conversion factor in the tables.

spfautsch

10-01-2019, 09:25 PM

I'm not sure it makes complete sense to me yet but thanks for looking into it!

I've already found by hand that 0x44(44) puts me 2% rich of target so that should be close enough. I was just laboring under the false assumption that this was a 16 bit scalar when in fact it's 8 bit and optimized somehow (for speed?)? I don't really want to delve into motorola machine code but why couldn't something like 0x4646 be valid?

I don't think it's a problem losing maximum possible value - the only reason to change this is to accommodate larger injectors so a tuner theoretically shouldn't need the ability for larger maximum PWs here.

Will give it a try tonight.

I was able to test with your fork of eehack last night and think I must have been mis-using the trims initially. It wasn't clear to me at the time that the ram table needed to be populated manually before enabling it. I know it would be an undertaking, but some of that stuff deserves a bit of a how-to write up in the readme or something. There's only so much intelligence that can be squeezed into the mouse-over tool tips.

Edit: Whatever the case, I suppose the parameter definition for 12680 should be setup as a two byte table instead of a 16 bit scalar.

kur4o

10-01-2019, 10:57 PM

It wasn't clear to me at the time that the ram table needed to be populated manually before enabling it. I know it would be an undertaking, but some of that stuff deserves a bit of a how-to write up in the readme or something.
It wasn`t supposed to but there was some bug in the patch. I am doing a reworked version that will fix some stuff and adds off idle cyl trim corrections.
Definitely a guide of how to will be needed.

Ok follow this guide.

Set the scalar to $4444

Set the conversion factor of the prime pw tables in the xdf to
(1.007028* X) + 0.000000

Now it will be pretty close.

It could be $xx xx for linearity in the calculations.

spfautsch

10-02-2019, 12:37 AM

It wasn`t supposed to but there was some bug in the patch. I am doing a reworked version that will fix some stuff and adds off idle cyl trim corrections.

You're really a glutton for punishment!

I'm not that worried about seeing what calculated PW the .xdf shows me - I'm more interested in how it works. But if it does maybe it's worth letting people know about.

Incidentally I found a 2002 ls6 bin here (http://www.gearhead-efi.com/Fuel-Injection/showthread.php?2741-Looking-for-2002-corvette-bin). I haven't been able to find a definition file to look at it with but maybe I have one in Jet DST. I think I deleted the virtual I had it installed on so I'll have to find the cd and build a new VM.

Also, the engine was exceptionally quiet last night for some reason when I first started it and I think I might have heard a faint pecking sound near the edit: right collector. I'm going to pull the clamp loose tonight and see if there's any carbon in any of the reliefs.

kur4o

10-02-2019, 10:55 PM

I already had posted a printout of the bin here here (http://www.gearhead-efi.com/Fuel-Injection/attachment.php?attachmentid=14618&d=1568235294)

You can search it for injection or injector to find the offset tables.

spfautsch

10-03-2019, 05:48 PM

Hot restart seems to be cured as far as I can tell. I didn't get a chance to test thoroughly last night or the night before, and a front moved in today so the summer like weather seems to have moved south (at least for now). But I should be able to tell from a warm restart in a few hours if it's right. How's yours - I thought you'd mentioned having sluggish starts after cooling for 4-5 hours?

Question on your fork of eehack - should the MAF indicator on the dashboard go "off" when switching to SD mode? I'm currently running a MAF tune and clicked the SD button before starting it this morning, but I was seeing airflow from the MAF and the green indicator was lit the entire time. The engine also stayed running and showed no signs of change when I toggled back to MAF at idle. I'm just curious if my log that was supposed to be for tuning VE is going to be useful.

I already had posted a printout of the bin here

Sorry, had completely overlooked that. Thank you! Interesting - GM's using some really low offsets in the 14v row compared to what the datasheet shows. 10v is pretty close though. Corrected high slope is also a good bit different, so I have to question the accuracy of the Ford Racing injector data.

I'm starting to think I just need to build my own injector flow bench and figure all this out on my own.

kur4o

10-03-2019, 10:37 PM

, but I was seeing airflow from the MAF and the green indicator was lit the entire time.

You have just found another bug in the patch related to manual transmission. It turns out that the routine that makes the switch between maf and speed density modes doesn`t run with manual bin. An easy fix for the next release.
So for now the sd/maf switch will be unoperational on manual bins till the next version is out.

If you are eager to test sooner there is only one byte that needs to be fixed for the button to start working.
AT $A0C9 change the stock value of $F8 to $54 in hex editor and save. Than save the bin with tunerpro to update the checksums and flash with eehack.

spfautsch

10-03-2019, 11:29 PM

I'll press on tuning SD with the MAF disable flashed. Hoping to have VE "close enough" by tomorrow. After that I may need some advice on the MAF pumpshot tuning.

Once you have the patches finished I'd be happy to test.

kur4o

10-04-2019, 12:02 AM

Here is the first beta source.

If you have a socketed pcm and wants to be a guinea pig go ahead.
My only concerns are for some typo errors in the patch section. Otherwise I am sure you will like the new controller interface. It is also resizable and can hide the less used stuff.

With delta ve control you can nail the overall off condition really fast.

kur4o

10-04-2019, 10:55 AM

I justflash a stock bin with the beta and evrything looks fine. There is no brick situation and the engine starts. I didn`t have the time to test functionality.
So you can safely flash with the beta. There will be no suprises.

I also noticed how to add info buttons to explain how to use the extra controls. Next on the list.

Now with stock bin I found that the engine starts harder, at the 3rd 4th attempt, but the cold idle was incredibly smooth and tamed. I haven`t heard the engine running that healthy from a long time.
So what was the difference. 4 less degrees advance, stock underspeed and maf tables and much more fuel being dumped. I guess I need to be back on the tunnig board.

Some log for comparison.

spfautsch

10-04-2019, 09:52 PM

I'll probably pass on testing until I'm done getting my tune at a respectable baseline - it's coming along really well so I'd rather not get distracted with other stuff right at the moment. I've been switching between your fork and an old version of the original that I've customized.

Now with stock bin I found that the engine starts harder, at the 3rd 4th attempt,

Are you stock cubes but bigger injectors?

I've been messing with that 12680 multiplier with the stock base pulsewidth table. For the sake of discussion let's call it a two byte table where both bytes must match (similiar to the target AFR tables). I followed your advice at first and used 0x44 (in both bytes). It seemed to be a little lean - hot restarts popped right off if I bumped the key and then hit it again after 5 seconds. So I went to 0x55. It hits every time but seems to be pretty rich there. I tried 0x4F and it still seems a bit rich so I'm working my way down from there. Whatever the case it's hitting on the third or fourth low res pulse every time now regardless of ECT.

kur4o

10-05-2019, 12:17 AM

an old version of the original that I've customized
You can share the new goodies and I can try to include them in the new release. The key shortcuts are very usefull, I still haven`t had the chance to expand them to more controls.

The 2bytes table whenever is changed, the prime pulse tables must be adjusted to compensate for the new offset factor.

With stock settings at $6666 the stock table will look like this

215.63 215.63
209.38 209.38
120.31 120.31
59.38 59.38
39.06 39.06
17.19 7.81
12.50 6.25
7.81 4.69
7.81 4.69
7.81 3.13
6.25 3.13
6.25 4.69
6.25 4.69
6.25 6.25
6.25 6.25

When the 2 byte table is changed to $4444
The table will look like this with new conversion factor (1.007028* X) + 0.000000

138.97 138.97
134.94 134.94
77.54 77.54
38.27 38.27
25.18 25.18
11.08 5.04
8.06 4.03
5.04 3.02
5.04 3.02
5.04 2.01
4.03 2.01
4.03 3.02
4.03 3.02
4.03 4.03
4.03 4.03

So whenever the 2 byte table is changed you need to readjust the prime pulse tables.
With $4444 scalar the resolution goes to 1ms intervals vs 1.5 ms at stock $6666 settings.

hot restarts popped right off if I bumped the key and then hit it again after 5 seconds
I have done this on a cold start with the same result. Adding fuel didn`t help at all, so my assumptions it is actually rich.
Hot restarts are tricky to tune and test due to some variables that can change on the fly. The main issue will be how much fuel left on the walls unburned with the engine switching off.

The prefect start will be at sub 0.3seconds only on the prime pulse fueling. Than the crank ve tables hit with 2 different afr tables based on low res pulses.

I found some 96 lt4 bin and explore the differences. The idle advance is set at 19* at cold coolant and gradually decrease to 14* while warming. On the lt1 f-body is reversed. Really interesting.

I have changed the injectors onlyThe ls` ones flows 24lb at 3bar, so no change with fuel flow. With the old some crap accell 24lb/hr I didn`t have any hard start issues even with completely stock bins.
So it could be the injectors needs some tweeking or the ls1 injectors patch I run needs some attention at starting routine.

I also tried the vette CL settings and the logs I made are on par with yours. Slight blm split and less crosscounts. The idle runs perfect in closed loop.
I have some theory how to tune the CL settings for more crosscounts, but need to test some of them to confirm how they work.

The off idle trims made huge difference at cruising around 1500rpm. I wonder how far we can go using them safely before the air stream equalize in the cylinders.

Stock settings are at 0-14% tps, <3000 rpm and no catprotect and PE.

On the lt4 bin the tps is at 24%.

I want to increase the tps as much as possible and add new conditions such as airflow or map thresholds.

spfautsch

10-05-2019, 05:06 AM

You can share the new goodies and I can try to include them in the new release. The key shortcuts are very usefull, I still haven`t had the chance to expand them to more controls.

There really isn't anything new or terribly worthy of sharing. I did recently fix a problem where the IAC override would set it to 4 (steps) and not change if switched to step mode. But that might have been a bug I created. Aside from the key accelerator stuff it was mainly making fields like knock count and stuff big enough this old blind guy could see them on the 11" screen.

With stock settings at $6666 the stock table will look like this

...

When the 2 byte table is changed to $4444
The table will look like this with new conversion factor (1.007028* X) + 0.000000

I think I'm pretty clear on that. I suppose what I'm confused on is if I were to set 12680 to 0x4c4c why wouldn't the prime pulse table look like this with 1.16ms intervals:

160.034 160.034
155.396 155.396
89.294 89.294
44.067 44.067
35.950 35.950
17.395 17.395
12.756 12.756
5.798 5.798
5.798 3.479
5.798 2.319
4.639 2.319
4.639 3.479
4.639 3.479
4.639 4.639
4.639 4.639

???

You won't hurt my feelings to tell me why I'm being a bonehead as long as I can learn something from it.

Hot restarts are tricky to tune and test due to some variables that can change on the fly. The main issue will be how much fuel left on the walls unburned with the engine switching off.

The prefect start will be at sub 0.3seconds only on the prime pulse fueling. Than the crank ve tables hit with 2 different afr tables based on low res pulses.

You'll get no argument from me that restarts in general are tricky with different injectors, displacement, etc. I tested the above multiplier tonight (0x4c4c) and had a very obvious flood at 73c restart (fired then died, required pedal to clear on 2nd attempt). This is why I was curious if we have any idea if injector offsets are added to the prime pulses. Then we get to the subject of having good injector data. My ecm is seeing 10v average during cranking. that's about 1ms of offset for the stock injectors. If my offsets are wrong, how far off are they?

So it could be the injectors needs some tweeking or the ls1 injectors patch I run needs some attention at starting routine.

You should put my experimental ignition system on it so you have another unknown! (joking)

I also tried the vette CL settings and the logs I made are on par with yours. Slight blm split and less crosscounts. The idle runs perfect in closed loop.

Mind sharing a screenshot of the BLM analysis?

I'm pretty happy with my VE table after tonight's log so I think I'm going to turn the MAF back on and fine tune a bit. Probably mostly on the damned prime pulse / restart issue. Beyond that I'd be happy to test any closed loop tweaks you would care to share.

I want to ... and add new conditions such as airflow or map thresholds.

If you can do that I can surely conquer startup fueling.

kur4o

10-05-2019, 04:59 PM

why wouldn't the prime pulse table look like this with 1.16ms intervals:

It is exctly like that. I have nothing to add. What I mean is that you can fine tune the tables from there.

So for now I want to end the disscusion on the pw scalar since it is all clear.

The 2 bytes table should be called. Multiplier to startup prime PW table.

The formula for the multiplier will be d={x*0.0152587891}/256

The multiplier decimal result must be entered in the crank prime tables conversion field with the following formula {x*d}.

Now you can start fine tunning with the new multiplier.

The InJ vol offset are for sure used during cranking. I improved that part of my patch with major improvement on the first start attempt.

I am starting to suspect that my ls1 inj flows more than specified at 100kp range. That explains the hard startup[rich condition] with stock settings.

I am getting rich right side at cell 6 only, 3-4 points split. I check logs with older settings and the split is also there at #6 cell. At other cells the split is negligible.
I also changed the off idle trims but it seems they don`t affect the split.

I experience zero split at idle so far. I tested it for a short period so it could be not very reliable.

I thought for a way how to find the break point for off idle trims condition.

Setting the trims on one of the side richer and look for the conditions where the blm splits equalize.
Than set the tps map thresholds.

I already made the map threshold patch but it interferes and interacts with the v5 patch so it is a little tricky to install. If you want to give it a try I will set you a bin. If it turns out bad I can change it to an air flow condition.

I suspect the $26fb table to be responsible for the crosscounts. It is also the cure for idle CL along with the swing voltage.
AT $26fb slightly increasing the values at cells 2-5[don`t touch the first cell- it is for idle]

y-stock ;f-stock ; my old settings
0.40 0.80 1.20
0.18 0.18 0.35
0.13 0.13 0.25
0.11 0.11 0.23
0.10 0.10 0.20

As you can see I thought more is better but not for the first cell. For the first cell less is better, almost twice less at stock y-body vs f-body.
At cell 2-5 increasing the value can lead to higher crosscounts. Field discovered by mistake. Not sure how good is this for engine running healthy but you can give it a try.

$270c is also different but not in the idle region. It also can attribute to the crosscounts but needs testing.
The main difference is at cells 2-5
vette; f-body
0.3 0.5

and the swing voltage
380 for vette
vs
460 for f-body

kur4o

10-05-2019, 06:28 PM

Your problems with hot restart might be due to crank spark advance. I ripped the settings from different calibrations.

Interesting I am running 2 degrees at 80*C and 7* at 20*C stock. The closest to cammed car is lt4 setting at 5* and 11*, all other calibrations have 8-10 * and 13*.

I suspect this might be culprit with my issues of first cold start.

So it could be hot restarts needs less advance and first cold start needs more advance.

spfautsch

10-05-2019, 10:30 PM

Interesting there's that many differences between the stock bins. On a whim I just tried the lt4 cranking spark table. Before this change pretty much all starts with anything under 0x50 in 2680 was somewhat slow and slobbery. With the lt4 startup spark it start and ran cleanly. Still hit on the 3rd or 4th low res pulse.

Unfortunately the weather has changed dramatically here so I almost need to wrap my engine and radiator in a blanket to get a hot restart at 80 ECT. Ambient is > 10c cooler than last weekend so air density way different also.

kur4o

10-05-2019, 11:08 PM

The lt4 bin also have a lot more fuel at startup. No idea why.

Here is printout for most of the popular combos.

spfautsch

10-06-2019, 05:34 PM

Curious what the 12680 multiplier is - the injector constant shows 27 lb/hr. I didn't know the LT4s got bigger injectors.

They did get completely different heads and redesigned intake with taller ports so that would explain the different individual trims.

kur4o

10-06-2019, 06:26 PM

RESERVED:2B41 fdb $34BC ; prime pulse table multiplier
RESERVED:2B43 fcb $C3 ; + ; initial prime pulse set
RESERVED:2B44 fcb $AE ; î
RESERVED:2B45 fcb $63 ; c
RESERVED:2B46 fcb $3A ; :
RESERVED:2B47 fcb $2C ; ,
RESERVED:2B48 fcb $1B
RESERVED:2B49 fcb $14
RESERVED:2B4A fcb 9
RESERVED:2B4B fcb 9
RESERVED:2B4C fcb 9
RESERVED:2B4D fcb 7
RESERVED:2B4E fcb 7
RESERVED:2B4F fcb 7
RESERVED:2B50 fcb 7
RESERVED:2B51 fcb 7
RESERVED:2B52 fcb $F4 ; ¯ ; first prime pulse adder
RESERVED:2B53 fcb $ED ; ý
RESERVED:2B54 fcb $88 ; È
RESERVED:2B55 fcb $43 ; C
RESERVED:2B56 fcb $37 ; 7
RESERVED:2B57 fcb $1B
RESERVED:2B58 fcb $14
RESERVED:2B59 fcb 9
RESERVED:2B5A fcb 9
RESERVED:2B5B fcb 9
RESERVED:2B5C fcb 7
RESERVED:2B5D fcb 7
RESERVED:2B5E fcb 7
RESERVED:2B5F fcb 7
RESERVED:2B60 fcb 7
RESERVED:2B61 fcb $F4 ; ¯ ; 2n prime pulse adder
RESERVED:2B62 fcb $ED ; ý
RESERVED:2B63 fcb $88 ; È
RESERVED:2B64 fcb $43 ; C
RESERVED:2B65 fcb $37 ; 7
RESERVED:2B66 fcb $1B
RESERVED:2B67 fcb $14
RESERVED:2B68 fcb 9
RESERVED:2B69 fcb 9
RESERVED:2B6A fcb 9
RESERVED:2B6B fcb 7
RESERVED:2B6C fcb 7
RESERVED:2B6D fcb 7
RESERVED:2B6E fcb 7
RESERVED:2B6F fcb 7

The multiplier is $34BC, I suspect the conversion is set wrong on the data I have. You can recalculate the tables with that scalar.

There are some extra tables added on the lt4 bin.

Mat adder to MAF, useful at cold start lean
and MAT spark correction. can save engine at high load and high mat.

It will be relatively easy to add them to 94-95 code. Will be there any benefit of them.

kur4o

10-08-2019, 11:08 PM

While investigating my hard start issue I measured my fuel pressure and what a surprise. The crap adjustable aeromotive unit holds different pressure during pump priming. On a random basis it opens at either 2.8, 3.0 or 3.2 bar. Interesting when the engine starts it frozes at 3 bar. There is also high frequence oscilation in the gauge pointer when the engine runs. It doesn`t move but it is kind of vibrating. Could this be the reason for lack of fuel rail damper.

The flow rate range from 24lb/hr at 2.8bar to 25.6lb/hr at 3.2 bar. A huge difference for proper fuel at starting.

spfautsch

10-09-2019, 12:06 AM

Just wanted to report back that I've think I've made some progress on the startup tables.

The LT4 cranking spark table didn't seem to solve any problems by itself. I'm currently still running it, but ended up scaling the BPW multiplier to the point the resultant prime table was about 12% rich of original when corrected for injector slope (a total of 22% with the adder for displacement). This got things to a point where it fires nearly immediately in most conditions. In warmer cells I continued to notice a start then die condition that I suspected was flooding. Reducing the initial AFR enrichment in the effected cells (table 0x12b10) seems to have helped. The only obvious remaining warm restart issue I've experienced is a point when it's been sitting for 2-1/2 to 3 hours and it will fire immediately and stay running but seems like it's struggling to do so for the first several seconds. As much as I'd like to develop a scientific method for tuning this, it seems like the cam overlap complicates things.

Anyone have any wisdom on how the IAC adders related to a/c work? I'm feeling surge at low throttle cruise when the compressor clutch engages. There's a scalar / multiplier at 0x2691, a pressure vs steps(?) 2d table at 0x2692 and another 2d table at 0x26a3 with unknown row labels. I'm seeing IAC jump about 20 counts when the compressor output is enabled at cruise speeds, but I'm not seeing anything close to 20 in any of the tables so wondering if 0x2691 is a multiplier.

Also wanted to mention I was shocked how much more fuel it consumed the two days running in SD mode. 18.5mpg makes me wonder if I wasn't tripping on something when I read the pump after the previous fill-up (29.4mpg).

Edit: Sorry didn't see your post about fuel pressure before submitting this, but that's another item I changed that I can't help but think has thrown my startup fueling off. I'm hoping to check mine against another shop's gauge this week to see if theirs is any more accurate. Mine reads 40 psi with the pump running.

spfautsch

10-10-2019, 11:16 PM

So back to tuning the prime pulse tables - last night I decided to go ahead and cook my closed throttle BLM split into my individual cylinder trims, which amounted to adding approximately 10% to the right side.

1 82 130
8 7D 125 increased to > 129 81
4 84 132 increased to > 141 8D
3 80 128
6 83 131 increased to > 140 8C
5 80 128
7 7E 126
2 83 131 increased to > 140 8C

No other changes were made save fiddling with some IAC adders relative to A/C.

To my surprise all startup fueling went out the window. Cold, hot, and everything else required cranking until the run flag was set. I just reverted the individual trim changes and startup went back to how it's been for the past several days - not quite perfect but very good. Just a note to anyone tuning on their startup fueling tables - individual cylinder trims seem to apply to the prime pulse and cranking afr tables.

Another thing I noticed when looking at the individual trims - what I gather from the disassembly it appears to my untrained eye that the 0x126db "low" limit might be ignored for transitioning to the off-idle trim table.

688A 18 CE 26 E5 @290 ldY #$26E5 ; bt8_CylBal(AtIdle)
688E 12 26 08 04 brset L0026, #%00001000, @291 ; at idle (or closed throttle) status bit
6892 18 CE 26 DD ldY #$26DD ; bt8_CylBal(OffIdle)

I don't see where bit 4 of L0026 is set - so maybe I'm just missing something obvious.

kur4o

10-12-2019, 01:33 AM

Great experimental finds. Even though I still insist that trims are not used until the engine run flag is not set. In your case @32*C that will be engine spining above 300 rpm for 4 low res pulses. I guess that can be reached pretty quick and after that there is a flood from the trims.

Actually the conditions are pretty starightforward. Not in PE or cat protect mode, engine on flag and tps thresholds. I also added a map threshold but that needs some patching.
L0026, #%00001000 is the ZERO tps flag.

To do more experiments on starting I strongly recommend playing with the beta I send you. You can preset the trims before starting with ign on, and change crank ve upto 5%.

NOw on my case.

I messed with so many things at once that didn`t make sense and the results are good so far.
First I fixed the startup fuel flow constant in the patch, Now there is separate fuel flow settings at start up. Next loaded some modified crank ve based on lt4

14
14
14
14
11
11
9
8
6
4
2
0
0
0
0
0
0

Loaded stock lt4 settings for the prime pulse tables and the multiplier. Also lower the trims by 1 on 3 cylinders.

ON the first cold try it started much better than before{I have to notice that the engine was air flowed and there was no residual fuel left over, cranking with inj fuse off to check the cranking fuel pressure}. Hot restarts upto 60*C were also right on.

Now when it starts the rpm goes upto 1500rpm, indicating there is much more fuel left over during cranking. Some of the older logs shows barely 1000-1200rpm rise immediately after starting.

I guess now it dumps much more at prime time and leans after that.

About the blms, it could be inbuilt limitation for some splits possibilities, based on headers and exhaust pipe configuration. I have seen numerous bins with differen swing voltages set on the left and right side. Maybe GM tried to cure the split by changing the per side swing voltage.

spfautsch

10-12-2019, 02:53 AM

Great experimental finds. Even though I still insist that trims are not used until the engine run flag is not set.

Is the run flag logged directly by eehack? I only see engine runtime, and just assumed the AIR output was enabled in conjunction with the run flag. Unfortunately I didn't log any of my hard starts with the "fat" cylinder trims.

L0026, #%00001000 is the ZERO tps flag.

I was hoping I could set the 0x126db value to use the closed throttle trims until about 7% tps, but it doesn't appear to work that way. ???

First I fixed the startup fuel flow constant in the patch,

Are you talking about the 0x12680 multiplier???

About the blms, it could be inbuilt limitation for some splits possibilities, based on headers and exhaust pipe configuration.

I think I'm going to swap injectors bank to bank because it's possible my BLM split might have appeared after going to the modified injectors. Yesterday I found a couple logs made before I replaced the stock multecs where the left side was 2 points lean of right, but that was almost 3 years ago. I've taken the modded injectors out of the fuel rails several times since first installing them and I never paid any attention to which one was where. If there's one injector that's flowing a lot more than the others at low pulsewidths that would seem rational. If it's more than one it's somewhat unlikely I've put those two (or three) on the left side in two or three different builds, but entirely possible. I was really hoping these cheap modded injectors would suffice, but if they're the cause for my BLM split I'm probably looking at a set of m-9593-lu80 or some LS7 injectors.

Edit: Swapping injectors was a bust - right side still leans out when it gets to temp. Took the intake off. I was shocked to find oil wetting the bottom of the throttle body and in all the intake ports in the heads. The pcv hose wasn't wet on the inside, nor was the right side valve cover breather tube. But the breather inlet in the throttle body (the hole that opens into the top cover) was slightly wet. Anyway, I'm finding nothing to indicate that the idle passages are clogged but I'm going to clean it thoroughly while I have it disassembled.

Looking at the miniscule size of the #1 feed hole I'm seriously debating enlarging it some. Seeing the size difference next to the #3 feed hole from the bottom really emphasized how big a difference there is here. This alone seems like it could account for at least 50% of my split.

Another interesting thing I noticed were traces of the crud getting sucked out of the idle feeds, being deposited on the inboard side of the intake runners. I'll try to post a picture or two tomorrow, but this reinforces the notion that adjacent cylinders are robbing charge from across the plenum.

spfautsch

10-17-2019, 05:05 AM

I've had some time to disassemble and clean things, and I think my oil issue was caused by the intake gaskets being positioned too far down on the front ports. When it goes back together I'm going to remove the plastic locating pegs and glue the gaskets to the heads with RTV overnight before final installation to make sure they're where they're supposed to be. The oil infiltration problem was primarily on the 1 & 3 ports where the gasket misalignment was most evident, and the rest could have just been normal PCV function.

So far, I have not seen a single LT1 inlet manifold that has had its idlespider properly cleaned during a socalled cleaning. This will definitely have an impact on your idle, as it does for us all.

My intake is the 5 plug design identical to the one posted back on page 5 by kur4o [link (http://www.gearhead-efi.com/Fuel-Injection/showthread.php?8105-95-LT-1-Idle-Cell-Comparison-Humidity&p=77919&viewfull=1#post77919)]. I blew some compressed air through the idle air plenum and was able to feel some coming out of all four of the tapped plug holes (the center hole connects to the EGR passage). There is a bit of crud in these passages (far more in the EGR passage) so I decided to soak the manifold in some diesel fuel before taking a pressure washer to it tomorrow. I wasn't able to find a disposable pan big enough to submerge the portions I wanted to clean so I flipped it over and started pouring diesel into the idle plenum. Immediately I noticed this was connected directly to the PCV inlet, so I capped it off. I then poured what I felt was an alarming amount of fuel into this plenum. Eventually fuel started to trickle out of the IAC inlet hole below (actually above) the primary throttlebody inlets. I was shocked at how much fuel this passage holds, and that the level of liquid while stabilizing only 1/8" below the port orifice holes was barely trickling out of the IAC inlet that by my estimation is at least 0.8" lower (intake inverted). I suppose it would take a couple x-rays to figure out what's going on here but it seems this plenum has a complex internal layout.

I'm seriously contemplating enlarging the #1 idle feed hole. All the others are 15/64" / 6mm, but this one is 5/32" / 4mm. I'm thinking I'm going to go to 13/64" / 5mm. Anyone have any thoughts on this?

Terminal_Crazy

10-17-2019, 09:53 PM

I'm seriously contemplating enlarging the #1 idle feed hole. All the others are 15/64" / 6mm, but this one is 5/32" / 4mm. I'm thinking I'm going to go to 13/64" / 5mm. Anyone have any thoughts on this?

I did & can't say whether it did or didn't make any noticeable difference.
Idle ports are all the same size now.
I'm using the Edelbrtock LT4 manifold to match the head ports.
Machining is not very good.
Flash left in idle ports and ports exit the runners in different places.

Notes are:

On Edelbrok 7109 Air Gap Manifold LT4
Idle air ports are drilled badly into the sides of the port wall near the head
No 1 drilled at 4.5mm
2-8 Drilled at 6.mm but only allow a 5.5 drill through into the port

Stock GM Manifold idle ports are
No 1 measures 4mm - smaller than the Edelbrock
2-8 measure 6mm but are drilled all the way at 6mm
All ports are drilled at an angle half way into the port runner floor

HTH
Mitch

kur4o

10-17-2019, 10:59 PM

When I made the ports equal diameter, I got all kind of surging problems with stock individual cyl trim settings. I had to rework the trims it all the way and now #1 gets much more air and needs more fuel. You get to the point where balancing cylinders with more fuel at #1 is not possible. The reason GM restrict the air flow at #1, and some fancy drilling sizes on the other ports. I also remember that not all ports were equal at 6mm.

spfautsch

10-17-2019, 11:08 PM

Odd. All others in mine are 6mm exactly.

I went up two drill sizes on #1 to 3/16". Hopefully it wont wreck the tune. I didn't see any other reason the left side would be getting less air than the right.

kur4o

10-18-2019, 02:13 AM

Prepare to start all over with cyl trims.

I made some progress on the first cold start. Still needs some fine tunning. While experimenting it turns out that running around 20-35*C coolant at startup it idles best at 14* advance and a liitle fatter mixture.[Tuned by sound and feel.]
I will let you know the best setting I came up when it gets fine tuned.

You can monitor the rpm rise after startup to tell if it needs more or less fuel. If the rpm rise alot like upto 1500rpm it has more unburned fuel left on the walls from prime pulses.

kur4o

10-18-2019, 09:27 PM

Another vote for the adjustent cylinder fuel robbing.
The lt4 settings are

Cylinder Multiplier
1 1.05 3 0.96
2 1.05 4 0.97
5 1.01 7 0.98
6 1.02 8 0.96

Somehow my trims closely resembles this pattern with the exclusion of #2 1.039 and #4 1.024.
Next on the list is to test it on running engine and starts to cut #4 to around 0.98 and see what happens.

The fuel robbing feature seems to be more pronounced on cammed car than to be related with ports size and headers lenght and style. Interesting to know the stock size idle air feed holes on a GM production lt4 manifold.

Can you source some pics on the vette air induction system like filter box, passages, maf mount and so on. Are you running some form of CAI kit. I suspect that the air induction system is the reason there are different scaling in MAF tables across platforms.

spfautsch

10-19-2019, 03:06 AM

I suspect that the air induction system is the reason there are different scaling in MAF tables across platforms.

I also suspect it, times ten. In fact it's widely published that OEMs generally mock up the entire front of the car on a flow bench to develop their MAF transfer curves. Google: "helmholtz resonator". OEMs have been using them for > 10 years to smooth airflow in the intake ductwork for this specific reason (more accurate MAF transfer curves).

Here's the stock y-body intake. I believe all MAF equipped LT-1s got this intake setup. I'm also fairly sure the ZR-1/LT-5 (read: 450 hp off the showroom floor) also got an identical airbox and MAF but with a slightly different shaped throttlebody bellows.

http://www.pfautsch.com/wp-content/uploads/1018191814-150x150.jpg (http://www.pfautsch.com/wp-content/uploads/1018191814.jpg)

http://www.pfautsch.com/wp-content/uploads/1018191815-150x150.jpg (http://www.pfautsch.com/wp-content/uploads/1018191815.jpg)

This might be closest to a ram air Pontiacs. The only anomaly in this tract is that the Y-body engine is slightly offset towards the passenger side of the car. Anything that came with an elbow in the air ducting will likely have vastly different MAF curves.

kur4o

10-19-2019, 11:10 PM

The recommendation from GM is at least 4 inch of straight pipe before the maf sensor, to get reasonable air flow readings. That is for the newer style slot type of mafs but I guess it can be applied to older setups too.

Thanks for the pictures. It looks like a funnel, before and aftre the maf and is vastly different from the f-bodies. The vette maf curve reads lower from 2-3% at low airflow to 5-6% at mid airflow upto 10-11% at high airflows, compared to stock f-body settings.

I am running the slp version of CAI and there is some problem with lower map readings at high rpm, high load at WOT, especially on third and forth gear I am wondering if it is airflow restriction of some kind. Have you measured the map drop at wot pulls under high rpm and load. What are the max AFGS you have reached with the new setup.

spfautsch

10-20-2019, 12:24 AM

I haven't been doing much logging or tuning at WOT with this setup so unfortunately not much to report. I will try to pick through what logs I have but as I recall the few times I've gotten into it hard while logging were mainly corner exits or two lane highway stops / takeoffs where I didn't have much WOT time due to traffic. The majority of the hard throttle passes I've made thus far were a few un-logged joy rides I owed a couple guys that borrowed me tools and such where it broke the tires loose around 5500 rpm in 2nd and immediately bounced off the rev limit.

What I've read on the subject of MAF tuning (primarily Banish's stuff) has me thinking it will be something I touch as a very last resort if the wideband is still showing lean during WOT passes after I've removed as much as I'm comfortable with from the injector constant. The accuracy of the instrumentation I have to work with is inferior to what GM was using in the 90s by orders of magnitude, and I have a lot of unknowns like the modded 42lb mystery injectors.

spfautsch

10-22-2019, 05:56 AM

Initial start + warmup after cleaning intake and enlarging #1 idle feed orifice didn't show any considerable improvement. But after an initial warmup I started working with individual trims and unlike before was able to see responses in closed loop BLMs after moderate trim changes. This is what I'm running currently for individual trims.

1 131 83
8 127 7f
4 134 86
3 126 7e
6 136 88
5 127 7f
7 125 7d
2 135 87

The last log I made (attached) had left trims at 131 and right at 135 albeit for a short time, and at higher air density / lower IAT numbers than previous logs. This is the best cell 16 split I recall logging with the current build. Idle felt and sounded good. Unlike 2 weeks ago I feel like there's potential for improvement simply from continuing to fine tune individual trims. I also removed a small amount from the prime pulsewidth multiplier to account for the trim changes and had good results.

I think there's a lot of knowledge to be had from putting an oxygen sensor in all 8 header tubes. There is definitely charge scavenging happening and I feel like we're just scratching the surface of learning the airflow dynamics happening here.

spfautsch

10-23-2019, 04:27 AM

Here's a log with a WOT pull starting at timestamp 1120.0. [link (http://www.pfautsch.com/wp-content/uploads/drivelog-2019-10-22-1.zip)]

As you've noticed it seems like it's making a small amount of vacuum at 100% TPS, but this was going downhill in 2nd so very low realized load. MAF peaked just under 320g/s @ 5900rpm. I could have pushed it another 1000 RPM but my second cup of coffee hadn't quite taken effect yet and I figured it best not to. Notice my wideband must have gotten too cold and jumped back to it's warm-up routine.

Results from logging today was promising. The idle split was back as bad as ever this morning, but BLM splits in the main driving cells were much better then before pulling the intake. I have no idea if enlarging the #1 feed hole or eliminating the oil getting sucked into 1 & 3 were both factors or only one. Whatever the case, this seems to be improvement in pretty much all the driving BLM cells.

http://www.pfautsch.com/wp-content/uploads/blm-analysis-2019-10-22-150x150.png (http://www.pfautsch.com/wp-content/uploads/blm-analysis-2019-10-22.png)

I have one disassembly question that I would fault no-one for ignoring. I've been tinkering with the IAC adders and multipliers trying to eliminate the surge when the a/c clutch engages and disengages. It seems like it's improved in all areas except idle, where the IAC adder seems too much and the RPMs will flash up slightly before the clutch locks completely. I'm trying to figure out what the row labels should be for the table at 0x26A3. The cell values seem to be multiplied by 0x2691 and I believe the lookup is comparing to VSS but not 100% positive. Here's what I've done with the disassembly.

855A 15 89 08 @32 bclr l_0089_, #%00001000
855D B6 01 56 @33 ldaA l_0156_NTRPMX_25/bit ; rpm scaled in 25 rpm/bit
8560 13 87 20 21 brclr L0087, #%00100000, @35 ; a/c request?
8564 B1 26 71 cmpA L2671 ; d8 in OE cal (5400rpm?)
8567 24 48 bcc @39
8569 F6 01 9C ldaB L019C ; one MPH/bit road speed variable
856C C1 30 cmpB #$30 ; 48mph?
856E 23 02 bls @34
8570 C6 30 ldaB #$30 ; if vss below 48mph load 48mph into B
8572 54 @34 lsrB ; else shift register B right
8573 54 lsrB
8574 54 lsrB
8575 CE 26 A3 ldX #$26A3 ; load pointer to 7 byte table to X
8578 3A aBX ; add registers b & x ???
8579 A6 00 ldaA 0, X ; load something into register a ???
857B B1 02 4C cmpA L024C
857E 23 26 bls @38
8580 7C 02 4C inc L024C
8583 20 56 jr @44

Disassembly isn't my cup of tea - if anyone has any pointers it would be greatly appreciated.

Edit: I see a major mistake I made around the bls jump at 0x856E - I'm thinking this means this tables rows top out at 48mph?

spfautsch

10-25-2019, 01:15 AM

From post #108 (http://www.gearhead-efi.com/Fuel-Injection/showthread.php?8105-95-LT-1-Idle-Cell-Comparison-Humidity&p=78045&viewfull=1#post78045)

Otherwise I am sure you will like the new controller interface. It is also resizable and can hide the less used stuff.

I finally took the time to download and build this. Nice work on the controller window! I was able to utilize it thoroughly today. I did make some modifications to get the overall build to my liking - if you're interested in incorporating them I can send you the sources I changed. First off I made the fields and fonts for the dashboard data I like to be able to see bigger. Next I added the log timestamp to the knock events. This may or may not be useful if you have a car with knock - it doesn't tell you which log the knock happened in so it can be a bit of an adventure if you're analyzing multiple logs. Lastly I added a setFocus() call to each of the individual trim button's toggled event i.e.

void controller::on_cylcutcorr_1_toggled(bool checked) {
if (checked == false) {
parameter_indicator_display(ui->lcd_cyl1,false);
return;
} else {
ui->slider_cylcorr->setFocus(); // <<< ADDED
drop_cyl_corr(1);
// int lcd_cyl1 = ui->lcd_cyl1();
int trim1_int = ui->lcd_cyl1->value() ;
ui->slider_cylcorr->setValue(trim1_int);
control->m4_cyl_corr (trim1_int);
update_m4_raw_display();
parameter_indicator_display(ui->lcd_cyl1,true);
}
}

This lets you jump directly to using the keyboard to adjust the trim slider with the left / right arrow keys after clicking one of the cylinder trim buttons. Keyboard = efficiency of motion!

The only thing that would be better is the ability to set the prime pulsewidth multiplier in a similar way. It's the only thing I feel I'm going to spend time fine tuning on since every time I touch individual cylinder trims startup goes to crap.

Spent about an hour today working on cylinder trims. Idle BLMs are still split by 4% but idle feels and sounds good open or closed loop. I was having a bit of part-throttle surging that improved immensely with a couple trial and error sessions with the new off-idle trim controls. I'm holding off on calling it a victory though because a weather system has moved in dropping air temps and bringing a bunch of moist air with it so BLMs are taking fuel out almost everywhere but idle. I hate winter! :-\

I'm thinking the off-idle trims may take even more time & effort than closed tps trims because even though my surge seems much improved, I'm feeling a good bit of NVH when opening the throttle slightly without load.

Whatever the outcome I feel like what's left are minor tweaks and fine tuning. And I'm pretty sure there are no other bolt-ons I can try that I don't already own except forced induction. Hmm... :-)

Anyway, thanks to everyone who chimed in to help. kur4o in particular - I still have a diy-ltcc board with your name on it if you'd like. I'm sure it will drive the capacitive discharge coils I have and by my butt-dyno estimation they seem to be just as beefy as the truck coils everybody wants (and they're cheaper).

spfautsch

10-31-2019, 01:47 AM

I hate to keep this thread alive, but I made another discovery today that is pertinent to the idle BLM splits that were chronically showing the right bank lean and the left bank rich.

So some background - it's been rainy and cold all week and all I've been able to do is work with startup fueling and idle trims. I've been noticing the wideband maxing out lean periodically when both left and right integrators synched up for a lean swing. I've also been noticing an intermittent miss at part throttle that also seemed to briefly peg the wideband lean (presumably from raw fuel getting in the exhaust). I began to wonder if my individual trims are too extreme and are causing these problems. Today I had a bit of time to experiment with trims in open loop while trying to keep my wideband matching commanded AFR. Oddly, the individual trims seemed to make very little difference when I added or removed from one bank. But while I was doing that I noticed that when I forced AFR above 14.5:1, after a few minutes the right side O2 would start trending down around 150mv and then eventually flatline at 40-50 (this is with BLMs reset to 128).

Then, as I was looking over kur4o's nice new controller window it occurred to me I've never tried moving EOIT. Sure enough, when I got to 129* things smoothed out somewhat. I then switched it to closed loop and to my amazement there was only a 2 point split. Toggling the EOIT override while in closed loop I witnessed the BLM splits come back and then go in unison with the override. So I suppose my lean condition on the right wasn't an exhaust leak or so much an actual lean condition as it is raw fuel getting into the exhaust during overlap. So I guess I'm going to try using 129* EOIT and start all over with individual trims and see what happens. I expect my off-idle fueling will get way out of whack due to loss of wall film.

spfautsch

10-31-2019, 06:04 AM

kur4o: after a bit of experimentation, I think your calcs are inverted on the display of the EOIT slider value. I think it should be:

void controller::on_slider_endofinj_valueChanged(int value) { //new end of inj
// upd. text display always
int degrees = (double)value * 5.625;
ui->display_endofinj->setPlainText(QString::number(degrees,'d',0)+"*");
// upd. mode4
if(ui->check_endofinj->isChecked() == true) {
control->m4_comm_endofinj(true,ui->slider_endofinj->value());
update_m4_raw_display();
}
}

Edit: disregard that - it's pretty superficial. I'd overlooked the ATDC label and I hadn't thought about EOIT terminology in ages.

So what I was assuming to be 129* BTDC was actually 596* BTDC. So my testing wasn't reducing EOIT but in fact increasing it by 56 degrees.

Whatever the case, this is promising news and falls in line with what the LS guys were saying about increasing EOIT from "60 to 61-64".

This has absolutely no relevance to intake valve closing on the cylinder with the active injector. I'll have to do some degree wheel work, but I suspect it's more closely related with having an adjacent (fuel scavenging) cylinder on it's intake stroke while fuel is being sprayed on the "current" cylinder.

Whatever the case, increasing my 0x12df2 table values in the 44c and greater cells from 0x60 (540* btdc) to 0x6a (596* btdc) has reduced my idle BLM splits from 10-16 to 1-4.

I'll post more analysis when I figure out the whys and hows.

Edit: After quite a bit of hypothesizing I'm still not sure why this works. The only thing I can come up with is that perhaps it moves injection to a point where fewer left bank cylinders are on their intake stroke while even bank injectors are firing. Fuel being robbed by another on the same bank won't cause a BLM split.

kur4o

10-31-2019, 10:39 PM

You still amaze me at discovering more stuff through experiments. I never thought about possible blm split by EOIT target. What you have done is move the target earlier to the actual intake valve opening. So there is more time for the fuel to turn into vapor. Stock settings of EOIT takes into account normal operating engine temp at 100-110C*, now when you run it at 80C* it takes considerably longer for the fuel to evaporate.

One reason this can happen is one of the head runs at substantial lower temp while the other is hotter. Not even temp distribution accross engine. The colder side can`t evaporate all the fuel and there is some left unused by the cylinder, during combustion the left over fuel evaporates and is robbed by adjacent across the board cylinders leading to one side running leaner and the other running richer.

Of course this is only speculation.
For the sake of confirmation you can heat the engine upto 100C* and confirm or deny the theory.

I also been considering improving the code of EOIT target by adding more tables, like MIN EOIT vs rpm or closed throttle EOIT table.
Think about it and let me know if there will be any benefit of doing it.

The setting in eehack is *ATDC, which is correct. If you need it at *BTDC just substract the result with 720*.

The shorcuts can get even better if you have some time to do it. Adding some more shortcuts like D+1...8 to set drive cyl and I+1...8 to set at idle cyl, D+0 and I+0 unset, and some more for enable and disable, maybe ctrl+I and CTRL+d might do it, I am sure you will find even some better combination.

Last time I played with the cyl trims controls it sucked really with that mouse and slider I even considered changing the design of it. But with some keyboard shortcuts it will play really nice.

I had a ps pump failure and abandon tuning so far, The last time I managed to run it I discovered that accidentally I have disabled the trims and they were set at 1 all the time, invalidating my last attempts. With 1 for trim the starting was a little weird sounding but still started almost good with no hesitation. When I enabled the trims it started again with healthy sound. Maybe your no start condition with off the chart idle trims can be the result of crank VE table not set right.

I also discovered that the injectors definitely flow more that my initial settings. I managed to increase the fuel flow scalar with much positive results. Also played with the warmup spark scalar table. At 20C* it idles best at 14 degrees and slowly recovering to 23-24 degrees at 50-60C*. Even 1-2 degree change at 30C* coolant change the sound alot. Still no clue how to tune it the proper way other than feel and smell.

spfautsch

11-01-2019, 12:45 AM

Another thing that may be a factor is that the exhaust system is asymmetric between the header collectors and the x-pipe. One side might be generating more exhaust scavenging than the other at idle speeds. In working out the different valve events looking for ideas I found 596 BTDC moves the injection event (at 800 rpm it's only 3 degrees for ~2ms) so it's almost out of all overlap scenarios. But this may be completely irrelevant. I'll definitely try getting it hotter to see if your idea of uneven cooling could be a factor.

The only reason I can think of for having the need for more EOIT controls is if this change causes a BLM split at part throttle, and it will be a few days before I'll have a chance to drive it. I'm hoping to rework the closed individual trims tonight, but I'll probably get roped into handing out candy.

I'll look at the possibility of adding those key accelerators - I haven't tried modifiers like ctrl, alt or shift so no idea how involved that, or using a normal key like 'i' and 'd' as a modifier key will be.

Terminal_Crazy

11-01-2019, 06:00 PM

Similar results to me then, extending the time the fuel sits in the port or forcing it further down the port so it has less chance of being sucked back out?

Have you tried a sweep test.
This was mine last year.
I think the LH inj pw staying the same is an issue I've had with flashing & logging with different versions.Never quite pinned it down.
EOIT can make quite a difference to the splits. I'm currently running at #63.
2018-12-13
Did another EOIT spread degree test with odd figures .........

Test B at 86+øC

Blm's running on pcm setting of &66 120 122
log went lean/weird around 74øC to 75.5øC 68C eoit = &50 80C=66
Left bank richer - rhs longer inj time ?????????????????????????????????

start 1200 Idle Fast
Time log ø table blm's blm's Inj PW SPLIT SPLIT

685 7170 146 66 120 122 1.79 1.79 pcm setting &66 + 2

832 8764 186 5F 115 125 1.79 1.79 +10
1110 11681 208 5B 113 130 1.77 1.94 +17
1225 12875 231 57 106 130 1.77 1.95 +24
1400 14711 253 53 104 130 1.77.1.92 +26
1540 16180 276 4F 99 134 103 139 1.75 1.95 +35 +37
1715 18010 298 4B 103 137 99 145 1.77 1.97 +34 +46
1890 19843 321 47 98 137 99 145 1.77 1.95 +39 +46
2030 21313 343 43 98 137 99 145 1.77 1.97 +39 +46
2135 22413 366 3F 97 137 99 145 1.77 1.97 +40 +46
2275 23881 388 3B 99 143 100 150 1.77 1.94 +44 +50
2590 27180 411 37 98 134 100 144 1.77 1.92 +36 +44
2730 28647 433 33 96 134 100 144 1.77 1.91 +38 +44
2870 30109 456 2F 96 136 100 144 1.77 1.92 +40 +44
3010 31574 478 2B 104 136 97 144 1.77 1.92 +32 +48
3150 33038 501 27 101 136 101 138 1.77 1.88 +35 +37
3255 34143 523 23 104 135 106 132 1.77 1.83 +31 +26
3465 36343 557 1D 104 132 104 132 1.77 1.82 +28 +28
3605 37807 579 19 104 132 110 132 1.77 1.85 +28 +22
3710 38908 602 15 99 132 99 132 1.77 1.86 +33 +33
3850 40373 624 11 96 139 104 132 1.77 1.83 +43 +28
4025 42209 647 0D 99 136 102 132 1.77 1.89 +37 +30
4200 44042 669 09 103 131 107 132 1.77 1.85 +28 +25
4340 45510 692 05 103 131 104 132 1.77 1.88 +28 +28
4480 46979 714 01 103 131 104 132 1.77 1.82 +28 +28

Restart at end of table
4585 48077 96 6F 107 118 111 120 1.75 1.66 *INJ PW swaps* +11 + 9
4725 49541 118 6B 116 121 116 120 1.83 1.56 + 5 + 4
4865 51005 141 67 118 121 119 120 1.77 1.60 + 3 + 1
4970 52105 163 63 118 117 119 120 1.77 1.63 - 1 + 1 <-<-
5145 53935 186 5F 117 117 119 120 1.77 1.66 0 + 1
197 5D 108 125 111 122 +17 +11
5285 55938 208 5B 107 130 102 139 1.77 1.82 +23 +37
5565 58321 231 57 107 129 105 139 1.77 1.79 +22 +34
5635 59054 253 53 104 129 101 139 1.77 1.85 +25 +38

Mitch

spfautsch

11-01-2019, 07:26 PM

I haven't tried a complete sweep test yet, finding that my splits changed immensely was mostly by accident. It's been ridiculously cold this week and I've had limited time to work. When I have it's in the garage with the door open so the gas fumes are tolerable, so I haven't been able to be as methodical as I'd like.

Last night I worked on closed throttle individual trims. One of my classmates showed up to trick or beer me as I was working so I had to stop, but I was basically back at stock trims with just a few minor changes and BLMs were 127 left and 129 right. I'd really like to get this nailed down so I can also nail down startup fueling, but as it works everything seems to be inter-related. Out of curiosity, I did briefly try moving EOIT to well after my exhaust valve closing @ 323 BTDC but the BLMs went crazy. Once I'm more satisfied with individual trims I'll give a full sweep test a shot.

My theory on why this is working is based on the idea that fuel scavenging inside the intake only edit: (or, primarily) happens while an injector is spraying, and once the fuel is down on the back of the intake valve and port bowl the scavenging effect is dramatically reduced because the air from the idle feeds can't reach it there. I don't think it's related to having more time for the fuel to boil into vapor, but I do intend to test because my theory could be complete nonsense. I attached the worksheet I built to help visualize the relationship of all the different events.

It's extremely puzzling that your LBPW never drops below 1.77ms. I'm curious - what values do you have in 0x126d5 and 0x126d7 (min injector pulsewidth)?

spfautsch

11-03-2019, 04:30 AM

Hey Mitch would you mind recapping your methodology when you did the sweep tests? Did you reset BLMs when you moved EOIT? How long did you let it idle before recording the BLM split, or did you average?

I did a sweep from 540 to 624 BTDC this morning and in the process noticed my BLMs seemed to inversely follow coolant temp. I have a 71c thermostat and fan on setting was at 82c so coolant temp was varying by about 4c. At 83c trims were about 2-3 and at 79c they were about 4-7.

So I decided to try bringing temps up by digging out my 82c thermostat and dropping it back in. I had to bump fan on temps up some so they weren't running on high all the time. After a short drive the coolant temp at idle was steady at 91.3 with fans contant on high. I did another brief (i.e. not well documented) sweep test and found nothing compelling. Splits averaged out to about 3 but maxed at 7. I'm planning on doing another thorough sweep tomorrow.

This is all with almost stock individual cylinder trims. Overall it's running much smoother than when I was running radical trims trying to compensate for a ~12 point BLM split.

In the midst of this I'm pretty sure my wideband sensor is toast. In closed loop it's reading pegged at 16.39:1 (5v) quite a bit. Last week when air temps dropped, on the last leg of the commute the wideband was flatlined at 0v (11.39:1) the entire 62 mile drive. I thought the fuse powering the controller might have blown, but it wasn't. After installing the 82c thermostat I took about a 10 mile drive and had a chance to open it up for a few seconds where commanded AFR was 12.5:1, and the wideband was flatlined through the whole PE event, then went back to reading really lean. Curious if anyone else has noticed this type of behavior in cooler air temps, or if I should start looking for a new Bosch sensor.

Terminal_Crazy

11-04-2019, 01:44 AM

Hiya
Car warmed up >86C
I started to reset BLM's after the first few, so went round the table again at the end.
I basically logged until the BLM's settled (about 1000 logs minimum between changes)

I'm currently waiting for my Y pipe for the long tubes as I've just found out my mate can only bend 63mm pipe on his pipe bender.
Then I'll be looking for a Dual Wideband.

Mitch

kur4o

11-04-2019, 02:56 AM

the wideband was flatlined at 0v (11.39:1) the entire 62 mile drive. .

It looks like a toast. Or it could be the controller. Are you running 4.2 or 4.9 bosch sensor. One of them was prone to early failures due to controller. I have 4.2 and it experienced alot of abusing[including coolant fouled and cleaned with acetone] and still works, not sure how acurate but it works. They should be pretty robust and last long since are oem used in turbo VAGs setup.

On the blm split. I am running different lenght tubes upto y and the sensors are at different lenght from the heads and still haven`t experienced heavy blm split. Actually they are pretty close off idle 2-4 points difference, with untuned off idle trims. At brief idle CL itempts I made I had 1-2 points only.

When you change settings look for at least 1 minute wait time for them to take effect, even changing the EOIT target takes some time, in the code there is smooth ramp out routine that makes it really slow.
If you find that there is a need for different target for idle and off idle vs rpm conditions, I will add the new tables in the code.

Spray pattern and lower than rated fuel pressure for the injectors could affect blms. Not sure how but I read somewhere that ls1 injectors had different spray pattern.

I have one theory how fuel robbing works. The path of the air is from the adjacent feed hole cylinder through the plenum, that way it can suck left over fuel. The path from other than adjacent cylinders will be too long for the fuel mix to travel. This effect can also lead to slowly filling the plenum with fuel vapors over time, making the mixture richer with longer idle events. Than can explain different behavour with supposedly prefectly tuned engine.

I didn`t get that deep in engine rotation, but if there are more than 1 cylinders sucking air at the same time like overlapping open intake valve events. Add to this the big cams overlap egr like effect, pushing small amounts of exhaust gas in the intake plenum and the picture gets worse.
If the cam you have opens the intake valve while the exhaust stroke upward cylinder movement is not finished, you are doomed for the perfect tune failure. Just get it close enough. You can always cure some blm splits with changing L and R swing voltages.

These are from stock ybody auto
L R
460.20 460.20
389.40 389.40
442.50 442.50
460.20 442.50
460.20 442.50

Vs stock ybody manual
L R
460.20 460.20
380.55 380.55
380.55 380.55
380.55 380.55
380.55 380.55

spfautsch

11-04-2019, 04:03 AM

I hooked up the serial cable to the wideband this morning and at cooler ambient temps it's giving an error 8 during warm-up. So it seems the sensor is boned. It's a 4.9. The troubleshooting guide says it could be getting too hot during PE events or the sensor is damaged. I think the heating element is burning out due to cold ambients.

Car runs a bit different with the 82c thermostat. I'm seeing a bit more knock above 2400rpm but that doesn't worry me a lot right now. I'm just concerned with what it will do during the summer months. The high today here was 16c and the fans basically ran on high all the time to keep coolant temp under 100c. Unfortunately no-one seems to make a 77c or 79c thermostat in this size so I don't think there will be a happy medium.

After some testing yesterday, I reverted to essentially stock individual trims. I added a couple points to #1 at closed TPS since I enlarged that feed hole, and added a bit to #6 which is quite a bit rich in the stock tune and I noticed my balance tests showed it seemed to be pumping the strongest at closed TPS. I don't know if anyone's noticed, but at part-throttle the base Y-body tune takes a lot of fuel out of #6 so it goes from way lean at closed throttle to way rich at part. Go figure. Low-speed cruising is good here, so I'm not going to make many changes. However, I'm somewhat baffled as to why if I have a 3 count split at idle, and add 1 to the lean side and remove 2 from the rich side, why that doesn't "cure" the split. But it doesn't. I suspect its just the nature of the cam.

I did another EOIT sweep from 523 to 624. I noticed another variable that probably explains why I'm noticing splits increase with idle time - IAT. After a short drive to warm it up for the test, when I pulled in the driveway with EOIT in the tune at 579 BTDC it idled for 2-3 minutes at 128/130 (timestamp 1500).

My process was to change EOIT then reset BLMs, then log for 100 seconds. I noted the timestamp at each change, and recorded BLMS and other relevant data after 99 seconds. During the log my IAT climbed from 32c to 44c. I suspect this is relevant so I'm hoping to reproduce today's test with a fan or something blowing cool air near the air box.

ts eoit lbl rbl iac ect afgs split
1500 579 128 130 27 91.3 8.11 2

1600 523 118 141 28 90.5 8.17 23 < iat @ 32c
1700 528 120 139 29 90.5 8.27 19
1800 534 126 134 29 90.5 8.19 8
1900 540 126 133 30 90.5 8.28 7
2000 545 127 133 31 90.5 8.37 6
2100 551 127 132 29 90.5 8.04 5
2200 556 128 133 30 90.5 8.20 5
2300 562 128 133 30 90.5 8.16 5
2400 568 128 134 31 90.5 8.20 6
2500 573 129 134 29 90.5 8.16 5
2600 579 129 135 29 91.3 8.09 6
2700 585 129 134 29 91.3 7.98 5
2800 590 128 134 29 91.3 8.13 6
2900 596 131 135 28 90.5 8.02 4
3000 601 131 135 28 91.3 7.93 4
3100 607 132 136 26 90.5 7.75 4
3200 613 131 135 27 90.5 7.86 4 < iat @ 44c
3300 618 129 136 26 90.5 7.78 5
3400 624 129 136 25 90.5 7.75 5

As you can see there's no magic value here - the only thing that seems relevant is that after 590BTDC the left side goes lean. Otherwise it seems like I want to be between 613 and 596. So bearing in mind that lower coolant temps need more time for evaporation, I did this with my EOIT table.

-40 180
-28 180
-16 180
-4 180
8 360
20 360
32 360
44 618
56 613
68 607
80 601
92 596
104 596
116 596
128 596
140 596
152 596

A short drive with these settings showed about a 5 point split in BLM cell 2, and less everywhere else. I think I'm going to drive this tomorrow or the next day weather permitting and see what happens. If this is as good as it gets I'm happy with it - it runs very well. My clutch isn't quite able to handle 2nd and 3rd gear under full throttle.

kur4o

11-05-2019, 12:10 AM

I think that moving injector spray time forward can compensate more for fuel robbing and change the fuel robbing dynamics, than it is temp related.
I guess when the manifold gets heat soaked, than the fuel will evaporate much faster and the eoit will be more related to IAT than coolant temperature.
So lets say all this variables should be taken into account while tuning cyl trims.
EOIT, coolant temp, IAT temp, spark.

Why don`t go the opposite strategy trying to tune the split. Add more fuel to the rich side or to adjacent cylinders implementing the fuel robbing theory.

With all that said, there will be multiple different cyl trims set settings that works best.

It will be lean cylinder makes less power, rich cylinder makes less power. All thatleads to higher map readings. The map will be good indicator for good cyl trims.
I lowered the map 2 points from no trims to my usual trims. I also experienced higher map readings in closed loop. I guess the leaner mixture needed more spark advance, but didn`t have the time to investigate further. The power steering pump decided to die on me, when it started to get interesting.

Did you cure the starting problems you had. While trying to tune my cold start problem with cyl trims disabled, it took 1 to 3 points on the second prime pulse to change from average start to hard start condition. I have 1 point averages set at 0.8 ms change in the pulse width.

Here is a reworked xdf that displays the prime pulse and the 12680 constant with the proper conversion, I also linked them together so you don`t have to make educated guess about the prime pulse.

Now I would call 12680 constant, the pulswidth resolution for prime pulse tables. SO if you want 0.5 ms resolution just fill the scalar vaule with 0.5 value, than save bin reload bin. Needed due to prime pulse values calculations in tuner pro to get updated with the new scalar.

Now since i have 0.8 and 1-3 points are alot for my 25-6lb/hr injectors, I suggest you go as low as possible, while keeping enough fuel for really cold startup situations.

I am running now
INITAIL prime pulse
156.9
140.0
79.7
46.7
35.4
21.7
16.1
7.2
7.2
7.2
5.6
5.6
5.6
5.6
5.6

and prime pulse adders

196.34 196.34
190.71 190.71
109.43 109.43
53.91 52.30
44.26 42.65
21.73 20.12
16.09 14.48
7.24 5.63
7.24 5.63
7.24 5.63
5.63 4.02
5.63 4.02
5.63 4.02
5.63 4.02
5.63 4.02

12680 set at
0.804663

You should also play with the 12df2 the smooth factor for EOIT change. Since you are going to have alot of oscillations in the 68-90*c region. Stock value is $10 lowering it or increasing it should make the transition to take place faster. Not sure which one.

spfautsch

11-05-2019, 01:28 AM

Why don`t go the opposite strategy trying to tune the split. Add more fuel to the rich side or to adjacent cylinders implementing the fuel robbing theory.

With all that said, there will be multiple different cyl trims set settings that works best.

It will be lean cylinder makes less power, rich cylinder makes less power. All thatleads to higher map readings. The map will be good indicator for good cyl trims.

Honestly, I'm not certain I'd attempted to take the absolute stock trims and just add the BLM split - that was with numerous versions of my own trims that probably weren't that good to begin with.

Did you cure the starting problems you had.

I haven't done any fine tuning on the prime pulse table. I've found the closer to stock individual trims the more predictable startup has been. But I've also been changing 0x12680 in a linear fashion (i.e. it's currently at 0x4f4f) and leaving the prime pulse tables at stock. My assumption was that this would give a 1.21ms minimum resolution. If I wanted more fuel I'd try 0x5050 and 0x4e4e for less. If you're absolutely certain this will give unpredictable / non-linear results please give me a clear answer and I'll rework the tables using 0.5ms (0x2222???).

Now since i have 0.8 and 1-3 points are alot for my 25-6lb/hr injectors, I suggest you go as low as possible, while keeping enough fuel for really cold startup situations.

12680 set at
0.804663

Just for the sake of clarity, that is what in hex bytes? I thought it was 0x3333 but that gives me 0.77819 (51 * 0.0152587891).

Thanks for the xdf. I'll take a look, maybe that will answer my questions on the scalar conversion.

Edit: ah... (x*0.0152587891)/256 - I think I'm clear on this now. So basically I need to determine maximum pulsewidth for cold starting and find a multiplier that gives me that divided by 256 (0xff), and 12680 can be any value between 0x0000 and 0xffff?

spfautsch

11-05-2019, 09:00 PM

Thanks for the updateded definitions, nice work! I noticed all I have to do to see changes after modifying + saving 12680 is to close + re-open the prime pulse table, or simply switch from hex view back to calculated values.

Here's another attempt at a prime pulse conversion tool. Assuming it's safe to use 0xff in the coldest cells of 12691, you can plug your injector constants in (as well as cylinder constant for displacement changes) and it will give you a base pulsewidth for 12680 that gives the most resolution in the higher temperature cells, along with the adjusted 12691 table as hex. It shows the results of the rounding errors in M19:N33.

Once you have a base pulsewidth and 12691 table in B20 and H19:I33 you can copy the hex and then paste-special (values only) back to the same cells to de-reference the formulas, then fine tune the hex values.

spfautsch

11-07-2019, 01:22 AM

kur4o: When you have a chance, see if this source will compile for you and check that the ui changes work and look acceptable on windows please.

Here's what I can recall changing:
- enlarged fonts on frequently viewed dashboard fields
- enlarged eoit field so hex value could be displayed in addition to degrees
- reworked all key accelerators in controller.cpp to use keyPressEvent (see bottom of file)
- added key accelerator "B" to reset BLMs button
- removed "I" as idle override accelerator key (for trims)
- added key accelerator info to tool-tips (all having accelerators except individual trims)
- increased the slider range on individual trims to 112-142 (factory y-body trim on #7 is 0x74 / 116d)

To set individual trims I made "I" and "O" the modifier keys e.g. hold down "I" and press "1" to set #1 idle trim. The right + left arrow keys can immediately be used to adjust the slider, then "S" to set. Note that when the user is not actively setting trims "S" is the accelerator for spark override. I had to add a state machine variable to allow using these modifier keys, and another to track when a trim is being "edited".

On the tuning front - haven't been able to drive it yet, supposed to rain tonight and tomorrow so fingers crossed for Friday.

Did some detective work yesterday and managed to find a 77c / 170F thermostat that would fit - a Motorad 2028-170. Seems to be working as expected. This was the first time I can recall bringing it up to temp at idle that didn't result in a large initial BLM split - 4 was about the worst I noticed after about 15 minutes of idling in CL. So there may be hope left for getting the split trimmed out.

I'm a bit concerned the t-stat may not flow enough to keep up with highway driving because the primary disc valve is only 31mm whereas the OE thermostat is 39mm. However, one application for the 170F t-stat is in the Ford 6.8L triton V10 so maybe my concerns are unwarranted.

spfautsch

11-08-2019, 12:15 AM

After more tweaking on idle trims it's much better. No idea how much the hotter thermostat is contributing, but I'm noticing trims at idle are a few points lean so I reduced EOIT to 0x68 / 135 ATDC and that lowered the BLMs by a point or two. It's never idled this smooth so I think I'm going to leave idle trims alone.

Startup fueling seems to be really good now. I ended up adding an additional 8% on top of the displacement adjustment and landed at 0.587ms BPW. I suspect the additional fuel it wants is a sign that my injector constant is lower than actual. Anyway, so far I haven't had a problem starting at all but haven't gotten very far into testing at different levels of cool down.

The 77c t-stat seems like it will be perfect. At highway speeds / loads the temp stays right around 89-91 without fans running. I hope it works as well when summer rolls back around.

Looks like the next area needing attention is off-idle trims. I'm basically running stock Y-body trims here and have a slight lean split to the right in cells 2 and 6 at low throttle openings - 4-6 points. I was able to stop and take some fuel out of the left bank and immediately noticed it caused really bad surge. Being able to adjust this on the fly is huge - I'd really like to buy kur4o a drink or two for this and all the other tuning tools he's added to eehack. Phenomenal work...

kur4o

11-08-2019, 03:52 AM

The next big thing after eehack release is the controller key shortcut. It is like a pro level factory tool now. Fast seemless and easy to control anything on the fly without going in deep staring analysis what`s where and than finding the buttons with that tiny mouse pointer. Great work.
It compiles without issues on windows, I even managed to add some changes already. I have added some info buttons already and made some final tweaks to the controller interface and will try to merge both versions.

The trim buttons worked perfect. I am thinking to move the I+S and 0+S to I+0 and O+0, since it is really unsetting the correction, and add *+9 for the enable button. Also plan to expand the controls to cyl cut and mafpumpshot with C+0..8 and M+0..8.

The open loop button might get better. As it is set now it doesn`t stop the blm learn mode off and doesn`t reset the blms to 128. BLMs and INTs even in open loop keeps adjusting if not set to 128.

The hex display on eoit is cool, I might rework some of the other displays like it. I still didn`t managed to get the cyltrims decimal conversion right. Can you take a look at it.

The smaller thermostat opening can be compensated with higher overal coolant system pressure. Find out what is the radiator cap pressure rating on the fords engines. That one is perfect for off hot summer usage and the 160 can be put when it gets really hot in the summer. I am suprised you even score a replacement due to the unique reverse flowing system in lt1`s.

The hottest cruising temperatures I got with 160* were in the 82-84*C with ambient air temp reaching 33 C*, high humidity and speed of 50-65 mph. That`s on average + 10-12 degrees above thermostat rating. A 77 + 10-12 will get you right where you want it.

By the joint tuning effort alot of uncharted stuff emerged and resulted to the perfect tune. Even the slightest possible variable was discussed and the vette CL settings were uncovered for the average f-body owner.

spfautsch

11-08-2019, 04:53 AM

I have added some info buttons already and made some final tweaks to the controller interface and will try to merge both versions.

Thanks, and please do. Most important to me is that whatever useful knowledge I can share, is shared freely to anyone willing to read.

I am thinking to move the I+S and 0+S to I+0 and O+0, since it is really unsetting the correction, and add *+9 for the enable button.

Sounds good. Bear in mind the modifier key for 'set' isn't needed. I'll try to PM you more details on this tomorrow but modifier keys '0' and 'I' aren't needed once in the process of adjusting a trim value, just 'S'. But '0' (zero) works fine for me. With 'enable' the modifier key will be needed. I realize there's a language barrier - just let me know and I can help figure it out.

The hex display on eoit is cool, I might rework some of the other displays like it. I still didn`t managed to get the cyltrims decimal conversion right. Can you take a look at it.

I'll give it a look. Honestly on any single byte field I like to do my tuning directly in hex, outside of TunerPro. That way I know the source of any rounding errors and can choose the 'sane' rounding manually.

The smaller thermostat opening can be compensated with higher overal coolant system pressure. Find out what is the radiator cap pressure rating on the fords engines.

Interesting. So higher pressure equals higher density even with liquids? I guess I should have paid more attention in physics class.

I am suprised you even score a replacement due to the unique reverse flowing system in lt1`s.

There is absolutely nothing special about the thermostat for the LT-1's reverse flow cooling system. The reverse flow stuff is cast into the block. The thermostat is exactly the same as every other thermostat in every other car running down the road regardless of vintage. I can explain in detail if needed, but long story short, an automotive thermostat is an automotive thermostat, end of story.

By the joint tuning effort alot of uncharted stuff emerged and resulted to the perfect tune. Even the slightest possible variable was discussed and the vette CL settings were uncovered for the average f-body owner.

I can take very little credit, but am immensely satisfied knowing this information is available for the next guy that decides to take on the task.

spfautsch

11-10-2019, 02:01 AM

Had a chance to do some tuning on the off-idle trims today. Unlike closed-throttle trims which seemed to be best when running very close to the fairly aggressive factory tune, it seems with this cam and the EGR it develops at low speeds a tamer set of trims is better.

closed stock open stock
1-83 82 83 88
8-7e 7D 81 7B
4-86 80 81 83
3-80 80 80 83
6-88 84 81 7A
5-7f 80 7e 7D
7-75 74 7e 7E
2-86 83 84 86

I tried running the stock trims, but when I attempted to adjust for BLM split (lean on the right side again) a very noticeable surge / hard miss developed at low speeds and throttle openings. Even at 75mph in 6th with the cruise control on the engine was noticeably "unhappy" at lower loads.

The open trims above are much smoother tooling around town, and even improved overall smoothness when lugging the car up a hill in 6th gear at 1300 rpm. Something I've been unable to do since putting the stock cam in storage.

Startup fueling has been all over the place, but it's getting to be the time of year when this one will spend most of the next 4 months sitting in the garage so I should have plenty of time to work on that. I'm hoping this winter it will spend markedly less time sitting on stands with a major powertrain component absent.

Terminal_Crazy

11-10-2019, 12:42 PM

spfautsch

11-10-2019, 07:10 PM

How did you calculate those Open throttle figues?

That was merely a wild guess based on the trims I was running before reverting to something closer to stock. It's more or less 128 / 0x80 / 1.00 with a small bit more fuel on the right bank and then slightly less on the back cylinders (7 & 8) and slightly more to the front (1 & 2).

Is there any easy way to display these in TunerPro in an easier to view format?
as in cylinder layout

Not that I'm aware of. Honestly I had committed the small block chevy firing order and bank assignments to memory before I had fully developed facial hair, so representing them in firing order makes perfect sense to me. But having been forced at gunpoint to work on a modern ford v8 I can understand how it would be less than intuitive. When the ECM tells you there's a misfire on #7 I instinctively want to look at the left rear cylinder.

spfautsch

11-10-2019, 08:25 PM

Anyone have an idea how the 12bf1 EOIT ramp out multiplier works? kur4o I thought you mentioned I should work on it when I'd tried decreasing EOIT with coolant temp.

In testing startup fueling I've been letting it come to temp at idle, and I believe I found something that could have been a factor with the really bad idle splits.

During warm-up when ECT hits 44c my O2s start to fluctuate - left moves around some but right goes dead lean. It stays this way until about 56c. This falls in line with the second EOIT transition - in the stock bin it's moving from 360* to 180*. Looking at some older logs, this lean condition didn't improve until around 65c. But the ecu had already switched to closed loop at 60c (which resulted in a massive split).

In looking at the disassembly I can't really tell what the ramp out multiplier does - I would presume it dampens the implementation of the EOIT value so as to not cause an abrupt shift in fueling, but I'm not sure what it's value represents. Whatever the case, it appears by what comments I'm seeing in the disassembly that the EOIT table data is interpolated only when transitioning to higher values (lower atdc degrees). It seems like the interpolation alone would offer adequate damping.

At any rate, I'm not sure there's any point to my questions - just looking for more information on how this all works.

I do however notice the possibility for a problem if the code transitions to closed loop at the "normal IAT" minimum closed loop temp (33c in base y-body tune). I'm moving that to 60c to match the "cold IAT" value in the 0x26e3 table,

kur4o

11-10-2019, 10:58 PM

Anyone have an idea how the 12bf1 EOIT ramp out multiplier works? ,

From what I gather from experiments it is some kind of factor for graduate change of values. For example if you command a change from $20 to $40 it will slow down the change to by $01 each second. so the transition will takes place 40 seconds. That is only example. I haven`t test the actual rate with $10, nor I know in which direction the increase of it will take place, slowing or speeding the change.

There is 1 way to test it. Monitor the eside word_a9 while changing the EOIT target from end to end on the controller window with ignition on.
To do it open raw window in eehack, and loop in 50ms interval the command e4 ... 03 00 a9 00 aa , you will get a response with the word_a9.
Check how much it takes for the change. Than flash a bin with new 12b1f value, for example $20 and do the same test. Compare the results if it takes more or less time for the change to occur.

On ybody calibrations the timers for closed loop operations are set much lower. You can try increasing them if the engine doesn`t warm up fast enough.

At startup I found the it warms up best at pig rich condition o2 above 900mv, when running close to stoich at warm up, it run rough and you can feel some misses from time to time. I also played with idle spark and make a curve to change the spark during warm up.
I end up with that spark coolant temp corr table.

5.0 5.0 5.0 6.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0
5.0 5.0 5.0 6.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0
4.0 4.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
0.0 -11.0 -11.0 -11.0 -11.0 -11.0 -11.0 -2.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0
0.0 -8.0 -8.0 -8.0 -8.0 -8.0 -8.0 -2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
0.0 -6.0 -6.0 -6.0 -6.0 -6.0 -6.0 -2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
0.0 -3.0 -3.0 -3.0 -3.0 -3.0 -3.0 -3.0 -3.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 -2.0 -2.0 -2.0 -6.0 -6.0 -6.0 -6.0 -6.0 -6.0 -6.0 -6.0
0.0 0.0 0.0 0.0 -8.0 -8.0 -12.0 -12.0 -12.0 -12.0 -12.0 -12.0 -12.0 -12.0 -12.0
0.0 0.0 0.0 0.0 -8.0 -8.0 -12.0 -12.0 -12.0 -12.0 -12.0 -12.0 -12.0 -12.0 -12.0

If you change that table beware that some xdfs can`t handle negative correction and zeroes the value when negative. I think I fixed it on the last xdf i posted here.

I plan to get the hotter thermostat also. Do you know a cross reference number for a Stant one.

Terminal_Crazy

11-10-2019, 11:15 PM

In testing startup fueling I've been letting it come to temp at idle, and I believe I found something that could have been a factor with the really bad idle splits.

During warm-up when ECT hits 44c my O2s start to fluctuate - left moves around some but right goes dead lean. It stays this way until about 56c. This falls in line with the second EOIT transition - in the stock bin it's moving from 360* to 180*. Looking at some older logs, this lean condition didn't improve until around 65c. But the ecu had already switched to closed loop at 60c (which resulted in a massive split).

Yes I've posted this previously a good while back.
Consistantly around 44C to 56C the RHS O2 flatlines.
The AFR gauge also leans out and car runs noticably rougher.
This is still in OL as I don't go CL until about 70C.

I've never found the cause. ( 3 sets of injectors have all done it).
at around 56C it seems to come back on line.

Another oddity I've mentioned before & nobody has answered...
Frequently noticed, I've had the car idling ( may not have been very hot, but certainly above 80C).
BLM's on one side would go up by say 5 points in several seconds.
The other side generally followed with 20-30 seconds

Mitch

kur4o

11-11-2019, 12:04 AM

Consistantly around 44C to 56C the RHS O2 flatlines.

Mitch

It is the transition region of the EOIT target and could be related for sure. Maybe setting the transition to happen at higher temperatures can cure it some.

I have also experienced some flatlines of 02s and tried alot of stuff fixing it with no positive results. I suspected exhaust leak and tried adjusting the maf curve also. It could be of the cold 02s not reading right but wideband agrees too.
I will try to dig some logs and check the coolant temperatures.

spfautsch

11-11-2019, 01:20 AM

kur4o your inbox is full. Thanks for the suggestions on monitoring the EOIT change - I'll try to give it a shot when I have a chance and report back.

On ybody calibrations the timers for closed loop operations are set much lower. You can try increasing them if the engine doesn`t warm up fast enough.

I just put my CL enable temp to 60c, and may go higher. I can't imagine why they would have even attempted to jump to closed loop at 33c, it seems absurd at that cold of an operating temp.

I also played with idle spark and make a curve to change the spark during warm up.
I end up with that spark coolant temp corr table.

I noticed this and will probably try to do the same once I have startup tables figured out. The reason behind this is that richer mixtures support faster flame front propagation (burn faster) so require less spark advance. The harshness you feel with lots of advance during warmup is cylinder pressure peaking sooner than is optimal (10-18 atdc depending on stroke and connecting rod lengths). This is another side-effect of the EGR developed by the cam overlap - at operating temps (and resultant leaner fueling) much more spark advance is needed because of the mostly inert (i.e. devoid of oxygen) exhaust gas diluting the charge.

I plan to get the hotter thermostat also. Do you know a cross reference number for a Stant one.

Sorry, I don't but will try searching for you. Unfortunately this is an alternate temp t-stat so Stant may not make one with this wax motor and in this size. If possible, try Amazon. Autozone sells this MotoRad piece rebranded as Duralast part # 15397. Or if nothing else works, if you PM me your address or that of a friend I'd gladly ship you one.

Another oddity I've mentioned before & nobody has answered...
Frequently noticed, I've had the car idling ( may not have been very hot, but certainly above 80C).
BLM's on one side would go up by say 5 points in several seconds.

I've never seen both sides jump like that, but have seen it on my right side. The only suggestion I can make is to keep in mind if you can hear any lope from your cam, what you're hearing is a misfire. And when the engine misses raw fuel gets into the exhaust which causes the O2s to read lean. And that causes the CL PID logic to add fuel when no additional fuel is needed. Perhaps the problem then "snowballs" due to the raw fuel.

I'm seeing this now with my EOIT at 0x68-0x69. I've found that when it's at 80c and higher the BLMs are adding 3-7% on both sides (134L, 137R is fairly common), and the raw fuel smell gets a bit worse. But as long as it continues to idle this smoothly at 800 RPM I honestly couldn't care less. As we all know, bigger cams like to run rich. I may just put my cats back on and call it good.

I've also noticed that the CL PID loop seems to be biased towards rich. In other words it is more eager to add fuel to compensate for a perceived lean condition, than it removes fuel due to a similar rich condition.

Edit: kur4o - I've also noticed what I think you've mentioned with tuning startup fueling. I've found if I shut the engine down when not at full operating temperature, and especially when not also in closed loop - the amount of fuel left in the ports dramatically changes how the next startup goes. I'm trying to develop a methodology for tackling this, but it seems like it's going to involve letting the engine sit idle (not running) for 2-3 days between tests, and bringing it up to full operating temp each time before shutting it down.

spfautsch

11-11-2019, 11:53 PM

LeMarky Dissod

11-12-2019, 01:19 AM

kur4o

11-12-2019, 01:45 AM

I've found a Stant 45967 and Gates 33747S that are close, and are shown on another vehicle application that uses the Motorad part (2004 Nissan 2.4L).

The only problem I see with them is they appear to have smaller primary disk openings and the bypass disk stem (on the bottom) is about 2.5mm shorter than the one called for. The Motorad is only about 1mm shorter. This means the bypass may not close completely when the thermostat is open. Probably wouldn't cause any trouble, but I'd prefer the Motorad piece.

Thanks for the part numbers. Does the stock gasket works fine with the 2028?
I also investigate the matter through rockauto and made a list of 2028 interchange part numbers with different open temp and applications. From what I gathered as info, The lT1 thremostat is the same overall dimension as the sbc 94-00 but flows more due to enlarged opening. Too bad delco doesn`t make 170* one.

The main question is, will the 2028 flows enough or should I keep the 160* one and do the cardboard thermostat mod. While searching I also found that the nascar guys target 200-210*F as the best temperature for the combustion process. Completely in par with the most fan settings 100*C-on 95*C-off pre obd2 emission standarts.
Somehow I nailed the temp down with 93*C low and 96*C high.

The reverse flow cooling system might change the Nascar equation and the optimal for lt1 could be 90-93 *C. STill needs experiments to confirm.

For startup test you can always clean the left over fuel with some starter time and injectors fuses off. Nailing down the pedal above 80% tps also stops fuel from being sprayed.

I make some test with hot restarts. First hot restart is always good. Second is almost good and on the third I have long cranking. The test was made with 4-5 seconds run time.

It could be subjective feel but when I adjust the fuel flow constant and return back to stock maf curve, the engine run much much better even though the wideband and spark didn`t show any deviations. I also witness some real low map values.

I dig up 2 logs where 02s flat drop from 50 to 53 *C, than recover with no apparent reason. I know for sure I have other logs too but can`t find them. In par with the EOIT target change from 44 to 56 degrees or 270* to 180* which interpolated is 225*-247.5* at 50-53*C. Really wierd.

spfautsch

11-12-2019, 06:56 AM

Thanks for the part numbers. Does the stock gasket works fine with the 2028?

So far so good - it fits around the t-stat housing and I'm not seeing any leaks after 150 miles and about 4 hours of run time. Unfortunately these do not include a new seal so if your current one is getting old I'd suggest getting a new one.

The lT1 thremostat is the same overall dimension as the sbc 94-00 but flows more due to enlarged opening.

This is why I like this particular Motorad part - the advertising on it says it has a 30% larger primary disk from the "standard flow" parts. I also looked at cap pressure and the 6.8L Ford V10 runs at 16 psi also so I guess the only other question is how much bigger is the water pump impeller? Again, I'm happy to be the guinea pig on this one - if it overheats next summer you can rest assured I'll be her complaining about having to replace whatever parts it breaks.

The reverse flow cooling system might change the Nascar equation and the optimal for lt1 could be 90-93 *C.

I've never read the official rules, but I think you may be making a bit of a leap in assuming they're not all running reverse flow cooling. I'm relatively certain I read somewhere that the reverse flow system used by the LT-1 was conceived by Smokey Yunick for just such an application. He was the DaVinci of cheating - err, "skirting the rules".

I dig up 2 logs where 02s flat drop from 50 to 53 *C, than recover with no apparent reason. I know for sure I have other logs too but can`t find them. In par with the EOIT target change from 44 to 56 degrees or 270* to 180* which interpolated is 225*-247.5* at 50-53*C. Really wierd.

Funny you should mention this. I just tested a cold start - ambients here have dropped to -7c and IAT was reading 9.5 in the garage with the door slightly open. Initial start was awful - cranking time was 1.7s. This jumped out as different because I reset BLMs after startup because they were ridiculously high (140s) from the last idle session yesterday. When ECT hit 41.8c right side O2 was reading 36mv steady, and 8 seconds later the left side followed. Idle quality went to crap. I shut it down b/c I wasn't wanting to suffocate on fumes waiting for it to get to operating temp.

kur4o

11-13-2019, 02:06 AM

Initial start was awful - cranking time was 1.7s

Is it dumping more fuel or it is starving. IF it needs more, bumping the key several times will do the trick and should start right away on the 2nd to 4th bump.

Decided to look at lt4 calibration to see how the EOIT is set.
The stock settings are

y-LT4 ;y-LT1 ;f-LT1
RESERVED:337A fcb $20 20 20
RESERVED:337B fcb $20 20 20
RESERVED:337C fcb $20 20 20
RESERVED:337D fcb $20 20 20
RESERVED:337E fcb $20 40 40
RESERVED:337F fcb $20 40 40
RESERVED:3380 fcb $20 40 40
RESERVED:3381 fcb $30 ; 4c 50
RESERVED:3382 fcb $40 ; 4e 50
RESERVED:3383 fcb $50 ; 50 50
RESERVED:3384 fcb $50 ; 50 50
RESERVED:3385 fcb $50 ;
RESERVED:3386 fcb $50 ;
RESERVED:3387 fcb $50 ;
RESERVED:3388 fcb $50 ;
RESERVED:3389 fcb $50 ;
RESERVED:338A fcb $50 ;

There is also baro correction applied

RESERVED:338C fcb $1D ; 1x10 baro correction for EOIT
RESERVED:338C ;
RESERVED:338C ; SOME TABLE RESULT BYTE_EC
RESERVED:338C ; Baro corr to end of inj
RESERVED:338C ; table value-byte_338d=x
RESERVED:338C ; x+end of inj
RESERVED:338D fcb $1D
RESERVED:338E fcb $1D
RESERVED:338F fcb $1D
RESERVED:3390 fcb $1D
RESERVED:3391 fcb $1D
RESERVED:3392 fcb $20
RESERVED:3393 fcb $20
RESERVED:3394 fcb $20
RESERVED:3395 fcb $20

The baro table runs from 53kp to 104.44kp and stock setting are -3 to 0* EOIT added to main EOIT table.
On lt1 and y and f body the baro correction is set to 0.

ANother fueling routine that left untouched and might play big role in blm splits is the transition fueling. There is some alient math there, that substract fuel on certain condition, possible compensation for the fuel left on the walls unevaporated.

spfautsch

11-13-2019, 05:52 AM

Is it dumping more fuel or it is starving.

I'm thinking it was starving, but probably not due to a calibration fault. I noticed when I shut it off I could hear the pump running quite a bit longer than usual. It occurred to me this morning it was probably sucking air (the fuel gauge has been reading near E for a day or two). When I peeked in the filler cap I could plainly see the pump's filter sock, and it seems like the pump might have popped out of the bracket. The y-body fuel tank design is one thing GM solidly botched.

After adding a few gallons and re-priming the system, cold start at 6.5c popped right off. So the O2 shift I mentioned on both sides may be invalid due to air in the fuel system.

The stock settings are

My stock bin had this for the 12df2 table:

20
20
20
20
40
40
40
60
60
60
60
60
60
60
60
60
60

There is also baro correction applied

,,,

The baro table runs from 53kp to 104.44kp and stock setting are -3 to 0* EOIT added to main EOIT table.
On lt1 and y and f body the baro correction is set to 0.

Very interesting find.

There is some alient math there, that substract fuel on certain condition, possible compensation for the fuel left on the walls unevaporated.

This is referred to as Tau in the Banish stuff I have.

spfautsch

11-16-2019, 04:44 PM

There is also baro correction applied

...

The baro table runs from 53kp to 104.44kp and stock setting are -3 to 0* EOIT added to main EOIT table.

I've been thinking about this - is this adder before the 5.625 multiplier? If not it seems somewhat pointless.

The boiling point is reached when the vapor pressure of a liquid matches the atmospheric pressure. Raising the atmospheric pressure will raise the boiling point. Conversely, lowering the atmospheric pressure will lower the boiling point of the liquid.

This makes sense, except that at closed TPS evaporation will be taking place inside the port at well below atmospheric pressure. Makes me wonder if there's a closed throttle check happening before the adder.

On the subject of EOIT ramp out mulitplier:

There is 1 way to test it. Monitor the eside word_a9 while changing the EOIT target from end to end on the controller window with ignition on.
To do it open raw window in eehack, and loop in 50ms interval the command e4 ... 03 00 a9 00 aa , you will get a response with the word_a9.

Will this work with engine running or must it be off? I initially overlooked the words "with ignition on", tried with engine running and got the below error.

14809

Thinking about it, I assume this can't be done with engine in operation because it's on the e side?

kur4o

11-16-2019, 05:11 PM

The boiling point lowers with lower atmospheric pressure. The only example I can think of is that you can`t boil eggs on a airplane`s high altitude since the water boils at much lower that 100*C temperature.
In high vacuum situation the fuel will boil and evaporate at lower temperature. I think that is also taken into account in transient fuel calculations and maybe that is the point of it. The evaporation rate of the fuel is not constant and that`s why the transient fuel routine only substract fuel, but never adds it. If that is somehow related with the camshaft events the lt4 transient calibration might be set different.

is this adder before the 5.625 multiplier?
It is added to the calculated EOIT table value. This multiplier is not used by the pcm, it is only to convert the hexvalues to human readable degrees.

I assume this can't be done with engine in operation because it's on the e side?

Only with IGN on, engine OFF the eside talks on the bus. With engine on the pcm is too busy to handle aldl requests.

spfautsch

11-16-2019, 05:54 PM

This multiplier is not used by the pcm, it is only to convert the hexvalues to human readable degrees.

So the -3 would be a raw value, and would change eoit by ~17 degrees. That makes sense.

Either way, with a base at 0x50 or 270 ATDC that seems like a fairly small amount of evaporation time for aluminum heads based on our sweep tests.

It does give me another variable to go back and look at though - I didn't pay attention to baro when I generatd my sweep test data.

spfautsch

11-18-2019, 08:52 PM

Seems like prime pulse tables are good - "good enough" at the very least considering no fine tuning has been done. Longest cranking time has been 1.0s and shortest 0.5. Using the spreadsheet (latest update attached) I found adding 2.5% on top of the displacement adder has been starting reliably from 7c up. Warm and hot restarts seem good, but it's hard to tell now because ambient hasn't been above 30c in nearly three weeks.

Thanks for putting your work on the 121F1 table up. I had to move it out of the garage for a minute yesterday and did so while it was cold and it was noticeably smoother and stronger slipping the clutch while backing up the apron and into the garage with this change. In the absence of an "official" spark vs AFR compensation table this does the trick.

Another tidbit I wanted to leave with - having complained about fuel smell... I had basically run it to empty last week and grabbed a can of the ethanol free fuel I bought a few months back and dumped a couple gallons in. It's been years since I've smelled real gas burning - while it still has a strong smell at first, once the engine has been running a few minutes, and especially in closed loop it's nowhere near as rough on the eyes. I'm anxious to get my wideband functional again so I can compare as I run the real dino gas out and then refill with more ethanol "contaminated" fuel.