That's it! I'm effing done with this ECU! Going to megasquirt!

[Turbo Nasty]

1990 corvette with built motor and 32lb injectors. I'm so done with the crap from trying to tune this stock ecu. I never got it to enrich AE properly, it will not cut fuel off the throttle no matter what I do, it randomly kicks up my idle really high sometimes when I start it, I have to turn it off and restart, but far worse than any of that it will randomly lean out my tune like it's got a mind of it's own and it stays that way until I reflash the chip. But now even that won't fix it! So I'm done with it! It's so lean it's not driveable. I'm going to get a megasquirt. I'm not sure which one I'm going to get yet. maybe you guys know one that's ideal for a 1990 corvette or have done one already. Thanks!

[Six_Shooter]

You'll probably have the same problems with Megasquirt. It sounds like you probably have a mechanical issues or a few.

[Turbo Nasty]

I've already looked into that. Fuel pressure is good, timing is good, all of the injectors appear to be working fine. I may send them back to get tested and inspected though just to be sure. But I think they are okay. This is a new engine as of early this year, new fuel pump and filter, injectors are flow matched bosch type 3 32lb rebuilt from fuel injection connection. And how do you explain my tune changing and going lean some days and then not being able to fix it unless I retune my fuel maps or reflash the chip? Then a few weeks later I have to do it again.

[Six_Shooter]

I don't think it's your "tune changing". I think it's something mechanical that's getting worse and re-uploading a bin resets any stored learn trims. Tunes don't just "change on their own", which again leads back to a mechanical issue. The fact that re-uploading the bin isn't fixing it now, is also a good indicator of this.

I've seen this same post so many times and in every case it turned out to be not ECM related.

[Turbo Nasty]

What do you think I should check out? The fuel pressure goes up when I rev up the engine so I think the fuel pump should be good. Injectors all sound and feel like they are working. I took the passenger side injectors out to see if they were clogged but they look fine. I'll check the others tomorrow.

[89S10_Project]

I think a good first step might be to disconnect power to the ECM (disconnect battery entirely) and let the ECU clear, then see what happens when you reconnect. This would clear out the stored learn trims, would it not?

Concur your problem is likely to be mechanical, not ECM. You're running a GM motor with a GM ECM. They WILL play nicely together, as has been exhibited thousands of times with enthusiast builds. It's time to start with the fundamentals and build from there. I'm sure you're an accomplished wrench, and this seems silly- but this is where the problems lie, 9 times out of 10. I speak from experience.

Don't be big-headed about it, don't take this as offense. Be honest with yourself and don't assume "well of COURSE I did "x" right, I've done it a lot of times before". *Check and verify*.

I am in a similar boat. I have a stock 350 in a commonly-swapped frame (S10) using GM ECM made for the purpose.. and it's running like crap. So I have backed off to ground zero and am checking everything anew.

[dave w]

If I were to change computers, I would use the '0411 PCM. I've designed and built a plug-n-play crank position sensor kit to fit the L31 Vortec distributor, which makes upgrading to the '0411 very easy and affordable. See attached pics. Basically, the '0411 is flashed with a modified 2002 Express Van L31 5.7 liter tune. The crank position sensor kit, with new distributor, is on par with the cost of a MegaSquirt system.

dave w

[Stokes1114]

If I were to change computers, I would use the '0411 PCM. I've designed and built a plug-n-play crank position sensor kit to fit the L31 Vortec distributor, which makes upgrading to the '0411 very easy and affordable. See attached pics. Basically, the '0411 is flashed with a modified 2002 Express Van L31 5.7 liter tune. The crank position sensor kit, with new distributor, is on par with the cost of a MegaSquirt system.

dave w

If you dont mind me asking, Are you going to be selling these crank sensor kits?

[dave w]

If you dont mind me asking, Are you going to be selling these crank sensor kits?Yes, I will be selling the crank sensor kits. The production crank position sensor housing is 3D printed with Black High Temperature Glass Filled Nylon material, which is similar to the distributor cap ABS material. The prototype, which is pictured in my previous post, was 3D printed with a material called All Temp. http://www.goengineer.com/3d-design/3d-printing/?gclid=CPyvzof349YCFRl2Xgod9pAPow

dave w

[johnny_b]

If I were to change computers, I would use the '0411 PCM. I've designed and built a plug-n-play crank position sensor kit to fit the L31 Vortec distributor, which makes upgrading to the '0411 very easy and affordable. See attached pics. Basically, the '0411 is flashed with a modified 2002 Express Van L31 5.7 liter tune. The crank position sensor kit, with new distributor, is on par with the cost of a MegaSquirt system.

dave w

dave w this is awesome! I do 3d printing as well.
Anyways couldn't this setup also be used in an lt1 with a modified intake for a conventional distributor? 0411 is multiple coils correct?
I'm very interested in this. do you have a separate thread regarding this project?

Thanks

[dave w]

dave w this is awesome! I do 3d printing as well.
Anyways couldn't this setup also be used in an lt1 with a modified intake for a conventional distributor? 0411 is multiple coils correct?
I'm very interested in this. do you have a separate thread regarding this project?

Thanks

I don't have a separate thread for the crank trigger kit.

If the LT1 intake was modified to accept a conventional distributor, the distributor crank trigger kit I've developed could be used. The 2001 / 2002 Express Van Vortec L31 5.7 liter uses a 4 tooth (4x) crank trigger wheel. When moving the crank trigger to the distributor, the number of teeth on the trigger wheel needs to double. That's why my kit has an 8x trigger wheel.

The '0411 coil on plug ignition system (distributor-less) uses a 24 tooth (24x) crank trigger wheel, which means a distributor mounted coil on plug trigger wheel will need 48 teeth. More Research and Development is needed to make a 48x trigger wheel work in the L31 distributor.

dave w

[Turbo Nasty]

I've had the battery disconnected many times, I just swapped the injectors from bank to bank because I have my wideband on the passenger side and narrow band on the drivers side and I wanted to see if the injectors are consistent and they are. So it's not the injectors. I tried several tunes I had in store today and each one definitely had their own personalities. Weird that a few days ago they all seemed to run identical. However even my known richest fueling tune the one from dyno day ran lean once I started driving it. My fuel pressure is good idling and revving up but I can't see what it's doing while driving. Is it possible to have good fuel pressure and have lack of flow getting to the injectors cause by a clogged fuel filter? I will be replacing the Fuel Filter tomorrow just in case. If that's not it I will have to suspect the Fuel Pump. I think I should mention I'm using an aeromotive fuel pressure regulator with guage right on the fuel rail.

[Turbo Nasty]

How is this helpful Dave W?

[Turbo Nasty]

If I were to change computers, I would use the '0411 PCM. I've designed and built a plug-n-play crank position sensor kit to fit the L31 Vortec distributor, which makes upgrading to the '0411 very easy and affordable. See attached pics. Basically, the '0411 is flashed with a modified 2002 Express Van L31 5.7 liter tune. The crank position sensor kit, with new distributor, is on par with the cost of a MegaSquirt system.

dave w
How is this helpful?

[dave w]

My post was centered on the topic or comments about changing computers, MegaSquirt computer vs. Factory computer. Generally speaking, the Factory LS1b or '0411 computer is one of the best supported computers, including support from EFI Live and HP Tuners. It's very possible there are several tuners in your area that have extensive experience with the '0411. Maybe there are tuners in your area with extensive experience tuning with Megasquirt?

I admit, my post did not mention any possible solutions that would be helpful to fix your engine.

dave w

[steveo]

Is it possible to have good fuel pressure and have lack of flow getting to the injectors cause by a clogged fuel filter?

on any sane fuel system, lack of flow would cause a pressure drop as fuel flow requirements exceed available flow, so no. you'd see it on a pressure gauge, although revving in your driveway is quite different from driving.

you could also have intermittent fuel pressure loss, might not be the pump, could be loose clamp in tank, cracked pickup hose, something like that.

i agree that switching ECMs right now would be a bad idea, in fact a different ECM might 'tune around' the problem more easily, but you'd still be burying whatever is wrong.

if 'resetting' or reflashing or whatever temporarily cures the problem, i'd say you have some kind of problem with closed loop. why don't you burn an open loop bin and see if it behaves itself?

it's also possible your ECM is flaky, but you'd probably see other evidence of that

i'd be interested to see some logs of when the weird behavior happens

[Turbo Nasty]

on any sane fuel system, lack of flow would cause a pressure drop as fuel flow requirements exceed available flow, so no. you'd see it on a pressure gauge, although revving in your driveway is quite different from driving.

you could also have intermittent fuel pressure loss, might not be the pump, could be loose clamp in tank, cracked pickup hose, something like that.

i agree that switching ECMs right now would be a bad idea, in fact a different ECM might 'tune around' the problem more easily, but you'd still be burying whatever is wrong.

if 'resetting' or reflashing or whatever temporarily cures the problem, i'd say you have some kind of problem with closed loop. why don't you burn an open loop bin and see if it behaves itself?

it's also possible your ECM is flaky, but you'd probably see other evidence of that

i'd be interested to see some logs of when the weird behavior happens

I thought the same thing about the fuel pressure. I used to play with closed loop but closed loop would make my car run lean and then I learned I needed a heated o2 sensor but that did not fix it in closed loop still ran lean so I just always leave closed loop turned off by setting the enable temp closed loop as high as it goes. But now it's running lean in open loop too!

I have a moates data logger but not sure how to record with it yet. I've been using it to see the data live on the tunerpro RT dash.

[Turbo Nasty]

My post was centered on the topic or comments about changing computers, MegaSquirt computer vs. Factory computer. Generally speaking, the Factory LS1b or '0411 computer is one of the best supported computers, including support from EFI Live and HP Tuners. It's very possible there are several tuners in your area that have extensive experience with the '0411. Maybe there are tuners in your area with extensive experience tuning with Megasquirt?

I admit, my post did not mention any possible solutions that would be helpful to fix your engine.



dave w

The MS2 should be a pretty straight forward install on my vette, it doesn't need any aftermarket cam sensor setup and I've built a MS1 for my turbo Miata before. The tuning part on that ecu was a lot easier than what I've dealt with trying to tune my stock vette ecu.

[Turbo Nasty]

From what I can tell my tunes are changing based on weather conditions of the day. It's finally getting cooler in AZ and it's messing up my tune. probably has something to do with the fact that tunerpro won't accept my bosch injector offsets so I'm using the stock offsets and now it doesn't know how to make adjustments because it's using bad information.

[Turbo Nasty]

Today I warmed up the car and it was lean cold, rich while warming up and just a little lean at operating temp. it was way lean cruising but if I stepped on the gas it was good 12.5-13.0 afr. But then it would start the lean out again at higher rpms 4500+ about 14.5 afr and I'd let off the gas.

[steveo]

From what I can tell my tunes are changing based on weather conditions of the day. It's finally getting cooler in AZ and it's messing up my tune. probably has something to do with the fact that tunerpro won't accept my bosch injector offsets so I'm using the stock offsets and now it doesn't know how to make adjustments because it's using bad information.

how do you mean? lets see your bin, xdf, and target offset data

[Turbo Nasty]

how do you mean? lets see your bin, xdf, and target offset data

I mean it's hot as hell in Arizona and that's the last time I tuned the car and it's stayed hot until recently and now my tune is way off. My tune was never really too good to begin with.

I did just have success richening it up. I got a wide open throttle run in an it was pig rich all the way! I did this by setting the 'Open Loop % Change AFR vs Temp' to all zeroes, and 'Open Loop % Change vs MAP' to all zeroes. The one that made the big difference was 'MAT Compensation Counts vs MAT' set to all zeroes. after the very rich WOT run I adjusted the affected temperature ranges to 40 down from the factory 128. it got my idle around 13.5 AFR and full throttle the same.

[Turbo Nasty]

Why so many crazy variables on this car!? I'm tempted to set everything to zero and just run off of the fuel map alone and retune the map for it The only noticeably affective variable was the 'MAT compensation Counts vs MAT'

[Turbo Nasty]

Here's todays bin.

[Turbo Nasty]

I'm using the $8D-1227730-V1.xdf.php

for some reason I was not allowed to upload it here. it said it was an invalid file.

[steveo]

Why so many crazy variables on this car!? I'm tempted to set everything to zero and just run off of the fuel map alone and retune the map for it The only noticeably affective variable was the 'MAT compensation Counts vs MAT'

this isn't at all normal, there are no 'crazy variables'. 8D is supposed to be really easy to tune, with a VE table in the ballpark, it should be drivable in all conditions. i tuned an 8D car for a guy once and had no problems, in fact on the stock bin it ran alright, just sucked a lot of fuel.

i know it's frustrating but do you want to try starting from scratch?

i looked at your bin and i think your VE tables are probably awful. i also saw a few tables for fuel compensation vs temperature and map were zeroed out. i think those are best left alone.

with your trouble with injector offsets, show me the injector data you're trying to enter and lets figure out why it wont take it. it should accept anything from zero to 7782 usec. just focus on getting the ~9-14v values correct, if you're outside that range, you're limping home anyway. i wonder if you're trying to put msec values in there or something.

[steveo]

i looked at your bin and i think your VE tables are probably awful.

i'd like to clarify that statement and just say that after tuning speed density for a while, you get to look at a ve table and go 'that shit will never work'. and that's exactly what pops into my head when i see yours.

was this developed from wideband data, trims, by feel, or what?

[steveo]

also how about a datalog and an adx? i want to troll it for sketchy sensor output.

if the forum wont accept a file simply zip the thing or rename it to .bin and tell us what the old extension was.

[steveo]

maybe this will give you some hope

i put a DGV weber carb on my beater mazda truck. it's pretty stock except for some crazy handmade exhaust.

it's a mechanical progressive carb, that means one barrel opens until half throttle, then the bigger barrel opens till wot, so a quiet and efficient cruising range with a massive (and fairly sudden) boost for heavy throttle.

with the jets that the mfr swore would run fine, it ran great on primary barrel, but when the secondaries open, it would bog and stop accelerating, at the same time it made a ton of noise like it wanted to smoke the tires.

they blamed the engine, so i checked everything.

distributor mech advance, vacuum, compression, ignition timing, cam timing, fuel pressure, exhaust restriction, new plug wires, the works, and it was all the same. i even re-gasketed the entire top end, since it was leaking a bit of coolant. all checked out.

i slapped the wideband on it, and it was at ~17:1 WOT.

jet kits are hard to come by for this thing, so i got fed up and grabbed a regular set of drill bits (can't find my numbered ones, maybe I sold them)

knocked the primary jet open with a 1/16th utility drill

i figure i probably made a #160 jet instead of the #140 that was in it.

runs perfect now. wideband avg 13:1 wot, 15:1 cruise.

i'll order a jet kit, but it's no rush.

moral of the story is, even with a computer, don't forget carb logic. sometimes, after you're 100% sure it's a fuel delivery problem, you just have to throw a good amount of fuel at the poor beast where it's hurting the most, then fine tune from there. a bit more gas never hurt anything, that's why you can pour fuel into an engine from a pint glass and it'll run.

[Turbo Nasty]

Here's the bin from dyno day last January. It ran great on this tune for two weekends or so and then the car wouldn't run right on it anymore. It was too rich on the street.

And here's the stock chip copy from my car too for comparison.

[Turbo Nasty]

also how about a datalog and an adx? i want to troll it for sketchy sensor output.

if the forum wont accept a file simply zip the thing or rename it to .bin and tell us what the old extension was.

I don't know how to start the recording with the datalogger yet. I will see if I can get the adx for you/

[Turbo Nasty]

I think I got'em

they were both .php before I zipped them

[Turbo Nasty]

Here's the injector offsets that my tunerpro RT rejects every time. For Bosch type 3, 32lb

9V
0.954994

9.5V
0.827989

10V
0.725101

10.5V
0.646214

11V
0.573173

11.5V
0.500671

12V
0.435207

12.5V
0.386359

13V
0.339858

13.5V
0.292894

14V
0.24897

14.5V
0.212892

15V
0.176967


I know these are not the direct values to put into tunerpro because tunerpro goes something like 14.0v, 14.2, 14.4 etc. I found a fancy calculator somewhere to convert the values for me. I think I have them written down here somewhere.

[steveo]

those are msec. a millisecond is 1000 microseconds (usec).
multiply by 1000. it may round to ~30usec

[Turbo Nasty]

those are msec. a millisecond is 1000 microseconds (usec).
multiply by 1000. it may round to ~30usec

Yes they are. I bet this will make a big difference in the tune. And by the way I am using a wideband AEM EUGO I think you asked earlier.

[Turbo Nasty]

If I were to change computers, I would use the '0411 PCM. I've designed and built a plug-n-play crank position sensor kit to fit the L31 Vortec distributor, which makes upgrading to the '0411 very easy and affordable. See attached pics. Basically, the '0411 is flashed with a modified 2002 Express Van L31 5.7 liter tune. The crank position sensor kit, with new distributor, is on par with the cost of a MegaSquirt system.

dave w

I appreciate the suggestion but would only consider it if that ECU is boost friendly

[Turbo Nasty]

Making progress on the tune today. I used an interpolation calculator to find my injector offsets and the tune is much happier now. I still had to get the MAT compensation adjusted to get my AFR's inline but that was about it. the tune is very consistent and predictable now and I hope it stays this way. I think it just needs a street tune for the fuel map or it needs the power enrichment reduced. It's running 13.5-15 AFR cruising but 10 at WOT

[Turbo Nasty]

Say you have a built engine like mine. You start with the factory tune and the only items that should need to be changed are Fuel/timing maps, injector offsets, cylinder volume, injector flow rate and initial spark advance. That should be it right?

[Turbo Nasty]

Not as much progress as I wanted today. Some things I noticed that are going on. If I use closed loop it stays active or it's fuel trims stay active even after I turn it off again. What happens when I use the stock injector offsets with closed loop is that it can't properly control the AFR and it end up going really lean. So in order to turn off closed loop I have to set the closed loop enable as high as it goes and burn the chip then shut off the car and restart.
When I use the correct injector offsets they are not compatible with my current fuel maps because those maps were tuned around the wrong injector offsets. So what happens when I use the right injector offsets with these wrong fuel maps is that it runs really lean everywhere except for WOT then it runs super rich. My fuel maps are already skewed in a bad way for some reason so I don't want to tune it out like that.

[Turbo Nasty]

Why is it that my fueling is so high on the lower KPA's? I know I must have made an adjustment to some kind of variable that did this but I do not remember what it was. It actually runs good this way but I know this is wrong and it wasn't always like this. I am posting todays .bin with the skewed map. notice it is really high on the left side lower KPA's.

Now I will post the .bin from xact dyno the day I dynoed almost a year ago. It ran perfect that day but if I use this bin today it is way too lean. What did I change that skewed my fuel maps so badly? I know my maps should look like this one from xact dyno.

[dfarr67]

You should be logging with a WB and a fuel pressure transducer.

[Turbo Nasty]

You should be logging with a WB and a fuel pressure transducer.

How am I going to plug that in to tunerpro? Can I splice that into my ALDL somehow? I already know for sure the issue is in the tune and not a fuel supply issue. It's a fuel control issue.

[dave w]

I wonder what is happening with the BLM averages and O2 sensor data? See sample screen shots from TunerPro RT and Log File Analyzer. It would be very helpful to post a .xdl file for us.

dave w