95 LT-1 Idle Cell Comparison - Humidity?

[kur4o]

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]

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]

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]

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]

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]

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]

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.