E38 LS3 427 Tune Help

[LastCall]

Working on an LS3-based 427/6 speed in a 69 Camaro. It's a very nice car, just wish is ran better.

There are several issues, but the few that have me stumped are the following:

LTFT are erratic - MAF-only tune. The LTFT might be pegged at +24% between idle and 2k RPMs. During this time, the O2's are not switching, just kind of flat lined at 550mv. Then they will start switching (look at frame 15k in the log) and the trims start coming down, all at the same MAF frequencies. Then I let the car idle and LTFT maxed out got a DTC for P0171/P0174, car smells rich though. All this on the same tune.

Hard start, hot or cold - tried adding a cranking idle air, (2 grams at a time until 22!), adding cranking fuel, reduced cranking VE, and reduced TB scaler (to 4200!) And there is a ton of min idle airflow. No change in start behavior. Cranks for a few seconds and then once started, runs pretty good.

One thing I noticed at the end of the day was that the charging system may not be up to par. Idling is about 12 volts, With the headlights and fans on, the voltage drops to high 10's/low 11's while idling. Voltage while cranking (measured in scan tool) was 8-9 volts. Seems really low to me, but not sure if that is the culprit of the hard starts.

Any pointers or help would be greatly appreciated!!

I did verify fuel pressure, it was 54-58 while cranking and at idle.

[joecar]

Does IFR table match FPR (referenced or not)...?

Try doing a full flash.

Clear trims after flash.

First, fix charging system... check voltsge drops, especially across grounds.

Also check engine grounds.

[statesman]

First, fix charging system... check voltsge drops, especially across grounds.

+1 That has to be fixed first.

[statesman]

LTFT are erratic - MAF-only tune. The LTFT might be pegged at +24% between idle and 2k RPMs. During this time, the O2's are not switching, just kind of flat lined at 550mv.

Check your O2 sensor heater fuse.

[LastCall]

Does IFR table match FPR (referenced or not)...?

Try doing a full flash.

Clear trims after flash.

First, fix charging system... check voltsge drops, especially across grounds.

Also check engine grounds.

Un-referenced reg, corvette filter set up.

Since this is a swap and PSI removed the VATS via HPTuners, I did not want to corrupt the ECM. Can I full flash over an HPT file and remove the VATS via EFILive?

I did clear the fuel trims after each flash.

Plan on checking the grounds and charging system today.

+1 That has to be fixed first.

That's the plan.

Check your O2 sensor heater fuse.

Didn't think about that, good idea, will do.

Here is a log from yesterday, I believe one of the issues is the charging system/battery. as you both have indicated. Looking at the log below, Control Module Voltage and Battery Voltage drop from 12.4 to 11.5 and at the same time 02 voltage drops from ~.580mv to .335mv (frame 562). When voltage goes back to 12, the O2s go back to .580mv range.

I did crank the car once (frame 243), voltage drops to 7/8 volts initially, then slowly goes to 9-10 at the end of the cranking cycle.

Did you guys see anything in the tune that would make for hard, extended starting?

[Doc]

Sounds like you are on an adventure that I am not all too familiar with. First off, where is the MAF? Physically in the intake stream? Could you post an engine bay pic? Don't have time to check the tune as I am on the dyno right now...Just guessing, bigger the cubes, the less startup timing is generally necessary than what the stock file would/could have. Physical placement of the MAF needs to be in a linear (as much as possible...lol) position for good steady readings. Erratic readings will force the pcm to try to rectify the problem with fuel trims. Not fun or effective. What happens if you unplug the MAF? AFA the HPT or "other" breeds of software cross breeding, generally speaking if you are dealing with stock OS's you can sometimes get away with it. I have just made a habit of completely starting over and crossing over information in a pure bread EFI Live calibration just to avoid any possible frustrating weirdness. Did a 68 Chev Panel Van with 6.2 E38 combo a few months ago where the customer did all the work and had a PSI harness. The customer provided the pcms (motor e38, tranny t43) which unbeknownst to him were different years. They did NOT get along at all. Finger pointing between the tuning and harness ensued. Calm heads prevailed and figured it out. PSI guys mentioned that they have had customers inadvertently plug in the MAF to the cam or crank sensor; can't remember which, but, if you try really hard and abandon common sense you can accomplish this. No start with that. Not your problem just wanted to pass on the experience.

[LastCall]

Sounds like you are on an adventure that I am not all too familiar with. First off, where is the MAF? Physically in the intake stream? Could you post an engine bay pic? Don't have time to check the tune as I am on the dyno right now...Just guessing, bigger the cubes, the less startup timing is generally necessary than what the stock file would/could have. Physical placement of the MAF needs to be in a linear (as much as possible...lol) position for good steady readings. Erratic readings will force the pcm to try to rectify the problem with fuel trims. Not fun or effective. What happens if you unplug the MAF? AFA the HPT or "other" breeds of software cross breeding, generally speaking if you are dealing with stock OS's you can sometimes get away with it. I have just made a habit of completely starting over and crossing over information in a pure bread EFI Live calibration just to avoid any possible frustrating weirdness. Did a 68 Chev Panel Van with 6.2 E38 combo a few months ago where the customer did all the work and had a PSI harness. The customer provided the pcms (motor e38, tranny t43) which unbeknownst to him were different years. They did NOT get along at all. Finger pointing between the tuning and harness ensued. Calm heads prevailed and figured it out. PSI guys mentioned that they have had customers inadvertently plug in the MAF to the cam or crank sensor; can't remember which, but, if you try really hard and abandon common sense you can accomplish this. No start with that. Not your problem just wanted to pass on the experience.

Thanks for chiming in Doc!

The MAF is located in the intake track, sorry couldn't get a photo. It is mounted horizontally in the middle of the 4" intake pipe. The MAF reading are not fluctuating too much, so not sure if that would be the culprit.

It is commanding about 4 degrees of timing will cranking, according to the log.

I have had issues applying a patch over the HPT patch, resulted in a no start condition. Had to flash in a different operating system and then back to the OS I wanted to run. When I did that it worked.

I'm picking up the RPM signal and not getting a cam sensor code, so I think I'm okay there. These swaps can get tricky, especially when multiple shops have touched the car and the hardware is pieced together.

I have a general question about trims, hoping you or someone else could answer.

If the O2's are not switching, are the trim values accurately representing the AFR (at stoich). I would think they would essentially become static at whatever value they are flat-lined at? From the reading I've done, a very small amount of fuel is injected/removed to get the O2s to oscillate. Based on the fuel added/removed, the ECM expects a corresponding MV value from the O2s. The difference between the expected change and actual change as reported in the O2 voltages determine the values in the STFT?

If I'm on based (probably am), let me know. Thanks!!

[Doc]

Good morning. Late night. Drinking coffee finally had a chance to look over the tune file. First and foremost I consider myself a technician with a full shop and lots of experience that is on a continuing education program in regard to using third party (no offense) hack tools that hack into the worlds largest properitary computer systems with windows based interfaces that give the best representation of what the end creators, in this case EFI Live, what they thought the actual tables made by the manufacturer really were / are and what they do. EFI Live does a great job in this regard. In EFI Live's and any other aftermarket software company's defense it's not like Paul and Ross have a direct line to GM Engineering head quarters and when a new platform comes out GM just hands over a full handbook of what they built. The car mfgs are extremely protective of their intellectual property and hold those secrets tight to the chest.

That being said I have come to the conclusion that in the spirit of trying to apply experience, reading what little actual manufacturer explanations provided at basic levels via tech manuals and via other calibrators experiences, I believe weather it is true or not that the first and foremost piece of the puzzle is the ecm having proper engine displacement and the best injector data you can obtain in order to get repeatable results.

In this, I noticed that your displacement value is still .77 or 6.2 liter. Rough math tells me that 427cu in. equals .875 so {B0104} should reflect this fixed value. MAF, VE, Virtual VE, Injector tables can all be altered to "make things run," but the end result usually is an un-balanced condition that manifests it'self in weird unpredictable, annoying ways.

It's like leveling a telescope or transit. All three legs need to be adjusted to the level of the earth. In this case our three legs are FUEL, AIR and Spark. One or two or even all three being out of balance forces you the calibrator to do some extreme tail chasing all over the place in order to try to get things in order.

Start off by ensuring you have the most accurate data available about the motor displacement, fuel injector data as they are static fixed values. After that using wideband and narrow band A/F values to determine the delta offsets via the MAF and or VE, Virtual VE tables and making the adjustments will usually result in very low fuel trim values.

I make that last statement assuming that all of those aforementioned sensors are in good working order. Simply put they are reporting to the ecm actual correct values so the ecm can make short term and long term corrections.

Improper setup of those sensors weather physical/mechanical/electrical will throw you into the bad data in / bad data to the ecm with hair pulling out results.

You did mention some electrical challenges. After having calibrated lot's of LSx transplant projects into Airboats, tons of old cars, steady regulated power and fuel are next on the check list. Ruling out, checking power, ground, fuel pressure and volume takes another set of possible bad actors off the list.

Thanks to all the hard work by the EFI Live team for a great product that gives you the calibrator a really good chance for success by virtue of the powerful scantool and editor.

Forgive me for stating this in this thread and not in the requested features but, I do have to take the opportunity to make a request for a "switch," in the scantool that can / could "turn off," the searching for modules that don't exist on these LSx transplant projects that are so popular and comprise nearly 50% of the work load I encounter. The waiting for the scantool to go through the motions of trying to find non-existent modules is painful when attempting to help make a customer's dream come true of having a LSx motor on their bar stool work.

Back to your situation, I think that if you change the B0104 and revert back on a lot of the little changes you have made like cranking fuel, etc you might just have a break through moment.

As far as the HO2s, if the car has as I suspect a decent sized cam then don't get too wrapped around the axle about them until you try the aforementioned suggestions. You have to remember that the HO2s were designed to operate in a stock environment, expecting them to report and work and express values you would come to expect on more stockish applications can be thrown out the window. For example, large cam that makes that great head turning sound and power can produce/report large weird looking trim values at low / idle conditions. Taking those values literally and trying to adjust the calibration accordingly can produce strange unwanted results while say for argument really lean + or rich - trim values reporting but the car runs smooth like butter has to command your attention that ultimately without too much nebulizing as to WHY and just give the car what it wants where it is stable and predictable is the way to go. That is until you have a customer with a loose understanding and phone/bluetooth data logger that asks about this "strange," looking data and asks you WHY?

We are technicians, calibrators on our best days. AFAIK, we are not engineers paid by the OEMs, they sign non-disclosure agreements keeping those corporate secrets in house.

Hope this helps.

[LastCall]

I appreciate the thorough response Doc, it is much appreciated. The info you have posted is very helpful.

Good catch on the displacement, I should have caught that before making any changes to the tune. Going to do that now.

I went through the injector settings and they look correct. They are 42# LS3 injectors and compared to an 08 LS7 tune. The only difference I see is Injection Timing RPM (B1206). Values are different from 0 to 1024 RPM. All the other settings match.

A few differences for cranking fuel I noticed comparing this file to a stock 08 Corvette file (what this tune was based on) was B0181 and B0182. Not sure how much these matter, but I'm going to try returning them to stock.

I have tuned a few E38/E67 combos with a cam, and familiar with the lazy switching at idle/low RPM scenarios. Sometimes even with just headers, the O2's can become lazy. I think EFILive may be missing a few parameters to properly adjust the O2 operating characteristics (proportional vs airflow, integrator delay, and limit/error tables). My research led me to the "other forum" where people were making changes to these areas to get the O2's to oscillate in low airflow conditions.

Do you usually disable the LTFT once you have the MAF/VVE corrected? How do you typically correct for trims at idle with EFILive software?

I verified the fuel pressure while cranking and idling. 54-58 cranking and 58 while idling stationary. I could drive around with the gauge as I had it plugged into the fuel rail.

[LastCall]

Hi Doc, regarding your on/off on the scan tool. In the properties menu, there is selection in the "Start Up" tab that looks for all OB2 info. Is that checked? There is also a "Printing" tab, not sure how that one works.