Did you ever just loose 10 degrees timing? Math?

[JeepsAndGuns]

You said you unchecked MPFI bit?

Still think 10 deg outside ecm / pcm.

Not sure about anyone else, but once I switched to mpfi, I checked the flag and have never unchecked it.

Mark, I just pointed out those same two fueling parameters in my thread just seconds before I read this one Those are the only ones I can think of that would have effected fuel, other than a couple AE parameters.

[EagleMark]

Duraspark module MSD distributor has been replaced with stock GM... still missing 10 degrees?

[1project2many]

Want to make sure we're speaking same language....
When you locked timing in code, you set all MIN the same, and MAX one degree different?

When this is done and engine is running, check DRP flag to see if it's being set (it should be) and advance flag to see if it's set or cleared.

Do you have an ECM bench harness for this computer? How about a scope? Can you set the bench up with a and current calibration for testing?

Also, if you want to forward cal I can take a look to see if anything jumps out.

[jim_in_dorris]

Mark, I may be way off, but there is one other common change to MPFI conversions that we haven't discussed. That is the jumper change on the memcal. Any posibility of putting an unmodified memcal in to see if the 10 degrees come back? It's the only thing we haven't looked at yet.

[comerz1]

Thought I would chime in here since one of the projects with this issue is mine.

jim_in_dorris, I picked up another 7427 ecm with memcal today to try exactly that, just didn't get a chance to do so yet. What will not having the jumper installed do to fueling? My understanding is the jumper makes the injectors fire every 4th drp, so will they fire more often without the jumper?

I will report back my findings on the memcal. Qucik question, the one I acquired today has a different BCC than my original, will that make any difference?

[EagleMark]

I beleive 1project2many said that the jumper was not needed? Or borderline? Others say it is needed for more then six saturated injectors. So I'll be looking forward to results using a stock memcal!

BCC if from 5.7L should be fine, just don't drive it if it's a 1995 that has PWM TCC. For that matter just use your Adapter and AUtoProm and bypass the chip. Then we have apples to apples comparison with/without jumper wire.

[comerz1]

That was my plan, just remove the memcal from the adapter, stick the unmolested one on, and see what happens!

[1project2many]

I beleive 1project2many said that the jumper was not needed? Or borderline? Others say it is needed for more then six saturated injectors. So I'll be looking forward to results using a stock memcal!

Boy... there really seems to be a lot of confusion when it comes to "jumpers." Any more of this and you'll find me on the ledge looking to jump, too. ;)

There are two jumpers (technically three) discussed with the MPFI conversion. The first jumper needed is at the sense resistor, not the memcal. This jumper eliminates the peak and hold function of the injector driver circuit. When using 8 saturated type injectors you need to short the sense resistor and a jumper is the recommended procedure. Since there are two resistors, there are two jumpers needed.

The second jumper mentioned is the mode select jumper on the memcal. Setting the voltage to zero on the mode select pin causes the TBI injector driver controls to use signals in hardware to trigger the injectors every fourth DRP. The easiest way to set the voltage to zero is to install a jumper to ground.

Most of the various forum discussions regarding GM ECM PFI and TBI mode select are related to the P4 ecm family. The 7427 is a newer generation of pcm and as such, has differences which may not be fully understood or utilized. One of these is the ability for software to control both the "on" and "off" times for the injectors. In the p4 the program would determine a duration of time for the injector to turn on. That value would be sent "away" to the injector control hardware. The control hardware would then enable the injector driver for that amount of time. As long as the hardware kept receiving distributor reference pulses and the injector enable time wasn't changed to zero, the injectors will keep firing. This is the way TBI mode is handled in the 7427 and setting the memcal jumper above simply alters when the TBI mode hardware triggers the injector drivers (both at once every 4th pulse vs every other driver / every other pulse).

The 7427 has an "also." The 7427 can also start and stop the injectors entirely from software. This is separate from and independent of the hardware control used in TBI mode. When CPI mode is enabled, this is how the injectors are triggered. Software triggers both "on" and "off." Based on reading on thirdgen, if you make a mistake selecting your option bits with CPI mode you will have driveability issues or the engine may not fire at all. Installing the memcal jumper and leaving TBI mode selected provides a hardware solution to this problem that bypasses potential problems from selecting the wrong option flags in the calibration.

[jim_in_dorris]

1project2many,
I have to reply here, because I understand exactly what you are saying here, and believe that to be the case. That stated, it is one of the few things that haven't been tried.. (Okay, I really don't expect that it will change anything, just throwing sh&& against the wall to see what sticks ). We have 2 completely different vehicles displaying the same symptom, obviously something is the same on both systems that is not common on tbi systems. What is it? I am not a master diagnostician, nor am I in possession of either vehicle, so all I can do is conjecture. I built a $0D TPI setup for a friend and did both the sense resistor mod and the mode sensor mod as well as the code changes discussed here. Unfortunately, 2 years later, the motor is still sitting on his garage floor while they are doing a complete body off, suspension change, 2/4 drop and boxing the frame. Then they plan on doing considerable body work, so I have no idea when the project will have hot water. I am following this thread because I don't want to have this problem when it finally gets hot water.

[1project2many]

I'm not saying "do" or "don't" to trying the stock memcal. There have been multiple posts lately in multiple threads that seem to confuse injector drivers and related modifications with the driver control system so I'm pointing out the differences.

Additionally, alhough I haven't even mentioned it, there's evidence that the MPFI conversion document the paople are using has incorrect settings as well. Selecting PFI mode at the memcal appears to bypass that problem but there remains a question as to what effects this has on the code.

I have not followed JeepsandGuns thread. I'm only addressing the problem presented here on this vehicle to reduce variables. There are tests that could be done with an oscilloscope that might speed this process along but apparently that route is out. So back out completely and start from scratch. Use pcm with jumpers on sense resistors. Install stock calibration adjusting only tables necessary to make it idle. Check spark timing at idle. If timing matches run full comparison between cals and look for differences. Check for options selected that shouldn't be, constants that aren't as expected, and especially look for corrupt bits in code. Gotta have a starting point.

[helo]

Duraspark module MSD distributor has been replaced with stock GM... still missing 10 degrees?

I've had 10* in BOTH directions unexpectedly/unintentionally. Source was things like altitude compensation bias, EGR bias, main bias. I'm certain it was merely an XDF issue or something when I saved one or more tables. I worked around it by setting most/all bias's to 0 (negative advance with EGR on or high altitude compensation would be pointless, right?)

[EagleMark]

The 7427 has an "also." The 7427 can also start and stop the injectors entirely from software. This is separate from and independent of the hardware control used in TBI mode. When CPI mode is enabled, this is how the injectors are triggered. Software triggers both "on" and "off." Based on reading on thirdgen, if you make a mistake selecting your option bits with CPI mode you will have driveability issues or the engine may not fire at all. Installing the memcal jumper and leaving TBI mode selected provides a hardware solution to this problem that bypasses potential problems from selecting the wrong option flags in the calibration.

So what happens when you do both the jumper in memcal and check CPI/MPFI mod?

I've also questioned some of the settings outlined, mainly all ASYC fueling settings?

But Async would be batch fire and Sycn would be bank fire? Why couldn't this run Sync now that it is MPFI? Sync could be an issue with CPI...

I've had 10* in BOTH directions unexpectedly/unintentionally. Source was things like altitude compensation bias, EGR bias, main bias. I'm certain it was merely an XDF issue or something when I saved one or more tables. I worked around it by setting most/all bias's to 0 (negative advance with EGR on or high altitude compensation would be pointless, right?)

CPI bins have no main spark bias. I found in code once where spark is added up the last thing done was SUB spark bias, this was a major concern with change to CPI/MPFI code, is the main spark bias still Subed out?

But you mention saving a table after changing a bias? One thing I did with AutoProm running live was take out 5 of main spark bais, spark at crank incresed by 5, I made no changes to Spark Advance table, but if you open spark advance table after... AFTER a bias change is shows up.

So if code strips out bias! Why did timing change when I reduced it by 5? Spark table still goes up by 5 and still 10 short to crank... hmmm... I getting my headache back, need more coffee...

[ony]

egale mark, look at my 6 cyl lean cruise file for a 747,it does not run at the same timming as it should, minis 5-10 it maybe a 6cyl thing unless something is on kpa instead of vacume.

[EagleMark]

In $42 there is no correction for main/warm spark bias in spark advance table, what you see is spark advance plus main spark bias. So comparing to data it all off by main spark bias as it's supposed to be. What you need to know is if distributor is set to zero, does timing at crank match data? If so spark table should be off by main/warm bias. 350 and V6 have a bias, BB/454 does not.

I fixed it in next release. You will see actual in SB or BB.

The difference in $0D is when looking at spark advance table it has main spark bias removed, what you see is actual... well until now...

[1project2many]

So what happens when you do both the jumper in memcal and check CPI/MPFI mod?

Unknown. Settings listed on thirdgen produced no injector pulses as code simply exited mpfi mode code without referencing DRP. Someone reported same setting with memcal jumper and engine ran. Appeared that memcal jumper forced injectors to fire even if code wasn't.