23037* Spark Advance???

[Jimmie]

I have a 1994 K1500, I built a new vortec head 350 and have been having issues with the tune since the engine swap. I'm using tunerpro rt, Ostrich 2.0 and aldu1. The issue I'm having is the datalog shows random 23037* spark advance, watching the display on tunerpro it shows maxed out on 50* at random times. I have installed new coil, new spark control module ac delco, another new coil, complete new dui distributor, upgraded from 6395 to 7427 PCM and still no change in weird SA glitch. I will be trying anything and everything this weekend to try and fix the glitch starting with changing the memcal. I'm looking for any advice in advance on what else could cause this so I can just go down the list and try everyting.

Thanks in advance
Jimmie

[dave w]

Possibly a similar gearhead-efi thread: http://www.gearhead-efi.com/Fuel-Injection/showthread.php?7226-Weird-OD-Spark-Advanced-FIXED-sorta

dave w

[Jimmie]

I would say this has to be close to the same issue after seeing the 23037 or so SA in your old data log, that's too close to be coincidence. So where do we go from here? You already tried setting the altitude spark bias to zero in the last .bin with no change in SA glitch. Only difference is I'm not running a holley TB but it is machined out with larger butterflies. Should we disable spark retard as a test to see if it corrects the issue? Also noticed that the SA issue only happens below 1000 RPM.

[steveo]

it's an issue with the ADX for sure. post the adx you're using lets see how messed up it is

[dave w]

Should be the A217 $0D TP5 v250.adx from this link: http://www.gearhead-efi.com/Fuel-Injection/showthread.php?303-16197427-PCM-Information-0D

Maybe compare the .adx from the link above with one being used by Jimmie, when it gets posted?

It's odd what Jimmie is experiencing, the .adx in the link above, has been used countless times without the weird SA.

dave w

[Jimmie]

Steveo, the ADX Dave posted is the one I'm using and I have not modified it. I can post it from my computer when I get home. Maybe I could re-download it, possible corrupt file?

[steveo]

i kinda half expected to see some kind of conversion problem but that part looks okay.

one odd thing (seen it before with GM) is that there's no need for this variable to be 16 bit at all, even the 8 bit output would give you 90 degrees of potential spark advance. no idea why they do stuff like that.

[Jimmie]

This is the ADX and XDF I am currently using. This is a rare glitch from what I can find and since I've replaced PCM and entire ign system, memcal tomorrow, this has to be fixable. Basically will not be the same system I started with so the glitch has to disappear somehow rite?

[Jimmie]

Here is a screenshot from tunerpro dash showing maxed at 50* SA but on the datalog I don't see the SA to match that high?? Tried to upload csv file but wouldnt let me.

[dave w]

The challenging question is there a problem with the ignition system or the data logging system.

Considering the entire population of $OD vehicles without data log information there are likely other $OD vehicles out there with the unusual SA that nobody knows about?

Considering the few $OD vehicles with data log information, it is statistically possible to capture data logs from some of the $OD vehicles with the unusual SA.

Could there be some original and / or updated factor $OD .bin files that have altitude spark bias set to 0 degrees? Most $OD .bin files have altitude spark bias parameters set to 9.8 degrees.

The overwhelming question is ... why does the altitude spark bias "fix" or have any affect on the the unusual SA?

Maybe there is an undated factory .bin file with altitude spark bias set to 0 degrees for your vehicle? If there is an updated .bin file available, it would be very interesting to see if there are any other changes besides altitude spark bias, especially the "undefined" parameters.

It would be a time consuming task to figure out what $OD .bin files were updated by the factory to with altitude spark bias 0 degrees. Possibly some research of old GM service bulletins?

dave w

[Fast355]

The challenging question is there a problem with the ignition system or the data logging system.

Considering the entire population of $OD vehicles without data log information there are likely other $OD vehicles out there with the unusual SA that nobody knows about?

Considering the few $OD vehicles with data log information, it is statistically possible to capture data logs from some of the $OD vehicles with the unusual SA.

Could there be some original and / or updated factor $OD .bin files that have altitude spark bias set to 0 degrees? Most $OD .bin files have altitude spark bias parameters set to 9.8 degrees.

The overwhelming question is ... why does the altitude spark bias "fix" or have any affect on the the unusual SA?

Maybe there is an undated factory .bin file with altitude spark bias set to 0 degrees for your vehicle? If there is an updated .bin file available, it would be very interesting to see if there are any other changes besides altitude spark bias, especially the "undefined" parameters.

It would be a time consuming task to figure out what $OD .bin files were updated by the factory to with altitude spark bias 0 degrees. Possibly some research of old GM service bulletins?

dave w

Zero'ing the bias increases the timing value. The only reason people see this glitch is because the ADX has a glitch and does weird things when the spark advance value goes negative. My suggestion would be put the bias back stock and ignore those glitchy readings or fix the ADX.

[1project2many]

one odd thing (seen it before with GM) is that there's no need for this variable to be 16 bit at all, even the 8 bit output would give you 90 degrees of potential spark advance. no idea why they do stuff like that.

But there is! If spark advance is less than zero you need all 16 bits to convey it. OP, what is your base .bin file?


My suggestion would be put the bias back stock and ignore those glitchy readings or fix the ADX.

Great idea. OP, please try this slightly modified ADX and report the results.

[dhworkin]

I noticed in your log16 that your at 50% throttle and not accelerating. I had the 23,000 SA and the throttle response problem. 121 log is with the 23,000 glitch that I had 14917. I raised the main spark tables to a higher value at lower RPMs and changed AE MAP and TPS and it helped, but it didn't stop the problem until I set altitude spark bias to 0. If setting it to zero gives it more timing, did it just get more timing that it needed and masked the problem? Here is the BIN and ADX I'm using. hope this will help. 14918.14919. this is a datalog after changing altitude spark bias14920

[Fast355]

I noticed in your log16 that your at 50% throttle and not accelerating. I had the 23,000 SA and the throttle response problem. 121 log is with the 23,000 glitch that I had 14917. I raised the main spark tables to a higher value at lower RPMs and changed AE MAP and TPS and it helped, but it didn't stop the problem until I set altitude spark bias to 0. If setting it to zero gives it more timing, did it just get more timing that it needed and masked the problem? Here is the BIN and ADX I'm using. hope this will help. 14918.14919. this is a datalog after changing altitude spark bias14920

I will say that is likely all you did. You added timing. I have been tuning these for years and never played with the altitude spark in any way. A positive bias allows for a negative value in the table.

[1project2many]

Here is the BIN and ADX I'm using. hope this will help. BJYK.bin.A217_0D_AEMx.adx.

This adx also limits spark to positive values from 0 to 50 degrees. It also attempts to interpret spark advance in the datastream as unsigned values. The .bin has a minimum advance of -9.8 which means the adx cannot display the minimum spark correctly. Try editing the .adx by selecting "signed" for spark advance and by lowering the bottom limit from 0 degrees to -10.

[Jimmie]

But there is! If spark advance is less than zero you need all 16 bits to convey it. OP, what is your base .bin file?



Great idea. OP, please try this slightly modified ADX and report the results.

Thanks, I will try this ADX and report the results back as soon as I get back home.

[Jimmie]

Looks like this took care of the weird SA glitch! I was hoping this would take the odd stumble out of it but at least now it will be solid data to go on. Thanks everyone for the great help on this!!!:happy:

[Fast355]

Looks like this took care of the weird SA glitch! I was hoping this would take the odd stumble out of it but at least now it will be solid data to go on. Thanks everyone for the great help on this!!!:happy:

The odd stumble is probably fuel related. Vortec heads and TBI right?

[Jimmie]

Yes, vortec heads, TBI, flat top hyper dual vr, rpm air gap intake, bored out tb. It is a dog below 2000 and stumbles like the key is turned off, shows lean off the scale on wb02 then after it downshifts or reaches 2000 rpm it screams to life. I have actually killed the engine by flooring it, but nothing seems to help very much.

[Fast355]

Yes, vortec heads, TBI, flat top hyper dual vr, rpm air gap intake, bored out tb. It is a dog below 2000 and stumbles like the key is turned off, shows lean off the scale on wb02 then after it downshifts or reaches 2000 rpm it screams to life. I have actually killed the engine by flooring it, but nothing seems to help very much.

Email me the injector size, fuel pressure, current base timing and the bin file and a datalog .xdl file running on the current tune as well.

When I get back into town this evening I will look at it. I have a few ideas for it.

[dave w]

Email me the injector size, fuel pressure, current base timing and the bin file and a datalog .xdl file running on the current tune as well.

When I get back into town this evening I will look at it. I have a few ideas for it.

Here's the latest information email be me: see attached.

dave w

[dave w]

Here's the latest information email be me: see attached.

dave w

Data logs using updated .adx with altitude SA Zero and altitude SA 9.8: see attached

dave w

[Fast355]

Data logs using updated .adx with altitude SA Zero and altitude SA 9.8: see attached

dave w

From first appearance looks to have way too much spark advance.

I will look at the actual bin when I get home.

My guess is that it does not have enough fuel in that area in the main fuel map. No amount of AE will permenately mask too little fuel.

[dave w]

BLM & SA updates:

[Fast355]

BLM & SA updates:

SA is looking much better overall.

Under 2,000 rpm still looks pretty lean in the upper KPA readings.

From what I remember the OP pulled the higher pressure spring out and put a stock spring back in giving him 13 psi.

I am tuning on a stock Vortec 350 with a performer rpm intake and cast iron exhaust manifolds that is running at 15 psi. I had to drop the injector flow constant to 50 lbs/hr even though the injectors calculate out to 68 lb/hr just to get enough fuel to the engine in the 1,600-2,000 rpm range. Not my prefered way of doing things. If the injectors were flowing enough fuel for WOT the flow rate constant could be programmed closer to actual rating. I would expect even a mild 350 Vortec to require about 22-26 psi with 350 injectors.

I had the VE table maxed out and was still seeing +20% fuel addition from the BLM and more still being called for from the Integrator value. I cut the injector flow rate in the PCM by 30%. The last log had fuel trims in the 105-118 range but I had 115 readings where the VE table had previously been maxed out.

Problem is just judging by the duty cycle at 3,000 rpm the injectors will be static before 4,000 rpm.

[dhworkin]

It is a glitch when the timing drops below 0 degrees. But what is causing the timing to drop into the negative? I can load a bin file without the WBO2 perameter changes , and there is no negative SA. I take the same bin file and make the WBO2 changes and the timing goes negative and it’s popping out the throttle body. I believe Jimmie, Mitch and myself are having the same problem.

[donf]

It is a glitch when the timing drops below 0 degrees. But what is causing the timing to drop into the negative? I can load a bin file without the WBO2 perameter changes , and there is no negative SA. I take the same bin file and make the WBO2 changes and the timing goes negative and it’s popping out the throttle body. I believe Jimmie, Mitch and myself are having the same problem.

It's a glitch in Tuner pro nothing else. You are confusing data logging errors with something else. I spent a long time figuring this out in my van thread. The same untouched bin using an OTC scan tool instead of Tuner Pro will log a negative number but not the crazy number. It would be physically impossible to retard the timing 26000 degrees. Remember GM never released its source code to the public. All this has been reverse engineered for free. There are mistakes. I do not doubt you have a popping problem, but you are looking at a distraction, not the cause. You seem to have isolated it in your case. Something is wrong with your wideband o2. Either in the bin or the way you have it hooked up. I would start by removing the wide band O2 completely and re attaching direct to the battery to eliminate any feedback problems you may be introducing with faulty connections or the way you chose to install. A lot of us have used the Wideband feature in tuner pro with good results. Since you are having major problems look there first.

[dave w]

I've been curious to understand why adjusting Altitude Spark Bias to ZERO will stop the negative numbers? Possibly source code that has not been reversed engineered?

dave w

[donf]

I've been curious to understand why adjusting Altitude Spark Bias to ZERO will stop the negative numbers? Possibly source code that has not been reversed engineered?

dave w
Because doing that just bumps the actual timing everywhere by the 9.8 degrees. What I did to figure all this out is set the whole table to a fixed number in tuner pro then measure actual timing with a timing light, while changing different stuff.