3.1 Turbo/Tuner pro HELP code59

At my wits end with this one. I've been using tunerpro for a couple of years for minor benign mods, while I never claim to be a professional. Regardless, before this current build, I used tuner pro on my Beretta using the stock code, the correct xdf and the correct adx. I did this all the time, and through minor modifications it ran great. I would just alter the bin, upload it to the emulator to get my fans coming off and on at the right times, etc. Then put the bin on my stock chip that I had thrown an atmel into, and moved on. And I'll lead with, to inform the reader, i'm not a complete retard at this, as a matter of fact I was messing around with my buddies trucks quite a bit last summer.

Fast forward.

for what it's worth, it's a 1989 Corsica hatchback. Five speed swap, LG5 turbo. LS springs, roller rockers and it's got a three-bar map in it. '730.

One day, I decided that I wanted to go out there and mess with the tune a little bit, I was still in the early stages. The dog chewed up my extension cord (bad LT batt.), So I took my prom into the house, plugged it into the mini programmer, pulled the bin, did the modification, put the bin back on the prom, went outside and stuck it in. the fans come on as if you're in diagnostic mode, and the check engine light flickers. No start. All I did was change fan temps because I put a 160 thermostat in. So now I'm confused, I go back in the house and mess with it some more. After a week or more of research, people are informing me through the posts that I read, to make sure the checksum is disabled, or set at AA. It was not set at AA, however I'm going to point out that I've done this process a million times it seems, and I've never touched the checksum or any of its parameters. I've never had a problem, and I'll defend that fact to my grave. But all the forums I'm reading are saying that that's impossible and I should have been disabling it or it would not work. I disable it. At this point I'm back outside with the emulator and a new extension cord, trying to figure this out. No check engine light at all/no run (but no fans) when it's set at AA. Fine, I'll throw in a two-bar map, emulate 8f, and see what happens. Same thing, fans come on, have to disable the checksum, wash/rinse/repeat. I have unaltered bins saved, same thing. Every single one. it doesn't matter whether I physically burn it onto a prom, it doesn't matter if I'm trying to emulate. You can set aa, and the fans won't come on and the light won't flicker but the car still does not start.

Fine, let's throw in a one bar, just to get it to idle, make sure that I'm not completely stupid, I disconnected the turbo. Same thing, correct xdf, correct bin, correct adx. Load her up in the emulator, upload it and of course wouldn't you know it. Flickering check engine light with fans. What is going on. fine even though I know it's not going to work I'll go disable the checksum, well maybe it's my xdf, but there is no checksum table anywhere on the OEM bin. I'm using tuner pro 5, ostrich emulator, and whenever I need to burn hard copies I'm using the mini pro programmer. My car is completely disabled, literally nothing I do will make it run.

Well wait, maybe it's your prom. Maybe it's the ECM!

I have spare parts, 6 ecms that I know are good, same problem. Also multiple compatible chips with Atmels. No go. Here's the funny part, I load all this up and with the key on I get acquisition. Every single sensor seems to be working correctly, temperature seems to be accurate. So it's not like I'm not able to communicate through acquisition. I've read a few things about settings in tuner pro, that I cannot find. And now I'm in this loop where no bins are working, don't know what to do. I know someone's been through this before, tell me what I'm doing wrong!

Burn the stock bin file, if it still does that then look into the eprom programmer settings or the programmer itself might be bad. However if all is fine with the stock file then you might have a checksum problem.

I know this might sound dumb, but I've done this myself...

Are you sure you have the ribbon cable installed correctly in the MEMCAL adapter?

About checksum. TP automatically updates checksum when you save a bin, so if you're burning EEPROMs, disabling Checksum is not needed. If you're uploading whole bins to an emulator, sometimes you can get away without checksum being disabled, however, if you're truly emulating, where only singular changes are being made checksum absolutely has to be disabled. That being said I have had a couple instances where it wasn't disabled, and emulating SEEMED to work, but wasn't really working properly.

Installing an unmodified MEMCAL, does the engine start?

Quote:

---

Originally Posted by Six_Shooter

I know this might sound dumb, but I've done this myself...

Are you sure you have the ribbon cable installed correctly in the MEMCAL adapter?

About checksum. TP automatically updates checksum when you save a bin, so if you're burning EEPROMs, disabling Checksum is not needed. If you're uploading whole bins to an emulator, sometimes you can get away without checksum being disabled, however, if you're truly emulating, where only singular changes are being made checksum absolutely has to be disabled. That being said I have had a couple instances where it wasn't disabled, and emulating SEEMED to work, but wasn't really working properly.

Installing an unmodified MEMCAL, does the engine start?

---

well, no, not exactly. The ICM fires it, when it's time for the .bin to take over, it chokes. This happens with checksum disabled as well. That's being said it has 36lb injectors on it, so the one time it started, it was rough and I'm 99% my limp mode for the 19lb'ers were running it. I didn't let it run long.

I double checked my cable, it seemed fine. I have a question. I've heard (read) about hitting the ecm with a bad checksum could brick it. Now I've tried multiple ecms, but they have all been hit with the same bin at one time or another, during this fiasco.

Is the checksum regenerated/updated when you make changes in emulation, i.e. like changes fueling etc?

No, bad checksum will not brink an ECM, at least not OBD1 ECMs.

No, checksum is only updated when the bin is saved. This is why checksum needs to be disabled during emulation.

So are you saying that it runs on base timing, but not when timing is being commanded by the ECM?

Quote:

---

Originally Posted by Six_Shooter

No, bad checksum will not brink an ECM, at least not OBD1 ECMs.

No, checksum is only updated when the bin is saved. This is why checksum needs to be disabled during emulation.

So are you saying that it runs on base timing, but not when timing is being commanded by the ECM?

---

Correct. that's what I believe is occuring. Where I'm getting confused is this happens regardless of what bin you use. Even a factory unmodified bin that never hits tunerpro, just straight to the prom, gets you a flickering light.

Same with emulation. The one exception; in beginning an emulation session, disabling checksum gets you no CEL, but what I perceive as the base timing only issue. With the flickering light, you don't even get the stumble.

I've got WB02 wired in, otherwise a fairly solid/stock wiring harness throughout the car. I would blame my chip burner, seeing as how you get the issue bypassing TP, but I can't get through with emulation either.

Possible electrical? what would (if anything) invoke this ECM behavior if it electrical and not in the bin/hardware?

First test, and best. Key off for 30 seconds or more. Then turn key on but do not start car. CEL shoould light for 5-10 seconds, turn off for several seconds, then turn on and stay on. If this does not happen then you shouldn't bother trying to start the engine. This test indicates the PCM is able to execute code, the code is working, and the checksum is not incorrect (or engineering mode is active). If the light behaves as expected you should stop chasing checksum/bin/ecm issues.

Quote:

---

for what it's worth, it's a 1989 Corsica hatchback. Five speed swap, LG5 turbo. LS springs, roller rockers and it's got a three-bar map in it. '730.

---

For my own curiousity, did you modify anything in code for the 3 BAR MAP? Using 3 BAR MAP with an unmodified 2 BAR calibration can produce bad results such is vastly incorrect timing and fuel values for the engine.

Apologize for the gap in time, I've been away for the Easter holiday..

I've gotten back at it, got the noid light set out. I get one flash of the noid light whenever you first turn the key on and crank. Injectors then have no pulse. I unplugged the injector harness at the back of the upper intake, and ran the noid light pre-injectors via a little engineering. Same result, one quick flash, nothing else there after. This points to a wiring issue, no?

I'm going to get out there again first thing in the a.m., start checking the harness over, with the multimeter. Its a healthy harness, and I see no breaks or shorts. I guess I'm not convinced it's still not in the .bin I'm running

Quote:

---

Originally Posted by 1project2many

First test, and best. Key off for 30 seconds or more. Then turn key on but do not start car. CEL shoould light for 5-10 seconds, turn off for several seconds, then turn on and stay on. If this does not happen then you shouldn't bother trying to start the engine. This test indicates the PCM is able to execute code, the code is working, and the checksum is not incorrect (or engineering mode is active). If the light behaves as expected you should stop chasing checksum/bin/ecm issues.



For my own curiousity, did you modify anything in code for the 3 BAR MAP? Using 3 BAR MAP with an unmodified 2 BAR calibration can produce bad results such is vastly incorrect timing and fuel values for the engine.

---

I would put the appropriate map sensor with the appropriate tune. My connector is modified to fit all maps.

Update. Strong voltage at the injectors, no short to grounds I can find.

Also, I discovered when you crank it for extended periods of time, every 5 or 6 seconds you'll get one injector pulse, this is straight from the harness, injectors disconnected. When you test it through the injector connectors themselves, same result. They're just not getting decent signal from the ECM.

Before I throw the ICM and CPS under the bus, not only are they both brand new, well fairly new, but I replaced the ICM with a known good one with the same results. And I'm getting RPM readings on the dash and through tunerpro, I'm just not getting injector signal.

On that note:

These past two days I've noticed something new, I simply cannot get my data stream working correctly. It connects and then disconnects endlessly, but when it connects you get a few seconds of data that seems to be correct. I've triple checked my tunerpro settings, my plug, connections in USB locations. It starts with an error 1 on tuner pro and will literally climb into the thousands cycling in and out of data.

Im still entertaining the idea of a bad part or wiring issue, but I am just convinced it's got to be something in the tune I cannot find any other issue, and I'm running out of things to test

Have you tried disconnecting the TPS on an engine start attempt?

Quote:

---

Originally Posted by Six_Shooter

Have you tried disconnecting the TPS on an engine start attempt?

---

Yes, and there was no change.

However, after working through my computer and deleting a wireless printer that was taking up all of my comm slots, (emulator and the extreme ALDL we're running on comm 17 and 18 on the laptop), I have recovered all data acquisition, and the emulator is working great from what I can tell.

A few quick mentions before I dive in again, I have a personal salvage yard. I went ahead and threw in three different coils that were known to be good and tested good, a different ICM, which would be the third one, but I have that kind of luck. and it's here. I wouldn't purchase these things to throw parts at it. I just have plenty of everything on hand.

I feel like I may have lost a few of the followers, so let's recap.

I have covered all hardware, twice over on this car. Healthy wire harness, ohm meter shows no shorts on any of the ECM wires (I've found), I'm so in-depth in this issue that I've checked most to make sure that it wasn't being shorted ANYWHERE. All six injectors ohm out perfectly, 36lb for reference. At the beginning of this issue I threw in a brand new crank sensor that I had on hand and never used. As well as an extra crank sensor wire that I had stored, ran it under the oil pan and just bypass the one that is looped through the valley. I've covered all of the wiring and hardware, and parts have been interchanged to no prevail, again because they're here.

So regardless of the ICM I have installed, whenever I have the noid light set hooked up, I will get a bright flash initially, sometimes the engine will stumble and try to start, but the recurring flashes are dim and intermittent. After doing a test for spark, currently the prevailing issue is I have spark but it is intermittent.

Ex; You'll see spark, and then for a revolution you won't see a few, and then for three revolutions or more it will be perfect. And then you'll lose spark for six revolutions and then it'll come back for two, you get the point.

Intermittent spark loss. Intermittent injector pulses. I don't want to be the person that dismisses all suggestions, but I've been through four known good ecms, a few different proms that are eligible to run my setup, with flashable chips installed. Matching map sensors to each mask, one bar all the way up to three. This issue prevails regardless. I'm becoming more and more confident in the vehicle should run with the tune setup, everything seems to be set right with timing. In which may force me to look elsewhere for assistance.

I have currently stopped experimenting with other masks, I'm on 59, 3 bar map. I think I've proven that this issue is going to prevail regardless what I do, tune wise.

I understand that the readers cannot be here to observe the vehicle, but I can almost bet a million that I've tried most suggestions. And I've built this car from the ground up, so me reaching out to people who don't know the vehicle shows how desperate I am. So I guess at this point if this forum will still benefit this issue in any way, it'll be someone having a similar issue in finding a short or something somewhere that is easily overlooked, that I've missed. I suppose it is possible that it's something I'm uploading inaccurately into the emulator, but I've even been on the phone with moates support. So I don't know, if anyone else has any more to add to the thread I'd be honored to read it. And as soon as I fix the problem, I'll update this post. Thx

Adam

Also, I would like to add, I recently installed wide band. SM-AFR DM-6. I have done my research, and I'm confident in where I pinned it in. But I would like someone to verify, the correct pin for the 7730. Regardless of my research.

The reason being is my narrow band reads back the voltage, with key on. I understand it's not going to be accurate until it warms up a little bit, but I get a response. Whenever I'm looking on tunerpro, under WBo2, it reads as following:

10.00 Actual AFR
4.80 Desired AFR

Desired will bounce around when I'm cranking, but I never get a change on actual. Am I waiting for this one to heat up to respond or do I have it pinned incorrectly?

Thank you

Just learning myself.

I have my WB on Pin F14, running S-AUJP. Also A3 can be used, option select in tune at least on the S-AUJP. I will be switching over to Code 59 when adding turbos.

So now I've got the dash pulled slightly out, as well as every single piece of conduit off the harness, it all looks healthy and beautiful and I've moved every square inch of it, it's all been wiggled and hand gone through. The short I thought I had, was really weak, I chased it from the ICM, up to where it meets the injectors, back to the ecm, back out across the engine bay and all the way to the fuse box, ecm fuse number one.

The back of one shoots up with a pink and white wire that goes to the cluster, as soon as I unplug the cluster I lose that faint ground. So I said awesome, I figured it out. left the cluster unplugged, same thing. It's got to just be a parasitic ground it's picking up through a circuit in the cluster, and I bet if I had another L body here it'd probably test the same.

Other than that, I can honestly say I've been through every wire behind the dash and under the hood. Tried another ECM today, no luck. Everything looks good on data acquisition, all sensors are responding. But even a test at the harness where the injectors plug in, one bright pulse followed by a few faint pulses and then all pulse is gone.

Spark is hit and miss, this has to be in the tune. I'm just inches away from blaming my emulator, as if it's sending trash to the ecm. I mean there is just simply nothing else I can think of otherwise I would try it in a heartbeat!

I never really believed anything was wrong with the harness until I found that parasitic ground, but no dice.

I've been into L bodies for almost a decade, 60°v6 motors for two decades, and I've been working on cars in general just as long. I can count on one hand how many times I asked for help, how many times I just could not figure it out. The situation has got to top every memory I have of a difficult repair. I'm about to have to tap out for a little while I just don't know what else to do, I've done it.

Thanks guys

The ECM is not even in the mix for spark while cranking, so you can rule out the ECM being the issue there.

The ignition will be base timing triggered directly from the ICM and crank sensor/dizzy reluctor. Once the engine goes above 400 RPM, THEN the ECM will take over spark timing.

You don't even need the ECM plugged into the harness and you will have spark in a properly running system.

If you have inconsistent spark you will have inconsistent fuel delivery.

Figure out your spark issue, then the fuel will probably follow in line.