2 bar tune for Turbo LT1 F-body possible with '93 ECM? DA3

[hotrodf1]

OK, so long story shorter, I guess the cap and rotor was the problem. Why it just failed to start all of a sudden, I'm not really sure. Still running the original GM Opti, but now it's the vented style. Cool. Also added a hotter coil for good measure (MSD 8226 replacement drop in). Was able to jump the spark further on my tester than with the stock coil, so it's actually measurably better. Good deal.

Anyhow, the patch that's in the code doesn't seem to work as intended. It doesn't seem to "set" the baro reading manually as we intended, and the ECM is still updating the baro reading later it seems.

There is a reason I thought it worked before. When I use the unpatched code, it will update the baro setting as usual - and come up with 54-55 kPa or so based on 2 bar MAP. Then I go and use the patched code it will stay that same reading for a while, but at some point just putting around it will update again. I don't know what triggers this, but it throws the tune way out. Like from idling at 13.9 or so to idling at 15.7. Bam, all at once.

TunerPro and Scan9495 both report the ECM sees 10.35kPa as the baro reading at that point. (This was what I was seeing when the car failed to start, which sent me down the path of thinking it was fuel related. THey just HAD to be connected right??? )

The car runs and drives even with the baro set at 10.35, which is surprising to me. But at least it can get home. It's actually not even that lean under boost when this happens, though not where i want to be. But using the patched code, the Baro does not update at key on, so it's kinda stuck at 10.35kPa.

I'm kinda thinking of trying the unpatched code a little more just to see what it does. Is it possible we fixed something that wasn't broke?

Or should I switch to unpatched code (to get back to 55kPa) , then go back to patched and try to see what circumstances are triggering the reset of baro reading? I think perhaps I can datalog for a bit, and should be able to see that in the log. Scan9495 seems to work better on my old laptop than TunerPro for some reason.

[steveo]

too bad that patch didn't work, if i have time i'll look and see what i missed

I'm kinda thinking of trying the unpatched code a little more just to see what it does. Is it possible we fixed something that wasn't broke?

there are always update conditions, and if you meet those conditions in boost somehow, it's going to reset to a false value. you definitely want to disable that routine

[hotrodf1]

Ok, it appears it's finding the right conditions during boost. I've seen it update to 65kPa, and then 69 kPa as well. That makes it like 10:1 rich at 4-5 psi. Still not at WOT. She's moving though. haha.

I figure I better take it easy before I do something I'm going to regret. I guess the next thing is to figure out how to get the baro update solved. I don't get why it updates at all different random times. Sometimes at 0% throttle and coasting down (not sure how else it could read 10.3kPa), and sometimes during boost.

[Xnke]

posts 18-20 explain why this was gonna be an issue, and how it was solved in the P66 code.

[hotrodf1]

Ok, am I reading this right?

A3CF : LDAA $00DA [load accum A] with value in MAP sensor input *****
A3D1 : CMPA $003F [compare A to memory] compare accum A (volts from MAP sensor) to stored value (current BARO value)
A3D3 : BLS $A3D7 [branch if lower or same] so if MAP sensor current reading is lower than the stored BARO reading, then disregard and skip to A3D7
A3D5 : STAA $003F [Store accumulator A into memory] So, if the MAP sensor reading is Higher than BARO, it stores this value to memory
A3D7 : LDAA $003F [Load accum A from memory] - now loads current BARO value to accumulator A

Can seemingly only make the stored baro reading higher. Course, this is one small snip of code.

[steveo]

yeah it basically does (pseudocode)

if(MAP > BARO) then BARO = MAP

i had patched that routine out already, to the best of my knowledge, though:

- @0xA3CF during the keep-alive (or other update) routine (which ensures map never exceeds baro), it compares map to baro, but doesn't actually store the value (i just changed STAA to LDAA, so it's loading the value again and continuing on)

[hotrodf1]

yeah it basically does (pseudocode)

if(MAP > BARO) then BARO = MAP

i had patched that routine out already, to the best of my knowledge, though:

If I turn the patch code off, the ECM then seems to initialize the BARO reading to current MAP reading before staring the engine. So it does do something, but somehow Baro still gets updated.

With patch on, the baro reading doesn't ever change (even if disconnecting the battery !?) unless it goes into the update routine. The "barometric equalizer" scalar in TunerPro doesn't appear to actually set that value in memory. Could the address somehow be off or something?

Sorry if all this is ignorant, I'm trying to kinda understand it.

[hotrodf1]

Have I found a spot that could be the issue?
Seems like we store a value to BARO memory location in this last line.

If you already disabled the other update routine, then this could be contributing?

Having a hard time understanding all this. . . .

OOPS, never mind, I see you already found what I saw.