Tuning available for 93-95 L32 and L82 PCMs

[badsam]

Sounds pretty cool...

Will this new tuning software work with OBD1?

[robertisaar]

one of the great (former) mysteries of the E-side...

it's actually a table used to filter and calculate the EGR flow due to solenoid combination changes.

anyways, i post this since i have roughly 98% of the E-side completely documented, the 2% that remains is either too insigificant to matter or is just too convoluted to understand. so not quite nAst1 completeness, but pretty close. i just wanted to post a pic as well to give everyone something to get confused over.

the T-side... less so. to me, the 4L60E control scheme is fairly difficult to understand, but most/all of the 4T60E derived stuff is easily recognized.

[SappySE107]

We gotta get you on some bmw computers.

[robertisaar]

i had someone try to contract me out to work on some BMW application... can't remember what it was, but it used an Intel processor that i wasn't familiar with(want to say the 8061?), anyways, from what was being implied, big money in that market.

[robertisaar]

oh, and now that i think about it, i should probably mention a few odd things i've come across....

if you watch your target AFR and it hasn't filtered up to your stoich ratio, there are situations where the INT will start moving around before you even get to a target AFR of 14.7... so while the base commanded AFR can be ~12.9:1, because of INT action, you can actually be around 14.7:1 due to INT dropping like a rock when it shouldn't be anything other than 128.

the second one is that the rich/lean DTCs can get set when in "closed loop", but with no INT movement and the O2 voltage being outside of the acceptable range.

anyways, i had both of those issues come up and had to develop patches around them. what a PITA. then i made another patch to only apply highway spark when highway fuel was active.... that one was more of a preference than anything. thanks GM, i love having to fix your mistakes.

[robertisaar]

the more i mess with this, the more i want to just rip all of the relevant info and just port the stuff to nAst1.

one particularly infuriating issue is shifting 3-4 with the TCC locked... i can either set the TCC relock delay to 0.0 seconds, keeping the TCC locked through the shift(works great with the 4T60 and older 4T60Es) and have it be rough as hell since i can't seem to get the TCC PWM to reset down to anything other than 100%, or set to 0.1 and higher and have the RPMs actually jump upwards through the shift. and with a really short relock delay, it feels bad, like slipping 4th clutches(until the TCC PWM starts getting high enough again), which is already a weak enough point on a factory 4T60E with over 200,000 miles.

the factory got around this issue by shifting into 4th long before TCC lock is allowed in 3rd gear unless you have the shifter blocking out 4th gear. same is done with the 4L60E calibrations.

and guess how much free space is on the T-side? a single 28 byte subroutine that isn't implimented in the algorithm(out of the entire 32,768 byte PROM). i already had to use 12 bytes of it to correct the transmission temp linearization subroutine. so i have 16 bytes to fix the issue..... and so far, i've gotten nothing working in that space.

of course, i already have this issue worked out in nAst1 on the bench........



on the E-side, i've had to patch quite a bit to fix factory flaws as well... the stuff i have documented includes DTCs 44, 45, 64 and 65 being able to be set when the PCM is in "closed loop" fueling, yet the INT isn't moving. don't ask how that flew past GM's code engineers back in the day, but it did. another one: the INT will start moving around (indicating closed loop fueling) WITHOUT the target AFR being stoich.... or even necessarily anywhere close to stoich. then one that's a preference for me... highway spark. originally could be added in without highway fuel being active. changed that and now it works as i want it.




anyways, that's my rant of the moment. i really need to finish up the T side of the XDF and get a 100% understanding of the transient fueling stuff... there seems to be 2 "levels" of it? anyways, that portion is one of the few significantly different sections of code compared to A1 on the E side.

[xLeGeNDx]

Interesting work, I would love to see something for the 94-95 DOHC without having to go full OBD II or OBD I

Swyped from my HTC One X using Tapatalk 2

[robertisaar]

94-95 LQ1 = 1FA(early calibrations) or 2E..... both of which have defintions out there for logging and tuning.

and it is OBD1, there is nothing OBD2 about that PCM.

[94Camaro]

soo..curious. this ever get sorted out or is it still in progress? Would be nice to be able to adjust what came in the car rather than make something else work (and would love to retain the full function of the engine as it came from the factory).

[95cutlassproject]

For the fwd i know for sure, i am not 100% on the rwd stuff yet. I have been doing tuning on and off with my 95 cutlass for about a year now.

[TechRod]

I have been driving my 3.4L/4L60E powered 81 Chevette (from a 94 Camaro) for 8 months on Madtuner's tune using the "OBD 1.5" PCM and I can say that I'm rather happy with the tune and shift points that he came up with.

[robertisaar]

the engine side is more or less 100% decoded.
the transmission side is less so, but still highly understood.

the same definition will work on FWD or RWD applications. keep in mind that there are actually 4 versions of the BINs out there though, i've only focused on version 4 since not only is it the latest revision, but it has most, if not all applications covered.

[3400-95-Modified]

Anyone crack a 94-95 Beretta bin? And from what I remember 94-95 Camaro stuff has been cracked for a while now... or at least I remember another member modifying his 95 F-body using OBDI hardware to "flash" it at Bfest 2010 in Chicago.

[robertisaar]

this works on all 93-95 L/N/W 3100 and F 3.4 cars.

madtuner also added in the capability to flash the 94-95 L99/LT1 stuff as well.

[3400-95-Modified]

well well well... I posted this to Bnet... I was completely unaware of this finding...