12P disassembly

[immortality]

I doubt Vl400 is going to release the info to you or re-write the code to just suit you unfortunately.

[34blazer]

Understood. Like I said, last ditch effort. Someone is helping me but having a commented disassembly would certainly speed things up a bit.

[antus]

Ive had a look at the code (its not mine to share) but there is no interpolation skip between the table resolution change. Rather, the code checks the map value first to see if its before or after the resolution change and then it calcs which table column its referencing (not a whole number so it includes interpolation). Then it uses the 3d lookup (which includes the data interpolation) referencing the whole table in one hit by column and row. Its not treated as two separate but side by side tables with an interpolation gap.

This glitch you describe has not come up before, and im not saying its not happening, but I do think its for a different reason.

Agree with the other comments on spark cut not being possible with software alone, other than cutting everything for an engine rotation which is a lot nastier than dropping individual cylinders like the 11P code can do with the hardware platform difference. I took a look at the linked thread to see what you guys are doing on that side of the world but I can only see stock decel fuel cut mentioned.

[34blazer]

The fuel delivery is inconsistent between 55kpa and 56kpa, right about the same spot the transition occurs between 10kpa to 5kpa increments.

In the link that I posted start looking at post #257.

[VL400]

Antus alerted me to this thread, also have not seen this issue before so would like to check what is happening. Have you got a log showing the inconsistent fueling at that load point?

It's been a really long time, but from memory when I tried to get spark cut working on the '808/'165 I found that changing the dwell time via software was not instant and it started to do some really odd things. Curious if you have you looked at what dwell does on a scope with the code mod you mentioned? I couldn't find the dwell patch code in your link.

[34blazer]

I took some screenshots of the code, but the gist of it was he had the EST signal held high and it caused the coil to misfire.

Yes I can get a fresh log for you to see. In the process of resetting the VE table since I switched back the double fire mode.

[34blazer]

...

[antus]

Yeah so thats the approach which we dont think the hardware allows to work like you would think from reading the code comments. I think we need to see a trace of the signal coming out of the ecm when running that code on the bench. Were talking years old investigation now and it looks like nobody saved the data, but we believe setting those registers to 0 didnt cut the spark and behaved abnormally. Have you tested it?

[34blazer]

I do not have the know-how to write that to the code. The person that wrote that patch had it working on his TT Camaro, and a few others have used it. Originally BL85c wrote it to be used in 6E, which is used with a 165 ECM. A friend of mine was looking into writing that patch into 12P but it's a slow process since he has no notes to use.

[34blazer]

I'm willing to test, though.