16197429 Documentation Project

**RobertISaar** · 08-25-2013

i took a few shots using the wife's T3(12.2MP DSLR) that way i could start labeling stuff as i discovered it's usage. i do need to take more though, especially at least one on the backside. there are a lot of tiny SMT resistors on the back.

i don't think i'll be able to get this down to the Ludis level of completion, but it will be more than what is known about it already.

that calibration..... happen to know if it was built off of the 93, 94 or 95 codebase? IIRC, they were all slightly different compared to each other.

**RobertISaar** · 08-25-2013

in my search for info:

http://www.retro.co.za/efi/DIY_EFI/1999/efi4-206.txt

found that the 94-95 LT1 PCM uses some common ICs as well. i've determined that the 129170 chip seems to be the injector controller, the LT1 uses it as well. based on the date code, i'm assumg it's first application were the cadillac V8s that ran SFI.

the LT1 also uses a crystal that causes the processors to run at 3.15MHz instead of the normal 2.1MHz that the P4s and this P66 unit does.

**The Dark Side of Will** · 08-26-2013

Originally Posted by RobertISaar

that calibration..... happen to know if it was built off of the 93, 94 or 95 codebase? IIRC, they were all slightly different compared to each other.

No idea... The .bins were on the DIY-EFI FTP site as of several years ago. I still have them as well, but would probably take me some time to dig them up.

**RobertISaar** · 08-26-2013

hmm..... weird. happen to remember a filename?

**The Dark Side of Will** · 08-26-2013

Would have had "Northstar" and "A" in one filename and "Northstar" and "B" in the other... might also search for variants of 16197429.

**RobertISaar** · 08-26-2013

hmm..... there are a couple of northstar A and B BINs in the gearhead-efi archive, but i already have both of them and a few more.

**RobertISaar** · 08-26-2013

discovered that the pink book doesn't necessarily even mention certain aspects that are relevant in the F1.

found a supplimental addition for it.

http://cache.freescale.com/files/mic...=Documentation

explains locations if not specific functions.... should make certain things easier to play with.

and i've come to a conclusion about the 6811 family: they kind of need all of those extra, custom-made chips for anything beyond basic functionality. for example, to operate SFI on an 8 cylinder engine using only the processor, the "simplest" way would be to use 16 TOC channels. 2 channels per injector would be needed because you would need to set up "time until turn injector on" and "time until turn injector off". and..... there are not 16 TOC channels to use, there are 5, at most. so what GM did was have some custom-made chips built that essentially replicated at least part of those functions. with the P4, nearly everything is external to the actual processor. with the P6, some of it moved onboard, but a lot of the complicated functions are still "outsourced" to the other chips on the board.

i can't think of a way to perform SFI on that many cylinders without a whole bunch of non-existant TOCs to do the work. i guess GM ran into the same problem, so they had something made up that is basically a programmable timer that sniffs on the address and data bus to see when it is being written to. so now instead of eating up TOC channels that didn't exist, you just deal with it all over the existing address and data bus.

there is such incredible complexity behind all of this. when GM went OBD2 and largely used the 68332, even more of the functions made it onboard via the TPU. this is why they can do stuff like SFI and coil-per-cylinder ignitions with relative simplicity. it all stays within a single IC until it needs to turn on an injector or coil driver and does it directly. its too bad you have to have an octo-core brain to easily follow along that assembly.