Well I finally got my TPI in my car many hours of sweat swearing and skin removal

now no more carb


any how I hooked up a computer to it today craig's software says "no link"

but it is passing bytes by F0h and 00h it doesnt look like there are any other ones

any ideas what is going wrong?


I have a stock 88 TPI off of a 305 on my 1986 350. using a 165 ecm with MAF, 24# LT1 injectors

stock 88 305 5spd bin with the injectors constants set to 24# and the rev lim upped

well there is the details now I need to get this software working so I can start playing


------------------
Objects in mirror are more stable than they appear
www.members.home.com/v8power
4BBL to TPI in the works
86 350 5spd

E.E.T (Electronics Engineering Technologist)

 

I'd heard of folks having problems with the 86 ECM's, any comments from anyone on this? Is the 86 stuff a problem? I want to know the deal! You ought to get '00' and 'FC'... Check for the 10K resistor across A&B...

 

my bad


I was shorting the connectors not using a 10k


one stupid thing a day isnt bad


but i am over my limit for this year

------------------
Objects in mirror are more stable than they appear
www.members.home.com/v8power
4BBL to TPI in the works
86 350 5spd

E.E.T (Electronics Engineering Technologist)

 

So to clarify, is the '86 ECM working now with the A-B correctly 10K'd? A validation of an 86 ECM workinbg properly would be good to have, one way or another.

 

Yes it is now working with the 10K but it was not when using a wire.

good job on th software it looks like it will be an invaluable tool once I learn how to use it

 

I'm using your prog on an '86 Corvette w/1227165; no problems. AKM interface cable; 486 laptop.

Excellent prog!...Looking forward to the next revision!...

 

after reading that again were you asking if my 1986 car was working? I own an 86 but the TPI is out of an '88

 

A validation of an 86 ECM workinbg properly would be good to have, one way or another.

Well, yes and no. My home grown MAX233 based I/F isn't converting the ECM data to RS232 properly, seems to be a MAX233 problem, so a new chip is on order.

I do know that data comes out of pin E of the ALDL at 160 baud, 5v (I scoped it to verify bit timing & level), so capture shouldn't be a problem as long as QBSER can handle that baud rate.

[What I say from here on is from information presented in the disassembled bin file, ie, not yet verified on the car]

Upon entering 10k mode, the ECM should begin broadcasting a 25 item diagnostic list unidirectional at 160 baud. On completion of the list, the ECM should listen for 50 msec for an inbound 8192 polling message. I haven't yet figured out what this polling message looks like. If it is accepted, though, 8192 baud communications continues until shutdown, and bidirectional communication becomes possible. This is where I'm trying to eventually go, since the Mode 2 table stuff is more interesting. Bidir should also allow attaining the holy grail - Mode 4 stuff.

I don't see any baud shifting in your source code, Craig, so I assume that the program is capturing the 160 baud table info. If so there is a *lot* more this ECM can do. I should be to the point where I can verify all this before too long. If I stumble across something, I'll pass it on. Until then take everything I just wrote with a grain of salt...

Jim
Red 86 4+3

 

You're right on in many respects.

My program doesn't dork around with the 160 stuff right now, goes straight to sending out the 8192 polling sequence, followed by acquisition at said rate. The polling sequence is sent, and if luck has it, the timing will mesh and the thing will beep and take off at 8192. Otherwise, it will continue to recycle trying to listen to the 160 at 8192 (meaningless).

Question is, are there problems getting it to 'kick off' with an '86 ECM? Is there some different polling sequence required or something? Is '86 somehow not capable of 8192? I don't know. I know it works great for '87 and such, that's all.

I agree about the Mode-4, that will be great. No one with working knowledge on the protocol has stepped up to the plate for discussion, though, and my MC68HC11 assembly skills are somewhat untouched. Anyone? Don't know how many times I've asked this. If someone steps forward with working knowledge of the Mode-4 communications, I will be working on an auto-tuning feature set for the '165 at LEAST.

Also, what's the story on '86 data ALDL pin locations? On the '87, 160 and 8192 are the same (M for me), bimodal, 160 default, 8192 after kickoff request. So question put forth: What's the 8192 and 160 pins for you, and what are your years?

The Mode-2 is OK, but I've been using Mode-3 (more random-access) to pull the EPROM image down. The Mode-4 is about all that still holds allure. That and successful 160baud decode pulldown though standard non-CTS serial comm.

I have been running the 233 without issue for a while, ran the 232 also without a problem. Haven't tried the 'simple' circuit. Found common/good grounds to be important. Found extended EPROM ZIF to be problematic.

-Craig



[This message has been edited by Craig Moates (edited September 17, 2001).]