Craig,
I have been using the 8.45 ecm software, and so far have had no problems at all, except one. I cannot download the prom. I can on the old E6, but the link wont stay active on 8.45.
I do now have a seperate on connector hookup beside the ALDL port, and am using screened cable. I have not had a data error in several hours of intensive use.
The software is very useful, although it took a while to sort through the pages.
Would it be possible at some time to add the IAC count (actual position) to page 4. Not greatly important, but could be useful.

Cheers for the great work

Richard

88 GTA, 350, flat tappet cam, ported plenum, A.I.R.delete, coolant bypass, hi flo cat, dynomax intermediate pipe, 80 flowmaster.

 

The IAC will be added to Page-4 for the next revision, no problem.

As for the EPROM dump, that's odd that it didn't work. One caveat I hadn't mentioned, here's how the EPROM dump 'should' be able to work:

1) Start program, set up defaults under option (8), and then go back to the main menu.
2) Go to one of the display options, like "(1)", establish connection, and make sure the data is reading OK.
3) Go back to the main menu, and select the dump option (what is it, (5)?)
4) Type in a vald DOS filename, and accept default addresses (for 165, need update for 730 defaults I think), and see if it rips.

I've had luck running it on my system, but that was during some interim work. Have to go back and see if I am now having a problem with the current release. You're using a 'stock' GM PROM code, modified or otherwise, right? Not Hypertech or something...?

Is anyone else having problems? Successes? It'd be helpful to benchmark this particular issue.

Thanks,
Craig

 

I've done as you described above, with v8.45; the first dump went on for about 20 seconds and stopped with a "timeout" error msg. I tried again and got a "full" dump. This is with a 7165 ECU.

Total time of 2nd try was much more than the first, so I would guess this "timeout" thing (is there a default that needs increasing?) needs work...

E6 just keeps on working!!!

 

Craig,
I actually was sort of doing what you said, already had a connection, then lost it trying to do a prom dump.
E6 prom dump worked with engine running or not running, didnt seem to care.
I will try again tomorrow and let you know. I am learning a lot about GM TPI from your software. The EPROM is a GM memcal for a 350 TPI, CSI, VATS, 3.27 gears.


Cheers

Richard

 

Yeah, I'm getting kind of exasperated with the fancy RS-232 protocol stuff I've been using. Thinking of going to a more 'raw' approach.

I found a nice document on RS-232 communications if anyone is interested, check it out:

http://webster.cs.ucr.edu/Page_asm/A...22/CH22-1.html

Lots of good stuff, lost to GUI nowadays...

-Craig

 

Craig,

That's a pretty good article as far as it goes, but keep in mind the 8250 hasn't been used since the IBM XT computer. The article doesn't even touch on the FIFO control register available in present day UART's.

Also keep in mind that QB45 has no provision to respond to interrupts, so you had better *be there* before the FIFO fills up. Logging to disk might suffer. I think it would preferable to modify the source of QBSER or even link an optimized assembly routine to perform the limited subset of UART interactions needed to read/write a car computer. Familiar with MASM?

Jim
Red '86 4+3


There's *more information* out there.

 

Haven't played with assembly since back with my old 6502 Apple. Should though. Programming's like an art, you either master it or you suck at it. I'm somewhere closer to the latter...

I'd be very interested to read anything relevant and more thorough if you wouldn't mind pointing me in the right direction. The old XT basis looks like it might give everything needed, and the new stuff's backward compatible, but if some of the new protocols make life easier...

 

Craig,

Sorry, I got some excellent references from a web search, then cut parts of text and printed - now I can't find the references any more. I'll work on it. Good assembly example code, too.

You may have no problem keeping up at 8192 baud, since you *know* when the data is arriving and are prepared to take it into an array right away. We're talking ~1 ms per byte and nearly 20 ms before the buffer overflows. That's a *lot* of PII time. I've re-written E6 to use only basic to do UART stuff. Haven't had a chance to check it out on the car, because...

I wired up the 2 transistor I/F and it syncs right up using 8.45 and 12v from the lighter. I found that 8.45 turns on both DTR and RTS, and substituting either for the 12v feed works fine. Now I'm building it all into a D-shell to replace the 4 transistor + MAX233 + 7805 monster I was so proud of until today...

Jim
Red '86 4+3

 

Yep, MAX233 was so clean and simple...

Let me know what you find. I'd enjoy playing around on a more low-level basis. I think you're right with respect to being able to capture the data as it comes, since the acquisition routine is a block to itself so to speak. When it comes to the 160-baud, I'm hoping this lower-level approach will help more with fuller control over the protocols and bit-splicing & timing issues.