Cliff Harris

I have a 1227165 ECM from an '89 Corvette, with the corresponding memcal. I have an '86 Corvette and find that this combo does not work for me, even though I reprogrammed the EPROM.

It occurred to me that the memcal is set up for a MAP car, which my car obviously is not. I see on the schematic that there is a connection on U11 pin 28 called "MAP", which is connected to a memcal line labeled CAL61 on pin 61. This pin is connected to U11 pin 20 ("TPS") through the 16 pin resistor network.

Do any of you guys out there know what resistance and/or voltage I should be seeing on this connection when it's a non-MAP car? I measure zero volts on U11 pin 28.

I'm asking because I don't have access to the ECM in my car and the '89 ECM is sitting on my workbench but it's stuck in limp mode and the microprocessor is held in RESET and nothing is working.

rodj

I don't see how a '165 memcal can be setup for MAP because the MAP spark control is different and IIRC is included on the memcal and not a standalone ESC unit as used with the MAF setup ?

Cliff Harris

Curiosity can be a powerful thing...

So I went out to the garage and pulled the ECM out of my car at 2:00 AM. i measured zero ohms across the 16 pin resistor network from pin 7 to pin 9, just like the MAP memcal. I did measure 0.54 V on the relevant pin when the ignition was on. How does it do that?

For the record, this is an early '86 with cast iron heads, automatic and 3.07 rear gear. EPROM is BUA 9340. 14 pin resistor network is 16055375, CTS8546. 16 pin resistor network is 160550376.

The '89 memcal's 14 pin resistor network is 16072665 and the 16 pin network is 16133420. The EPROM is APYP.

Cliff Harris

As for the MAP vs. MAF setup for the ECM, it's all in the code...

This ECM was used on many MAP cars, throttle body injection cars and cars with various numbers of cylinders from 4 to 6 to 8. Here's a list I got from somewhere:

http://home.earthlink.net/~cliff_har...age/usage.html

I thought the '89s were MAP cars, but I just read somewhere else that they are MAF. Is that correct?

bjankuski

1989 is a MAF car not a MAP car. There is no reason a 1989 chip should not work in a 1986 car. The 1989 chip is not the reason the car will not run correctly.

Cliff Harris

I went to bed and was lying there and it hit me. I had typed 0.54 V and thought, "Gee, that number sounds familiar ;-)" and still didn't make the connection that it IS the TPS voltage. The TPS and MAP inputs are connected together. I guess the chip just wants to see something other than zero volts on the MAP pin. After I hooked it up on my test bench it showed the right voltage.

I'm still stuck in limp mode, though. I can't figure out what inputs I need to make it work correctly. My test bench is mostly done but still missing all the pulse inputs except I do have VATS working. I don't have any of the pots hooked up yet. Maybe those floating inputs are messing me up. The quad drivers went into ~FAULT mode when nothing was hooked up to the outputs. It shouldn't need any pulses to work because obviously there are none when the ignition is turned on but the engine isn't running. I'm seeing ~RESET pulsing away at 100 KHz, which doesn't give the microprocessor enough time to do anything. Now I'm starting to wonder if ~RESET is coming out of U12 instead of U1. Maybe it's both. One thing I discovered in my career is that when things don't work you start studying the circuit really closely and questioning everything. Good learning opportunities there. It's too bad Ludis isn't around to ask why he drew the schematic the way he did. He obviously had access to information that we don't.

I did put this ECM in my car and it ran REALLY badly. The guy I bought it from told me it came out of a running '89. He didn't say how well it was running...

Caboboy

This sounds to me like somethings backwards. It couldn't be the memcal because it has **** and can only go in one way. Is there an adapter being used?

Cliff Harris

No adapters. I unsoldered the original EPROM and programmed and put in a new EPROM I bought online. Stock BUA1728 program with no changes.

Since this was bugging me, I hooked up the TPS, CTS, MAF and O2 sensors on my test bench. I grounded the EST input. Still the same. Stuck in ~LIMP mode. I REALLY REALLY want to see something come out of the ALDL port. That won't happen until the microprocessor gets released from ~RESET mode.

One thing I have not done is try running it with the memcal from my car (or the whole ECM, for that matter). Since the ECM is now sitting on the floor on the passenger side, that will be easy to try. Since the microprocessor is not running, the program doesn't matter, but maybe something is significantly different with the resistor networks. I'm kinda leery about taking the ECM out of my daily driver and plugging it into something that is problematical. There are a lot of wires there. I might have something hooked up wrong. I don't want to end up with TWO bad ECMs...

Cliff Harris

It worked with the memcal from my car. Now the question is: What's the difference?

Cliff Harris

I figured out how to attach pictures to a post, so here's a pic of what I'm working with:



The brown board on the right will have the pulse generating circuits on it someday (VSS, EST & ESC). The VATS pulse generator is already in place.

The green board has all the status LEDs on the left, standoffs for the brown board, pots for various things (TPS, MAF, O2, CTS, VSS & EST - now wired) and switch inputs on the right (ignition, Park/Neutral, OD, A/C clutch and high A/C pressure). There will be a pushbutton switch in the bottom right corner to simulate knock pulses. The green board will be the cover of a black plastic box that is not shown.

I managed to break the lead off my fuel pump LED and I haven't determined the voltage for the IAC LEDs. I used bipolar red/green LEDs so I can see if the IAC is stepping out or in.

Once I get all this working, the plan is to use the Moates Ostrich I just bought to REALLY futz with the code. I bought a used HP 1651B logic analyzer ('80s vintage - but that's the same era as the ECM), but I can't find a System disk for it so I can't boot it. Whenever I get THAT set up then I can see exactly what the code is actually doing. The trouble with this scenario is that I can't find a 68HC11 inverse assembler for the HP, so I guess I'll have to write one. The fun never stops...

Cliff Harris

I found out that Corvettes went to MAP in '90.

I put the original APYP EPROM back in the carrier and everything is now working. I guess I did something wrong when I burned the BUA 1728 EPROM I was using.

Found some interesting stuff. The ECM stays on for about 10 seconds when you shut off the ignition to allow for MAF burnoff and IAC reset. Then it goes into standby mode.

sabercatpuck

If you are still looking for an inverse assembler I am most of the way through coding one. I am posting my progress on the delcohacking website. It is based in excel vba code. It isn't complete yet but it is getting to be functional. At this point it is specifically geared to the CSV's I get from my hp logicwave pc based logic analyzers I have, but it wouldn't be to hard to use it with others since all I need is the address in column a, the data in column b, and I think I remember the column c as the E pin from the processor. It goes low when it is pulling the first byte of an instruction so it makes a nice way to do error checking and find interupts.
http://www.delcohacking.net/forums/viewforum.php?f=4

tequilaboy

27C128 vs 27C256 eprom differences?

Is one of your eproms a 27C256 per chance? I've seen a couple MAF memcals with 27C256 eproms.

If programming a 27C256 eprom with a 16 k bin, you need to either stack the bin, or offset the programming start address so the code is at the bottom. This could be your issue.

Cliff Harris

All I have is 27C128 EPROMs. I bought some from somebody on eBay and they turned out to be about 10 miles from me.

I know about the methods of making 27C256 EPROMs work. I just don't need to do that.

My ECM test bench has gone into permanent hold mode. It actually has everything hooked up except I haven't designed a circuit to generate distributor reference pulses. I have generated the pulses with a pulse generator though, so I know it works.

Some of my status LEDs are displaying the opposite of what I think they ought to. I thought I was very careful to make sure what state they should be in, but oh well.