It’s a 2732 eprom and I broke the 3rd pin down on the right hand side (looking at the top of the chip with the notch to the top). Any chance that this thing will still read correctly???

I don't know if this helps anyone but here it is anyway:



BTW, is there some way to insert just plain text into a message here, I couldn't come up with a better way so I actually took a screen shot of it, put it up on a web site as a picture and embedded it in my post.

I don't know enough about how chips work to help you on that.
Just figured misery loves company.
You probably said the same thing I did when I broke the $50 ignitor on my $100 heater last night
(Trying to fix it up and caused fatal damage )
Now I'm stuck doing anything for a few days till I get a new one.
Sh%t happens.

Well, let me put it this way... my truck was my only moving vehicle that I had to drive tomorrow and it was doing weird things with the O2 and egr (I think that the O2 is coolant fouled, long story) so I was going to disable the egr and closed loop... now I'm not sure what I will do. I have extra 2732's floating around, but of course it's a '92 with a '299ecm in it which there is almost no info available about...

So yea, I uttered a few choice words with this one...

Pin #22 (A9)? That's an address line. I doubt the chip will work at all and if it does, there will come a point when the ECM tries to assert that line to access an area of memory and the ECM will probably crash. On the other hand, my '8746 WILL run w/o a prom, just REALLY, REALLY poorly.

If you have the equipment to do a read & burn of a new chip, you can probbaly get it in the programmer and rig a sewing needle & length of wire to make contact with the missing pin.

Well, it did read and verfy alright, but looking at it with the only ecu/xdf that I could find for the $99/'299 there are some really weird values in a few places, like the stoich AFR reads 8 and the egr temp on is comming out as 248...

Yes, only some of the data will be messed up. Each of the "A" pins form the address.

Address = A11 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0

Each line has either 0v or 5v applied to it. Since A9 is broken it can not have this voltage applied and usually a floating line like this defaults to a 5v level. Sometimes it defaults to a 0v level.

Every time the ECM tries to apply a level it does not happen so the wrong data is returned only sometimes. That wrong data is going to cause odd things to happen. Overall, you will get strange results and operation. It needs all of the "A" pins to work properly.
In terms of reading and verifying. Yes, it will still read. What are you verifying against?? The odd values are due to the A9 pin being broke. Those locations are in the affected address zone where A9 was a 5v when it should have been a 0v. The chip will always read properly. It is like someone giving you bum directions. You can always read the directions......but the directions are wrong and cause you to go to someone's house instead of the intended speed shop.

Is there a little bit of pin left at the case to solder a 30 ga. wire too?
If not, might as well order a DIP socket and some EEPROM chips. Note, ATMEL discontinued the AT29C256 in a DIP package.
The pick you posted is a 4kx8 chip and so are the 2732. I would erase & burn a 2732 and drop it in. The timing of those EPROMs didn't change all that much in the late eighties & early 90s. I have been told the 299 ECM runs the same 68HC11ish processors. Anyway, I would use a 2732 first. You can't break the ECM by trying it. Worse that will happen is that it will act broken like it is now. It should be fine though.


If you really need to drive it. I would pull the chip and run it in limp mode. I think that is safer than running that bad chip.

PhLaXuS, thanks for the idea... I put the chip in my burner, bent up a little make shift pin out of a resistor lead and it read with what look to be reasonable values in all the registers...

now onto other "issues"

I’m really confused… the pic I posted was a 2732 and you wrote “and so are the 2932.”

The pin that I broke off was off the factory 2732(I believe that it is actually a 2732A) and I’ve got a stack of 2732a’s here.

You said “Anyway, I would use a 2732 first.” Again, I’m confused, what else am I going to use??? I have looked for some 2532 (supposedly pin compatible) but I don’t have anything to use besides a 2732A so I’m not sure what you’re getting at.

I was confused too. I thought you mis-typed and wrote "2932" when you meant "2732".

Never heard of a 2932 eprom. I did some checking and I don't think they exist. That must be GMs part number and not a normal vendor part number.

If it is a 24 pin chip then it probably has the normal 4kx8 pinout. I have never seen an oddball 24 pin eprom pin configuration. In that case, use one of the 2732A chips you have. There is nothing to lose by trying this except for maybe destroying a 2732A chip. You could probe the VCC and GND pins in the ECM to make sure they are the same as the 2732A (that you posted above) so that you don't destroy a 2732A chip. If it was my stuff I would program a 27C32A and drop it in. Turn the key on and check for a flashing SES and turn off if a problem existed. I really think it will be fine though.
The only difference from the 2732 series is the programming voltage that I know of. 2732=25v, 2732A=21v, 2732B=12v. The part the ECM does (reading) is all the same for each chip.

As for trying something else. I use 29C256 and 20C010 in just about everything. I am so accustomed to that so I kind of said it without thinking that not everyone is setup that way.

EDIT - I just realized that you said the 2932 is pin compatible with the 2532. Those ARE NOT pin compatible with the 2732 from what I remember. So don't try the 2732 if your ECM really takes a 2532 pin-out chip. I will look for the pin differences.

2532 pin-out
You could bend the pins of a 2732A & re-wire the few pin differences with 30 ga. wire if you are desparate. That is, pins 20, 21 and 18 of the 2532.


EDIT - link to 2532 datasheet: http://www.jrok.com/datasheet/TMS2532.pdf

Note the differences with the enable signals.

Also, http://xtronics.com/memory/GM-eprom.htm shows that your 16146299 has a 27C32A chip in it. They could be wrong, but I am guessing it does. The 2532 chips have been out of production for a long time. I can't see GM using them in a 1992 vehicle.

When you read the chip with the programmer....did you set it for a 2732 or a 2532? I am betting you used 2732 and it read properly. It should not read properly if you set it for 252. That is a dead give away what type of chip/pin-out it is. If it is a 27C32 pinout then drop the 27C32A in it's place.
Out of things to say. But I bet it is a 2732 because GM was F*ed if they put a 2532 pinout in that ECM.

One more: http://www.tunercat.com/tnr_desc/ecm_sup.html#trucks says it is a 2732A also. They are usually always right. Try the 2732A.

Are you using this PCM to control the 4l80E in the Camaro also?

Crap, one typo at the beginning and the rest turns into a big pile of confusion.

The vehicle is an LO5 powered ’92 K1500 Blazer, ‘299 ecm, $99 mask with a 2732a chip

Yeah, in the electronics world part numbers are close in general and used over and over in different fields.

You will be fine using the 2732A. Set the programmer accordingly. Good news is that you can use up the 2732A chips you have.

EDIT - I just realized that you edited the orig. post. This will confuse a newbie/other people even more if they read this post looking for info.

_I think_ that I edited that while you were making your post there, that was also about the place that I decided that this thread has gotten so confused that there was no saving it.