ben73

Ok this is really weird...
I have used 2732A's, 28c64's, 27Cxxx and 29Cxxx's before in a few different GM ECM's without a problem.

For many months in my own 7747 I used 28c64's instead of the original 2732's by making an adaptor as described by Ken73 in this thread:
https://www.thirdgen.org/techbb2/sho...ighlight=28c64
Basically, it looks like this:





I have now attempted to use this mod (again in place of a 2732) in 2 different 85 TPI (1226870) cars with the same WEIRD results! The cars start and run fine, but anytime between 5 mins and a month later, they give a code 51 and go into limp mode! When i pop the chip back in the programmer, the contents have changed!!!!! I have tried with 28c64, 29c256 and 27c256, all failed eventually!

Should i be grounding the 2 unused pins at the top end of the larger chips?? Are they seeing a floating voltage??

Can anyone take a guess as to what is going on here??

Thanks,
Ben.

Grumpy

Obviously the 6870 doesn't like EEPROMS, since it's a MAF type system, so it should be updated to a MAP or MAP/MAF hybreed system. Perferrably the later.

ben73

Initially I thought it was an issue with EEPROMs also, so i tried a 27c256 and it did the same!
I'd really like to find the problem.. Last resort will be to convert to a '7165.

RBob

iTrader: (1)

CMOS devices are required to have all unused inputs tied to either ground or VCC. Otherwise they oscillate like a leaf in a hurricane.

RBob.

ben73

Thanks RBob. I suspected something like that, but its not something I knew for sure.
Thank you, I'll try it out!

ben73

RBob,
One quick question. I looked up the datasheet for the 28c64 at http://www.atmel.com/dyn/resources/p...ts/DOC0001.PDF

One of the overhanging pins is WE (Write Enable).
The datasheet says on pg3, under "WRITE PROTECTION" :

So i am guessing I should be connecting WE to Vcc to ensure the chip contents are stable???
Is this correct?
Much appreciated.

RBob

iTrader: (1)

Yes, ~WE needs to be tied high (VCC).

RBob.