Possible bad ECM input needed.
-
Cliff8928
- Administrator
- Posts: 1775
- Joined: Wed Mar 03, 2004 10:03 pm
- Location: Long Grove, IL
- Contact:
Re: Possible bad ECM input needed.
We can try that at Bfest...Asylum wrote:Sometimes if you tap on ME, (with a big enough bat) I'm able to act normal for a while!Cliff8928 wrote: I could tap on it and it would act normal for a while.
My bad ECM would cause the engine to bog down with ANY throttle input when it was acting up.
Cliff Scott
-
RobertISaar
- Registered User
- Posts: 98
- Joined: Tue Nov 27, 2012 7:14 pm
Re: Possible bad ECM input needed.
heating/cooling cycles and vibration take their toll on even the best solder joints.
electrolytic caps also have a finite lifespan.
more often than not, those two causes are responsible for most internal ECM failures. outside of that, you can unintentionally damage them in any number of ways, but that is outside of their normal operating range.
at this point in time, I would be worried about the PROMs starting to lose bits as well. I've actually pulled a few BINs from PROMs that were experiencing "bit rot". one bit being off can not only cause the checksum to fail and set a DTC, if it happens in the code section, one bad instruction will almost always cause the bad opcode vector to be called, which forces the ECM into limp-home mode. at least LHM is fairly foolproof, it's an almost/completely analog circuit, so long as you have functioning injectors and reasonable signals coming from the crank, coolant temp and MAP/TPS sensors, it gets the job done.
electrolytic caps also have a finite lifespan.
more often than not, those two causes are responsible for most internal ECM failures. outside of that, you can unintentionally damage them in any number of ways, but that is outside of their normal operating range.
at this point in time, I would be worried about the PROMs starting to lose bits as well. I've actually pulled a few BINs from PROMs that were experiencing "bit rot". one bit being off can not only cause the checksum to fail and set a DTC, if it happens in the code section, one bad instruction will almost always cause the bad opcode vector to be called, which forces the ECM into limp-home mode. at least LHM is fairly foolproof, it's an almost/completely analog circuit, so long as you have functioning injectors and reasonable signals coming from the crank, coolant temp and MAP/TPS sensors, it gets the job done.
