I have been dealing with this for a week -No luck at all.
Full story follows but my main question is: what does the wire between ICM and ECM do? Adjust or fine-tune timing (or dwell -for old school)? Could I have a bad ECM? Is that common? Are there any sensors that would cause my problem and NOT record a code?
------------------------------------------------
I'll start from beginning:
No CEL >>>
I bought this 95 accord that had been sitting for a couple years. Got it started and it was firing on all six cyl and seemed to be running fine. Left for a day and got back and #3 was missing = compression was ZERO. Pulled valve cover off and found the intake rockers under the cam! Removed the cam bolts and was able to get them put back where they belong. Now the car is running fine...
UNTIL I went to wal-mart & got a $20 oil change and added some teflon (slick 50) for good measure (I always put it in any used car I buy -Long story).
Drove home -on the way the rear exhaust fell off so I had to stop on the freeway and tie it up with a neck-tie, which was the only thing I could find in the car -I know the exhaust system shouldn't have anything to do with my ignition prob but just being thorough in my explanation.
Parked car and a couple days later it would start for a second then stall.
I was thinking it was fuel issue so I checked the solder joints on main relay -didn't see any cold joints but hit them all with soldering iron anyway. Relay tests good and was not the problem. The fuel looked yellow so I drained tank and put new premium in. Removed injectors & verified they were working.
was assuming fuel prob because it was starting for a second, then stalling immediately after. But I was wrong.
Found that I was getting ONE revolution of spark, then none so I replaced the coil and ICM with junk yard replacements. Still no spark so I got another ICM -No spark.
Now it almost never even does the single spark so I'm at a loss!
I have not tested continuity from ICM to ECM but I am assuming since it happened overnight that the wiring is fine. I did pull the carpet back and removed nuts that hold ECM in place but it seems like a real pain to get to the connectors. I have traced the yel/grn wire from ICM to harness near firewall thoug and verified continuity...
It seems like the I4 and the 2.7L V6 from 94 to 97 have the same ignition setup even though the V6 has coil mounted under rather than inside distributor so I believe that any info from similar 4 or 6cyl ignition systems will be relevant to my problem, so if you have experienced similar trouble with 4cyl please don't hesitate to reply to post. I will appreciate any help I can get!
I dropped by a friend of a friend's (mostly Honda work) shop, and asked if it was likely that I got two bad ICM's from the junk yard and he said he didn't think so -I know that is just speculating... He also told me that he usually just replaces the whole distributor when he encounters this type of problem (probably referring to coil-in-distributor setup) -which I can understand so not to have customers coming back mad a few days later -but I can't really afford to go buy a new distributor and coil!
Thank you for bearing with me through my long story!

 

Do you have power at the coil? And just to be sure are you saying you have injector pulse?..it seems so but I want to verify as that will rull out some items.

 

yes, have power at coil and ICM. Injectors are receiving pulse and spraying fuel.
Thank you! ~DX

 

You will need to check for a signal at the yellow/green wire going to the ICM while cranking. Best to do with a scope...if you don't have that you will need to start checking inputs into ECM.

 

The wire between the ICM and the ECU is grounded by the ECU to charge the coil. When the ECU switches that wire from ground to open, the coil fires. The signal will look like a square wave on a scope.

Recheck for any stored engine codes.

How are you testing for spark?

Do you happen to have a 12V test light?

 

Been traveling a lot. Hope to have some time tomorrow before I go back on the road to scope the wire between the ICM & ECM.
For spark testing I have an adjustable-gap tester (one end to ground, other to plug or coil wire, gap is in a threaded plastic frame that allows you to adjust the gap by turning the contact).
I have a cheap test light that is computer safe and indicates 12v or Gnd with red & green LED's. It works well for what it is.
Will see if any codes are stored.
Thanks for all the help.

 

Have square wave on wire between ICM & ECU when cranking engine but no consistent spark (only the occasional single spark on the first crank).
Being that the ECU pulls that wire to ground to make a spark, should I get a spark every time I ground the wire (ignition ON and that wire disconnected from ECU) -Would that be a way to test the ignition system???
And since I have the square wave, does that indicate that the coil and/or ICM is bad? While it seems improbable that I have pulled three bad ICM's and two bad coils at the junk yard, I guess it's not impossible. I will feel like a real dunce if I have wasted so much time on this because I was too cheap to spend $40 buying the parts on FeeBay!
Or is there anything else that could prevent spark?
Thanks!

 

no codes>

 

The square wave verifies the engine computer is giving the proper signal.

If the signal wire to the coil is switching between 0 and 12V you may have a bad distributor cap/rotor.

 

^That is the green wire @coil.