madking411

Not sure what has been going on with my hatch, 91 Civic Si, OBD1 converted w/a Boomslang harness and chipped P28.

Last time I drove it, a few weeks ago, it was perfectly fine and I parked it inside the garage, and came back to it and it all started from there. I hit the key to start the car back up, and it would not start. I was wondering what could've gone wrong, from the cam timing, to the fuel, spark, etc. All of that checked out okay. So it led me to look in the dash, and notice that the CEL turned on when I put the key to the ON position, and it would stay on solid with no reaction from the main relay or fuel pump. I then started to play with the main relay, and wiring to it, and came to the conclusion that the relay had gone bad... Well I was wrong. I ordered a new main relay, installed it, and the problem still persists.

HERE, is where things start getting really crazy. With a new main relay in the car, what it does/doesn't do is what I cant figure out;
When you turn the key to the ON position, and do not try to bump the starter, you must let the car sit in the ON position from 3-7 seconds with the CEL solid on before the main relay will click, the fuel pump will run and then the CEL will turn off, and actually start the engine up.
OR
If you turn the key to the ON position and attempt immediately to start the car, before the main relay clicks/runs the fuel pump, the CEL will simply stay on solid, not run the fuel pump, and the car will allow the starter to run until the battery dies.

So, from there, that is where I was stuck. I then pulled up the Honda diagnostic flow chart for the main relay control circuit, and went through the process of each step...



All the steps on the flow chart checked out OK except for step 7, which states there should be about 10V available...I only saw 7.5-7.6Volts. So, it then made me think that there was high resistance on the blu/wht wire and the wiring between the ignition switch and fuse box, and between the fuse box and main relay. I checked for resistance, which there was none on either of these sections of wire. I even checked the wire between the ecu and fuse box, which also has perfect continuity. So, according to the dealership service manual in this case, the wiring doesn't show me anything out of the ordinary.

After that, I re-connected everything, and went to start the car up to see if maybe one of the connections was previously loose...Same result.

So, next I began to assume that maybe there was a loose connection somewhere, and started to mess around with some under-dash connectors, and wiring to the fuel pump. There was a certain point where I was able to move some of the wiring under the dash, and it would repeatedly run the fuel pump, and make the car behave ''normally''. The problem seemed to jump from place to place in the car. I even got to the point where I was thinking that my ecu was fried, because I decided I would try it for giggles, just to put my old OBD0 ecu in the car and as soon as I did, the KOEO position ran the fuel pump every time like normal. Reinstalled my OBD1 ecu in it, and it just sat there with the CEL on and no main relay/fuel pump reaction. One of the connections under the dash that I would fiddle with that I could get the main relay to react with was this one: Im not sure what all wires that connection includes, but it seems like that's where my problem could be.



And so, I checked that connection, re-attached it in case it was loose and attempted to re-start the car. Same result.

I've taken the car out after getting it to crank up, a couple times and after driving for 15-20mins, it will also throw an intermittent CEL code for #8, TDC sensor.

So, at this point I am in need of some outside help, if anyone has seen anything like this before and/or may know what the problem is or could point me in the right direction it would be much appreciated!

F22Master

This seems very likely. Maybe you have a capacitor going bad in your ecu.