After failing smog repeatedly, I took a two year break from tinkering. Today, I reentered the garage and what i went through was too good not to share, which is not to say that i enjoyed myself; although now, as i sit here on the couch, the puzzle of troubleshooting in my mind is indeed entertaining.

And here it is: 1991 civic si. running a d16a1, which is a jdm obd1 vtec engine. I have been using a p28 ecu, as the engine is identical to the d16z6, or so i was told. When i acquired the engine it came with a p08, which is used to run the d15b vtec engine.

Today, I installed a wideband, the Innovate MTX-L. To do so I swapped out the p28 for the p08.

The p08 uses a one wire O2 sensor. The p28 uses a four wire O2 sensor. I connected the narrowband output of the MTX-L to the signal line, calibrated and went for a spin. It idled great at 14.7 and runs great, too. While driving it did not go rich once, nor did it lean out much, except of course on decel. This was very exciting until i checked the ignition timing, which was retarded, although it was not running like a retarded engine. Idle was smooth at 750rpms. I jumped the service connector and adjusted the ignition timing, shut down, disconnected SC, and restarted, but the timing returned to its retarded state. Why?

I assumed, Oh it must be the ecu, so i put the p28 back in, but this is not as easy as it sounds. The p28 looks for a four wire o2 sensor; it expects to see the narrowband signal and that the heater control circuit is working. This required a bit of wiring, which I'd be happy to describe if anyone is interested.

Using the p28 the car did not run as nicely. AFRs were good, power was the same, but it was somehow not as smooth at idle, let's say that it was fussy; might i add that it has always ran this way using the p28. After a quick spin I checked and adjusted timing, which took. Anyone know why? Does the wiring of the engine harness differ between these two ecu's, besides the o2 sensor and the purge solenoid (the p08 doesn't use one). Additionally, the CEL behaves the same for both ecus; it turns on and off at start up, and goes solid when the service connector is jumped.

 

PROCEDURE.

Above all get procedure right, otherwise, what you get is wrong.

Apparently, when checking the base timing the SCS must be jumped. I was just jumping it when adjusting.

Here is the correct procedure.

Check ignition base timing with the engine at operating temperature. Turn off engine. Short the diagnosis connector for the ECU, which will tell the ECU to leave the engine at base timing. Check timing now, this is the ignition system's base timing. You are not seeing the BASE TIMING unless the SCS is jumped.

 

I would be curious to see pics of the engine since a d16a1 is a dohc Integra engine...

 

2nding this, the info you provided isn't accurate so I'd like to see pics as well.