(I attempted to update my original thread of the same name but was informed it is 3106 days old lol. It can be found with a simple search)
TL
R - the cause of this 15 year headache has been a faulty sensor ground (D22) that tested fine but would intermittently short while the car was running.
Back from the dead! Hard to believe but I've been dealing with this problem for over 15 years but I believe the solution has been found. I've been living with this nagging issue the entire time and only after a pet peeve quota was reached would I hunker down and try to figure out the cause, fail, then go another few years and repeat the process using some other theory. Anyway, long story short it was a ground wire (D22) that provides logic ground to the throttle position sensor, the O2, the EGR etc. The conversion harness supplied by Rywire back in 05 had a single purple ground wire that was to be tapped into for all those sensors. Only after inspecting every single wire by hand did I reach the culprit. Where the ground wire goes through the firewall the insulation had been worn just enough to provide contact thus grounding the circuit improperly. Back in 08 (and 10, 12, 16) I had checked the TPS with the engine off and key on and the numbers came back good! What caused the issue did not present itself until the vibration of a running engine caused the wire to tap randomly into exposed metal which is why it was so hard to nail down. A new ground with solder, shrink wrap, and Super 33 tape solved the problem. I replaced any connections that looked suspect while I was there but only D22 had obvious damage. It's only been a week but I cannot remember a time the car has idled smoother, decelerated as linear or general driveability has been as pleasant. Still not new car smooth, but a Caddy by comparison to the way it was.
PS. the method I used to find the problem was to disconnect one sensor at a time until the symptom went away. After 15 years and multiple attempts to diagnose the problem the flowchart/helms way did not work in this case, it was too random vs a simple dead sensor. Unplugging the TPS left me with a high idle but smooth as glass decel with an obvious CEL for TPS but also O2. Going back to the wiring diagrams I saw that both sensors shared the D22 for logic ground and thus my discovery. I knew it wasn't the O2 bc she would act up on cold starts before the O2 even came onboard closed loop. Wasn't a vac leak because it was never consistent (and I'd regasketed the entire intake side over the years). To summarize, I hate wiring and it is occasionally the work of Satan.
Mak
__________________
'85 CRX Si Original EW3 @ 254k...suck it Detroit
'89 CRX HF-Z1 ~ When haste is needed
'97 Civic DX Hatch ~ Formerly 600hp, now 0hp, soon 115hp.
Linear Mode
