Hello, my fellow Honda-Tech's.

I have a bone stock, manual, '97 CX hatchback (D16Y7) with an intermittent stumble at idle, while holding a cruising speed and light acceleration but not while under heavy acceleration.
Mostly when hot, but sometimes lightly while cold, it sometimes just randomly does what feels like a misfire. You can see the idle drop depending on how bad it acts up and sometimes completely stalls and will have an extended crank.
I have also noticed that when I take my foot 100% off the gas for deceleration, it completely disappears.


Some details:
- only does it while motor is at normal operating temperature.
- doesn't do it all the time, it's random. sometimes it'll stumble and stall at idle and fight me for 5 minutes or sometimes I get 5 minutes of it running perfectly.
- P3600 code will pend for a while before becoming a confirmed DTC. (I can't find any information on a P3600 DTC.)

New/new(ish) parts:
- distributor ICM
- distributor ignition coil
- distributor cap and rotor
- ZFR5F spark plugs
- NGK spark plug wires
- primary o2 sensor
- valve adjustment done 45k ago (i know it's probably due again)

Tests I've already conducted/looked into:
- 2 separate MAP sensors both have 5v reference, ground and respond well to me sucking on the sensor (verified with live data)
- IACV handles cold idle and slowly brings it down perfectly
- primary o2 does its job and responds quickly/accurately (verified with live data)
- checked and wiggled every harness connection point I could find under the hood and checked grounds. everything looks good
- all injecotrs ohm'd out within 0.1 ohms of eachother
- all injector plugs fired in a perfect rhythm when tested with my Noid light.
- different gas from different gas stations (always 87)
- TPS has correct voltages at the sensor and responds on the scanner to every little input

Tests I have not done for various reasons, usually because no equipment:
- fuel pressure
- output voltage from distributor
- actual decent data logging with an oscilloscope or advanced scanner (mine is a bluetooth scanner that works well but isn't very fast to update)


Alright so, I'm kind of stumped now. I don't know what else to check right at this moment.

Any thoughts? Input? Anything is helpful. Thank you all.

 

How consistent are your ECT and IAT readings once the engine is at operating temperature ?

 

My immediate thought is that it's related to fuel delivery (a lean stumble). Can you capture a datalog of the narrowband o2 output and STFT leading up to / including the stumble? If your STFT begins to drift downward leading up to the event, the ECU is observing a rich condition and steadily pulling fuel to achieve what it thinks is stoich.

 

Alright, my apologies for disappearing there for a little while. Life got busy.

I am unable to datalog with this crappy *** app I use - for some reason it says it saves logs then when I go to fetch it, it finds nothing.

I can tell you that I have watched the live data while the car stumbles and the STFT and o2 seem to be reacting normally. The o2 will start fluctuating between 0.8 and 0.9 volts and the STFT will slowly go negative every second. Like, -1.5% more every second until the car either stalls or stops acting up.
I've seen it go as far as -28% STFT.

So I believe the o2 is reading correctly and the ECU is leaning hard in response to it. I just don't know why it keeps randomly going super rich.

Also, the parts cannon is still going as I bang my head against the wall... new ICM in the distributor... that was an expensive one but car still acting up just the same. Oh and it does it while the entire car is cold now too. Much less when cold but still does it.

 

I've watched these specific sensors, along with others while driving and I did not observe any strange fluctuations or anything out of the ordinary at all.

They appear to respond in real time to temperatures and the data never swings hard or fast. They read correctly when drivetrain is 100% ambient temp too.

 

So just some theory crafting for whoever may stumble upon this thread:

- It would *appear* the motor is randomly going super rich and the ECU is leaning it out hard to compensate, thus causing my problems.
So I'm guessing it has to be fuel delivery or ignition related, right? Could 1 or 2 injectors be randomly getting stuck open causing the rich condition then causing the entire motor to lean stumble because the ECU is leaning out?
How about the probability of a failing ECU? I've heard it's possible but extremely rare.
Is there perhaps a certain wire I should check continuity/resistance on? (maybe the ECU controlled ground for the injectors pulse-width?)
Has anyone heard of the crank sensor maybe failing intermittently causing issues? (or the wires leading from it to the ECU?)

I'm just thinking out loud here to give readers an idea of what I'm going through/thinking about. You know, sometimes obvious crap eludes even the best techs and all it needs is a 2nd set of eyes lol

 

This is the reason why I asked about the IAT and ECT values... if the ECU believes that the engine is cold, even though you have been driving for some time and it is actually up to operating temperature, heavy amounts of fuel is being added for compensation. For instance, if the ECT is reading 70-80'F while the engine is at operating temperature, the computer will add between 25 and 30% more fuel... then the O2 sensor detects a very rich condition and begins to pull fuel out accordingly. Do you see the dilemma there ?

It IS possible that you are having injector troubles caused by the ECU. The best way to rule this out is to try a known good ECU to see if the problem continues or is eliminated. It could also be a failing FPR, fuel pressure dampener or a stuck bypass port in the fuel pump randomly failing and causing the issue that you describe.