Josh-Toker

Hello Honda-Tech, My name is Josh, 25 years old, wrenching on Honda cars since i was 14 years old. I have recently encountered a very annoying problem on my 95 Honda Accord EX with JDM F22B engine, thats the SOHC B1. My car has a very intermittent stumble, don't really wanna call it a miss, just seems to studder or ca ca cut ca cut at a very light throttle position usually 2500 RPM and below, but i have to have my throttle slightly toe-ed or it will not do it. As soon as it begins to do this, If i full throttle her WOT, then she immediately ceases to stumble, wakes rights up and rips with no loss of power. It will also never cut out or miss a lick at idle and always starts up without hesitation.

I have checked all wiring between my ECU and DIST, INJECTORS, TPS, HO2S. All checks out with full continuity and 0-1 ohms of resistance. I also built the entire engine bay wire harness and even the harness right up to the ECU. Replaced TPS recently and properly calibrated to .50 V at Idle and 4.45 at WOT, although i heard 4.8 was correct for WOT mine goes to 4.45, heard this is fine. Also have new plug wires, cap, rotor, ICM, also use an External Coil (Blaster 2) and MSD CDI Controller. Had this setup on previous F22B1 engine with no mishaps. Tried known good set of injectors with no change whatsoever, also even tried known good ECU with no changes whatsoever at all, problem carried over. Problem persisted for months on and off sometimes not studdering for weeks at a time.

After about three months got first CEL code 1. So i checked my HO2S wiring again and find no wiring issues whatsoever at all, so i replaced HO2S with a Bosch replacement from Napa, problem is so intermittent hard to tell if i fixed it, well a couple days go by and same ****, so i swapped the Bosch out and put in another HO2S sensor i had laying around in my parts bin, problem didn't resurface for another couple months. Problem has recently gotten much more frequent, it does it 3 out of 5 times i drive it. Recently got a Snap On Vantage and probed my HO2S, here is the procedure for the quick test.

According to the Snap On Vantage, KOER, Full Operating Temp. Channel 1 hooked to HO2S Signal Output (white wire) @sensor connector by backprobing. COM hooked to a known good ground. @ 2500 RPM HO2S Voltage should cycle between about 850mv and 250mv. O2S cycle varies with feedback system. HO2S should cycle 3-5 times in 5 seconds or 6-10 times in 10 seconds. Snap throttle several times while observing HO2S graph. With each snap, HO2S voltage should quickly rise (almost vertical). Voltage should go above 850mv and remain high momentarily, until rich air/fuel mixture clears out. Next maintain 2500 RPM for 10-15 seconds. Quickly release throttle while observing HO2S graph. Should see quick drop below 200mv.

Here are my results:

@2500 RPM HO2S Voltage = 0.010mv or ten milivolts = FAIL
Snapped throttle HO2S Voltage = sharp rise to about 200mv = FAIL
@2500 RPM/15 seconds quickly release throttle = quick drop to 0.050mv = pass.

So it failed 2 out of the three checks, basically my HO2S sensor appears only to reporting lean conditions back to the ECU, which as far as I understand would make the ECU increase the duty cycle to the injectors. On a further note recent inspection of my spark plugs revealed that indeed things have been rich as they were covered in black soot, also have had people tell me my car dumps a little black smoke when im getting on it. What im getting at is could my entire problem be that my HO2S is causing my car to run at its richest causing a misfire or incomplete burn at a specific RPM window. Being a pilot I know that when fail to lean the engine once i have climbed up a several thousand feet she will stammer or cut-out a little, and i respond by leaning her out and she smooths right up, infact if i monitor the exhaust temp and maintain 1250 to 1350 degrees Fahrenheit she runs like a top. Which leads me to my next question would It be worth my while to just plan on adding a permanent EGT gauge to my setup, so i can keep an eye on my mixture. Planning on getting P28 black box with hondata S3 and a Ryewire harness, although i built my own harness and its pretty clean I have to say im in love with Ryewire. Also i did not shield my HO2S wires in my harness with foil as the original honda harness had done, im guessing to block eddy field currents. I asked a guy at PFI speed performance in Ft. Collins, CO and he said that would not cause my problem. Any thoughts and all feedback are much appreciated. As much as a pain in the *** this has been i am equally excited to get to play with some tools and learn something new to add to bank of Honda know-how.

MAD_MIKE

Bosch O2S are inherently crappy, usually lazy when new.
Purchase a new Denso, Delco, or NTK unit.
If your O2S looks like the top one, its gonna be lazy its a crappy design.


Even if your 'spare' parts bin unit looks like the lower one, OS2 become lazy over time and will no longer properly switch between rich/lean. But it will not be lazy/slow enough to cause the CEL to come on. It is always best to replace the primary O2S when doing plugs. Consider it the nth plug.

Josh-Toker

The bosch sensor i had before that i bought new was exactly like the one pictured above, the current sensor i have now is a stock honda sensor and is like the one pictured below. I have performed another diagnostic test provided by my Vantage. The In-Car Propane test. Same circuit and probing method as before, just this time at idle and warmed up, add propane to artificially simulate a rich condition and monitor scope graph as before. With Propane added scope should show reading above 850mv. So i added propane to my PCV hose to the IM, immediately i notice the scope indicate 940mv, so for the first time my HO2S reported an expected rich condition the ECU. Now i create a vacuum leak to simulate a lean condition, immediately i notice it drop to 0mv to 0.050mv as expected. Now im slightly confused, how can it not catch the rich condition when the throttle is opened and fuel is added at 2500RPM but it can sense the rich condition with added propane at an idle. I left the propane on for a few minutes while monitoring the scope, it says it should produce a nice solid flat line above 850mv, now i get a almost flat line except for very minor glitches its flat. Also when i turned the propane off the car died immediately as if the ECU was attempting to en lean the mixture to point it slowed injector pulse down far enough it was primarily running on propane. So does a lazy HO2S become more lazy to CO and HC than Propane. I do not want to purchase any parts unless im 100% that it is gonna solve my problem. The only test my Vantage shows left is a bench test and it says it is not as accurate as the In-car propane test. Any ideas, could i be experiencing electrical noise as my engine is revved because i have unshielded HO2S wiring all the way from sensor to ECU even though i have been told this cannot cause this condition. Or is for some exotic reason my engine actually running lean some how, I have no vacuum leaks, I can close the throttle plate with a CD Jewel case wrapped in electrical tape and she dies instantly. So could my MAP instigate all these problems further up the stream, and im noticing the results, if i unplug the MAP it dies instantly, if i unplug the TPS, it will seems to continue running fine, i can even rev passed limp mode limit and on further which surprised me,, how can it rev fine like that without TPS input, though im not sure that this has anything to do with my issue it made me curious. It did throw an immediate code 7 though, so the ECU noticed it had lost TPS input. Im to the point that im considering running new shielded wire from ECU to HO2S for process of elimination, also when testing it earlier i noticed that while holding her at 2500 RPM it seemed to cut at that RPM the most, i would notice a loss of a couple hundred RPM though my Throttle position had NOT changed. As if it is all cylinders losing fuel or getting Too much fuel in response to the lean readings from the HO2S.

Josh-Toker

[IMG] [/IMG]

Some of my build pics.

Josh-Toker

[IMG] [/IMG]

Some recent pics with a shot of the spark plugs, these plugs were put in nearly 5 thousand miles ago, they look rich to me.
BTW EGR Has been blocked off with a 1/4 steel plate. Had it like this on my last set up never had any issues, i leave the EGR plugged into the ECU and disconnect its vacuum source, never thrown and EGR code. I have not used Nitrous since I noticed this issue, I also do not believe that it caused me problem as I also had that on my last set up as well and it never bothered anything, i have the Zex set-up, i have to arm it then it engages at WOT, it dumps NOS and GAS into my intake piping. Although I'm looking forward to solving this issue so i can confidently spray it once in a while, I do miss the fourth gear pulls.