• NAXJA is having its 18th annual March Membership Drive!!!
    Everyone who joins or renews during March will be entered into a drawing!
    More Information - Join/Renew
  • Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

how smart is the OBD-l ECU?

Shorty

NAXJA Forum User
NAXJA Memorial Lifetime Member
I'm trying to wrap my brain around what might be a very simple question:

Will the ECU ignore an O2 sensor that reads rich or lean if it sees an injector circuit fault at the same time?

back story with specifics:

-- 1994 XJ with 2.5L low output 4 cyl.
-- injector circuit malfunction code (27) and dead miss last Sunday
-- gas dripping out of drain hole in the muffler (stuck open injector)
-- used Matco version of Snap-On's "the brick" code reader to determine that injector 3's circuit is F.U.B.A.R.
-- unplugged the faulty circuit's injector and still had a dead miss (duh)
-- dripping gas stopped and I reduced the chances of polluting my "new" catalytic converter until I could dig deeper
-- changed injectors two days ago and solved miss :clap:
-- at no point during the 6 days of using this as a daily driver, did I get a rich mixture code (pig-rich with injector stuck open) nor a lean mixture code (unplugged injector after determining the correct one)

I guess the real question is:

If the ECU knows that there is an injector circuit malfunction, does the ECU ignore the O2 sensor being out of range?

anyone with the smarts to know how the ECU "sees" the problem is welcome to chime in!

she's fixed and running fine, I'm just trying to wrap my brain around why I never saw any codes for rich or lean mixtures.....


 
It's possible that once it set a fault for the major issue of the injector circuit that it just went open loop, in which case, yes, it would completely ignore all O2 sensor inputs and just go back to it's base fuel map.
 
any clue where I could find out if that were true? It's been bugging me since "day one" why I couldn't see the rich mixture code!

I run an auto parts store that helps people understand trouble codes and such-- I hate not knowing what the ECU will ignore vs. what you "should" get along with the primary code.
 
I'd put my money on it being so stupid it didn't notice either, actually.

My 91 has an issue right now - it's either a bad CPS or a bad TPS from what I can tell. It will randomly stutter at mid throttle range and lose power. Backing off the throttle and then going back to where I was will make it go away momentarily even if I never slow down significantly, so the RPMs never really change much (manual trans.)

No codes. I only get a code when I completely unplug the TPS, and it goes away if I turn it off and then start it back up again.
 
I'd put my money on it being so stupid it didn't notice either, actually.

My 91 has an issue right now - it's either a bad CPS or a bad TPS from what I can tell. It will randomly stutter at mid throttle range and lose power. Backing off the throttle and then going back to where I was will make it go away momentarily even if I never slow down significantly, so the RPMs never really change much (manual trans.)

No codes. I only get a code when I completely unplug the TPS, and it goes away if I turn it off and then start it back up again.


That really sounds like a bad TPS since it is dependant upon throttle position and not engine rpm which would be more cps related.
 
that's what I was thinking but some friends have told me they had exactly the same symptoms and it ended up being a CPS. I have a new one sitting around waiting for me to install it, just keep forgetting to do so until it's acting up mid-drive.
 
I'd put my money on it being so stupid it didn't notice either, actually.

so it saw the injector fault, but didn't see either the rich, nor the lean exhaust because it was too stupid???

It was "smart" enough to show injector #3 as having the problem, but could only report an "injector circuit malfunction" as the code. I believe the logic built into the ECU is far greater than the codes it can put out-- whatcha think from that angle?
 
Wouldn't be at all surprised, I have seen the hardware capabilities of a platform get completely ignored by the software crew far too many times to discount it.

This is making me want to start digging into an OBD-I ECU. Damnit, I have enough projects on my hands already!
 
.This is making me want to start digging into an OBD-I ECU. Damnit, I have enough projects on my hands already!

get on it, damnit!! my curiosity is far greater than your roof, or your impending inspection, or anything else you've got to deal with......

:shhh:
 
I can't answer your question directly, but once did have a faulty splice in the + of one injector on an 93, which caused a miss when it warmed up and the injector failed to fire. This returned a harness and injector line fault, but never an O2 fault, so it would seem certain that it did not register lean.
 
I'd put my money on it being so stupid it didn't notice either, actually.

My 91 has an issue right now - it's either a bad CPS or a bad TPS from what I can tell. It will randomly stutter at mid throttle range and lose power. Backing off the throttle and then going back to where I was will make it go away momentarily even if I never slow down significantly, so the RPMs never really change much (manual trans.)

No codes. I only get a code when I completely unplug the TPS, and it goes away if I turn it off and then start it back up again.
TPS dead spot. You got an analog meter to test the TPS with?

What happens is the ECU sees you accelerating, then there's a drop off because of the dead spot, so it thinks you snapped the throttle closed and goes to decel mode for a half instant. At least that's what it sounds like.

You can take apart the TS and move it's "wipers" to a different path on the circuit traces. Makes it good as new once you get off the worn "tracks"

This is making me want to start digging into an OBD-I ECU. Damnit, I have enough projects on my hands already!
You should tackle the renix box first ;)
 
The OBD TPSes seem to be glued shut but once I verify my new TPS fixes it, I'm gonna take it apart with a dremel and see how bad it is. That's exactly what I figured is wrong with it, since the problem spot is right at the throttle position that gets used the most.

I only have a 2.5L TBI RENIX ECU on hand, but I've already cracked it open and taken a look inside something like a year or two ago. IIRC they put all custom part numbers on the chips, that makes it exponentially more difficult to RE and I will not be doing so unless I find a cross reference for the AMC/chrysler part numbers to "real" standard electronics industry numbers.
 
The O2 sensor doesn't read fuel content of the exhaust gasses. It reads O2 content. With a misfire the PCM could interpret that as a lean condition though. Being that the injector was literally pissing gas I would also change the oil.
 
The O2 sensor doesn't read fuel content of the exhaust gasses. It reads O2 content. With a misfire the PCM could interpret that as a lean condition though. Being that the injector was literally pissing gas I would also change the oil.

understood
shoulda
done that

I appreciate the help, I'm just stuck on why I didn't see ANY code other than injector malfunction!!

I'm thinking Matthew Currie and I have started to show why the OBD-l ECU is as good as it get at onboard diagnostics :gag:
 
I would have expected to at least have seen a misfire code as well. Who knows. Without having access to the code and dissecting it or having a chrysler engineer from back then its hard to determine the rhyme or reason.
 
I would have expected to at least have seen a misfire code as well. Who knows. Without having access to the code and dissecting it or having a chrysler engineer from back then its hard to determine the rhyme or reason.
OBD1 does not have a misfire code. You won't see that until 1986. After OBDII there is a misfire code, but this is extrapolated, it's not directly brought on by ignition or injector misfire itself. It's cleverly guessed from anomalies in master timing.
 
OBD1 does not have a misfire code. You won't see that until 1996. After OBDII there is a misfire code, but this is extrapolated, it's not directly brought on by ignition or injector misfire itself. It's cleverly guessed from anomalies in master timing.

:gee:
 
I doubt an OBD1 system would be that smart. We see that kind of stuff all the time on newer vehicles, say a pid called INJ-F. Injector fault. Should read no all the time, if it toggles to yes then the computer goes to open loop fault mode.

I work on some transmissions that will throw a code with the merest hint of a botched shift, and others from the same manufacturer and era that won't throw a code while revving its guts out at a stop in Drive.

Ford has implemented a manual trans clutch protection strategy after a few too many warranty claims. Rpm is compared to input shaft speed and if mismatched for too long, timing is pulled and the throttle plate closed.

Wandering off topic a bit, I get a kick out of what they can do with computers lately, and often don't.
 
... or worse, can do with computers lately, and try to, but fail.

For example, the traction control and ABS systems on some modern vehicles that are known to freak out and either hamstring the engine or the brakes under slippery conditions acceleration or braking. I know a few people who have had the brakes go to the floor because their ABS flipped out, thinking they were braking on ice, and stopped them from stopping. Nearly caused an accident.

They pulled the ABS fuse after that and haven't had a close call since, driving in the same conditions...

Another guy tried to pull out of a parking lot in slushy conditions and was nearly T-boned by a truck because his traction control thought he was slipping, and refused to let him spin the wheels till he got moving.
 
Back
Top