Jump to content
Ultimate Subaru Message Board

Recommended Posts

For the last month, the Check Engine light on my 94 Legacy 4wd AT sedan has been illuminating most times that I drive. The light is never on for the first minute after starting the engine cold but rather usually comes on before I have driven a mile or two and then will latch on solidly until I turn off the ignition. If after the light comes on, I turn off the engine within the next 10 to 20 miles, it will usually turn back on within a minute of restarting the engine. However at least some days after running beyond that mileage, it stays off. I am certain it is not due to a mechanically intermittent connector or wire nor does it seem to be heat or cold related.

 

Unfortunately I cannot read out any engine fault codes via the flashing CE light process as the light never turns on at all although I did perform both actions described with either the two black or two green connectors at:

 

http://www.surrealmirage.com/subaru/engine.html

 

That includes the test with the two green connectors mated that should have turned on the light but did not. And have disconnected the battery to bleed off any old codes or weirdness in that system. I also have the Engine and Electrical sections of the original shop manual that includes the schemtaic. Thus verify the wire jacket colors on the noted connector were correct to positively id what I was doing. I have been working on complex hardware electronic repair and testing for three decades so am confident I performed the reading functions correctly as the process is rather simple. Of note test codes used to be readable as recently as 18 months ago as a Subaru dealer did so and then a few months later a smaller repair shop complained that they could not read out my codes but at that time I suspected that was due to their method. So lacking ability to read codes, this is why I'm hoping someone might recognize the above symptoms and narrow what I might focus at.

 

My old sedan has 274k miles and does have some other problems. A right rear cam seal is leaking minor oil and the front left cam seal is leaking a slight amount so I need to add a pint of oil about every 1000 miles. Two of the spark plugs holes stripped a few years ago and helicoils were installed by some pros. At the time the dealer worked on the car, a thermostat and water pump had gone bad. Sometimes after driving down a steep mountain slope, upon leveling out, oil seems to briefly move into a cylinder and then smoke comes out of the exhaust for a short time. Otherwise it has been relatively healthy and in fact I always pass the smog tests here in California easily. If I don't receive any pointers, will probably bring it down to someone with experience, maybe a dealer, and have a general tuneup to eliminate obvious problems. The car is however beyond the end of its expected life so I am not willing to put much into repair or mainenance.

Link to comment
Share on other sites

i'm not sure what would cause that. if it runs fine is it an option to just leave it like that for now? black tape or unplugging the wires will rid you of the light.

 

the rear cam cap oring is really easy to replace. two bolts, a couple minutes, and done. the front obviously requires timing belt removal.

Link to comment
Share on other sites

On both the Green(test), and the Black(read) connector sets there is one wire that is from ECU, and it's mate is a ground. That's it......just grounding each pin, signaling to the ECU to test or read codes.

 

 

Is it possible that the grounded side of the connectors is not actually grounded? And if there is a break in the ECU grounding harness....it could contribute to causeing the code in the first place.

 

The ground side of each connector is the Black/blue stripe IIRC. Whatever the color it is the same in both sets......the signal wires are the different colors.

 

Try connecting the signal side directly to a ground, instead of through it's mate. See if you get the codes.

 

 

*I have a long set of wires with mates for the Read and Test connectors on one end and alligator clips on the other. I got tired of subaru code retrieval yoga.

Link to comment
Share on other sites

I am not a mechanic so take the following story as just my experience:

 

My previous car was a 91 Chevy Camaro Z28. Every time I drove the car for a sustained period - at least 10 minutes - at highway speed, the check engine light would come on. The car never missed a beat and other than the light on the dash you would never suspect that anything was wrong with the car. This was a pre-OBD2 computer in the car. No code was ever stored in the computer. After shutting off the car the light would be out and not return until the next excursion on the highway.

 

When I related my story to a mechanic friend he told me that older computers could easily get soft errors that would not be stored. I was told to drive the car to his service center with the light on and to not shut the car off when I arrived. He was able to diagnose a pin sized hole in a vacuum line by pulling the soft code from the computer.

 

Steve

Link to comment
Share on other sites

grossgary, yeah I bought an oring a few months ago but haven't bothered to put it in yet since the leak is trivial.

 

Gloyale, thanks for that useful advice. On my page 19 schematic the single contact green connectors called Read Memory are (B54) BR black with small red stripe and (B55) RB red with small black stripe. B54 goes through some bulkheads into page 21 where indeed it shows a ground. B55 just goes to C12 on the Multipoint Fuel Injection Control Unit that is nearby under the dash with a mob of yellow connectors going into it.

 

The single contact black connectors called Test Mode are B53 also BR that routes directly to B55 plus B52 Or orange that routes to C13 on the MPFI CU and elsewhere. Besides verifying the wire colors on those connectors matched the shematic including their routing to the correct MPFI CU pins, I also ohmed out that B53 connects to B55. However did not consider directly grounding B54 as you suggest but will do so today since indeed an open in that wire network to the ultimate grounding point would cause such a fault. If the light remains off, it could mean the MPFI CU input circuit for the grounding has been fried.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...