Jump to content
Ultimate Subaru Message Board

EGR puzzle for the late night crew...


Recommended Posts

Okay... I had a CEL telling me by code to check my EGR solenoid, which I did. I replaced it, and when I did, the brittle hose cracked. i drove it maybe 20 miles as necessary, with the CEL coming on only toward the end of this. Since then it has sat. I figured I'd need to fix up the plumbing, so I cruised by the shop today. they didn't have an exact match for the hose I needed, so I took the next closest thing, assuming it would work.

 

Looks like it was about 3/32" too big. Fit a little loose, but I sinched it down. Then, after I cleared the computer (using those fancy green connectors by the hood hinge (it's a 92 Loyale by the way), the CEL came right back up when started. Now I am concerned,... Is it the loose-ness of fit, that is disturbing the vacuum and making this thing light up, or did I fry another EGR by running it with a bleeding hose? I hope it ain't the latter, as that would make me feel more than a little bit dumb.

 

I haven't gotten round to rechecking the code that the EGR is sending out. I'm working with not a lot of daylight once I get home from work, so I thought you all might be able to enlighten me a bit as to the source. Will a more appropriate size hose fix me up, or am I looking at another EGR replacement?

 

As always,

Thanks

belizeanbus

Link to comment
Share on other sites

The CEL and associated code don't actually have anything to do with te EGR. It actually refers to the EGR solenoid which is located in the intake manifold. They are a very common failure and a quick search , or even browsing through the first 4 pages of this section of the forum will yield at least a few links with some info on EGRs, Purge control solenoids and CELs in general.

Link to comment
Share on other sites

If you need an EGR solenoid, check with Whosiwhatzit, in Klamath Falls. He has one that I was going to buy from him. I found one yesterday in a JY locally, but was still going to buy his and put it on a shelf for a spare. But, if someone is in need of one I hate to have one sitting on a shelf collecting dust.

Link to comment
Share on other sites

don't get me started on EGR CEL issues...

 

 

the solenoid itself should be fine.

 

 

My car (88 GL) likes to throw the code no matter what I do*. lol, if I unplug the solenoid, then start it, the CEL goes OFF for a while :confused: then comes back on.

 

eventually when it warms up here I need to trace all the wires back to the ECU and see what I find.

 

another thing you can try if you just want the light to go off... tape a 33ohm resistor into the plug instead of the EGR solenoid. In theory all the ECU sees with either that or the real solenoid is a 33ohm resistance :brow:

 

 

*hook up correctly, clear codes according to the FSM, it still comes back on.

Link to comment
Share on other sites

Oh, lordy lord lord lord lord ....

 

Let us know when it warms up there and you do check everything back to the ECU. I'm afraid I'll still be wondering about this too.

 

I'm starting to suspect .... everything.

 

Have you looked at the tech note I quoted elsewhere, from the Usenet alt.autos.subaru group, reporting a manufacturing defect in the EGR solenoids? The author -- a computer guy with a good reputation going back years -- took one apart and found that inside, where the coil wires should have been soldered, they were just wrapped a quarter of the way around the contact point and the whole thing shrinkwrapped. He said after a few years, enough oxygen gets through the plastic to corrode such a 'connection' (and he speculated on how many solenoids made this incompetent way are out there going bad).

 

He suggested either carefully disassembling it and soldering the connections, or just swapping til the light goes out, basically.

 

I thought it was bad when I took apart a wall in my house, and found out the carpenters had just stood some of the studs up and forgotten to nail them to the top plate (in earthquake country). But forgetting to solder electrical connections worries me almost as much.

Link to comment
Share on other sites

Nice to see another Californian on the board to sympathize with me over the EGR stuff. So, the thing is, I don't much care about the CEL being on, but this is a new-to-me Soob, and it passes emissions just fine on all counts but one. Here in Cali they fail you if the CEL is on.

 

So: can you trick the system? I notice that one of the sets of wires that is used for code reading/resetting the CEL, if left connected, shuts down the CEL when the car is running. Can I just plug them together and head to the smog-shop? Or are they on to that trick?

 

The resistence of the the EGR Solenoid that was on there originally was way off. Obviously bad. Seems it had been swapped out before, because of a jimmied electrical connection. The new EGR Solenoid (used actually), I didn't test the resistence on, though I'll get to this as soon as I don't have 45 thousand things to do (can't wait for the weekend). It kept the light off at first though... now it comes right on everytime she's started. I prefer a simpler solution to trying solenoid after solenoid until one works. That's a lot of cash to be throwing here and there. But one way or another I need to get this thing to pass smog.

 

belizeanbus

Link to comment
Share on other sites

So: can you trick the system? I notice that one of the sets of wires that is used for code reading/resetting the CEL, if left connected, shuts down the CEL when the car is running. Can I just plug them together and head to the smog-shop? Or are they on to that trick?

 

you might be able to stick a 33ohm resistor in the plug. I can test that theory tonight when the car with a working EGR circuit gets back.

 

you can get the resistor at radio shack for cheap.

Link to comment
Share on other sites

Well, if anyone is going to trace this problem back to the ECU, please share what you learn. In this case, I swapped the EGR Solenoids between my two loyales. They both work in one, but the other throws code 34 with either.

 

The resistor idea is a good one, but that leaves the Solenoid unplugged and non-functional... which I would assume might lead to some performance issues and maybe emissions problems that would keep it from passing smog... am I right? I don't know much about auto electrics...

 

What about leaving the green wires connected? Under the hood, by the hinge, there are four wires, two with white connectors and two with green. With the green ones connected, the check engine light seems to stay off. Is there a problem with just leaving these connected?

 

Any other interesting ideas are certainly welcome.

Thanks.

Link to comment
Share on other sites

If they are both the same fuel system (SPFI) the computers should be swappable, and if the issue is the computer itself, you could swap the good one between cars to pass emissions

 

or unplug your CEL :D

 

I think in your case, there is probably a problem with the computer (that is how it flows in the T-shooting chart in the FSM) - maybe start looking for a cheap JY replacement

 

I had a T-code for the CAS and no IG-pulse after I did my conversion, turned out to be a bad computer!

Link to comment
Share on other sites

That last response has a couple of the type of answers I'm looking for. Thanks... Perhaps someone could spare a moment for a couple more or less inane questions...

 

1. where is the computer located? Is it in the engine compartment or is it the "box" under the dash that flashes codes?

 

2. how to unplug the cel?

 

3. What should a used 'puter cost... I was just researching some other threads and was seeing some big numbers being thrown around... I'll check the marketplace and the local boneyards, but I'd like to have an idea going in...

 

First, I'll just swap them between cars to see if that solves it... Both are SPFI, one '90 and one '92. the '92 is throwing the code.

Link to comment
Share on other sites

That last response has a couple of the type of answers I'm looking for. Thanks... Perhaps someone could spare a moment for a couple more or less inane questions...

 

1. where is the computer located? Is it in the engine compartment or is it the "box" under the dash that flashes codes?

 

2. how to unplug the cel?

 

the computer is the box with the flashing light under the steering column. If you do replace them, make sure it's bolted back in - apparently if it isn't grounded it will fry.

 

For the CEL, you can take the front of the dash off and unplug the harness going to it, or take that bulb out. Either way involves taking the front of the dash off (not hard).

 

I may trace the wires back to the ECU next week when I get back home.

Link to comment
Share on other sites

85Sub4wd -- would you explain what this means and how you checked?

 

> a T-code (trouble light code)

> for the CAS ( )

> and no IG-pulse ( )

> after I did my conversion, turned out to be a bad computer!

 

My 88 spfi 4wd got a new computer just before I bought it -- the shop had said that "the computer is not sending a ground signal to the sensor" -- which only fixed the problem for a few days, long enough for me to buy the car and drive it about 50 miles.

 

So -- if you know how to check this problem, I'm guessing that an "IG-pulse" is something to do with the reference ground signal provided by the computer?

 

Anyhow, if you can spell it out -- I've only owned a Subaru for, um, nine days now.

 

----BUT, I took the boiled down summary info I've collected here to the local Subaru shop that's now working on it for me, and they were impressed. They didn't know some of this stuff.-----

Link to comment
Share on other sites

85Sub4wd -- would you explain what this means and how you checked?

 

> a T-code (trouble light code)

> for the CAS ( )

> and no IG-pulse ( )

> after I did my conversion, turned out to be a bad computer!

 

My 88 spfi 4wd got a new computer just before I bought it -- the shop had said that "the computer is not sending a ground signal to the sensor" -- which only fixed the problem for a few days, long enough for me to buy the car and drive it about 50 miles.

 

So -- if you know how to check this problem, I'm guessing that an "IG-pulse" is something to do with the reference ground signal provided by the computer?

 

Anyhow, if you can spell it out -- I've only owned a Subaru for, um, nine days now.

 

sorry I was vague,

I used fragments of the 1989 Factory Service manual, which can be found here (look under EA82)

 

http://www.finleyweb.net/default.asp?id=142

 

it goes through how to troubleshoot the computer in the second set - sorry I did not think to post the link sooner - realize that your car will probably be slightly different, but it should still be very close

 

the "IG-pulse" is the 0-5v (I think that is right - it may be 0-12v) signal that goes to the ignition amplifier transistor thingy I can't remember the proper name of mounted below the ignition coil - that bracket must be grounded well

 

the CAS stands for Crank Angle Sensor which is the distributor - however on SPFI cars, it optically senses the engine speed and position, rather than time to fire (that is determined by the computer - gives MUCH better advance curve than normal disty) - my computer was bad, and it kept throwing a false code that my CAS was bad, when it wasn't - got three and tried them before I found the REAL problem (the computer)

 

whatever you do, whenever you mess with the computer (pulling it or sensors), make sure the battery is disconnected so that you do not fry it, and make sure the retaining bolts for the computer are TIGHT before reconnecting the battery (if you have removed the computer) or else you will fry the computer - I fried two that way before I figured that out - that said, make sure your computer case, engine, and related components are WELL grounded - this is critical to proper operation

 

Good Luck, Subarus are great cars, and I hope you find the gremlin so you can enjoy yours!

Link to comment
Share on other sites

In repsonse to the above... how would you know if you had fried your computer?

 

Next... perhaps someone could explain the solution I have found to my 92 thowing the 34 (EGR) code... I swapped computers between my 90 and my 92 and now both are happy. No codes. The computers have different part numbers on them, which I could report if anyone is interested. I initially put the 90 one on the 92 (the car which had been throwing the code and on which I had replaced the EGR Solenoid). The previous EGR Solenoid really was bad... I checked the resistence... The codes disappeared and that was that. I figured if the 92 one, which had been throwing the code, in the other car it would still throw the 34 code, no matter what the condition of the the EGR circuit.... But it works fine...

 

So yeah... I can't explain it but maybe someone else can...

 

And now I'm thinking, maybe the the 90 one (which is now on the 92) is fried and that's why it's throwing no code... so, how would I know?

Link to comment
Share on other sites

In my case, I knew it was fried because neither one would send the ignition pulse signal to the coil, and both were giving me a bad CAS error, but I was using what were known to be good CAS sensors. I swapped the third computer in and it ran fine, started right up and all.

 

The later model computers seem to have wider operating parameters, and better adaptability. I had an earlier model one (from an 88 or so) in my car for a while and was running rich - went through the whole TB/injector clean, new O2 sensor, and problem persisted. Swapped in one off a '92, problem solved. - I think they improved the later model computers.

Link to comment
Share on other sites

Anyone got part numbers to compare?

 

THis one came off my 1988 SPFI GL 4wd wagon (said to have failed and not be giving a good ground signal to the EGR solenoid):

MECF-022-4E (printed) with stamped numbers: 4668 on the narrow end.

 

On the flat side is a label with a bit "42" and "Hitachi Ltd. Tokyo Japan"

and printed in boxes on the label are the numbers

22611 AA392, MECF-022, then a stamped 8809.

 

This is the label on the box from which a replacement Subaru ECU came:

 

Subaru Parts, FUji

Unit AY-EGI CONT

22611AA398

QTY 1 94.2 S

 

And another label:

M80650515

 

(This could still be a very old ECU, the tape on the box was so dried out it was coming off).

Link to comment
Share on other sites

About frying comptuers --

 

 

I do see warnings here and there about being very careful to always disconnect the battery before removing either any sensor or the ECU itself, to avoid frying ECUS -- someone posted he'd fried three before figuring that out!

 

ANyone got experience or references on doing that the right way?

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...