Jump to content
Ultimate Subaru Message Board

93 Legacy ej22 hunts then dies at idle


Recommended Posts

Determined today that the MAF is NOT the problem on the subievan. So here we go... diagnosis time.

The motor is new, as are most peripherals. I had suspected the MAF because it and the throttle position switch are the only old parts.

The engine starts when cold, but once it has run briefly it will start hunting at idle and will die if not kept above 1600RPM with the pedal.

It runs decently if moving,  but stopping for signals will put in into death spiral with hunting getting bad after about 20 seconds sitting.

Two locals think its either throttle position switch or coolant temp sensor. 

All hoses new and tight. Air control valve was new subaru installed less than 2000 miles ago.

Your insights would be greatly appreciated! 

Link to comment
Share on other sites

Bennie, Thanks, your thoughts are interesting, but all injectors are new Subaru, as is the Air control valve. (2000 good miles out of all parts).

Everything was working. no adjustments have been made.

The Coolant sensor may also cause a rich condition, as the ECU always thinks the coolant is cold. This would affect all 4 cylinders equally.

Yes am going to have to learn how to read codes. Do not have a functioning CEL on the dash, so need to dig into the ecu stuff under the back seat and figure that out. Have found the code reading tutorial so, that's a start. 

Thanks,

N.L.

Link to comment
Share on other sites

My normal 1st thing to check (including the TPS and IAC) is the CTS/coolant temp sensor as it can exhibit these symptoms. Sounds like it's running rich? The CTS doesn't always throw a code. 

Note: Don't confuse the gauge sensor w/the actual CTS sensor that tells the ECU what's going on....Google it if unclear.

Link to comment
Share on other sites

8 hours ago, wtdash said:

My normal 1st thing to check (including the TPS and IAC) is the CTS/coolant temp sensor as it can exhibit these symptoms. Sounds like it's running rich? The CTS doesn't always throw a code. 

Note: Don't confuse the gauge sensor w/the actual CTS sensor that tells the ECU what's going on....Google it if unclear.

Thanks,   Yes, it is running rich.

The coolant temp sensor is on the left side on the back of the coolant manifold- correct?

Gauge sensor is to the right of it also on the coolant manifold?

 

 

Link to comment
Share on other sites

3 hours ago, FerGloyale said:

Nuetral switch signal?  Often causes idling issues when rolling to a stop.

Also check air intake boot and all PCV hoses for any metered air leaks.

 

Thanks!

It actually runs OK for about 20 seconds before trying to die at stop lights.

Yes, have done so, all new and all tight with clamps. (had post MAF leak issue when it was first built, and it is similar except for now it is running OK at first before gaining a bit of heat).

This is a vanagon conversion, so the subie AT issues do not apply.

 

Best,

 

N.L. 

Link to comment
Share on other sites

My ej22 was acting up weeks ago when the rain was constant in socal.  It had similar issues of running fine with gas and idling poor.  Turned out there was water in the spark tube and it corroded the connector so it didn't clamp on the plug tightly.  Worth a check if you haven't done so.

Link to comment
Share on other sites

17 hours ago, NotLurking said:

This is a vanagon conversion, so the subie AT issues do not apply.

 

Best,

 

N.L. 

 

Yes I know it's a Swapped harness.  And that can be part of the Nuetral switch issue.  Assuming you've still got the VW trans, what was done to connect it to a nuetral switch signal to tell the ECU wether it needs to keep the engine idleing? It needs to see a signal to allow fuel cut when in gear, but still rolling at zero throttle (for engine braking)

TPS, VSS + Neutral switch position are what let the ECU wether to fuel cut for engine braking, or keep engine idling because you are about to be at a stop in neutral.

You can use the AT/MT discriminator wire at the ECU to tell the ECU to be for "AT" where it will not be as sensitive about keeping idle.  Engine braking may suffer slightly though.

Link to comment
Share on other sites

12 hours ago, FerGloyale said:

 

Yes I know it's a Swapped harness.  And that can be part of the Nuetral switch issue.  Assuming you've still got the VW trans, what was done to connect it to a nuetral switch signal to tell the ECU wether it needs to keep the engine idleing? It needs to see a signal to allow fuel cut when in gear, but still rolling at zero throttle (for engine braking)

TPS, VSS + Neutral switch position are what let the ECU wether to fuel cut for engine braking, or keep engine idling because you are about to be at a stop in neutral.

You can use the AT/MT discriminator wire at the ECU to tell the ECU to be for "AT" where it will not be as sensitive about keeping idle.  Engine braking may suffer slightly though.

 

FerGloyal, thanks for persisting.

Apologies for dismissing you without truly taking time to understand you. In truth, I agree, that providing the ECU with as many windows into what is going on around it the better things should perform power and efficiency wise. 

The vehicle was running well previously with good torque and power maybe not the best mileage, but at >4,000 lb. loaded it's a lot more to push around than a legacy. It is equipped with an original stock rebuilt  VW 3 spd. Auto trans with fresh torque converter. This problem has worsened over time, and (relatively) cold ambient air temps seem to make it worse. 

I think your points are good. This is the first I have heard about either the AT/MT discriminator, or the Neutral switch, and I need to confirm whether either is connected....

(Do you have links to instructions for setting up either the AT/MT discriminator, or the Neutral switch in a vanagon with stock auto trans?? )

You have touched on one of the disappointing things I learned this afternoon. Realizing that reading the codes was essential, I decided to look at ECU connected things for the first time to find out what was being thrown. Seems the trouble is that my experienced conversion mechanic didn't connect the CEL to the ECU. so... there is currently no way to extract the codes. (supposedly, the ecu "had a lamp built into it", but no.). Spoke with mechanic, asking where to pick up the CEL signal led to getting into the harness, which was made and installed by  mechanic. That led to the discovery that the VSS wire was labelled but not connected to anything. Both of these issues existed during "normal" operation, but you're right, they need to be addressed.

For the VSS, I know that kits are available to run off of the CV joint, so that's an easy step to complete. 

Regarding the CEL readout signal, do you know where to pick up a signal? (sigh, yes I really need to own a Legacy manual with schematics and troubleshooting info.)...

I did learn about how to put the system into a test/code mode etc., and noticed that it smoothed out quite a bit at idle and was less prone to dying in that mode.

Is this similar to a "Limp home" mode? would any fault be smoothed in this mode, but power and efficiency sacrificed? 

 

Update: MAF and Coolant temp sensor replacement with new subaru parts had no effect.  (well, at least I have spares now for the travel kit).

 

Reading other sites/posts lead back to something I fixed 4 years and a scant 2000 miles ago: replacing the ACU. I tried cleaning it but it was erratic at idle even after. Replacement was the key. I have ignored it because "It's 'new' it must be OK" illogically echoes in my head. Next step- remove and clean the ACU and see what happens. Could be that irrational "logic" got in the way of getting straight to the problem. Why would this new ACU  be dirty with so little mileage? Theory: For this install, the oil pan has to be shortened an inch, however, everyone says "don't worry you can still fill 'er up, there's plenty of space above the oil". OK, but... If the oil is too high it could be atomizing and much more than usual getting moved through system with crankcase vapors. Could be wrong, but I think that CC vapor does circulate through that part. Will look at plumbing tomorrow to determine.

 

To be continued....

Link to comment
Share on other sites

16 hours ago, nvu said:

My ej22 was acting up weeks ago when the rain was constant in socal.  It had similar issues of running fine with gas and idling poor.  Turned out there was water in the spark tube and it corroded the connector so it didn't clamp on the plug tightly.  Worth a check if you haven't done so.

NVU, Thanks for weighing in.  This has been worsening for a few months to the point where it's now unsafe to operate in traffic. Your experience is interesting, But I'm not sure what you specifically refer to. Do you mean the spark plug connector? At the plug end? that could progressively worsen.  I'll R&R the plug ends a few times.

This makes me think of another possibility... Fouled plugs. Will also take a look there. Probably worth doing the old test of running on 3 then switching to the next etc. to see whether one of the cylinders is dragging the rest down, could be an intermittent fuel injector. (in the old days, a clogged idle jet would be the culprit).

 

N.L.

Link to comment
Share on other sites

It's water getting past the rubber boots and collecting in the hole.  Just pull the plug wires and look at the clamp inside them.  Might as well pull the plugs and inspect while you're there.  Might be a good idea to blast the holes with compressor air before taking out the spark plugs.

JDM_SUBARU_EJ16_EJ18_EJ20_EJ22_SOHC.jpg

I found a picture, see the wavy areas where the plug wires are.  Water likes to collect up there and eventually makes it inside the hole.

Edited by nvu
Link to comment
Share on other sites

2 hours ago, nvu said:

It's water getting past the rubber boots and collecting in the hole.  Just pull the plug wires and look at the clamp inside them.  Might as well pull the plugs and inspect while you're there.  Might be a good idea to blast the holes with compressor air before taking out the spark plugs.

JDM_SUBARU_EJ16_EJ18_EJ20_EJ22_SOHC.jpg

I found a picture, see the wavy areas where the plug wires are.  Water likes to collect up there and eventually makes it inside the hole.

 

NVU, That's a good point. (and just yesterday, while changing out the coolant temp sensor, It was apparent that little bits of roadway debris collects on top of the engine).

There should be (could be) a wheel well filler or even a mud flap to keep crud and water from flinging into the engine compartment. Hmm, could probably sell a few of those if they existed!

Thanks, I'll be sure to pay close attention to those areas!

 

N.L.

Link to comment
Share on other sites

11 hours ago, NotLurking said:

 

 

Regarding the CEL readout signal, do you know where to pick up a signal? (sigh, yes I really need to own a Legacy manual with schematics and troubleshooting info.)...

I did learn about how to put the system into a test/code mode etc., and noticed that it smoothed out quite a bit at idle and was less prone to dying in that mode.

Is this similar to a "Limp home" mode? would any fault be smoothed in this mode, but power and efficiency sacrificed? 

 

 

 

CEL wire can be found at largest of the 4 ECU connectors.  Pin 19, which is 1 off of center on the bottom row.  Red w/Yellow stripe wire.  It is the ground for the lamp, so lamp must have 12+ with key on and ECU grounds to turn it on.

Test mode does lock in some parameters so cancels the closed loop normal operation.

Seems you should put the ECU into AT mode if it isn't.  IIRC this means grounding pin 20 of the second largest connector (3 in from end of bottom row).  Sorry can't find the color.

Link to comment
Share on other sites

6 hours ago, FerGloyale said:

 

CEL wire can be found at largest of the 4 ECU connectors.  Pin 19, which is 1 off of center on the bottom row.  Red w/Yellow stripe wire.  It is the ground for the lamp, so lamp must have 12+ with key on and ECU grounds to turn it on.

Test mode does lock in some parameters so cancels the closed loop normal operation.

Seems you should put the ECU into AT mode if it isn't.  IIRC this means grounding pin 20 of the second largest connector (3 in from end of bottom row).  Sorry can't find the color.

FerGloyal- Thanks, the CEL location is very helpful!  that's my next task, hooking up a CEL light and checking codes....

the AT mode info is also excellent to have, and I'll make sure that gets grounded as well.

will follow up with findings.

 

N.L.

Link to comment
Share on other sites

13 hours ago, heartless said:

you can find wiring schematics here: http://jdmfsm.info/Auto/Japan/Subaru/

Sadly, 1990-1994 are all lumped together in one folder under the Legacy/Outback section, but it is better than nothing at all, and since very little changed during those years it should help.

Thanks, Heartless, that's good info to have whether it's perfect or not. I also ordered a Haynes manual to not fly blind so much...

UPDATE: Have sorted the wiring and now have CEL lamp wired into dash, and have read codes (throwing 24 & 42) The Air Control Valve and TPS are the culprits.

Will remove and clean the almost new ACV in hope that that clears things up. Also noticed that the little hose that connects the ACU to the Throttle body is kinked. It's like 2" of hose between hard fittings that force it into an "L" shape. Will use a longer hose and put a loop into it to prevent kinking on re-install.

Suspect that swapping the idle switch has its adjustment goofed up now. Will attempt to properly adjust with feeler gauge before buying a new one :

Spec I found here: 1&2 terminals- continuity at fully closed throttle, continuity at 0.7mm open throttle, and open circuit at 0.9mm open throttle as measured at throttle stop.

 

For FergLoyal: the  Neutral & inhibitor switches (code 51), parking switch (code 52), and VSS (code 33)  are all waived codes under this install per CA. law 

I'm going to hook up VSS with aftermarket kit and see if it helps operation at all. It should, otherwise why would Subaru have used it? (Vanagons don't have a VSS input source as-built).

  • Like 1
Link to comment
Share on other sites

Do you know for sure what year and model your harness and intake are from?

 

The reason I ask is that there are actually 2 different IAC (Idle air control) valves used in the first gen legacies.

90,91 Manual trans models used a 2 pin type (no center pin in the 3 pin connector) That can only actively open the valve

All AT models and all 92-94 (and 95-98 2.2+2.5) use the 3 pin type that can actively open and close the valve.

You mentioned having replaced the IAC, are you sure you got the correct one?  And did you get it well sealed to the intake when you did?

Link to comment
Share on other sites

  • 4 weeks later...
On 2/18/2019 at 5:18 AM, heartless said:

fwiw - the haynes manual is pretty useless aside from the most basic of things..

the link I gave you will have much more useful information in it (and more accurate)

No kidding!!! my mechanic said theHaynes info on how to set the TPS is B.S.  (It did not work as he predicted). 

They focus on pins 1&2 and the initial on-off, which seems sensible, but...

The Factory manual completely ignores pins 1&2....

Got it set up by "osmosis", and it seems to be working fine now. No bucking, no stalling.

Link to comment
Share on other sites

On 2/18/2019 at 1:12 PM, FerGloyale said:

Do you know for sure what year and model your harness and intake are from?

 

The reason I ask is that there are actually 2 different IAC (Idle air control) valves used in the first gen legacies.

90,91 Manual trans models used a 2 pin type (no center pin in the 3 pin connector) That can only actively open the valve

All AT models and all 92-94 (and 95-98 2.2+2.5) use the 3 pin type that can actively open and close the valve.

You mentioned having replaced the IAC, are you sure you got the correct one?  And did you get it well sealed to the intake when you did?

Thanks for the note, yes, I'm sure it is all from a 1993 AT car, as I bought the whole (barely) running car and salvaged what I needed. the almost new IAC was not dirty, and moved freely. Did not test resistance. Yes it's all tight.  THE ONE THING about the IAC- it has coolant lines in and out. Simple, right? Well, Subaru no longer makes the tight little coolant elbow that goes between the IAC and the throttle body. Mine had a piece of straight hose which had become kinked... Should have used a longer piece and looped. Better solution: I found a Gates elbow in 5/16" that worked great once both ends were shortened appropriately. This part is considered a heater hose. Suspect this was what was throwing off the IAC outputs and causing codes.

So, the takeaways:  1) Fault codes must be hooked up and functioning for easy (er) diagnosis. 2) A functioning multimeter is also essential. 3) Aftermarket Manuals seem spotty at best, so beware. If unsure, ask the bros on US.org  4) Out of all who "guessed" at the problem The mechanic was right (it was also on FergLoyal's shortlist). He reminded me of this. I reminded him that guessing was still for fools!  Lessons learned. 

 

Thanks everyone for your help!

 

N.L.

  • Like 1
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...