90 Legacy 2.2 Start/Stall Problem

[matt4irish]

My friend's 1990 Subaru Legacy L Wagon 2.2 has developed a problem that is perplexing us. He was driving the car when suddenly it began to buck and then stalled. Was able to restart a couple times and limp the car home. However, once home it rarely starts. For the past several weeks, we have tried to assess the situation and repair to no avail.

 

First, he replaced fuel filter. We put a fuel pressure guage on and the guage shows 30psi when the key is turned and this pressure is maintained during cranking. The car seems to have spark in all cylinders. It turns and turns, but no start.

 

If it is left unattended for a few days, the car will start, run for a few minutes, but will ultimately warm and stall. When it stalls, it bucks roughly first, then dies.

 

We checked the temp coolant sensor and found it had rotted. We replaced the temp coolant sensor and the car started, ran for a good 10 minutes, but as soon as we gained the confidence to take it for a test drive, we made it down his drive and the car started to buck and stalled. Had to tow strap back to his garage and the same symptoms reappeared. Crank and crank, but no start.

 

The cranskshaft sensor shows varying voltage while cranking (believe this shows it is functioning). We can't seem to get a test on the cam sensor, but have replaced with another used unit, still nothing.

 

We swapped the MAF out of my fine running 96 Outback (it was same part #) and no change.

 

We think the car may actually be flooding itself out, but this is only due to the fact that at one point we pulled the #1 plug out and saw it was wet (after repeated cranking).

 

Any advice is very much welcomed. Thanks!

[Manarius]

I'd get a JECS MAF from another 90-91 Legacy, not one from a 96 Outback - that makes a big difference. Before that though try to clean the 1990 MAF with some rubbing alcohol and a Q-Tip (Be gentle about it). If that doesn't do it, I'd go with either a vacuum hose knocked loose or off completely, acting up TPS sensor (just clean it), or a plugged up PCV system.

[matt4irish]

I'd get a JECS MAF from another 90-91 Legacy, not one from a 96 Outback - that makes a big difference. Before that though try to clean the 1990 MAF with some rubbing alcohol and a Q-Tip (Be gentle about it). If that doesn't do it, I'd go with either a vacuum hose knocked loose or off completely, acting up TPS sensor (just clean it), or a plugged up PCV system.

 

Original Subaru MAF Part #s were identical on 90 and 96. Actually 90 Subaru MAF looked super clean.

 

We have checked vacuum hoses thoroughly, but will check again.

 

Swapped the PCV valves too, no changes.

 

This one has been a real bummer.

[Manarius]

Are they the exact same make (JECS)? The 90-91 ECU's are VERY temperamental.

[matt4irish]

Are they the exact same make (JECS)? The 90-91 ECU's are VERY temperamental.

Will check when I can get back to his place.

 

Wonder if this could be ECU issue?

 

No Engine Codes thrown, but OBD-1 often doesn't give much insight.

 

Thanks . . . all suggestions welcomed!

[Manarius]

Will check when I can get back to his place.

 

Wonder if this could be ECU issue?

 

No Engine Codes thrown, but OBD-1 often doesn't give much insight.

 

Thanks . . . all suggestions welcomed!

Have you reset the ECU after you installed each of these new sensors?

[matt4irish]

Have you reset the ECU after you installed each of these new sensors?

This may be a really ignorant question, but how do you reset the ECU?

 

We have disconnected the battery for 2-5 minutes each time. Does that do it?

[Manarius]

This may be a really ignorant question, but how do you reset the ECU?

 

We have disconnected the battery for 2-5 minutes each time. Does that do it?

2-5 minutes isn't enough. One can either play the battery dance (what you attempted to do) by pulling the negative terminal for 30-45 minutes or pull the ECU fuse for 30-45 minutes.

 

After that, put the car into neutral or park, turn off ALL accessories and start the car. DO NOT TOUCH THE GAS! Let it run for 10 minutes or until it gets to normal operating temperature. Once one of those two happens, turn off the car. The ECU has been reset at this point and the car is driveable.

 

I would reset the ECU first before you go and replace any more parts.

[matt4irish]

2-5 minutes isn't enough. One can either play the battery dance (what you attempted to do) by pulling the negative terminal for 30-45 minutes or pull the ECU fuse for 30-45 minutes.

 

After that, put the car into neutral or park, turn off ALL accessories and start the car. DO NOT TOUCH THE GAS! Let it run for 10 minutes or until it gets to normal operating temperature. Once one of those two happens, turn off the car. The ECU has been reset at this point and the car is driveable.

 

I would reset the ECU first before you go and replace any more parts.

Thanks . . . will do! Will let you know!

[Cougar]

From your posts it sounds like the engine is getting fuel and spark too the plugs so I would suspect the ECU is okay. I would check the codes for any clues to this problem and if you haven't checked the compression in each cylinder yet I would do that.

[Skip]

How old is the timing belt?

 

could it have a few teeth missing

 

allowing it to jump back and forth between

in time and out of time?

 

Might want to pull an outside outer cam cover off

and have a look.

[Legacy777]

The problem really sounds like the issues I had with mine when the MAF was going out.

 

I'd also check the wiring going from the MAF to the ECU.

[matt4irish]

From your posts it sounds like the engine is getting fuel and spark too the plugs so I would suspect the ECU is okay. I would check the codes for any clues to this problem and if you haven't checked the compression in each cylinder yet I would do that.

 

No ECU codes thrown. Compression has been between 165-180 in each of the four cylinders.

[Cougar]

If you are sure the compression is ok then I would suspect that the problem may be with one of the engine position sensors. I think they can go bad and not show a code if the output is low. You may be able to check the output with a digital meter using the AC volts function to see if there is a significant difference in the readings while cranking the engine.

 

If the sensors are ok then the suggestions by Skip and Legacy777 may be what you need to check.

[OB99W]

May be of interest:

http://endwrench.com/images/pdfs/Fuel.pdf

[OB99W]

[...]The car seems to have spark in all cylinders. It turns and turns, but no start.

 

If it is left unattended for a few days, the car will start, run for a few minutes, but will ultimately warm and stall. When it stalls, it bucks roughly first, then dies.

[...]

We think the car may actually be flooding itself out, but this is only due to the fact that at one point we pulled the #1 plug out and saw it was wet (after repeated cranking).

Okay, it has spark, but have you determined that the spark is sufficient or there all the time? Besides the crank and cam sensors already mentioned, perhaps the ignition itself is failing as it warms up; that could make sense given the wet plug(s).

 

You could attach a timing light to get an idea of what's going on. Most timing lights will trigger even during cranking, and if it appears to be sporadic or the timing seems wrong that might provide a clue.

[matt4irish]

Long delay in evaluating this as the friend lives in the country and the car's been buried in snowdrifts.

 

I replaced fuel pump today just to be sure (remember, I said the car was flooding, but I had a line on a cheap fuel pump). No change.

 

Car started and sputtered like crazy for a few seconds before dying.

 

Riddle me this . . . could the timing be off?

 

I know that's an obvious question, but take it one step farther.

 

If the timing were off slightly, could that cause the car to act flooded?

 

And then, if the car were allowed to sit for a protracted period (days and days), could it restart and idle strongly for a few minutes only to sputter and die when put in gear and then accelerated?

 

Really confused by this!

 

Thanks in advance!

[matt4irish]

HELP!!! See last post regarding timing question . . . looks like this belt has been on 6 1/2 years/80,000 miles. Can't visually inspect until weekend, but are my questions in prior post possible?

[Manarius]

If it was timing, the car should run horribly all the time. However, stranger things have been known to happen (the guy with the chunk off his crankshaft comes to mind).

[Kilroy]

It really does sound like a MAF or ECU problem. It's exactly what happened when my MAF started going south. You can pick another ECU up at a wrecker for about $10 too.

[Skip]

Matt4

I agree with Manarius, it would run bad always, not just after a cold start/ wrm up if the valve timing were off.

 

You sound like you know something about engines,

remember the carb days ?

If the choke sticks on, the engine will run fine cold then

as it gets warmed up it will tend to "load up" and

die from gas fouled sparking plugs.

 

With this engine there is no choke as the

fuel injection thermosensor tells the ECU the engine is warm

and to adjust the air fuel mixture (like the choke slowly opening on a carbed car).

 

Possibly the thermosensor is bad but not bad enought to trigger the engine check light.

 

The ECU never adjusts the fuel mixture as it warms,

it "loads up" (wet plug) and stumbles DW (Dead in the Water)

 

Or

Left field off the wall hit -- the fuel pressure regulator is

KaPoot and the injectors are feeding way too much fuel

all the time and when it gets warm

too much fuel fouls the plugs

[matt4irish]

OK . . . here's the latest. Pulled timing belt covers, no issues. Car actually started today and idled fine for awhile, but started to buck and backfire at about 1800 rpm. Though it was cold (30 degrees) and windy (30mph) and I was outside, I got a chance to really check things out and a couple things jumped out.

 

First, the hose from the crankcase that goes to the T or Y valve that connects to the air intake and PVC valve was severely cracked. Probably lets air suck in. Doubt that's enought to cause the issue, but will get a better hose in there!

 

Second, I noticed that one of the fuel lines was seriously "smashed". This metal fuel line runs under the intake manifold on the passenger side (right under the two pipes of the manifold that curve downward toward the fuel injectors). It makes a 90 degree bend behind the manifold and connects to a small rubber hose that goes into (or comes out of) the fuel pressure regulator. Honestly, it looks like a beer can would if you stepped on the middle of it. Not sure how a line like this could get smashed. Wonder if it collapsed in some way.

 

Finally, I pulled apart the IAC and gave her a good clean.

 

Car wouldn't restart, but we may be on to why!

 

A few questions:

 

1) Do you think a heavy restriction in this line could be the issue? Maybe in the "smashed" state not enough fuel really flows, or some sediment or something broke loose and lodged across this restriction.

 

2) Could this theoretically "unbalance the fuel pressure", making it appear normal when connecting a fuel pressure guage to a T fitting by the fuel pump, but leaving the passenger side cylinders (1 & 3 I believe) starving while resulting in lots of fuel (loading) to the drivers side cylinders (2 &4)?

 

3) Or could it simply be a volume game. That's why the car will occasionally start and idle for awhile, but stall under a load (starving in cylinders 1 & 3 near the fuel pressure regulator)?

 

4) Is the crankcase hose at all likely to be a cause?

 

Not looking forward to getting to this as I imagine the whole intake has to come apart to replace the fuel rail (or fuel line). Welcome any input. You all have been great help!

 

:banghead:

[Cougar]

I assume you are going to fix the damaged items you have found so far. Have you tried spraying some starter fluid into the intake to see how the engine responds to that? You may need to flush out the fuel system as it may be contaminated.

[Skip]

I just went out and took the below picture.

 

Hope it helps.

As best as I could, I followed this line and it is the

"return line" to the fuel tank and if restricted woulld affect all injectors making the FP abnormaly high.

 

Please re read my earlier post and note the fuel injection

pressure regulator thought.

 

for your questions

1) maybe see above but pretains to all injectors

 

2) why didn't you just pull the plugs and see which where

wet fouled (gas soaked) and which (as you ask) were lean and dry (pure white)???

 

3) see number two

 

4) Leaks are never good but a lean condition would be

present see number 2 again

[Skip]

And as a final thought.

 

You seem to have a firm handle on doing this

you way.

 

I will humbly resign from this thread and

allow you to your own/ and other devises.

 

My best of luck to you and sorry i could not help.