90 Legacy 2.2 Start/Stall Problem

[matt4irish]

Skip,

 

Please don't resign...hope my slowness in this isn't frustrating you. The primary issue is that the friend's house is 25 miles away and he has he car outside. The weather here stinks, so it takes extra motivation to get there!!! Plus, I need to keep the home life in check too!

 

Your picture is exactly the correct fuel line . . . the blockage is about 6 inches further downstream after the line makes a 90 degree turn toward the front of the car passing under the intake for cylinder 3, then 1, I believe.

 

You are correct, I should pull the plugs 1 x 1 and check.

 

I think your theory about abnormally high pressure is probably right on and that high pressure may be most noticable at injectors 1 and 3 since they must be just upstream of the FPR. I have seen the #1 plug very wet in the past. Need to pull the other 3 when the car stalls.

 

In the meantime, do you think I could cut the damaged part of the line and put in a securely clamped rubber fuel hose across the damaged area under the intake or would there potentially be too much heat?

 

I think to replace this line, I would otherwise have to pull the entire intake.

 

Thanks again, you are a TERRIFIC help!

[matt4irish]

Some progress today. Routed a rubber fuel line hose from the Fuel Pressure Regulator output around the rear of the engine compartment to the return line near the fuel filter. The car started and idled great and for a LONG time (15 minutes). Still bucked and sputtered when we pressed on accelerator, but the car was much harder to stall. Finally after about 20 minutes of idling, we were able to accelerate crisply to 2500 rpms twice in succession without bucking. The car stalled a few minutes later. Would not restart.

 

Starting to wonder if Skip's fuel pressure regulator might be the next part we consider eliminating by substitution.

 

Any thoughts?

 

Thanks again.

[Cougar]

To see if the engine is stalling due to a lack of fuel try spraying some starter fluid into the intake. If the engine responds to that then you know you are not done with the fuel system. Perhaps the fuel pump is getting weak after running for a period of time or there is a electrical problem with the wiring to it.

[matt4irish]

To see if the engine is stalling due to a lack of fuel try spraying some starter fluid into the intake. If the engine responds to that then you know you are not done with the fuel system. Perhaps the fuel pump is getting weak after running for a period of time or there is a electrical problem with the wiring to it.

 

 

Would you believe this car is still dead? I've been away for awhile and my friend had a new baby and took in a foster teenager . . . man that's a double whammy, but a great excuse for not doing any more diagnosis.

 

So, tonight he went out and started the car. She started right up and ran for about 10 minutes, came up to temp just fine, and then did the old sputter and die. He actually pulled all 4 plugs. The plugs ALL appeared DRY.

 

Then, he pulled each injector connector and measured the resistance on each injector. The values are as follows:

 

Passenger Rear (3) -12.8 ohms

Passenger Front (1) - 16.4 ohms

Driver Front (2) - 17.4 ohms

Driver Rear (4) - 12.7 ohms

 

These are a little off the spec. of 12 ohms, but not dramatically. We had swapped fuel pumps, so I think the pump is good, but can't tell you for sure that pressure is maintained at the stalling point as the pressure guage is at my house (27 miles away). We will likely this next, or anything else anyone suggests.

 

Leaning toward ECM, but not ready to go there yet. If so, it is the 22611AA380 ECM. Anyone got one close to Indiana?

 

Thanks in advance for any and all ideas!!!

 

Matt:

:-\

[Cougar]

It sounds like the fuel pump circuit is cutting out. Try checking the voltage on the pump leads to see if the pump is getting power when this trouble happens. The fuel pump relay could also be causing this trouble.

[matt4irish]

It sounds like the fuel pump circuit is cutting out. Try checking the voltage on the pump leads to see if the pump is getting power when this trouble happens. The fuel pump relay could also be causing this trouble.

 

Strange thing is that after it stalls, a no start condition remains. It will crank and crank, no start, BUT, we can hear the pump whir when the key is turned before the cranking.

 

It is a real puzzler.

 

We really appreciate your continued guidance!

[Cougar]

As I previously suggested did you try spraying some started fluid into the intake after this trouble happens to see if that gets things going? If that works then you need to check to see if the injectors are working. Then check the fuel pressure and see if that is good.

[matt4irish]

Will have him try to spray into the intake to see if it will fire.

 

In your experience, will one bad injector take the whole system down?

 

And, what is the best way to check injectors? I noted the resistance for each injector, should the higher readings concern us?

 

Aren't the injectors and their firing controlled by the ECU once the car warms and the ECU takes over?

 

Forgive my plethora of questions . . . kind of new at tracing "lack of fuel" issues beyond the dead fuel pumps I have seen in the past.

[Cougar]

One bad injector will not kill the engine. You did good in checking the coil resistances and it is a pretty good indicator that it will work. Listening to them while they are working is a good check also.

 

The injectors are always controlled by the ECU whether if it is in open loop or closed loop mode. They are usually turned on by the ECU completing the ground side of the circuit. A good way to test the operation is use a LED test light called a NOID light. Using a analog meter to measure the voltage across the injector coil will work also, better than a digital meter.

 

Questions are good and that is how we can learn.

[matt4irish]

One bad injector will not kill the engine. You did good in checking the coil resistances and it is a pretty good indicator that it will work. Listening to them while they are working is a good check also.

 

The injectors are always controlled by the ECU whether if it is in open loop or closed loop mode. They are usually turned on by the ECU completing the ground side of the circuit. A good way to test the operation is use a LED test light called a NOID light. Using a analog meter to measure the voltage across the injector coil will work also, better than a digital meter.

 

Questions are good and that is how we can learn.

 

Your answers are helping and I do feel like I'm learning. He's not home tonight so he couldn't spray in throttle body, but this process made me think of another question.

 

It does seem fuel flows easily during idle and then something happens. I'm confused a bit about the function of the fuel pump relay (FPR). Is there any time that the ECU sends a signal to the FPR and tells it to cut the pump?

 

It definitely seems like something happens that changes the fuel . . . either the supply is cut (like if FPR circuit opened) or the injection process stops (not sure how or why).

 

Having a fuel pressure guage on while we duplicate this will help, but I know we tested in past and the fuel pressure remained stable at 30-32 psi.

 

Feel like we're getting warmer, but would love your thoughts.

[Cougar]

The fuel pump relay is just like any other relay. It is there to handle the switching of the power to the circuit. It takes a small amount of current to control a higher one. If there is a problem the relay can be turned off by the ECU. The ground side of the injectors is also controlled by the ECU. So things can be shut down that way also. I'm not really sure what side is turned off when there is a problem seen by the ECU. By checking voltages you can see what is working and what isn't.

[matt4irish]

The fuel pump relay is just like any other relay. It is there to handle the switching of the power to the circuit. It takes a small amount of current to control a higher one. If there is a problem the relay can be turned off by the ECU. The ground side of the injectors is also controlled by the ECU. So things can be shut down that way also. I'm not really sure what side is turned off when there is a problem seen by the ECU. By checking voltages you can see what is working and what isn't.

 

 

He just ran out and it started right up, ran 10 minutes, then stalled under load. He sprayed carb/choke cleaner in throttle body (no starting fluid on hand) and tried to start . . . nothing. Sprayed again, nothing. Believes there is still spark. It almost seems that car might be developing these problems when ECM moves from open loop to closed.

 

Does that seem plausible and how would we know when it ECM switches modes?

[Cougar]

I'm not sure the engine will run on the cleaner. I would get some starter fluid and check it.

 

You may have something there about the ECU. I'm not really sure how you could determine when that happens. There may be some action with a valve or solenoid you could monitor with a meter and see when the ECU goes to closed loop.

[matt4irish]

I'm not sure the engine will run on the cleaner. I would get some starter fluid and check it.

 

You may have something there about the ECU. I'm not really sure how you could determine when that happens. There may be some action with a valve or solenoid you could monitor with a meter and see when the ECU goes to closed loop.

 

Cougar . . . Great news!!!

 

Replaced the ECU tonight, and guess what . . . the car started and ran and ran and ran! Looks like the problem all along was an ECU!!!!!!!!!!!!!!

 

Thanks for all your help and persistence.

 

I'm a big fan of car-part.com. That's where we found the appropriate ECU and ordered. Once it was replaced, the car ran fine. Only $25 +$10 shipping from a yard in PA!

 

Hooray for ultimatesubaru.org & car-part.com!!!

 

Thanks again,

 

Matt:headbang: :headbang:

[2.5GL]

...

[Cougar]

Right on! Glad you got it fixed and running well. Another Soob owner is down the road and smile'n. Your welcome for the help and thanks for the feedback.

 

Drive on Matt.