Cougar

I would check the fusible links for a loose connection problem. It may also be an intermitant relay or a bad connection somewhere.

Here is a link to the site: http://techinfo.subaru.com/html/index.jsp

I had this kind of problem with my '88 GL-10 turbo wagon. Replacing the fuel pump fixed it. I suspect it was your pump also that really caused the problem.

Now you know about the Subaru experience Holly. Enjoy the new ride and drive on. Welcome to the forum.

Good story but pretty scarey.

After reading the original post again I now think that there is a headgasket problem as was pointed out by another poster.

An intermitant relay is possible, especially since you say the problem clears after restart. This would point to dirty contacts in the relay. You could try bypassing the leads of the relay to verify this. I think the relay has two set of leads that run through it.

As far as I know, it is not a problem to drive it this way but should be repaired as soon as possible.

You may be able to prove out the injector by tieing a resistor to the connection that runs back to the ECU for the ground connection. I would guess using around a 10 ohm resistor may work. Touching the other end of the resistor to ground should make the injector turn on if the resistor value is not too large. If you are not seeing drive signal pulses at the injector though you will still need to check other signals and votages at the ECU. Something is missing. I would check the CAS and the ignition signals going to the ECU as I mentioned earlier. As another poster mentioned, it may be beneficial to go to your local library and see if you can get a copy of a wiring diagram for your model car. This kind of problem is one of the most difficult to troubleshoot I know of for these cars, since it requires input from other areas to the ECU to make the injector work.

I agree about the fusible links. They are in the black box mounted on the coolant reservoir. There could be a loose connection problem also with one of them. The smaller wire that ties to the positive battery post is what supplies the power to the fusible links and the fuse panel. Make sure that has good connection at the battery. If the fusible links are ok then check for internal corrosion of the wire at the battery connection. Having a test light to check for voltage would be real helpful with this problem.

Subpennyman, Did you read my last post to you on this in the other thread? Here is a copy of it in case you missed it: Yes, a scope would be better to see the pulses. It's just that not many people have one on hand or know how to use one so I didn't think about it. By all means use it if you would like to. You shouldn't need to have the injector tied to the lead to watch for the pulses. Since I don't know what the second set of higher resistance leads are for I thought it may be good to leave them in the circuit. Having them disconnected may change the test results. It sounds though that the ECU is not generating the pulses to open the injector. I think the ECU makes a ground connection to turn on the injector. The ECU needs to see the ignition pulses from the distributor to do this I believe. It may also need to see the signal from the crank angle sensor. I'm not sure what lead carries the ignition signal to the ECU but you may be able to find it using the ohmmeter. It would also be good to check the crank angle sensor signal also. The sensor is located near the flywheel I think.

Ok, help is here. Lets get to work. Do you have a fuse number for the fuse that is blowing? When you mention cluster, I assume you mean instrument cluster. Do you have an ohmmeter to work with on this and know how to use one?

A timing belt is used to drive the cam shaft(s), usually there is two of them, from the crank shaft. They are located in the front of the engine. If one breaks it will disable that side of the engine. If the distributor is run from that side it will mean no spark to any of the plugs also. It sounds like you may not be getting spark so I would check that first. It could also be due to no voltage getting to the ignition so check that if you don't see any spark.

For your coolant leak I would check the intake manifold gaskets and the hoses connections for a small leak under pressure. The pros have a chemical they can put in the coolant and use a UV light to look for signs of an external leak. I think there is a small hose near the water pump that is easy to overlook.

Have you done the normal tune up procedures? You might also try plugging the vacuum line to the brake booster to see if that helps. It may have a small leak.

I agree with MilesFox about adding the new battery. You should always change out the battery with a new alternator unless the battery is inexcellent shape. I used to own an '88 GL-10 turbo and had a similar problem of studdering while going up a particular hill. After the car had died on the road the fuel pump was replaced and that problem went away. If you have not checked the fuseable links for a loose connection yet I would do that. They are in the small black box mounted on the coolant reservoir. It may be the real cause of this and what caused the alternator problem also. Another thing to look for is a leak in the intake manifold gaskets.

I would check the amount of voltage that is going to the motor while it is running and make sure it is at least 12 volts. If it is, and the motor ground is ok, then a new antenna may be needed or the original one repaired. It may be stuck or the coiled line is broken inside it. This is a common failure.

Yes, a scope would be better to see the pulses. It's just that not many people have one on hand or know how to use one so I didn't think about it. By all means use it if you would like to. You shouldn't need to have the injector tied to the lead to watch for the pulses. Since I don't know what the second set of higher resistance leads are for I thought it may be good to leave them in the circuit. Having them disconnected may change the test results. It sounds though that the ECU is not generating the pulses to open the injector. I think the ECU makes a ground connection to turn on the injector. The ECU needs to see the ignition pulses from the distributor to do this I believe. It may also need to see the signal from the crank angle sensor. I'm not sure what lead carries the ignition signal to the ECU but you may be able to find it using the ohmmeter. It would also be good to check the crank angle sensor signal also. The sensor is located near the flywheel I think.

I assume you saw my post #28 about the resistance value of the injector. I would now see if the pulses are getting to the injector. It sounds like you have an analog meter which will help show the pulses if they are there. Place the probes across the coil leads that showed the low resistance and the cable attached to the injector and see if the voltage fluctuates while cranking the engine. If you don't see any change then something is not working in the control box.

One other thing to check is make sure there are no blown fuses. Have you done that?

It sounds to me that it is the purge valve relay but I'm not sure. This won't effect the injector.

I checked my manual for the resistance value. It stated that it should read between .5 and 2 ohms. So perhaps yours is ok. I still don't understand what the other coil is for unless it is for some sort of feedback signal to the ECU. Again, the injector in my manual only has one pair of wires to it. Another thing my manual mentions I was unaware of is the vehicle specification code. My manual only shows codes up to four. A fifth one may have been added later. This code will basically come on when there are no other codes in memory.

I don't see how that could prevent the injector from getting fuel into the engine. I am assuming this relay is for the fuel cannister to send excess fuel back to the tank.

Wow, those resistance readings are not anything like I would expect. Does your manual say anything about what the readings should be? It sounds like the injector is bad but I would try to verify that against a new one if possible. To fix your ohmmeter you need a new battery that is used for the lower scale readings. The battery voltage is too low to set the zero on it.

As far as the ECU codes, I think that five pulses are normal and it means there are no troubles reported. Not certain on that though.