Strandead Please Help!!!!

[hooziewhatsit]

get part 2 of the EA82 manual here: http://www.finleyweb.net/default.asp?id=142

 

in the second half of it there are a lot of flow-charts showing troubleshooting guides for different components including the injection system.

 

Give it a look then let us know how it goes.

 

good luck.

[NoahDL88]

just a quick and easy check, you still have a good timing belt on the drivers side yeah?

[CITRUSHARK]

The car starts and runs. It got me home after installing a new injector. But it is starting to do the same thing which is stall out after about 30 min. The ground that the mechanic put on the injector circuit is causing the injector to fail. I must find all the grounds that go to the ECU for the injector circuitry. I have used the link that you mentioned before to help. And the link contains pin specifics for the ECU. But it shows the specifics for the 1989 model. I need to get the detailed pin specs for the 1987-1988 ECUs.

get part 2 of the EA82 manual here: http://www.finleyweb.net/default.asp?id=142

 

in the second half of it there are a lot of flow-charts showing troubleshooting guides for different components including the injection system.

 

Give it a look then let us know how it goes.

 

good luck.

[oddcomp]

grounding out the injector to any part other than the ecu would result in large amounts of fuel being dumped in unmetered bad performance way bad mileage

and injector failure since they are not intended to stay open like that

 

if what you say is correct... i really am amazed it ran at all it should have just flooded out with the injector staying open cause of the grounding

wow

..

[hooziewhatsit]

And the link contains pin specifics for the ECU. But it shows the specifics for the 1989 model. I need to get the detailed pin specs for the 1987-1988 ECUs.

 

shouldn't all the pinouts be the same? I just used that same manual to help debug an 87 SPFI wagon I'm having CAS issues with. The wire colors may not match, but the pinout should be the same.

[FirstSubaruGLwagon]

I had a very simular experiance with an old volvo once.

I went though that car and tested every thing,, replaced 3 relays and finally replaced the main computer... After all that work and expence , turns out it was a really bad tank of gas I had filled up on at a local Gas&Go.

[CITRUSHARK]

They are very different. For example, the crank angle circuitry is right next to the fuel injector circuitry on the 89 schematics. But it is on the seperate end on the 87-88 circuitry. among several other differences. I have done alot of research on this. My only recourse is to find the ground that the ECU is relying on to provide ground for the injector.

shouldn't all the pinouts be the same? I just used that same manual to help debug an 87 SPFI wagon I'm having CAS issues with. The wire colors may not match, but the pinout should be the same.

[oddcomp]

They are very different. For example, the crank angle circuitry is right next to the fuel injector circuitry on the 89 schematics. But it is on the seperate end on the 87-88 circuitry. among several other differences. I have done alot of research on this. My only recourse is to find the ground that the ECU is relying on to provide ground for the injector.

 

its a common ground shared with the rest of the ecu

only difference is it runs thru a few fet transistors that act as electrical relays to pulse the injector open and closed in regards to engine load rpm maf reading

ect

 

the grounds for the ecu usually run up into the harness and ground out on the motor itself because of all the sensors connected

it tends to screw up the sensor readings if the ecu grounds/returns are on the different part of the car than the sensors

unless there is a dedicated ground wire run to the sensor and there usually is but it also usually all connects to one point on the motor

 

then you have the main battery to engine ground

and a aux ground between the motor and the frame

i usually add a few extra grounds just for fun

[CITRUSHARK]

Well I have located the two wires that come from the injector to the ECU. Both of them are reaching the ECU. For whatever reason, the ECU IS NOT grounding the injector but it will provide voltage to it. That sounds messed up.

 

 

The car will run fine, real good, with the injector wire grounded with the bypass wire to the engine. But it will not start without it. For whatever reason, the ECU has decided not to give ground to the injector. One of the master mechanics that I spoke with said that it could be something in the distributor.

 

The ECU is only giving me a code 14(improper reading from injector). It is not giving any codes for the crank angle, or anything else.

 

I suppose I could check resistance, voltage, etc. for all the eletrical and mechanical components on this car. But there is obviously nothing too wrong with anything if I can start and drive down the road just fine. If there is something on the car that is telling the ECU not to provide proper signal so that the injector will not fire, it is because of design flaw. Afterall, if I cheat the ECU, the car will run just fine. but the injector will eventaully burn out. And if there is another component that needs to be replaced, it will be expensive, hard to find, and take 1 to 2 weeks to get here. Afterwhich, I have to live with the fact that if I am going to rely on this car to get me anywhere in the future, I know that I will have to go thru hades again just to get it running if something else goes wrong.

 

So far I have gotten alot of good responses. But still no definitive answer as to why the ECU WILL NOT GIVE THE INJECTOR GROUND. If I cannot find an answer without replacing the distributor to see if that MIGHT fix it, I am just going to junk this car.

[CITRUSHARK]

Thanks for the info, I will look into adding some grounds. It makes good sense that all of the grounds for the ECU would use the same ground. I have located, and tested the ground that runs in the same wiring harness of all the relays.( the one bolted on the right side of the engine). could it be the transistors in the ECU that is causing this? Is the ECU BAD? And if so, why isn't the ECU giving me a code for it?

 

 

its a common ground shared with the rest of the ecu only difference is it runs thru a few fet transistors that act as electrical relays to pulse the injector open and closed in regards to engine load rpm maf reading ect the grounds for the ecu usually run up into the harness and ground out on the motor itself because of all the sensors connected it tends to screw up the sensor readings if the ecu grounds/returns are on the different part of the car than the sensors unless there is a dedicated ground wire run to the sensor and there usually is but it also usually all connects to one point on the motor then you have the main battery to engine ground and a aux ground between the motor and the framei usually add a few extra grounds just for fun

[oddcomp]

Thanks for the info, I will look into adding some grounds. It makes good sense that all of the grounds for the ECU would use the same ground. I have located, and tested the ground that runs in the same wiring harness of all the relays.( the one bolted on the right side of the engine). could it be the transistors in the ECU that is causing this? Is the ECU BAD? And if so, why isn't the ECU giving me a code for it?

 

does your car have some for of tachometer reading for engine rpm?

actually now that i think about it ,, the ecu should be getting a tach signal otherwise it would not know its supposed to fire the injector

in turn means that if it knows it's supposed to fire the injector but thinks there is a problem it will throw a code.. if that makes any sense

 

as far as the cause to the problem goes.. i would almost bet money that the injector driver inside the ecu has failed "or"

the wire between the ecu and injector has failed

 

reason being is the fact the ecu is programmed to monitor things like how much voltage / amperage a item uses to function

correctly

so if a injector dies and it shorts out the ecu would see it as a injector failure due to the high demand being put on the injector driver

and it triggers a error code

same thing goes if the injector burns out and no longer provides any resistance at all

the ecu would "see" this and determine the injector load is to low and also signal a injector is fubar code

 

what it cannot tell you is if its a fault in..

the wiring

the injector

or the injector driver mosfet inside the ecu..

any one of those things going wrong will put it into a error code + not run or not run right

 

my best suggestion after you trace down the wiring harness and look for

corroded wires/mouse chewed wires

is to go find/borrow a spfi ecu from someone and slap it in place

and see what happens when the motor is wired up correctly

 

really as far as i know .. what you describe when you say the motor runs with the injector grounded out to the motor.. really baffles me and my current knowledge of fuel injection

 

realisticaly <sp?> grounding out the injector should be the same as flooding a carbureted motor out .. to dam much fuel and not enough air to support combustion

so the motor either does nto run or it runs.. not so terrific

[Cougar]

It sounds to me that the transistor switching in the ECU to provide the ground for the injector has failed. I would try another ECU to see what happens.

 

I have repaired a couple of ECU's that had the same problem with the fuel pump ground circuit. It turned out to be a small TO-92 style driver transistor had failed in both of the units. I found these by doing a diode test with my DVM meter.

[CITRUSHARK]

Thank you both ODDCOMP And COUGAR. It's helpful to get some info about the ECU Possibility.

 

I have located, and tested, both wires going to the ECU. They are both defenitely making a clear path to the ECU. Just as another test, I ran a new wire from the injector to the ECU. And as expected, it made no difference.

 

I have also checked the wiring at the harness, and cleaned all the grounds that I can find. I do have a Tachometer, and it works fine. So it does seem that the ECU probably is defective from what you are both saying.

 

So I'm off to see what I can do about getting a ECU. Please let me know if you can point me in the direction of one.

 

Thanks again,

CITRUSHARK

does your car have some for of tachometer reading for engine rpm?

actually now that i think about it ,, the ecu should be getting a tach signal otherwise it would not know its supposed to fire the injector

in turn means that if it knows it's supposed to fire the injector but thinks there is a problem it will throw a code.. if that makes any sense

 

as far as the cause to the problem goes.. i would almost bet money that the injector driver inside the ecu has failed "or"

the wire between the ecu and injector has failed

 

reason being is the fact the ecu is programmed to monitor things like how much voltage / amperage a item uses to function

correctly

so if a injector dies and it shorts out the ecu would see it as a injector failure due to the high demand being put on the injector driver

and it triggers a error code

same thing goes if the injector burns out and no longer provides any resistance at all

the ecu would "see" this and determine the injector load is to low and also signal a injector is fubar code

 

what it cannot tell you is if its a fault in..

the wiring

the injector

or the injector driver mosfet inside the ecu..

any one of those things going wrong will put it into a error code + not run or not run right

 

my best suggestion after you trace down the wiring harness and look for

corroded wires/mouse chewed wires

is to go find/borrow a spfi ecu from someone and slap it in place

and see what happens when the motor is wired up correctly

 

really as far as i know .. what you describe when you say the motor runs with the injector grounded out to the motor.. really baffles me and my current knowledge of fuel injection

 

realisticaly <sp?> grounding out the injector should be the same as flooding a carbureted motor out .. to dam much fuel and not enough air to support combustion

so the motor either does nto run or it runs.. not so terrific

[CITRUSHARK]

That would be great if I could fix the ECU. I have a little experience with electronics. But I am unclear as to how to locate and test the TO-92 transistor you mention. Could you please give me a good description as to what it looks like and what readings I should be getting from them?

 

Thanks for your help,

CITRUSHARK

It sounds to me that the transistor switching in the ECU to provide the ground for the injector has failed. I would try another ECU to see what happens.

 

I have repaired a couple of ECU's that had the same problem with the fuel pump ground circuit. It turned out to be a small TO-92 style driver transistor had failed in both of the units. I found these by doing a diode test with my DVM meter.

[Cougar]

The transistor I replaced was in the fuel pump circuit. Your problem may have different components involved.

 

You should use a digital voltmeter to do the checking with if you do this. You need to trace the wiring from the ground side of the injector to the inside of the ECU and see what the contact is tied to. It may be going to a power transistor lead.

 

Transistors are basically back to back diodes and need to be checked for forward and reverse conductivity. Using a digital meter to in the diode check mode, the display will usually show between .5 to .7 volts in the forward bias direction. The reverse direction will usually show 'OL' or over limit.

 

You should be able to do a search on the web to find out how to check transistors. Here is one link for your info.

 

http://www.elexp.com/t_test.htm

[CITRUSHARK]

Thanks Cougar. I will research how to check the diodes and such. I found an ECU for $125 here in Florida. However, not knowing if there was/is something else that caused the ECU to malfunction, I am leary to spend any more time and money right now. I had intended to use this sooby and keep it as a backup for the future. However, Due to the loss of time and money during my sooby downtime, I am going to get other means of transport for now.

 

I fully intend to fix and keep my GL. It will probably be 2-4 weeks before I purchase or fix the ECU. So I hope everyone who has helped me will keep an eye on this thread.

 

Thanks everyone who posted on this thread!

 

TTFN,

CITRUSHARK

 

I

The transistor I replaced was in the fuel pump circuit. Your problem may have different components involved.

 

You should use a digital voltmeter to do the checking with if you do this. You need to trace the wiring from the ground side of the injector to the inside of the ECU and see what the contact is tied to. It may be going to a power transistor lead.

 

Transistors are basically back to back diodes and need to be checked for forward and reverse conductivity. Using a digital meter to in the diode check mode, the display will usually show between .5 to .7 volts in the forward bias direction. The reverse direction will usually show 'OL' or over limit.

 

You should be able to do a search on the web to find out how to check transistors. Here is one link for your info.

 

http://www.elexp.com/t_test.htm

[Cougar]

That isn't too bad of a price but you may be able to get one for half of that if you do some more looking. Check the Marketplace section of this site and place a request for one.

 

I wouldn't worry about damaging the new ECU, I doubt there is a problem with anything external to it. It just failed internally, it happens.

[CITRUSHARK]

That is good to hear that I needn't worry too much about damaging the ECU. I am working with a couple of cats here at USMB. I have sent them a response to the info they have given me about some parts. I am hoping that I can get enough replacement parts so I can use my sooby every day, and not worry about it breaking down in the future. So many owners here talk about having 300,00 miles on their sooby. I can't wait to be able to drive mine.

That isn't too bad of a price but you may be able to get one for half of that if you do some more looking. Check the Marketplace section of this site and place a request for one.

 

I wouldn't worry about damaging the new ECU, I doubt there is a problem with anything external to it. It just failed internally, it happens.

[CITRUSHARK]

I am wondering if a damaged ECU would cause the voltage meter in the dash panel to fluxuate? It has a new alternator, so I don't think that the alternator is causing that. I have also checked all the grounds that I can find, and checked the wiring harness, and connectors. I am also wondering if the fuel injector being grounded out like I mentioned could causde this?

That isn't too bad of a price but you may be able to get one for half of that if you do some more looking. Check the Marketplace section of this site and place a request for one.

 

I wouldn't worry about damaging the new ECU, I doubt there is a problem with anything external to it. It just failed internally, it happens.

[Cougar]

I wouldn't think the ECU is causing the trouble. Check the connection between the alternator output and the positive battery post. Also check the fusible links for a bad connection. How much fluctuation do you have? The problem could also be in the wiring to the meter. I would check the voltage with a DVM to verify the trouble area.

 

As far as your injector goes, the way I understand the problem is, the ECU is not making the ground connection to it. The circuit is open ended. You made a ground connection manually and proved that because the injector worked when the return end was grounded.

[CITRUSHARK]

One day it would be fine all day. The next day it would bounce between 8 and around 15 volts. It would do it if I was at a stop light, etc. It wasn't doing it only if I hit a bump or something. I checked the wiring at the alternator; checked the wiring going to the distributor, and the battery. I also checked the fusible links. I am not able to check it while it is running right now. So I am not able to trace the problem. It does not do it when it is not running(at all).

 

My main concern is putting in a new ECU. Of course I won't be able to do anything with it until it is running, so I guess I will have to put in the new ECU and then trace the voltage if the problem still exists. I am just concerned that the fluxuation in voltage will ruin the new ECU.

I wouldn't think the ECU is causing the trouble. Check the connection between the alternator output and the positive battery post. Also check the fusible links for a bad connection. How much fluctuation do you have? The problem could also be in the wiring to the meter. I would check the voltage with a DVM to verify the trouble area.

 

As far as your injector goes, the way I understand the problem is, the ECU is not making the ground connection to it. The circuit is open ended. You made a ground connection manually and proved that because the injector worked when the return end was grounded.

[Cougar]

It shouldn't hurt the ECU. I would recommend you get a voltmeter if you don't have one as it will help find the problem area. You could use a test light probe also. Place the tester at a point you want to test and start tapping around suspected trouble areas with a screwdriver handle. You should be able to track the problem down that way.

[CITRUSHARK]

Thanks Couger. I do have a volt meter, and I will keep you posted

 

Thanks Again:)

 

It shouldn't hurt the ECU. I would recommend you get a voltmeter if you don't have one as it will help find the problem area. You could use a test light probe also. Place the tester at a point you want to test and start tapping around suspected trouble areas with a screwdriver handle. You should be able to track the problem down that way.

[Cougar]

Your welcome.

[CITRUSHARK]

Well I've got good news and bad news.

The good news is that while tracing the circuit in the ECU, I found the cause of the gound circuit malfunctioning. Although there were no scorch marks, a break in the negative circuit occurred.

 

After repairing the circuit, I was able to start the car without the rigged up wire from before.

 

Now for the bad news.

 

I drove for about 25 mins. just fine. After which, the car did the same thing as it was doing before; it just died.

Just like before, it wouldn't start after a few atempts. I gave it a few minutes, and it started. got it on the road again, and it died again. So I did the same thing... Gave it a few tries, no luck..... Gave it a few minutes, and no luck. Gave it a few more minutes... and i t started again.

 

So I spent the next 2 hours driving it around the neighborhood. And since then I have had no problems with it.

 

Could that be caused by a failing fuel pump?

Your welcome.