1G MFI/MPI Relay

[meyer795]

Hello,

First of all thanks Steve for all the pics and time you have taken to help your anonymous cohorts.

I checked my 1990 talon TSI for voltage coming to pins 3, 9, and 10 and I found that 3 and 9 have 12 volts.
Pin 9 does not get any voltage. I do not have the automatic tranny so I am not sure which sensor to check next? Your drawing is obviously for an auto... any thoughts would be appreciated... Thanks !!


Best Regards,
Jesse

 

[yokotabrat]

Which pins have 12V? In one sentence you say 3 and 9 do have 12V, but the other says pin 9 doesn't. Assuming you are looking to see where pin 9 on the mpi relay goes, on a M/T car, it goes to the starter relay. You should only see 12V there when the car is cranking. The more important question is what do you see for voltage on pins 7 & 8 of the mpi relay when you are trying to start the car?

 

[steve]

Any thoughts would be appreciated.

Depends on what your trying to solve. (Please tell us)

If the problem is the fuel pump then the output on pin 2 is important and you would be looking at pins 3, 7, and 9 to figure it out.
If the problem is your not getting power to the MPI circuit then the output on 4 and 5 is key and you would look at pin 10 and 8 for clues.

As Bert points out pin 9 only has 12v on it when the car is cranking. Assuming the car starts the ECU will pull pin 7 low to keep the pump running.

 

[meyer795]

Which pins have 12V? In one sentence you say 3 and 9 do have 12V, but the other says pin 9 doesn't. Assuming you are looking to see where pin 9 on the mpi relay goes, on a M/T car, it goes to the starter relay. You should only see 12V there when the car is cranking. The more important question is what do you see for voltage on pins 7 & 8 of the mpi relay when you are trying to start the car?

I confused the pins in my last post sorry. The pin 9 has no voltage which sounds like its supposed to. If I understand correctly, I need to test whether the voltage gets through the MPI relay while the key is on. I am new with the DMM, so I am wondering: do you just plug it in and shove the point of the electrode in the back of the connector?

The car seems to be getting fuel (fuel pump whines and inlet pulses) but has no spark and no ECU boot. I fried the shit out of my ECU beyond cap replacement, and I got a matching one from a guy of craigslist. I opened the replacement and it doesn't look or smell fried. After putting the new one in I noticed its not booting, but the MPI does click... I will try to check the MPI relay across my battery this afternoon.

I'm gonna go out there right now and start shoving the DMM into the backside of the wire connector and see what happens.

Thanks for your responses guys!!!

 

[steve]

The car seems to be getting fuel (fuel pump whines and inlet pulses) but has no spark and no ECU boot. I fried the shit out of my ECU beyond cap replacement, and I got a matching one from a guy of craigslist. I opened the replacement and it doesn't look or smell fried. After putting the new one in I noticed its not booting, but the MPI does click... I will try to check the MPI relay across my battery this afternoon.

Sounds like the fuel pump side of the relay works so you should focus on the MPI side.
Since pin 10 has 12v all the time just go to the ECU connector with the ECU disconnected and ground pin 63 (or 66) and see if that turns on power to pins 102 and 107.

It's common for leakage to damage the traces controlling the activation of the MPI relay.
If grounding the pin turns the power on then you know the problem is the ECU and not the relay. (which would be the first place I would have been looking since I've yet to see a bad MPI relay but lots of bad ECUs)

 

[meyer795]

Sounds like the fuel pump side of the relay works so you should focus on the MPI side.
Since pin 10 has 12v all the time just go to the ECU connector with the ECU disconnected and ground pin 63 (or 66) and see if that turns on power to pins 102 and 107.

It's common for leakage to damage the traces controlling the activation of the MPI relay.
If grounding the pin turns the power on then you know the problem is the ECU and not the relay. (which would be the first place I would have been looking since I've yet to see a bad MPI relay but lots of bad ECUs)

Alright, I did that. Just in case i did it wrong heres what I did...

Shoved a little wire into #63 and bent it down to touch the middle steel area. Then I shove the voltmeter into 102 and 107 while grounding the other end. 0 volts. Bad MPI then?

 

[steve]

Just to make sure, what color wire is 63?

 

[meyer795]

i forgot to mention I turned the key and still nothing.. I was rough on it while i took it out so I'm wondering if the wiring between the two is the problem. I could test it across the battery by using some wire right?

 

[steve]

I could test it across the battery by using some wire right?

Yes, just like I said in post 5.
So what color wire went to the pin you grounded. Since the ECU was disconnected the ignition switch isn't involved. It's normally the input to the ECU to turn on the MPI relay but there is circuitry between the two.

 

[meyer795]

The #63 wire was red with a black marking. It was the same as the wire looked coming from pin 8 on the MPI relay.

I tested the MPI across the battery today and it works. For some reason its not getting across the MPI ..... I looked for power at the MAS, CAS, and TPS and there was nothing. It is very strange.. Unless the wire is cut somewhere I am at a loss.

The only other thing I can think of is the MPI is not sending enough current through it. What should the resistance be between 10 and 8? Mine is at 6.7 MegaOhms. is that correct?

Thanks

 

[steve]

The #63 wire was red with a black marking. It was the same as the wire looked coming from pin 8 on the MPI relay.

That doesn't sound right. Sounds like you flipped over the connector and forgot to flop where pin 1 was when you did so. The Red with Black is the AC clutch. Pin 63 and 68 are black with blue stripe wires.

http://www.dsmlink.com/images/forums/1GECUPinout.pdf

I tested the MPI across the battery today and it works. For some reason its not getting across the MPI ..... I looked for power at the MAS, CAS, and TPS and there was nothing. It is very strange.. Unless the wire is cut somewhere I am at a loss.

The only other thing I can think of is the MPI is not sending enough current through it. What should the resistance be between 10 and 8? Mine is at 6.7 MegaOhms. is that correct?

Thanks

That also doesn't sound right but it may be because there is a diode in series with the coil winding and you had it reverse biased. If the relay clicks when you put 12v across those pins and you read battery voltage at pins 4 and 5 while the battery is connected then the MPI side is good.

Your not going to see power at the MAF and CAS until you get the MPI relay switched on and you need to use the right wire.

 

[meyer795]

That doesn't sound right. Sounds like you flipped over the connector and forgot to flop where pin 1 was when you did so. The Red with Black is the AC clutch. Pin 63 and 68 are black with blue stripe wires.

http://www.dsmlink.com/images/forums/1GECUPinout.pdf



That also doesn't sound right but it may be because there is a diode in series with the coil winding and you had it reverse biased. If the relay clicks when you put 12v across those pins and you read battery voltage at pins 4 and 5 while the battery is connected then the MPI side is good.

Your not going to see power at the MAF and CAS until you get the MPI relay switched on and you need to use the right wire.

once again, steve, your right. I was checking the wrong wires. The ECU has power coming to it but still no boot. Got it off craigslist so I'm ####ed out of $75 dollars. Damn. Well I learned a lot about car elec systems thanks so much. I hope this thread helps others in our position.

 

[steve]

The ECU has power coming to it but still no boot. Got it off craigslist so I'm ####ed out of $75 dollars. Damn.

This doesn't surprise me. How many bad ECU's do you have now?
Pack them up and send them in. I'll fix you up.

 

[meyer795]

I have two bad ones now. The first one is burnt halfway through the board. Idk why the other one doesn't work.

I just bought another one from the 1991, plugged it in and it fired right up. Opened it up and theres acid on the board!!! Its a curse


Is it possible that my car is burning them up??? I don't know what to do anymore...

 

[steve]

I just bought another one from the 1991, plugged it in and it fired right up. Opened it up and theres acid on the board!!! Its a curse


Is it possible that my car is burning them up??? I don't know what to do anymore...

There are things in the car that can like a bad ISC but you have to remember that all of these ECUs have a problem with the caps leaking. Unless the caps have been changed before they leak or have been changed and the damaged repaired the are going to have problems.

My wife sometimes kids me and asked if I've fixed all of them yet but I've only made a dent.

 

[meyer795]

Here are some pics of the burnt one.



Steve,
I am going to send you the one that works but is leaking. I rinsed the acid with isopropanol. The ecu causes clicking when its in the car.

 

[steve]

That's the +5v regulated filter capacitor and it looks like it leaked, shorted and blew up.
It's possible that in the process of doing so it spiked the CPU and interface chips. Look to telltale bubbles or cracks in the tops of the black square chips. The bigger is the CPU and the smaller interface logic. On EPROM ECUs there are two smaller square chips.

I don't believe that damage has anything to do with your car. It has to do with not getting the caps changed in time. It also doesn't have anything to do with the MPI relay except that a dead ECU may not turn the relay on.

 

[meyer795]

I got a working one which I was refurbished. Now the car is making a screeching noise when I start it (starter?) and then it runs for awhile before it just suddenly dies. Do you think that is related to the ECU or something different. Thanks in advance

That's the +5v regulated filter capacitor and it looks like it leaked, shorted and blew up.
It's possible that in the process of doing so it spiked the CPU and interface chips. Look to telltale bubbles or cracks in the tops of the black square chips. The bigger is the CPU and the smaller interface logic. On EPROM ECUs there are two smaller square chips.

Can I send that one to you with my other two to look at? I bet you can fix the first two.

 

[steve]

I got a working one which I was refurbished. Now the car is making a screeching noise when I start it (starter?) and then it runs for awhile before it just suddenly dies. Do you think that is related to the ECU or something different. Thanks in advance

Can I send that one to you with my other two to look at? I bet you can fix the first two.

The screeching may be the alternator belt.

Not enough data to comment on why it dies.

Sure send them all in.

 

[meyer795]

The screeching may be the alternator belt.

Not enough data to comment on why it dies.

Sure send them all in.

Steve,

It is not the sound of a belt screeching, its more like a grinding screech. Also today I noticed my MAF translator is stuck in red LED mode, which apparently is indicative of WOT+AUX.



UPDATE: I just went to start it this morning and it fired right up. Drove it to autopart store to get a CEL code but they didn't have OBDI, on the way home the CEL went off and the MAFT is back in green light mode. Strange.... maybe the switch just needed to warm up ?