Resolved 1G MPI fuse draw

[JFZERO ECLIPSE]

Searching for an answer to my mystery battery drain. Unplugged the negative from the battery and put a test light in series with it. Following luv2rallye's tech page and I have 1 nasty draw.

As I pulled all the fuses from inside the car first nothing turned the light off. Went under the hood, to the A/C and main fuse blocks. Still nothing, last but not least I went for the fuses on the battery+. Found it!

MPI fuse.

After pulling the fuse and reconnecting the test light, it lights up and then dies to nothing. Even pulling the relay it does the same thing lights up and then after about 6-7 seconds it dims down and goes out.

Putting the fuse back in it lights up like Christmas .

I need a little help here.
Car is 100% bone stock aside from a CD player that I installed, (I even unplugged it to make sure it wasn't the problem) and removed some of the emissions related hardware that shouldn't have any ill effects.

Not sure where to go next.

 

[Boostdriven]

I wonder if you're having the same problem my brother had with his 92 Talon, if he didn't drive the car for few days the battery would die completely. We narrowed it down to MFI relay not shutting off after the ignition was turned off. He also had had another issue along with that one where the fuel pump would turn on as soon as you turned the keys to on position and would stay on until you turned the key off.

Pull the left foot kick panel on the passenger side off and you will find the MFI relay there. If you car has sat there for a while and you feel the relay and its warm then that's your problem, that's how we figured it out. Of course I did some more testing just to be sure.

There should be 9 wires going to that relay. Part of that MFI relay turns on power to your injectors and few other things and the other part turns on your fuel pump. There should be a blue wire and a white one with red stripe. Those 2 wires give a ground signal from the ECU when you turn the ignition key to on position. The other wire there should be, 2 red, 1 black/white, 1 black, 1 black/yellow, 1 black/green, 1 black/red or brown. With ignition OFF, take your volt tester and connect one wire from it to constant power source. Use the other wire to test if you're getting the ground signal from either of the two wires (blue, white/red) going to the MFI relay, if you get a voltage reading on your tester then you have the same problem that my brother had.

Solution: contact Thomas at ECMTuning via email and he will explain to you how to get a work order set up so you can send off your ECU for repair. I think my brother paid like $60 dollars including shipping it back, the problem was fixed and the car still runs to this day. This was over a year ago.

Hopefully this help.

 

[JFZERO ECLIPSE]

I'll have to swap out the MPI relay with a known good one, sad thing is the ECU I have now is a fully rebuilt unit courtesy of Thomas Dorris as well as ECMLink V3 full installed

Well I found the cap that goes on the fuses on top of the battery... Going across it goes 30 amp-30 amp-20 amp-60 amp

The 20 amp is the MPI fuse. Well previous owner decided to put a 40 amp fuse in there. I wonder what kinda damage I can expect, I'll change out the MPI relay for sure later and I'll get back with my findings.

 

[JFZERO ECLIPSE]

Well an update to my battery drain.

I basically have pulled the MPI fuse if I know I will be anywhere longer than 5-6 hours for just in case measures.

But bone stock ECU plugged (won't fire car up anymore, but just for testing purposes) and factory relay again plugged in and with MPI fuse in = drain.

ECMLink V3 ECU plugged in (what I've been driving on) factory relay plugged in. MPI fuse in = drain.

Bone stock ECU plugged in, MPI relay unplugged, MPI fuse in = drain.

ECMLink V3 ECU plugged in, MPI relay unplugged, MPI fuse in = drain.

Eliminate the ECU so
ECU unplugged, MPI relay plugged in, MPI fuse in = NO DRAIN!

What is my next step?

I have a hard time believing both ECUs are crap. So rule them out, my guess is new MPI relay?

 

[steve]

I have a hard time believing both ECUs are crap. So rule them out, my guess is new MPI relay?

Just because you don't like where the diagnosis leads doesn't mean you can ignore the results. Look and see if both ECUs are pulling the pin to the MPI relay low and activating it. If the pin isn't low with the ECU plugged in then you can rule them out.

 

[JFZERO ECLIPSE]

And can you explain exactly which wires to check? And use what part of a multimeter (buying one this weekend)

 

[steve]

Have you looked at http://www.dsmtuners.com/forums/newbie-forum/263224-mfi-mpi-relay.html

Your going to use voltage measurements and resistance measurements mainly when working on a car.

 

[JFZERO ECLIPSE]

Didn't make it to grab a multimeter just yet, I lost my original a while back. Are the auto and 5 speed relays the same?

From what I read on that thread to wiring to the relay is different as you stated in post 5 "This is the circuit for an 1G automatic car taken from the 1G tech manual. The manual cars have the starter relay and clutch safety switch replacing the AT Inhibitor switch for providing power to the second fuel pump coil to activate the pump while cranking the engine."

I swapped a relay a while back from a 90 5 speed car into my 91 auto car and it wouldn't fire up so I assumed the relay was bad.

 

[steve]

The 90 and 91 relays are different and the way the fuel pump side of the MPI relay is hooked up to the ignition switch differs between auto and 5 speed cars.

The key connection for this problem is the wiring to pin 8 on the MPI Relay from pins 63 and 66 on the ECU.

The ECU either floats (acts like it's disconnected) or pulls the pin to ground depending on if if wants the MPI relay turned on.

Pin 10 on the 91+ MPI relay gets 12v from the MPI Fuse on the battery.

Since the relay coil is just a bit of wire you see 12v at the other end of the wire too. When you first turn the ignition on the ECU switches ECU pins 63 and 66 to ground which are connected to Pin 8 on the Relay, completing the circuit, current flows in the coil causing the relay to turn on.

After you turn off the ignition switch the ECU counts down 10 seconds and turns off grounding pins 63 and 66 so the voltage at the MPI Relay pin 8 floats back up to 12v and the relay turns off.

If you don't see battery voltage at pin 8 when the ignition is off and about 0 volts when the ignition is on and the ECU is plugged in, you have a problem. I'm expecting you'll see 0 volts all the time with your ECUs plugged in and 12v at the pin all the time when they aren't connected based on what you have said so far.

 

[JFZERO ECLIPSE]

Hum. So to sum up, 90 relay into 91+ no go.
But 5 speed relay to auto car?

I'm going to buy a multimeter tomorrow and hopefully get back with my results.
Like I said before car runs and drives perfect. Just a massive drain if I leave the MPI fuse in....
Drain stops when I either a. Pull the MPI fuse or b. Unplug the ECU. Unplugging the relay does nothing.

But again I'll come back with results hopefully tomorrow, I'm selling my crotch rocket and gotta do a pile of other stuff tomorrow. Shouldn't take more than 15-20 minutes tops to check though.

 

[1990dsmkid]

Come on now Jimmy boy.....lets get this problem figured out. Could have asked me to come down and help....I have s multimeter and this would be solved already

 

[JFZERO ECLIPSE]

Well get you butt down here and help!!!

I hate electrical garbage.... Rather replace a tranny than work with electrical...

 

[steve]

Unplugging the relay does nothing...

Then perhaps you might teach me something new because I can't imagine a case without really hacking up the factory wiring where that would be true.

With the MPI relay unplugged, the MPI circuit ends pretty quickly. Almost as quickly as when the fuse is pulled.

 

[JFZERO ECLIPSE]

I completely said that wrong .

When I unplugged the relay the light still stayed on. I wasn't clear by the way said it... But again I have absolutely no skills in electricity....

From my understanding it goes battery to fuse, fuse to relay, relay to fuel pump, and then relay to ecu, then the ecu send power back to the relay?

Also as stated above you said the 90 relays won't work in 91+ cars....

But another question is..... Are auto and 5speed relays interchangeable due to the starter relay and clutch safety switch replacing the a/t inhibitor switch?

 

[JFZERO ECLIPSE]

If you don't see battery voltage at pin 8 when the ignition is off and about 0 volts when the ignition is on and the ECU is plugged in, you have a problem. I'm expecting you'll see 0 volts all the time with your ECUs plugged in and 12v at the pin all the time when they aren't connected based on what you have said so far.

1990dsmkid came over today with a multimeter, here are my findings....please tell me if I did anything wrong...

Ecu plugged in, relay plugged in, fuse plugged in, pin 8 with voltmeter 12v reading with key off....
Ecu plugged in, relay plugged in, fuse plugged in, pin 8 with voltmeter .10v reading with key on...

Ecu unplugged, relay plugged in, fuse out, pin 8 with voltmeter .23v key off or on don't matter....

Ecu plugged in, relay plugged in, fuse in, negative off the battery I'm getting a current draw of .94 amps.

Am I missing anything??

 

[steve]

1990dsmkid came over today with a multimeter, here are my findings.

Ecu plugged in, relay plugged in, fuse plugged in, pin 8 with voltmeter 12v reading with key off.
Ecu plugged in, relay plugged in, fuse plugged in, pin 8 with voltmeter .10v reading with key on.

Once more test would have been, did it return to pin 8 to the 12v reading with key off after turning it on? It takes the ECU about 10 seconds to turn off.

Measuring the output voltage at pins 4 and 5 would make sure that the relay isn't shorted. It should be 0v when off and 12v when the ignition switch is on.

Ecu plugged in, relay plugged in, fuse in, negative off the battery I'm getting a current draw of .94 amps.

Am I missing anything??

How and where are you measuring the currrent? Did you have the meter connected between the negative cable and the battery?

Any aftermarket alarm or radio equipment. It doesn't seem like the ECU is keeping the MPI turned on.

 

[JFZERO ECLIPSE]

Once more test would have been did it return to pin 8 with voltmeter 12v reading with key off after turning it on?

Measuring the output voltage at pins 4 and 5 would make sure that the relay isn't shorted. It should be 0v when off and 12v when the ignition switch is on.

I will try this in the next day or so when I snag a multimeter from harborfreight. I only had car in the off position and took the reading, then disconnected the negative from ground, turned car to on position, then regrounded multimeter and took the reading.

So hook multimeter to pin 4 then ground and do the same with pin 5....
I will post readings of it off and on.

How and where are you measuring the currrent? Did you have the meter connected between the negative cable and the battery?

Any aftermarket alarm or radio equipment. It doesn't seem like the ECU is keeping the MPI turned on.

Yes sir, negative pulled off battery, postitive from multimeter on battery and negative from multimeter on the unhooked negative cable that was off the battery. I put it on the battery and it was around 1.12amps when I first hooked it up, but then after the isc adjusted and as about 10-20 seconds passed it finally settled down to .94...

No aftermarket equipment of any kind, car is 100% bone stock, electrically speaking.... The only things that are aftermarket on the car is a ecmlink v3 software inside a 5 speed ecu. The car has a factory security system if that matters as well

 

[steve]

I will try this in the next day or so when I snag a multimeter from harborfreight. I only had car in the off position and took the reading, then disconnected the negative from ground, turned car to on position, then regrounded multimeter and took the reading.

So hook multimeter to pin 4 then ground and do the same with pin 5....
I will post readings of it off and on.

It you look at the wiring diagram you'll notice that pin 4 and 5 of the MPI relay are connected so one reading is enough. Red probe to the pin, black probe to ground.

Yes sir, negative pulled off battery, postitive from multimeter on battery and negative from multimeter on the unhooked negative cable that was off the battery. I put it on the battery and it was around 1.12amps when I first hooked it up, but then after the isc adjusted and as about 10-20 seconds passed it finally settled down to .94.

If the ignition switch was off, I don't think the ISC should have done anything. It requires the ECU to be running for that and the ECU shouldn't be running with the ignition off.

 

[JFZERO ECLIPSE]

It you look at the wiring diagram you'll notice that pin 4 and 5 of the MPI relay are connected so one reading is enough. Red probe to the pin, black probe to ground.

I sent my fiance out to harbor freight to pick up a small multimeter while shes heading to the gym... once she gets back home I will test the relay again.
EDIT: Blah today is Sunday and places close early.... I'll get one tomorrow.

If the ignition switch was off, I don't think the ISC should have done anything. It requires the ECU to be running for that and the ECU shouldn't be running with the ignition off.

I'm pretty sure ~80% that the ignition switch was off when I disconnected the negative... I always thought that when the battery is connected the ISC resets itself...

Also another question when doing the ISC test via ohms. I have a tan one installed and I know that it should float around 30ohms when they are in good condition.... if it shows nothing it means the coil is shot for that particular winding..... when testing the first one we tested was 60 ohms and the 2nd,3rd, and 4th were ~30 ohms... Is this a problem?

 

[steve]

I'm pretty sure ~80% that the ignition switch was off when I disconnected the negative. I always thought that when the battery is connected the ISC resets itself.

It's been a long time since I've disconnected the battery on my DSM so I don't remember 100% but I don't see how it would, the ECU doesn't turn on with the ignition switch off and the ISC can't reset itself.

Also another question when doing the ISC test via ohms. I have a tan one installed and I know that it should float around 30ohms when they are in good condition.... if it shows nothing it means the coil is shot for that particular winding..... when testing the first one we tested was 60 ohms and the 2nd,3rd, and 4th were ~30 ohms... Is this a problem?

Every multimeter I've ever used reads 0 ohms with the two probes touching (a short) but the all read something unique for infinite ohms (an open).

If you measured 60 ohms between the center pin and one of the end pins that coil is bad. If you measure the same display as when the meter isn't connected then that coil is open and bad. If you measure 0 ohms the coil is shorted and it will damage the ISC driver in the ECU.

 

[JFZERO ECLIPSE]

It's been a long time since I've disconnected the battery on my DSM so I don't remember 100% but I don't see how it would, the ECU doesn't turn on with the ignition switch off and the ISC can't reset itself.

Disconnecting the negative and putting it back on does yield the ISC positioning itself. At least on my car but may be connected to this particular drain issue.

It you look at the wiring diagram you'll notice that pin 4 and 5 of the MPI relay are connected so one reading is enough. Red probe to the pin, black probe to ground.

I connected multimeter to pin 5 and ground I get a reading of 12v whether the car is in the off position or in the on position.

BUT I traced back the wires under the factory loom just a hair and for some reason pin 10 and pin 5 have been connected via a small piece of wire. IDEAS?

I could tell that it is NOT factory due to black tape being in there.

 

[steve]

Disconnecting the negative and putting it back on does yield the ISC positioning itself. At least on my car but may be connected to this particular drain issue.

I connected multimeter to pin 5 and ground I get a reading of 12v whether the car is in the off position or in the on position.

I traced back the wires under the factory loom just a hair and for some reason pin 10 and pin 5 have been connected via a small piece of wire. IDEAS?

I could tell that it is NOT factory due to black tape being in there.

This is starting to add up.

There shouldn't be anything between the input to the MPI relay and the output. Wired this way the MPI relay is always turned on and would explain the current draw. This also explains why your ISC rehomes since the ECU is always on.

The question is why is that wire there? Does the MPI relay not work, did the PO have a bad ECU?
The way forward is to remove the wire, see if you no longer get power out of the relay, and then track down why.

 

[JFZERO ECLIPSE]

Alright, I will remove the wire after work, and then conclude my findings. Anything specific I should check via multimeter aside from removing the negative from battery and checking for amp draw?

I will also try to start it up and see how it runs...

This is starting to add up.

There shouldn't be anything between the input to the MPI relay and the output. Wired this way the MPI relay is always turned on and would explain the current draw. This also explains why your ISC rehomes since the ECU is always on.

The question is why is that wire there? Does the MPI relay not work, did the PO have a bad ECU?
The way forward is to remove the wire, see if you no longer get power out of the relay, and then track down why.

Well I went out and removed the wire first and foremost I then pulled the negative off the battery and rechecked my amp draw. My original amp draw was .94 amps, now it is .01 amps, that's as low as the tester will read.

Next test was will it start the car, YES.
Now the factory ECU that I got with the car ran and drove fine for about 2-3 weeks and then all it would do one morning was crank, crank, crank.

Put in my other ECU with v3 ECMLink and car fired right up.

Now my biggest concern is could that draw have damaged my new ECU?
Also, I wonder if that could have caused my old ECU to shit the bed?

 

[steve]

Now the factory ECU that I got with the car ran and drove fine for about 2-3 weeks and then all it would do one morning was crank, crank, crank.

Put in my other ECU with v3 ECMLink and car fired right up.

Now my biggest concern is could that draw have damaged my new ECU?
Also, I wonder if that could have caused my old ECU to shit the bed?

While I can't say 100% that leaving the MPI relay turned on 24/7 won't ever damage your ECU considering you were pulling battery cables and perhaps working on other engine circuits that were powered when they shouldn't be, the draw was normal for when the MPI circuit is powered up and by itself shouldn't cause the ECU any problems.

It sounds like your original ECU may have suffered from capacitor leakage and the first user indication was that the MPI control failed and the car wouldn't start, that wire was a hack to work around it. Untreated, the leakage continued to damage the ECU until something else critical died and it wouldn't start the car again.

 

[JFZERO ECLIPSE]

That would be a good possibility, the guy that I bought it from was well let's just say he didn't even know how to change his own oil. So I'm betting a shop did that to get it back to running.

Normally it would take 8-10 hours to fully drain my battery to where it just clicks once when I try to start it, and now it hasn't been fired up for a tad over 36 hours and it fired up perfectly.

I thank you for your expertise in this matter and marking it resolved!!!

My next thing I have to fix is a major boost problem. Drove the car the other day and I accidentally drove the car with the ISC unplugged about 6-7 miles. After I got about 3 miles out the turbo started sounding very loud (much more-so than normal) and I lost all boost.

I have to perform a boost leak test first and foremost, and I do know the turbine housing has a huge crack in it. So I know a turbo upgrade is in my near future.