Fuel Pump Power Issues.....

[femmeDSM]

I am having a hell of a time trying to get my stupid fuel pump to power on.

I bought the car with a botched re-wire and a bad fuel pump that was keeping the car from starting. I pulled all the old wires and re-did the fuel pump rewire correctly, and replaced the fuel pump with a Walbro 190 -- still won't turn on or prime w/ the key in ON.

The ONLY way I can get the pump to turn on is to take the 12v power wire running from the battery and touch it to the wire going to the Fuel Pump.

A list of things that I have checked/replaced so far:

- New Fuel Pump

- Re-Wire (double, triple, and quadruple checked -- everything is hooked up perfectly)

- Fuel Pump will NOT turn on using the check connector in the engine compartment.

- Tested voltage to MPI Relay, per FSM instructions. Tested out fine.

- ECU is new

- I can hear the MPI Relay click ON while cranking, and OFF a couple seconds after I try to start.

- I have good spark, and good compression, engine will turn over but will not start.

- When I revert the wiring to the stock setup (i.e. eliminate the 12v power wire from the battery, etc.) I get the same results.

- Tested voltage on the black/white power wire from pin 3 on the MPI relay -- anywhere from 9.5v to 12v (however, it was dark and cold when I did this, so I am planning on re-testing this tonight in case I got a bad ground).


I am at a loss here -- Is my only option now to trace that black/white power wire all the way from the MPI to where it connects at the FP rewire relay? Honestly, if that's my best option, I would almost rather just hook my FP up to a switch in the cabin. If all else fails, that would (SHOULD) work, right?

I am at that point where I have tested so much crap that my brain is starting to hurt.

I hope I am just overlooking something stupid, here........

Rep points will be given for ANY and ALL input!

Thanks much.

 

[femmeDSM]

I suspect it has a 9 pin serial port and a 25 pin parallel port. I haven't seen a laptop with a 25 pin serial in over a decade.

Yeah, I thought about that after I posted..... I feel stupid now. . Thanks for clearing that up. Being in the computer field, I should have known that

I'll make you up a cable for a PC this weekend.

Thank you! I wish I still lived in Chicago -- I would have to take you out for lunch and/or a drink. haha.

------------

Also, a (not so) short update on status....I was able to do a bit of testing before I left for the office this afternoon. I pulled the fuel line off of the rail, and with the pump running (connected to the 12v hot wire), plenty of fuel came out into the bottle I attached to the line.

Not sure what that verifies, other than the fact that the fuel pump is actually doing what it's supposed to be doing, there is nothing blocking the line, and that the fuel filter is functioning. I guess that leaves the injectors and their respective electrical components up for review. As far as testing injectors, the only times I have done this has been with the engine running, using a noid light. Not sure how the process would work with the engine not able to turn on.

Or could the MPI Relay not be communicating with the injectors.....? After all, given the way we are turning on the pump, we are bypassing the MPI Relay.

Tonight's plan still includes starter fluid into the intake, and a continuation of trying to get the relay to work...

I will post results on these as soon as possible........*sigh*
It would be nice if I could just make a LITTLE progress, here.......

 

[steve]

Or could the MPI Relay not be communicating with the injectors.....? After all, given the way we are turning on the pump, we are bypassing the MPI Relay.

The MPI relay is really two relays in one metal box.
One for powering the ECU and half of the engine bay, the other for powering the fuel pump.
I had a customer drop off a couple that had been opened for testing so I grab a picture of the "guts"

The turbo cars injector circuit works with battery voltage being switched by the MPI relay to the injector resistor pack when you turn the ignition switch on. There is one resistor for each injector and the power feed is common. From each resistor the voltage goes to its injector coil pin 1. Pin 2 on each injector then goes to the ECU. The ECU has four power transistors, one per injector, used to pull the voltage from the injector coil to ground causing current to flow and opening the injector.

So you can check to make sure you get battery voltage to pin 1 on each injector when you turn the ignition on. You can measure the resistance (2-3 ohms) of each injector between pins 1 and 2 to look for open coils. You can force an injector to fire by grounding its pin at the ECU connector. In normal operation the ECU gets crank/cam position (timing) from the CAS and with the air related signals from the MAF decides when and for how long to open an injector and when to charge and fire the coils.

Steve

 

[femmeDSM]

Short update -- finally got a chance to spend a little time with the car today, and tested the FP Relay. I hooked everything up as you suggested, and when I connected Pin 30 to Pin 86 -- relay clicked, but battery voltage did NOT show up on pin 87... I reversed the ground and MPI relay wires to see if I could get anything new to happen, but no luck there either. One kind of dumb question that I'd never really thought about before, though -- should the relay be clicking or doing something when I'm cranking the car? It's not...

ALSO, finally tried the starter fluid, and.........She fired right up! Bittersweet victory, though, as of course, she only ran for about 1-2 seconds - just long enough for the idle to barely settle down at ~800, and to see a large puff of white smoke come out of the exhaust pipe. haha. It was definitely a good feeling to see her run, if only for a few seconds

I ran out of daylight as I got to testing resistence on the injectors, so I will have to lug my big light down there tonight and do that. I did notice that when I pulled the connector caps off the injectors, two of them look severely chewed up. Not the best of signs, I suppose........

I'll report back once I have numbers on injector resistence.....Just had to come inside to warm up.......One thing I don't miss about living in Chicago is the weather

 

[femmeDSM]

OK, checked resistence, and either I am doing something wrong (highly likely ), or I have found the problem.

The number will jump to about 18.00 ohms immediately, then it will bounce around randomly before finally settling down at 0.00 ohms.

Good grief.

 

[steve]

I hooked everything up as you suggested, and when I connected Pin 30 to Pin 86 -- relay clicked, but battery voltage did NOT show up on pin 87... I reversed the ground and MPI relay wires to see if I could get anything new to happen, but no luck there either. One kind of dumb question that I'd never really thought about before, though -- should the relay be clicking or doing something when I'm cranking the car? It's not.

Sounds line time for closer look at that relay.

Here's what should be going on.

When you turn the ignition on the ECU pulls pin 8 on the MPI relay low, switching power from pin 10 to pin 4 & 5. That powers up the rest of the ECU, and half of the engine bay. This seems to be working on your car.

When you turn the switch to start, battery voltage from the switch is fed (assuming the Clutch Safety Switch is working and the clutch is pushed in or the gear selector is in park or neutral on a ATX) to pin 9 on the MPI relay, switching power from pin 3 to pin 2 on the FP side of the MPI relay. Pin 2 is connected to the check connector in the engine bay and to the fuel pump. This is the black/white wire we are using to trigger the bypass relay.

Assuming a typical Bosch type relay pin 85 and 86 are connected to the coil. One side gets connected to ground and the other to the trigger wire from the MPI relay. When battery voltage is present on the trigger wire the relay should click and connect the 30 and 87 pins. The convention is that 85 is the ground, 86 the trigger, 30 the input and 87 the output.

Once the engine is running the ECU pulls pin 7 on the FP side of the MPI relay low to keep the coil energized and continue switching power to pin 2.

So, you should see battery voltage on the black/white wire everytime the engine is cranking or running, or when you jumper the check connector to the positive terminal on the battery.

You should have battery voltage on your wire from the battery all the time.

The simple test or that relay is ground pin 85, connect the battery to pin 86 and make sure that it clicks. Then measure the resistance from 30 to 87, it should be 0 ohms meaning that there is continuity between the two. Once you remove the battery from pin 86 the relay should click again and the resistance between 30 and 87 should be infinite.

If it doesn't do that then it's bad.

When you tested the relay seems to have energized but it didn't seem to switch pin 30 to pin 87. Is this relay new?

OK, checked resistence, and either I am doing something wrong (highly likely ), or I have found the problem.

The number will jump to about 18.00 ohms immediately, then it will bounce around randomly before finally settling down at 0.00 ohms.

This doesn't seem right for just measuring the two pins on the injector. I might do that it you tried doing the harness.


Sorry this was so long but I think it help to know what should be going on.

Steve

 

[femmeDSM]

Thank you! You're right -- it definitely helps to know how things *should* be working.

The relay that I bought (yes, it is new) is this one: http://www.radioshack.com/product/index.jsp?productId=2062477

I hooked everything up per these instructions (followed them to the letter):
http://www.plymouthlaser.com/rewire.htm

I will test it again using your instructions tonight (if I don't freeze to death ) -- hopefully this is just an issue of a bad relay... I have been noticing some more weird electrical things going on, though. I.e. sometimes when I open the doors, the door locks will try to go into a loop of locking/unlocking repeatedly. It also did this once yesterday while I was cranking the car over...

The MPI relay black/white wire is still getting ~10.5v while cranking (both at the MPI and the pump) -- is there still a voltage problem here, or should this be enough voltage to turn on the pump? I am unsure exactly how much voltage the pump needs to turn on. Perhaps you can clarify.

I will also try to test the injector resistence again...It was dark and cold during my last attempt, so I probably screwed something up. The fuel delivery problem is making me mad, though...Especially now that I know for sure that the car WILL start if it is getting fuel -- it's making me impatient. . Is there some special way that I should be testing the resistence? I just turned my multimeter to Ohms > 20M, and tested like I test everything else...Jeez. I feel dumb sometimes ...haha. I don't know what I would be doing if I didn't have your wise advice

 

[femmeDSM]

UPDATE:
Re-checked resistance on the injectors, and they appear bad. I got readings from 0.2 to 0.9 from each one (numbers still jumped back and forth a bit, but this time stayed consistently within that range).
Also, we got a little crazy with the multimeter and decided to test voltage at the actual injector wires while cranking -- consistant at 10.5v on each injector.

Also, tested the relay and (lo and behold) it was bad! What are the odds, seriously? We bought a new one and tested it -- it works, but the pump still does not turn on while cranking

At this point, should I just go ahead and order new injectors? I've had my eye on some 550's, so now is as good a time as any, I suppose. (I know I will need an S-AFC if I decide to do this).

Also....given that both the MPI relay wire and now the injectors are getting 10.5v, is this a normal amount of voltage to expect? Or could the MPI relay itself not be getting/sending enough power? How would I check that, if so?

I know we're super close to figuring this out -- any day now! .

Thanks

 

[steve]

Is there some special way that I should be testing the resistence? I just turned my multimeter to Ohms > 20M, and tested like I test everything else.

UPDATE:
Re-checked resistance on the injectors, and they appear bad. I got readings from 0.2 to 0.9 from each one (numbers still jumped back and forth a bit, but this time stayed consistently within that range).
Also, we got a little crazy with the multimeter and decided to test voltage at the actual injector wires while cranking -- consistant at 10.5v on each injector.

Were you still using the 20M ohm range? You should be using a 10 or 20 ohm range since the injectors are going to be somewhere around 2 to 3 ohms.

Time to check the relay one more time. If your getting 10.5v at the Black/White wire from the MPI relay that should by plenty to activate the new relay.

Steve

 

[femmeDSM]

The only resistance settings that my multimeter has are: 2M , 20M , 2K , 20K , and 200K. Unless I'm missing something ... Regardless, I checked the resistance @ the injectors on each setting, just to make sure, and I got different variations of "0" on each one (either 0.00 or -0.00) except for 20M, which was the ~0.9 reading...

Also, re-checked the relay (again!) and it is in fact working now. After replacing the original relay with the new/working one (last time I buy automotive parts from Radio Shack...) I had accidentally reversed the Fuel Pump wire with the MPI wire when I hooked it back up....Switching those back to where they belong fixed the problem. I further verified that it's working properly by removing the fuel line from the rail and cranking the car -- lots of fuel came out. So my fuel pump power problem is solved!

This doesn't leave many other parts that could be bad....I've never heard of a "bad fuel rail", so simple logic allows me to deduce that the injectors are shot...........

Yay! Never thought I'd be so happy about bad fuel injectors. . I'm just glad I found the problem (er....at least I *hope* I did....).

 

[steve]

The only resistance settings that my multimeter has are: 2M , 20M , 2K , 20K , and 200K. Unless I'm missing something ... Regardless, I checked the resistance @ the injectors on each setting, just to make sure, and I got different variations of "0" on each one (either 0.00 or -0.00) except for 20M, which was the ~0.9 reading.

Your meter has a 3 digit resolution and resistance ranges of 20,000,000, 2,000,000, 200,000, 20,000 and 2,000 ohms. Your trying to measure 3 ohms, so you need to use the lowest range to get any resolution. (I'm surprised there isn't a 200 ohm range)

It's also important to note what your meter displays as an open (infinite resistance, probes not touching) and a short (0 resistance, probes touching).

In general while troubleshooting we are looking for one of three cases. Short, Open, or out of range. While it's possible, I have a hard time with all your injectors being shorted (reading 0 on the lowest range). On most of the of those resistance ranges, 3 ohms is going to look alot like 0 ohms.

Another way of check the injector is to jumper the power from the resistor (one of the black with color stripe wires to the connector) to one of the pins and ground the other. With the ignition turned on the injector should click each time you ground the other pin. If they don't they could be bad or they could be stuck from old gas. Since the fuel pump hasn't been working up to his point that's possible.

If you get new bigger injectors, your also going to need someway to correct for them being bigger like a SAFC, MAFT, etc.

Steve

 

[femmeDSM]

Yikes, I feel like a total tool ..... I guess I should have figured that the "M" = "Million".....haha. Thanks for clearing that up.......

ANYway! I put it on the smallest setting (2K) and re-tested (yet again ) This time, I got a reading of 0.003 for each one.

I will try turning them on via the jumper and report back. I think you're right in assuming that the injectors could be stuck from all the old gas. Due to the fact that it's been sitting w/out starting for almost a year now, I was originally planning on thoroughly cleaning the injectors anyway. I have done this before on my GST as described in the VFAQ here: http://www.plymouthlaser.com/injectest.htm

Could I go ahead and do that now, just to cut out the possibility that they're stuck?

 

[femmeDSM]

OK, disregard the latest readings.....I gave up and got one of those nifty auto-ranging multimeters....The REAL (absolutely, positively, for real) readings are as follows:

#1: 2.6 | #2: 2.3 | #3: 2.5 | #4: 2.5

I don't know what the hell I was doing before....Chalk it up to a blonde moment, I guess.

 

[steve]

OK, disregard the latest readings. I gave up and got one of those nifty auto-ranging multimeters.

Nice tools always make the job easier.

It seems that the injectors windings are measuring ok, so the next step would be to try pulsing them. I'd also try using that new meter to see if the ECU is trying to fire them.
If you set the meter to measure voltage, ground the black probe and connect the red probe to one of the injector pins on the ECU you should see about battery voltage when the ignition is on and it going almost to ground (0 volts) when the ECU tries to fire the injector.

The VFAQ for cleaning looks like you need a few additional hands and I'd be worried about the injectors popping out of the rail when you pressureize it.

Steve

 

[femmeDSM]

Since we've finally tackled the fuel pump power problem, I figured I would continue my quest for a running car in another thread, so this one doesn't continue to stray in several different off-topic directions and become confusing for people who might need to search the boards for these problems later. haha.

Here is the continuation thread : http://dsmtuners.com/forums/showthread.php?p=1731313

-----

Steve, you have been a tremendous help, although I'm sure you're getting tired of me by now