Newbie ForumBeginner/newbie/general DSM questions. first mods, how to run 10's when you haven't run 12's yet, any tech question that doesn't fit in another tech forum. New Members must limit their tech posts to this forum and sub-forums.
I can't get it to run and I found out I wasn't getting fuel. I checked the fuel pump with the fuel pump check plug next to the battery, and the fuel pump works. Checked the MPI relay and also tried a different one, and still wouldn't start. MPI relay ok
My next test was to spray fuel into the throttle body, did that and it started up for a few seconds. It ran fine for those few seconds, so I have narrowed it down to a fuel problem. If the MPI relay is ok, i'm thinking its a bad ECU? Is this correct or can it be something else???
Also, the car is a 94 Turbo AWD with an EPROM ECU. If the ECU is bad, do i have to put an EPROM ECU back in it? Can I use any ECU from a 90-94 turbo AWD? The caps in the ECU look good and the ECU looks to be in good shape in general.
Can someone please help?!?!
Advertisement
To browse the forums without the advertisements above, Login/Register
Get a Noid light and see if the injectors are firing first. Just because it's not getting fuel doesn't mean that it is the computer. Did this just happen or was there a problem that has been getting worse. When say the fuel pump works, do you mean that it actually pumps fuel or it just runs? A pump can run and not actually flow fuel.
Step 1 - Check injector pulse with Noid light
Step 2 - Check fuel pressure with guage
I haven't tested the injectors but I'm not getting fuel to the fuel rail. When I check the pump with the fuel pump check connector, I loosened the bolt on the top of the fuel filter. The pump turned on and pump fuel out of the filter.
The fuel pump isn't turning on when the key is in the ON position or when the engine is cranking.
i had a similar problem back when i owned a honda. pull the ecu out and let it sit for 30 seconds and plug it back in. it will reset the computer. if that doesnt work, then i idont kno because it worked for me
So you know for a fact that the pump does not run with the ECU? Just making sure.
Spraying fuel all over the engine compartment is not the correct way to check fuel pressure. Buy a guage or find someone that has one before you burn your car to the ground. Just because fuel comes out does not mean that the pump has enough pressure to supply the injector with enough fuel for the engine to run.
Step 0 - Consult a service manual/wiring diagram
Step 1 - Check fuel pressure WITH A GUAGE
Step 2 - If no fuel pressure, check for power/ground at pump
Step 3 - If no power/ground at pump, check for power/ground at ECU
Step 4 - If power/ground are ok at ECU and not at pump, check resistance of wiring
Step 5 - If power/ground are not ok at ECU, check all power/grounds for the ECU
Step 6 - If all power/grounds for the ECU are ok, then replace the ECU
The MPI relay is grounded by the ecu which closes the fuel pump circuit allowing power to the pump. Consult wiring diagram and probe that wire. If there is no ground, resistance test from connector at relay to the same wire at the ecu plug. If there is no resistance between the wire or minimal resistance (>.1ohm), that means there are no shorting/breaking of the wire, and that the ecu isn't functioning properly.
ok so the update is that the ECU is working properly. I sent it out to Steve to change the caps and he said the fuel pump runs properly with my ECU in his car.
So not I'm thinking its probably a harness problem?? If it is, am I going to have to switch out the whole thing? I hope this fix isn't a major project!
tempature sensor grounds ecu so if that doesn't work your car won't start, so be aware of that. There are also a few other sensor that will keep the car from start however, I can tell you are getting spark because your car started. you can tell if you ecu is bad take it out and open it up and smell it if it smells like rotten eggs it's toast
I have checked that the pump works through the fuel pump check connector. It turns on through this connector but not the key. That is the only testing on that I done on the fuel pump so far. I'm still reading through the manuel you see what the next tests i should do.
I have checked that the pump works through the fuel pump check connector. It turns on through this connector but not the key. That is the only testing on that I done on the fuel pump so far. I'm still reading through the manuel you see what the next tests i should do.
Now have you checked the control relay? since you got it to power up from the check point...
The control relay? You mean the MPI relay? I have checked that and even tried another one and both of them test out OK, but still don't turn the fuel pump on.
The control relay? You mean the MPI relay? I have checked that and even tried another one and both of them test out OK, but still don't turn the fuel pump on.
Same thing...lol
So does the relay trigger, power coming to the coil side of the relay (pin 7).
I assume there is no power coming into pin 3 on c-63 plug (mpi relay plug)
Does your Cam Angle Sensor work? I'd try swapping a known good one in and see if that helps with your problem. When the CAS spins it does two things 1) it tells the ECU where the camshaft is in its rotation in order for the spark to be fired at the correct timing. 2) It tells the ECU when the engine is turning in general. This kicks on the fuel pump side of the MPI relay and starts it pumping. Have you ever noticed if your car dies while the ignition key is still on that the fuel pump still pumps for about 2-3 seconds after it dies? The CAS is telling the ECU that the engine is not turning and therefore the fuel pump stops.
If you have a logger you can test some things. Go to the "actuators" section and click on fuel pump relay test. It should kick on and then turn off after about 5 seconds. If it does, then the circuitry in the ECU should be good (which you already verified) and the ECU is just not getting the signal to close that circuit.
I'm betting you don't have spark or a fuel injector pulse either. All of which would be related to a bad CAS.
If I were you, I would post 6 more posts in our tech section to make yourself a "Proven Member". Then, look through the classifieds for a used CAS. They're always floating around in there cheap as hell too!!
____________________________
Kletke - 16g @ 27psi on Meth
I do have spark to all cylinders and good spark to, so that would mean the CAS is working correctly right? I do not have fuel to the fuel rail, but the car did start for a few seconds when we sprayed fuel into the throttle body.
it may not be your ecu. theres probably a problem with your wiring
Not ECU...
I believe he's already stated he has shipped it to steve and he replaced the caps and tested it in his very own car.
Well... there are two sides to the CAS according to the FSM they are called the Crank Angle and TDC Sensor. There are 4 wires that go to the CAS:
pin 1 White TDC (to ECU pin 22)
pin 2 Black Crank Angle (to ECU pin 21)
pin 3 Red +12v Power
pin 4 (Larger)Black Ground
To test it you should get four pulses per revolution off pin 2 at the CAS and 2 pulses off of pin 1 at the CAS. This is to give the ECU two signals a 4 pulse beat for the crankshaft and a 2 pulse beat for the camshaft. Since the crankshaft spins twice as fast as the camshaft.
Now... The part of the FSM I looked into isn't clear on the logic of the ECU as to which pulse signal controls pin 56 that goes to the fuel pump side of the MPI relay to ground it out to pull down the relay and kick on the pump.
____________________________
Kletke - 16g @ 27psi on Meth
Right. Just because you're getting spark, doesn't mean your getting spark at the right time. The CAS could be 180* out.
By the way, ohm tests suck and I'll tell you why.
Imagine in a wire there are four strands of copper. If three of the four strands of wire are broken, yet the one wire has the same resistance compared to the four combined - your ohm test is inconclusive. Amperage will be more capable of flowing through four strands, then one. That is why you always do voltage drop tests for anything automotive related.
Also, I just found this while searching on your particular issue trying to futher understand the CAS operation.
"The Crank Angle Sensor and the Top Dead Center Sensor are located on the right side of the cylinder head. These sensors are incorporated into one unit, being a disc and a pick-up unit. The disc is affixed to a shaft that is coupled to the camshaft and the light-transmitting unit is mounted stationary to the units housing.
Crank Angle And #1 TDC Sensor
The disc contains 4 large slits around its circumference to indicate the crankshaft angle. An additional light-transmission slit located inward from the edge is used to indicate number one cylinder's top dead center position. The pick-up unit assembly uses two luminous diodes and two photo diodes, in order to detect the two different slits. There is a very slight clearance between the luminous diodes and the photo diodes, and the disc rotates within this space.
As the shaft rotates the slits at the discs edge pass between the light and the optical reading part of the unit. The light emitted from the luminous diodes pass through the slits to the photo sensing diodes. When the photo diodes receive the light, they become conductive and generate a signal, which is sent to the Control Unit.
The Control Unit is able to detect number one cylinder TDC by the signal generated through the single inner slit on the disc. The sequence of the fuel injector firing is based on this signal.
The slits at the outer circumference of the disc represent the exhaust stroke and serve to detect the position of the crankshaft (and, therefore, the piston) relative to top dead center.
Based on these signals, the Control Unit is able to determine the fuel injection pulse width, ignition timing, and also the intake air flow rate for each revolution of the engine."
So, if one of the diodes inside the CAS is bad it is possible for you to get spark, possibly not at the correct timing, and not fuel... Then again, someone like steve might know more about the logic inside the ECU and how it uses the two signals.
____________________________
Kletke - 16g @ 27psi on Meth
The fact that the pump doesn't turn one while cranking is a clue here.
On 1G's the fuel pump side of the MPI relay has two coils to close the contacts. One is ECU controlled and the other in the starter circuit. The whole fuel pump power circuit isn't that complex and can be debugged with a multimeter.
The fuel pump is powered from the IG1 terminal of the ignition switch as is the coil, power transistor, and Check Engine Light. You should have battery voltage (B+) at pin 3 of the MPI relay when the ignition switch is in the RUN or START positions. Both the relay contacts and the first winding is connected here. The other end of this winding is pin 7 of the MPI relay and runs to the ECU pin 56. You should see B+ here too except when the ECU is trying to run the pump. It does so by pulling to ground (about 0v) causing current to flow through the winding. The ECU watches the CAS to decide if it should be running the fuel pump.
The second winding gets it's power from the starter relay branching off from the wire to the starter solenoid and running to pin 9 of the MPI relay. The other side, pin 6, is always connected to ground #7 so that anytime there is power on pin 9 the fuel pump relay is activated. The source of this signal differs depending on manual vs automatic and if your car has the factory theft alarm but if the starter cranks you should see power on pin 9.
The output of the relay on pin 2 runs to the fuel pump and the check connector under the hood. You should see B+ on pin 2 whenever the fuel pump should be on. if you power the pump from the check connector you should also see the voltage here and if you don't the harness is damaged.
The last part of the circuit is the pump ground. Inside the tank the negative side of the pump is connected to the mounting hardware. There is a wire from ground #9 on the rear bulkhead below the hatch latch to the top of the fuel pump assembly.
Since he's not seeing the pump run at all, the problem would seem to either be in the power side of the fuel pump circuit, not in the ECU or sensors. Since the pump runs from the check connector it also looks like the problem is before that connector which leaves us with the harness, relay and power to the relay.
If you have power at pin 2 of the MPI relay when you crank the engine, but not at the check connector, then there's a break in the wire somewhere between the two. I commonly find breaks in wires at splices where wires connect. Wires like to corrode at these points. There may be a splice where the check connector is branched off.
Ahhh... I did not know that the starter circuit also powered the fuel side of the MPI relay. I thought it only looked at the CAS to tell it when to kick on.
Great info here...
____________________________
Kletke - 16g @ 27psi on Meth
Last edited by ->PrOjEcTGS<-; 11-08-2009 at 11:54 AM.
Reason: spelling
Ahhh... I did no know that the starter circuit also powered the fuel side of the MPI relay. I thought it only looked at the CAS to tell it when to kick on.
Great info here...
Yeah my starter is tempermental, but when it doesn't crank, my pump always kicks on.
Yeah my starter is tempermental, but when it doesn't crank, my pump always kicks on.
Right on... Now that I look at the schematic again it makes more sense. Damn dotted lines....
My next question is a hypothetical.
What if the starter is cranking the engine over but pin 9 of the MPI relay doesn't get a signal from the starter relay to engage the fuel pump. Looking at the schematic, if the CAS is spinning and the ECU is getting a signal from that shouldn't it pull the coil down at pin 7 of the relay and engage the fuel pump? This is of course given the fact that all the rest of the wiring from the relay to check connector is good and the relay is getting power from IG1? If this is true, then he would have power to the relay and a signal from the ecu at pin 7 to close the relay and engage the pump. But, in his case it doesn't, so what are the chances that both the coils are bad in the relay. Slim... This should indicate a problem with power to the relay or a short in the harness.
OP, to troubleshoot this ^ I would pull the cover off of the MPI relay turn the key to the "RUN" position and manually engage the fuel side to see if your getting power from the relay to your pump. Then if it doesn't kick on when manually engaged I would then start checking for power at the relay via the pinouts mentioned by steve.
____________________________
Kletke - 16g @ 27psi on Meth
What if the starter is cranking the engine over but pin 9 of the MPI relay doesn't get a signal from the starter relay to engage the fuel pump.
Looking at the schematic, if the CAS is spinning and the ECU is getting a signal from that shouldn't it pull the coil down at pin 7 of the relay and engage the fuel pump?
This is of course given the fact that all the rest of the wiring from the relay to check connector is good and the relay is getting power from IG1? If this is true, then he would have power to the relay and a signal from the ecu at pin 7 to close the relay and engage the pump. But, in his case it doesn't, so what are the chances that both the coils are bad in the relay. Slim... This should indicate a problem with power to the relay or a short in the harness.
Yes, If the engine is cranking and the ECU is getting CAS signals the ECU will pull pin 7 low to run the pump. I can't remember if the ECU gates any of this by the state of the ST input to the ECU or not.
I pretty much boiled the available information down to this point already. If the contacts were bad in the relay it could also cause this. So like I said there looks to be a fault in the power side.
Quote:
Originally Posted by ->PrOjEcTGS<-
OP, to troubleshoot this ^ I would pull the cover off of the MPI relay turn the key to the "RUN" position and manually engage the fuel side to see if your getting power from the relay to your pump. Then if it doesn't kick on when manually engaged I would then start checking for power at the relay via the pinouts mentioned by steve.
There is no reason to pull the cover off, just test the relay. You can here it click if it activates and you don't risk damaging it.
