Injector Voltage

[whitetiger]

I have no injector spray....

injector 1 12v on both
injector 2 12v and 1v
injector 3 12v and 1v
injector 4 12v and 1v

ECU wires 51-52-60-61 all have 12v+
mpfi relay good
mpfi fuse good
new fuel pump
new filter
rail has fuel pressure


how do I fix?

 

[steve]

I have no injector spray....

injector 1 12v on both
injector 2 12v and 1v
injector 3 12v and 1v
injector 4 12v and 1v

ECU wires 51-52-60-61 all have 12v+

I know you have been struggling with this for a while. If Squirrels were eating your hoses and wires like you've said then share what you've done to fix them. Right now I'm not confident in any of your assesments as to whats ok and what's not. No offense intended.

Here is the best (only) '90 wiring diagrams I've found.
http://www.ecanfix.com/~mdhamilton/1990schematics.html
Here's a '91+ for comparison.
http://www.taboospeedshop.com/wiring.htm

I also know you were reading the other thread with 4g63tsiawd, so know that you can't have +12v at the ECU and 1v at the injector. Your either measuring wrong or something is wrong with your harness. I would hope since you own a car audio store that you understand the basics of electronics and can read circuit diagrams. The injector circuit is real simple until you get inside the ECU. Hopefully the diagrams above will help you.

We knew in 4g63tsiawd's thread that his car would run if he gave it fuel directly. That told us his CAS, ignition, and ECU were working. From what you've posted we don't know that to be the case with your car.

So lets start back at the beginning and see how you came to where your at.

Steve

 

[whitetiger]

I know you have been struggling with this for a while. If Squirrels were eating your hoses and wires like you've said then share what you've done to fix them. Right now I'm not confident in any of your assesments as to whats ok and what's not. No offense intended.

So lets start back at the beginning and see how you came to where your at.

Steve

None taken.

Mechanically the car appears fine: no shaft play on the turbo, clean oil, all drive train fluids were full, brakes checked out, and so on for the initial run down before buying the car - I was satisfied the car would need plenty of work - but I couldn't pass it up having owned a 91 Eclipse some years back. This AWD Turbo has to be even more fun!

I knew the car had been sitting for a few months, so basics had to be done: I lubed the cylinders overnight, drained and filled fluids.

I checked the vacuum diagram and replaced 3 hoses as the diagram in the manual said. easy enough.

I bought a new ECU, do to a shorted battery. The previous owner parked the car because they didn't use a battery tie down so the battery bounced hit the hood and grounded out while driving. There's a big ARC burn on the underside of the hood.

Checked all air ways, clear and clean after removing the nest in the filter can.

I found a great pin out for the ECU, pulled the entire wiring harness for the ecu, I cut the harness open and traced EVERY SINGLE WIRE from the ECU and made sure they had continuity and no frays. It took about 4 hours, as I had to repair multiple breaks due to the rodents. (The CAS plug, cruise, Timing adjust ground, coil pack plug, O2 sensor, MAF sensor, and may some others). When I finished I quadruple checked my work (I didn't want to have to redo any of this) then loomed and taped the harness back and re-installed it.

Dropped the fuel tank for cleaning, added new filter and pump, pulled fuel rail, and injectors, soaked injectors in seafoam overnight, purged the fuel lines and put it all together. I had a fully operational fuel system - I checked the return side of the rail for pressure by removing the return hose, placing a catch can and rag and bumping the key... I shoulda got a bigger catch can, I had all kinds of fuel supply.

New plugs and plug wires - of course. I grounded one plug at a time and bumped the motor, each plug was nice and strong with spark.

I worked over all the grounds for 100% clean metal to metal contact.

That's all I can think of now.

Oh yeah, I forgot to tell you guys. The car ran for about 5 seconds after putting the freshly cleaned injectors in. (they had B-12 Chemtool and SEAFOAM all over them).

I'm so excited to be this close, hopefully this is the last step and I'll have this motor running.

I'm really stooped on the whole injector puzzle here, please help.

 

[steve]

Oh yeah, I forgot to tell you guys. The car ran for about 5 seconds after putting the freshly cleaned injectors in. (they had B-12 Chemtool and SEAFOAM all over them).

Now that's interesting. You didn't put anything in the cylinders just whatever was in the injectors and the car fired?

One thing you can do to verify they work is ground the ECU side of the injector yourself. You should hear the injector tick as it opens and again when you disconnect and it closes. As long as the power is supplied via the stock harness (use a jumper to go from the connector to the injector and another to ground the other pin) and you don't hold them open for more than a few seconds you shouldn't hurt the injector.

The fuel pump is only run while the ignition key is in start position. If the car starts the ECU will pull another coil in the relay to keep it running.

Do you get any CEL codes?
You might want to make a datalogger cable so you can see what the ECU is reading from the various senders. You need a laptop with a serial port, a male DB9 connector, a 1N914 switching diode, a 10K resistor and some wire. There are posts by me on how to make one for a PC.

 

[whitetiger]

saying driverside is the right side #1, is the right or left wire on #1 supposed to be ground? and should the be 12v on all 8 wires?

and as far as the car firing, I cleaned the surface where the injectors meet the intake with B-12... I'm sure some went into the cylinder. I'm not a fan of priming the motor manually so I haven't done it again.

CAS? Knock sensor? ECU faulty? resistor box? I don't know what to check next.

CEL is never comes on

 

[steve]

saying driverside is the right side #1, is the right or left wire on #1 supposed to be ground? and should the be 12v on all 8 wires?

Remember the diagrams above? According the the '90 turbo ECU diagram and consistant with the 91+ diagrams, the power feeds to the injector pin 1 from the injector resistors should be black-yellow (1), black-blue(2), black-white(3), black-red(4). The leads that the ECU uses to pull pin 2 of the injector coil low are yellow-blue(1), yellow-black(2), light green(3), light green-white(4).

When the connector is off the injector only one pin will be at battery potential. The side to the ECU will be floating since they go to open collector transistor drivers. I can't tell you which pin is which, I assume when your looking over the top of the engine from the bumper the left one is pin 1 but that's only because of convention in places the read left to right.

Number 1 cyl is the driverside closest to the timing belt.

and as far as the car firing, I cleaned the surface where the injectors meet the intake with B-12... I'm sure some went into the cylinder. I'm not a fan of priming the motor manually so I haven't done it again.

CAS? Knock sensor? ECU faulty? resistor box? I don't know what to check next.

Darn, sounds like you might have been buring what was in the intake, rather than something coming out of the injector.

The CAS most noticably controls spark even though it's the main timing reference for everything, it's just easier to see if it's sparking to check the CAS.

Getting 12v to the injectors goes throught the resistor so it would seem their not open.

ECU is still a unknown. Until hear or have proof that it's firing the injectors, it can't be ruled out. But I'd stick to checking that the injectors open first.

If you have a scope the next thing I'd do is connect to the ECU and watch to see it switch it's side of the injector to ground when you crank.

It's possible that your Engine Coolant Sensor is bad. It's reported to cause no starts but mine hasn't ever failed so I don't know if it floods the engine or interferes with the injectors firing. (* Tells the ECU it's really, really cold out which floods the engine *)

 

[whitetiger]

I used my volt/ohm meter to check the injector wires at the ECU. I have voltage flux from lowest of 0 to highest of 12v - but most of the low voltage is in the 2-4v range. I assume this to be the ECU pulsing faster that what my meter reads. my meter is somewhat slow on response.

injector 1 has 12v and 12v
injector 2 has 12v and 1v
injector 3 has 12v and 1v
injector 4 has 12v and 1v

in this case the 12v side is ECU side? and the 1v is what?

can I put 12v on the 1v wires and see if the ECU will ground and give me injector spray?

 

[steve]

You were checking while the car was cranking?

What are you trying to tell me with:

injector 1 has 12v and 12v
injector 2 has 12v and 1v
injector 3 has 12v and 1v
injector 4 has 12v and 1v

Where were you measuring these voltages and under what conditions?

Steve

 

[whitetiger]

key on

I want to know if the 12v measurements are the ecu side of the injectors?
Is the resistor box wired between the ecu and injectors?
and the wires measuring only 1v - where do they go? are they the pins 54 and 55 on the ECU?

as I understand I need 12v on all injector wires, correct? then the ECU grounds the loop to make injectors fire, correct?

forgive me to getting really back to basics... I just don't want to overlook any thing.

maybe the parts store gave me the wrong ecu. this ecu has no EPROM. I think the old one did. but can't be sure. reason I say this, I called the parts store today, and ask for an ECU for my car (90), they said they couldn't get it but they could get me one for a 91.

how can I tell if it's the proper one?

 

[steve]

Have you looked at the wiring diagrams I posted earlier, it's all there.

The voltage goes from the battery, to the MPI fuse, to the MPI relay, to the Injector resistor, to the injector coil, to the ECU. Normally the ECU lets it's side of the injectors float. That's why you should see 12v on both sides of the injector with the key on. Nothing is effectively connected to the ECU side of the injector coil until the ECU wants to fire the injector. During cranking and idling the ECU side of the injector goes low (close to ground, it might read 1v on your meter) for 2 milliseconds pulses each time the ECU wants the injector to fire.

Again if you disconnect the connector from the injector the only pin on the wiring harness at 12v is the one to the injector resistor. You shouldn't see any sigificant voltage on the other pin. There may be a difference in ground potential between where you are grounding your meter and the ECU's ground reference.

The Squirrels didn't eat your engine grounds did they?

Steve

 

[GTM]

Now that's interesting. You didn't put anything in the cylinders just whatever was in the injectors and the car fired?
...

The fuel pump is only run while the ignition key is in start position if the car starts the ECU will pull another coil in the relay to keep it running.
...
Steve

Actually most systems will have the fuel pump come on for a second or so when turned to on if the pressure has dropped below a specified value. When you change the fuel filter you can prime the system by just turning the key on and off several times until the pump stops. Then you can engage the start without having to go through excess cranking.

I've not followed all the activity so not up to speed on all the events. Sometimes it's helpful to pull the plugs, defeat the fuel pump and crank for a few seconds to be sure the system is not wet with gas and oil. The fact it started and ran for a few seconds might suggest the plugs got wet once the engine started.

Another option if not tried is to disable the fuel pump and bleed the pressure down. Then try starting fluid to start the car. I've used Chemtool for decades to start engines but also know it's possible to flood them so if there is some question this condition could exist then starting fluid might be a better choice.

Cheers,
GTM

 

[steve]

maybe the parts store gave me the wrong ecu. this ecu has no EPROM. I think the old one did. but can't be sure. reason I say this, I called the parts store today, and ask for an ECU for my car (90), they said they couldn't get it but they could get me one for a 91.

how can I tell if it's the proper one?

'90 AWD Fed Turbo MT is MD145903 it will say 4478 on the white edge connector
The FWD ECU will work.
'90 FWD Fed Turbo MT is MD128625 it will say 4474 on the whie edge connector

Anything else is the wrong ECU. A later ECU won't keep the car from starting but will cause tach problems in a '90 even after you switch the the 2 pins that are different between 90 and later cars.

Let me know if you still have the old ECU.

Steve

 

[whitetiger]

Ecu 6578e on the white plastic inside the casing

and the case says md166260 ( I had to peal the REMAUFACTURED sticker off to see this)

 

[steve]

Actually most systems will have the fuel pump come on for a second or so when turned to on if the pressure has dropped below a specified value. When you change the fuel filter you can prime the system by just turning the key on and off several times until the pump stops. Then you can engage the start without having to go through excess cranking.

Come on man, you know me as well as you can over the internet. If the 1G DSM worked that what I would have said so. On a 1G the system is primed by turning the key to start.
Not only can you hear the pump turn on if it did, I've had a scope on the fuel pump power line and checked and read the ECU code that turns the pump on and off and it doesn't have any feedback on fuel pressure.

I've got to go back to writing ECU code. Maybe you can help him out. I'll look in tomorrow and see if it's still a problem.

Steve

 

[steve]

Ecu 6578e

MD166262 E2T36578 6/90-6/92 Federal MT 1G DSM Turbo AWD (That's 91 and 92 model years) '90 cars were build in '89 and up to beginning of June 90 when they switch to making '91s

Steve

 

[whitetiger]

E2T36574

1926 M



will this ECU work or no? do I have to switch pins? is this my problem?

 

[GTM]

Come on man, you know me as well as you can over the internet. If the 1G DSM worked that what I would have said so. On a 1G the system is primed by turning the key to start.
Not only can you hear the pump turn on if it did, I've had a scope on the fuel pump power line and checked and read the ECU code that turns the pump on and off and it doesn't have any feedback on fuel pressure.

I've got to go back to writing ECU code. Maybe you can help him out. I'll look in tomorrow and see if it's still a problem.

Steve

I don't mind being wrong, thought I had this discussion with my son a year or so back and we sometime later heard the pump come on when the key was swicthed to on. Unfortunately I don't know where his manual is and I couldn't read those diagrams so I'm at a loss.

I think he has to determine what his problems are and I made some suggestions which may be valid or may not. I am familiar with the damage rodents can do to some plastics. Had a customer tow their new Z car from Grand Canyon to Los Angeles claiming it was covered under warranty. It's just a matter of eleminating some things and re-checking his work.

Cheers,
GTM

 

[whitetiger]

I found this to be a FWD 91-93 ECU

and further investigation into using this ECU leads me to believe that this may indeed be my problem.

http://www.ecanfix.com/users/mdhamilton/ecuwiring.html

at this link I found that pin 6 and 14 can be swapped but noticed that pin 110 is different between each ECU meaning that my 90 has a {MPI relay} not a {"1g" terminal for ingnition switch}

so maybe the MPI relay is not getting it's signal from the ECU
and obviously the MAS and Idle switch are backwards.

being new to this I'm not sure if this would keep the injectors from firing... but I'm kinda pissed that the parts store opted to give me the wrong ECU.
I don't really look forward to do this conversion: http://www.ecanfix.com/users/mdhamilton/ecuconversion.html

 

[GTM]

I found this to be a FWD 91-93 ECU

and further investigation into using this ECU leads me to believe that this may indeed be my problem.
...
being new to this I'm not sure if this would keep the injectors from firing... but I'm kinda pissed that the parts store opted to give me the wrong ECU.
I don't really look forward to do this conversion: http://www.ecanfix.com/users/mdhamilton/ecuconversion.html

I would exercise option #1, call the store and indicate they gave you the wrong ECU and take it back. You have had enough grief with this unless there is some advantage to having the later unit. If the don't have the correct unit and can't get it for some time you determine if you want to go through the mods to make it work. If you are not comfortable with those mods then you might be faced with getting your old unit and your money back to find another source.

Cheers,
GTM

 

[steve]

I found that pin 6 and 14 can be swapped but noticed that pin 110 is different between each ECU meaning that my 90 has a {MPI relay} not a {"1g" terminal for ingnition switch}

so maybe the MPI relay is not getting it's signal from the ECU
and obviously the MAS and Idle switch are backwards.

Pin 110 is labled differently but it's the input from the ignition switch on position to all 1G ECU's. It's used to bring the ECU out of it's standby state. With the ignition switch off the ECU draws power off pin 103 to keep the RAM in the CPU chip valid. When it sees 110 go to 12 v it pulls pin 63 low turning on the power side of the MPI relay. This supplies power to ECU pins 102 and 107 booting the ECU as well as providing power to the MAF, CAS, ISC, Solenoids, and Injectors. You'll see the CEL light come on for a few seconds and go off when the ECU boots.

The MPI relay has a second side independant from the first. That's a dual input (coils) relay for the fuel pump.
One of those inputs is from the starter circuit and the other from the ECU.

When the starter engages it provides 12v to one of the coils. That turns the fuel pump on. When the ECU sees that the engine is running, it pulls the other coil low to keep the fuel pump running when you release the ignition swiitch from the start position. The ECU monitors that the engine is running and will turn off the fuel pump shortly after the engine stops running for safety. The output of the fuel pump side of the MPI relay goes to the fuel pump check connector in the engine compartment and to the fuel pump. You can measure it to verify power when it should be running or apply battery voltage to it to pressure test the fuel system for leaks with car off.

The IPS and MAF reset signals are swapped in position between '90 and later but neither will cause the engine not to start from what I've seen. They cause drivability problems if swapped since the ECU doesn't see that the throttle is closed (IPS) and can't reset the MAF airflow counter between shifts.

Something that occurred to me is to check the TPS voltage. I believe that if the ECU sees the throttle being wide open dusing starting it disables the injectors to allow you to clear a flooded condition. (* Nope, only DSMLink does that *) The output of the TPS is pin 19 on the ECU (pin 2 on the TPS) and it should be close to 0.5v with the throttle closed. If it's closer to 5v (I forget the trigger voltage) the ECU thinks the throttle is WOT.

So, having the wrong year ECU doesn't normally keep the car from starting but it will cause problems with the tach functioning on a '90 car like I said earlier. This is caused by the tach signal being inverted on a '90 and by the difference in how the signal is terminated in the later ECUs at pin 109. The later style ECU don't mind the inverted signal but their termination messes up the tach gauge reading the signal at times. I haven't had a '90 car to work on to get to the root cause.

 

[whitetiger]

with the key on in the RUN posotion
I tested the injectors again... 2.6 ohms each. I slid the injector boot back to expose the injector wires, tested both sides of all 8 wires, 12v on each.

Taking what you said about manually pulling the injectors low to teh ECu side - I pulled the injector plugs off - retested all 8 wires... the 12v wires going to the resistor box and the ecu wires tested 0 since the ecu floats ... NOW I knew which wires were what...

I pugged them back up, leaving the boot off to expose the wires I was working with. I got a length of wire and attatched it to my meter probe so I could watch the voltage on the ECU side of the injectors, the other end of the wire I contacted to the ecu side of the injector plug... taking my meter probe I tapped it to GROUND, the voltage dropped but no injector tick. I aslo tried the same test without the meter probe (direct ground) still no ticking injectors.

I was close enough to hear them and I was it was quiet around me when I tested.

 

[whitetiger]

I also tested the TPS with the key on = .5v closed and 5.04v at WOT at pin 19

what's next?

I'm not getting a CEL at all - ever - key on, off, acc, or at start position

 

[GTM]

I also tested the TPS with the key on = .5v closed and 5.04v at WOT at pin 19

what's next?

Have you tried pulling your hair out? Hope a little levity is not out of line.

While waiting for Steve to get back with some more specific info I'm trying to figure out why you would have 12v on both sides of the injectors if that was while cranking?

Don't mean to be mucking up Steves work for it's pretty obvious he's very sharp on this stuff.

Did you call the store that sold you the wrong part?

I would be interested in knowing why the engine crankshaft sensor would not play a roll in starting. It would seem to me that the later ECU which you have, would be looking for this input and not finding it would possibly cause the problem you are having. Maybe when this gets solved I'll be up to speed.

Cheers,
GTM

 

[steve]

At this point I'm worried about the injectors not ticking when you manually ground the ECU side manually completing the circuit. To check it, I grabbed one of my old 450cc injectors, a 20W 8 ohm resistor, a couple of meters, my bench battery and hooked it up. The bench battery is on it's last legs and only puts out about 9v right now and drops to about 5v under load with this test circuit drawing about 500mA, but that was enough to fire the injector with a clearly audible metalic click. So I wonder if yours are either gummed up, if we have some other problem or you have the worlds quietest injectors.

 

[whitetiger]

I tested the voltage with the injectors plugged up while a friend tried to start the car...

here are my findings:

battery voltage 12v
battery voltage during crank 11.47

all injector voltage 12v
all injectors during crank 12v~5v

I assume the pulse is faster than my voltmeter and that's why 5v is as low as my readings go.

still no appearant injector tick and no gas going into the motor.