'84 Carrera won't start

[logician]

Time to play Click and Clack. I picked up what appears to be a completely unmolested '84 Carrera recently. But car won't start. Here's what I know:

0) Car does not show any signs of catching. Not the slightest hint of a cough.

1) Car cranks really well.

2) A timing light on the #1 plug wire shows it's getting spark.

3) When I jump 12V directly to the fuel pump fuse I can hear the pump running. Don't know if it's pumping, though.

4) When I crank the engine, I can see 12V being supplied to the fuel pump fuse.

5) Tank indicates 1/2 full.

6) I *don't* smell any gas in the exhaust after cranking for a while (30 seconds). Implies zero fuel being injected.

7) The car came with a spare DME relay that looks new, but I have no knowledge of it's condition. Swapping that DME relay in doesn't help.

My conclusions based on this info:

a) The DME relay is working to supply power to the fuel pump.

b) The fuel pump is running when I crank the engine, but I don't know if its actually pumping fluid.

What are the likely problems that would prevent the car from starting?

Thanks!
-Juan

[timmy2]

First guess, plugged injectors.

[logician]

Clogged injectors doesn't seem like a good match. Unlikely that all six injectors would be clogged so completely that there would be no smell of gas at all.

[timmy2]

Well, logic would dictate removing/loosening fuel lines to check for fuel flow from the first connection as the line comes into the engine bay all the way to the injectors ensuring there is flow at all components. Eliminate everything in between and if the injectors are good it is something else.

[timmy2]

How long has the car sat without running?

[logician]

Not sure how long it sat without running. But my guess is measured in low years.

[cabmandone]

Put a fuel pressure gauge back at the fuel rail test port. Get a noid light and test for injector pulse. Get and I can't stress this enough a KNOWN new good DME relay. Is the engine original to the car or was it put in by the P.O. could have crossed the Speed and Reference sensors when installed which will keep it from firing.

[COLB]

Is the engine original to the car or was it put in by the P.O. could have crossed the Speed and Reference sensors when installed which will keep it from firing.

Despite my best efforts, I did this when I replaced the sensors.

A two minute swap of the connections, and it fired right up. Definitely worth trying since it is easy and free.

[ischmitz]

It is stuck injectors. They tend to stick close (yes, all six of them) after about 3 month.

To confirm introduce starter fluid into the air intake to see that the engine catches and fires.

Next, have someone crank while you tap real good on the injector body. That'll most time free them up. If not, the next step is to shock them open. Do it an injector at a time: Connect one side to GND and then very quickly strike the other side with +12V. This will get them to work again.

Report back,
Ingo

[SiberianDVM]

Originally Posted by COLB

Despite my best efforts, I did this when I replaced the sensors.

A two minute swap of the connections, and it fired right up. Definitely worth trying since it is easy and free.

Don't beat yourself up, it's easy to do since they are interchangible. I put colored tape on mine after I screwed them up the first time.

[ischmitz]

Originally Posted by cabmando

Put a fuel pressure gauge back at the fuel rail test port. Get a noid light and test for injector pulse. Get and I can't stress this enough a KNOWN new good DME relay. Is the engine original to the car or was it put in by the P.O. could have crossed the Speed and Reference sensors when installed which will keep it from firing.

If he has spark it clears the sensors (he has confirmed that). If the fuel pump runs during cranking it clears the DME relay (check with DVM on fuse).

what's left is: No fuel

- no fuel pressure (clogged filter, etc.)
- injectors stuck (very likely after long sitting)
- bad DME with fuel output stage (due to short or random)
- bad DME (P&H driver due to mis-wired tachometer)

Ingo

[mysocal911]

"To confirm introduce starter fluid into the air intake to see that the engine catches and fires."

The easiest test to do at this point. Carb/brake cleaner also works.

[logician]

Great suggestions everyone. I'm trying to queue up a bunch of things to try when I go to my shop later today.

Question: how does the DME determine if the engine is turning over and so to enable fuel? Some cars do that via the air box, so if the computer sees any airflow, it knows the engine is turning over. Is this what the 911 DME does? Or does it use other means (the crank sensor, etc?) I'm trying to understand if a bad air box could be a scenario.

(Missing OBD2... it seems unfathomable these days that anyone would design a computer controller without any diagnostic ability.)

[mysocal911]

Originally Posted by logician

Great suggestions everyone. I'm trying to queue up a bunch of things to try when I go to my shop later today.

Question: how does the DME determine if the engine is turning over and so to enable fuel? Some cars do that via the air box, so if the computer sees any airflow, it knows the engine is turning over. Is this what the 911 DME does? Or does it use other means (the crank sensor, etc?) I'm trying to understand if a bad air box could be a scenario.

(Missing OBD2... it seems unfathomable these days that anyone would design a computer controller without any diagnostic ability.)

The DME ECM uses the reference sensor to indicate that the engine is cranking.
Subsequently the speed sensor is required to actually develop spark and injection
pulses.

[mysocal911]

Originally Posted by ischmitz

It is stuck injectors. They tend to stick close (yes, all six of them) after about 3 month.

To confirm introduce starter fluid into the air intake to see that the engine catches and fires.

Next, have someone crank while you tap real good on the injector body. That'll most time free them up. If not, the next step is to shock them open. Do it an injector at a time: Connect one side to GND and then very quickly strike the other side with +12V. This will get them to work again.

Report back,
Ingo

Follow this advice!

[rick-l]

To clarify what was said above the fuel pump runs whenever the key is in the start position and if the DME senses the engine is running it continues to provide a signal to the pump relay.

An 8051 microcontroller whas very high tech in the early 80's. OBDII in 1980? What kind of a PC did you have in 1980?

Originally Posted by logician

Great suggestions everyone. I'm trying to queue up a bunch of things to try when I go to my shop later today.

Question: how does the DME determine if the engine is turning over and so to enable fuel? Some cars do that via the air box, so if the computer sees any airflow, it knows the engine is turning over. Is this what the 911 DME does? Or does it use other means (the crank sensor, etc?) I'm trying to understand if a bad air box could be a scenario.

(Missing OBD2... it seems unfathomable these days that anyone would design a computer controller without any diagnostic ability.)

[cgarr]

Originally Posted by rick-l

An 8051 microcontroller whas very high tech in the early 80's. OBDII in 1980? What kind of a PC did you have in 1980?

I had an 8086 running IBM-DOS

[rick-l]

The IBM PC was introduced on August 12, 1981

[cgarr]

Originally Posted by rick-l

The IBM PC was introduced on August 12, 1981

When was that microprocessor designed?

[cabmandone]

I must have gotten CRAZY lucky. The engine I put in my car last fall had sat for 4 years. Once I sorted out the bad DME and got my speed and reference sensors wired properly from PO miswiring, it fired off and ran like a top. My suggestion of checking fuel pressure at the regulator along with the other things I suggested is how I figured out it wasn't fuel related.