New AFM & Now Car Runs About 10 Seconds

[dcroop]

"Stuttering"and "missing"on hard acceleration used be occasional enough to overlook, and recently has become frequent enough to be a nuisance, so I bought a rebuilt AFM in hopes of eliminating the issue. Other than that one increasingly pesky thing, this car has started, idled, run fine.

Swapped in the the new AFM tonight and the car starts and will only run for about 10 seconds. Can't even drive it.

Seems like most common cause of this symptom is faulty FI coolant temp sensor. If that is my issue, what I don't get is why it didn't cause the same problem with the old worn out AFM. Only one thing has changed.

Any ideas? Thanks in advance.

[76was124]

Assumingi you have a good AFM rebuild, I would make sure your boot is back on the AFM and intake correctly to be sure there are no air leaks. (Sometimes the bottom lip of the boot doesn't go on correctly and has a gap). Lack of air flow through the AFM could keep the flap from staying open enough to keep the fuel pump running. I would also double check your timing.

[KevAndAndi]

+1

Can you provide more information on the rebuild? Who did it, and was it a complete one? Did they test and calibrate the AFM?

With your old AFM, was the fuel pump energized as soon as you turned the key to the ACC position, before starting? (That's not supposed to happen.) If the AFM/fuel pump relationship was working properly before (and the fuel pump was energized only upon engine starting and running), then it is possible that the flap is not moving to energize the FP, possibly due to an air leak, as suggested by the previous post.

Keep in mind that one "air leak" could be the AFM's air bypass screw, which is not an owner-adjustable item. If the rebuilder did his work correctly, that screw should have been calibrated using test equipment and then plugged to prevent tampering.

In my case, there was a major air leak where the AFM meets the big air hose, the air bypass screw had been tampered with, and the AFM flap was bent outward and didn't move inward during idle. For this reason, the PO bent the arm in the AFM in order to override the flap and "permanently" energize the FP (which should be an emergency workaround only, but is too often deemed a perfectly fine fix).

To simulate the bent-arm workaround, turn the key to the ACC position and move the AFM flap inward with your finger. Do you hear the fuel pump running? If so, you find yourself at the point where folks make the mistake of bending the AFM fuel pump actuation arm to "solve" the problem, whereas the real solution is to get the flap moving inward.

[bradartigue]

You need to get a voltmeter and test the AFM and the sensors. If the new AFM is bad, send it back. Test the ECU as well.

[dcroop]

Thanks for the responses. So:

There are no air leaks. The fuel pump comes on as it should and not with the key.
This is a completely rebuilt AFM done by Fuel Injection Corp purchased through one of our Fiat parts vendors
I tested the AFM through the ECU with ohm meter and got the following:
Pins 6-7 - 114 (>600 = OK)
Pins 7-8 - 315 (>1200 = OK)
Pins 8-9 - 201 (>350 = OK)
Tested the FI Coolant Temp Sensor and got:
1073 (should be 2000-3000 at 68 degrees) engine was cold & it's about 86 out - so I don't know if that means it tested OK or not
Tested thermo time switch - it came on for about 1 sec

I didn't test the O2 sensor but it was replaced when new engine was done maybe 6000-7000 miles ago. Injectors were also tested good at that time.

I also reinstalled the old AFM, and as expected, the thing starts and runs just as it always has. I called my mechanic/ Italian car expert here in Raleigh and his first guess without seeing anything is dud AFM.

[focodave]

Sure sounds like the newly rebuilt AFM is not working.
Perhaps it was rebuilt by the "new guy" at FIC on a Friday afternoon at 4:00pm.

Unless, of course, there are some very unusual and highly unlikely coincidences occurring when you are switching components and un-plugging/re-plugging electrical connectors.

[narfire]

Stuttering"and "missing"on hard acceleration used be occasional enough to overlook, and recently has become frequent enough to be a nuisance

Sounds like the strip the pins run on in the AFM are kinda worn...in my opinion.
If you are up to it, carefully pop the top off the AFM with a box cutter and have a look at the strip the pins run on.
By loosening the nut/screw on the shaft one can move the arm up and down and with a VERY carful bend of the arm, one can get the pins to run on a fresh part of the strip. We are talking 1/2-3/4 of a mm here but it can sure make a difference.
I just tried a search for past posts and no love. Perhaps someone else can find it.....
Chris

[bradartigue]

Send the thing back.

[redcars]

As for your missing on hard acceration. I had that on my 82 and it was restricted fuel supply. Mine was a bad hose from tank to pump. If the pump cannot get enough fuel the presure drops. Low presure=no fuel from the infectors. A clogged fuel filter would give the same results or any restriction after the pump. This can be check out by running a fuel pressure gauge.

[KevAndAndi]

I agree with Brad that you should send it back to the vendor. There's no reason to put yourself through a bunch of troubleshooting with what should be a "like-new" AFM, particularly given that installing the old AFM confirms the "quits after 10 seconds" problem is caused by the new one.

Which brings me to the question: Why not send your old AFM to FIC directly and have them rebuild it? I'm delighted with the job they did on mine. And they provide an 18 month warranty in the unlikely event that you run into a problem with that rebuild.

[lanciahf]

"Car Runs About 10 Seconds" This will happen if the signal from the coil gets disconnected. If the coil signal is missing the motor runs on the cold start valve and if no signal comes from the coil then the main injectors will not fire.

Easy check.

[KevAndAndi]

By the way, regarding the resistance values for the coolant temperature sensor: Here's a graph I put together using the upper and lower bounds at each of the three temperature values that are always cited. There's nothing really scientific about my graph; the curves interpolating between the data points were drawn in Excel "freehand," but they probably correspond pretty closely to the "real" technical data on the sensor.

[KevAndAndi]

Double post deleted.

[dcroop]

Thanks to everyone for the great responses. I'll post an update once it's resolved.

[bradartigue]

"Car Runs About 10 Seconds" This will happen if the signal from the coil gets disconnected. If the coil signal is missing the motor runs on the cold start valve and if no signal comes from the coil the main injectors will not fire.

Easy check.

It does work on his old AFM.

From the post above he tested the new AFM and it had failure states. Sounds like a bad unit.