About to run my '87 Carrera off a cliff - still surging and backfiring

[Tippy]

Thanks Anthony, I'll try.

[chamilun]

I had same issues on 82 sc w/ MSI ignition: turns out the wires were backwords

[Hoss911sc]

Does the engine run smooth above 3800 rpm? I had a similar issue at lower rpm (in the wet) that turned out to be the Oxygen Sensor. It is ignored above 3800 on my car ...

[Tippy]

It hasn't used an oxygen sensor in years and never had a problem running that way.

[scarceller]

If the tach drops like a rock to 0 RPMs I suspect the speed or reference sensor. If the DME misses (does not see) a few speed pulses it will re-boot it self. When it does this the engine quits for a second or so and the tach just drops.

The connectors for the speed and ref sensors are on the bracket on intake runner #3, on this bracket you have 3 connectors, they are from top to bottom:
Top - CHT Sensor
Middle - Speed Sensor
Bottom - Ref Sensor

With engine running you can unplug the bottom Ref Sensor and engine should still run without ref pulses. But if you unplug the center one the engine dies without speed pulses. One test is to unplug the bottom one (Ref Sensor) and see if engine dies.

Also measure the resistance of the sensors (both sensors are the same) they should both measure about 1000ohms (+/- 15%).

If you suspect the sensors you can purchase the BMW sensors from our host as they work just fine in our 3.2L at 1/2 the cost. Just search for part # 12-14-1-708-619-M14

[scarceller]

Could your idle switch be sticking closed? Unplug the idle switch, when you do this the car will immediately run off the PartThrottle maps and you'll notice a high idle in the 1000-1200 RPM range. Don't worry to much about the hi-idle simply take it for a spin and see if things are better. If the idle switch sticks and does not immediately open when the throttle blade starts to open then you'll be stuck in the IdleMaps even though the car has transitioned to part throttle.

You need to be 100% certain that idle switch is functioning and not sticking as well as properly adjusted. The switch must be closed with TB blade closed and it must open when the blade moves ever so slightly open (like 1mm movement).

[Tippy]

Originally Posted by scarceller

If the tach drops like a rock to 0 RPMs I suspect the speed or reference sensor. If the DME misses (does not see) a few speed pulses it will re-boot it self. When it does this the engine quits for a second or so and the tach just drops.

The connectors for the speed and ref sensors are on the bracket on intake runner #3, on this bracket you have 3 connectors, they are from top to bottom:
Top - CHT Sensor
Middle - Speed Sensor
Bottom - Ref Sensor

With engine running you can unplug the bottom Ref Sensor and engine should still run without ref pulses. But if you unplug the center one the engine dies without speed pulses. One test is to unplug the bottom one (Ref Sensor) and see if engine dies.

Also measure the resistance of the sensors (both sensors are the same) they should both measure about 1000ohms (+/- 15%).

If you suspect the sensors you can purchase the BMW sensors from our host as they work just fine in our 3.2L at 1/2 the cost. Just search for part # 12-14-1-708-619-M14

Originally Posted by scarceller

Could your idle switch be sticking closed? Unplug the idle switch, when you do this the car will immediately run off the PartThrottle maps and you'll notice a high idle in the 1000-1200 RPM range. Don't worry to much about the hi-idle simply take it for a spin and see if things are better. If the idle switch sticks and does not immediately open when the throttle blade starts to open then you'll be stuck in the IdleMaps even though the car has transitioned to part throttle.

You need to be 100% certain that idle switch is functioning and not sticking as well as properly adjusted. The switch must be closed with TB blade closed and it must open when the blade moves ever so slightly open (like 1mm movement).

Regarding the reference sensors, I've thought of the same thing but of course the fact this acts intermittent makes it impossible to figure out without replacing (I've tested within specs at the DME harness - but will retest). Before, it would run great once at part throttle right above right-off-idle (and above to WOT) but it would cut-out at right-off-idle.

Like there is an enrichment problem.

Now, it is cuttnig out under boost while @ WOT too.

As far as the idle switch, I replaced with a new one already. I need to adjust it better but I know it is working well. Closed, the timing is 5 degrees. Open, it is at 20 degrees (aftermarket ignition settings).

I noticed there is a "tach" pin in the DME main harness.

What I have been pondering is:

1. Is the DME losing power intermittingly?
2. Is the ignition side of the DME acting up causing the tach dropping like it should show?
3. To what you said, is the speed reference sensor acting up?

That is what I have been trying to figure out but without another DME to try, I am kind of stuck.

[scarceller]

You also mentioned ignition switch, this is very possible. I had a faulty ign switch and it caused the DME Relay to lose power momentarily and the DME would reboot. A quick test is to start the car and at idle with the handle of a screw driver reach behind the dash where the Ign Switch is and tap it a few times to see if the DME Relay acts up.

Also, if you can get the tach to drop to 0RPMs predictably I'd suggest holding the DME Relay in your hand when the tach drops. If the DME Relay is loosing power and deactivating you'll feel it click. Alternative is to put a test light on the Relay's Fuel Pump output line and watch the test light when the tach acts up.

There are 2 common reasons the DME re-boots:
1- The Relay disengages and the DME looses power
2- The DME does not see speed pulses for a given period of time. The DME Software looks to see if 30 pulses or so are missed.

You need to figure out if this is the problem.

But keep in mind you may also simply have a faulty DME, here you need to put your DME into another car (Never put another DME into the problem car).

[scarceller]

One more possibility: you mentioned a new AFM? is this a brand new AFM? Have you tested the AFM output signal to be sure it has no dead spot(s)?

[Tippy]

While the engine was running, I beat on the DME, yanked on the ignition switch wires, and slapped the DME relay around.

I couldn't get it to hiccup.

I even yanked around on the fuse panel in the front trunk.

It only cuts out under load but all of sudden it doesn't like to rev under no load now too.

I am really thinking the DME went wonky......

[Tippy]

Originally Posted by scarceller

One more possibility: you mentioned a new AFM? is this a brand new AFM? Have you tested the AFM output signal to be sure it has no dead spot(s)?

Doing that today after work. It is used.

[scarceller]

AFM Signal test is very easy: put a DVM on the signal line then turn key to 'RUN' engine 'OFF' then push the barn door in the AFM open with your finger if you can reach it or a pencil (eraser side) push the door very slowly while watching voltage it should sweep cleanly 0-5vdc with no dead spots. It may start slightly above 0v maybe around .5vdc but that's OK and it may not hit 5.0vdc but should get close. The area to really look at is the 1.0 to 4.7vdc area pay close attention to 1.0 to 2.5vdc as this is typical cruise range and the AFM spends a lot of time here and this is where the carbon often gets damaged in the carbon track.

So is the AFM new?

[scarceller]

And one more possibility I've seen is a faulty injector! If a injector has a intermittent short or has some of the winding shorted then the resistance for that injector drops near 0ohms, if this happens you loose fueling completely. The DME detects over current and shuts down the injection pulses. I don't suspect this is your issue but it could be worth checking the resistance of each injector. If the car has issues free reving in neutral you could also unplug 1 injector at a time and see if it can free rev, idea is to eliminate each injector in sequence. The injectors are all wired together in parallel so 1 faulty injector can take them all out.

[Tippy]

Originally Posted by scarceller

AFM Signal test is very easy: put a DVM on the signal line then turn key to 'RUN' engine 'OFF' then push the barn door in the AFM open with your finger if you can reach it or a pencil (eraser side) push the door very slowly while watching voltage it should sweep cleanly 0-5vdc with no dead spots. It may start slightly above 0v maybe around .5vdc but that's OK and it may not hit 5.0vdc but should get close. The area to really look at is the 1.0 to 4.7vdc area pay close attention to 1.0 to 2.5vdc as this is typical cruise range and the AFM spends a lot of time here and this is where the carbon often gets damaged in the carbon track.

So is the AFM new?

It is used. I did the 9-volt battery test on my old one. It had a flat spot right at the "idle" area but figured it was where the idle aiflow range is supposed to be.

Originally Posted by scarceller

And one more possibility I've seen is a faulty injector! If a injector has a intermittent short or has some of the winding shorted then the resistance for that injector drops near 0ohms, if this happens you loose fueling completely. The DME detects over current and shuts down the injection pulses. I don't suspect this is your issue but it could be worth checking the resistance of each injector. If the car has issues free reving in neutral you could also unplug 1 injector at a time and see if it can free rev, idea is to eliminate each injector in sequence. The injectors are all wired together in parallel so 1 faulty injector can take them all out.

That is why I wanted to oscilloscope EVERYTHING that is cyclic to see that one moment of interuption.

Good ideas. Thanks.

[Tippy]

This is what is circling my mind:

1. AFM dead spot (thought on original, replaced with another used one)
2. Speed and reference sensors acting up
3. Dying ignition switch
4. Bad grounds
5. Bad component(s) driving ignition in DME
6. Bad component(s) driving injectors in DME
7. Bad solder joint unseen to the naked eye in DME
8. Bad dizzy cap causing arching (replaced new)
9. Bad microswitch (replaced)
10. Dying coil (replaced new)
11. Blown piston rings (compression test confirmed good)
12. Cam timing (runs good once past light part throttle)
13. Intermittent power loss to DME relay (DME relay new)
14. Intermittent injector harness
15. Intermittent injector cutting out (fairly new Siemens 55lb.)
16. Tach (aftermarket VDO, not factory VDO) shorting DME
17. Fuel pressure higher than stock so DME is cutting injector pulse-width too much
18. Alternator overcharging
19. Burnt valve (compression test good)
20. Dizzy dust cap

All I can think of off the top of my head that I have thought about or tested already resulting in a no fix.

[Determined]

bump to subscribe

[ant7]

Originally Posted by Tippy

This is what is circling my mind:

1. AFM dead spot (thought on original, replaced with another used one)
2. Speed and reference sensors acting up
3. Dying ignition switch
4. Bad grounds
5. Bad component(s) driving ignition in DME
6. Bad component(s) driving injectors in DME
7. Bad solder joint unseen to the naked eye in DME
8. Bad dizzy cap causing arching (replaced new)
9. Bad microswitch (replaced)
10. Dying coil (replaced new)
11. Blown piston rings (compression test confirmed good)
12. Cam timing (runs good once past light part throttle)
13. Intermittent power loss to DME relay (DME relay new)
14. Intermittent injector harness
15. Intermittent injector cutting out (fairly new Siemens 55lb.)
16. Tach (aftermarket VDO, not factory VDO) shorting DME
17. Fuel pressure higher than stock so DME is cutting injector pulse-width too much
18. Alternator overcharging
19. Burnt valve (compression test good)
20. Dizzy dust cap

All I can think of off the top of my head that I have thought about or tested already resulting in a no fix.

hi Tippy,
well, i can fully understand why your frustrated with this, you seem to have done most checks, Another one i can think of [havent seen this mentioned] that may be a cause is distributor, the wiring from the internal coils could be intermitently shorting etc.
I have known this on other older cars but may not be an issue with the 3.2 but again, worth a try if its just a case of swapping it out for known good one.
hope this helps.
Anthony.

[scarceller]

After more thought and seeing all the things you already verified, I suspect faulty DME or faulty speed or ref sensor.

Get the DME into another car to test it.
The sensors are sometimes intermittent failures so best to just replace them both if you suspect them.

[Tippy]

Ant7 - the dizzy is passive. There isn't any electrics within it.

Thanks.

Scarceller - I haven't verified everything I posted in the list but they've simply crossed my mind.

Some, I don't have the equipment to test them and am trying to rule out EVERYTHING within my knowledge and abilities before I pay someone else to figure it out.

It's a pride thing I guess....

[Tippy]

At this point, I'll take any suggestions.

I know electronics can create weird situations.....

Scarcellers and others, I need to review some of the suggestions and rule them out.

Appreciate everyone's inputs!