interesting Steve. thanks for sharing.

My 3.2 had the labels but over the years the BG and DG (?) wore off.

Here's a wild A$$ Guess and if it is right I will expect a prize. You wouldn't think your sensor that is now 300k along with the inductance of the sensor, the filter capacitance on the input and some gain from the FI2438-03 (whatever that is) is producing an oscillator that is initiated by the unusual high RPM?

Steve, I remember you and me talking about this incident. Essentially, what you are descrbing is a design flaw of the 3.2 DME. It has inbuilt current limiting. But in case of just one of the six injectors shorted the total current only goes up marginally, enough to hurt the system but not enough to trip the current-limitig circuitry.

But this here seems a little differnt.

James, I agree 100% with your idea, that is why my first attempt was to swap in a known good DME. And it did exactly the same. So my conclusion is that my case is NOT the DME.

Cheers,
ingo

Rick, that is somewhat similar to what I am thinking - the input stages of the DME for the two sensors are amazingly tollerant to waveforms and levels. So even a sensor that is pratically shot might still provide a marginal signal that runs the engine fine. But the resonance frequency of the entire layout has now drastically changed with the faulty sensor.

My plan is to get a new sensor and replace the DME relay (just to be sure) and see where that leaves us. Unfortunately (for you guys) I am taking a week off so the suspense might kill you. But I will report back once we put the new parts in.

Ingo

Do one thing at a time so you'll know what fixed it.

Well here is a thought for you. I happen to have 6 factory injectors a DME and two DME relays sitting around from my PMO conversion. I could ship them to you and you could essentially replace the ignition system and see which was causing the problem.
I am leaning toward a faulty injector that is pully too much current but only when forced to stay open such as WOT. It trips the DME but then resets and fires off.
The DME is from a 1984 and has tested good. The injectors were fairly new before removed and there is one new and one used DME relay. If you really want to go crazy I can send along a coil and distributor.
Pay the shipping and send them back in a couple of weeks when done testing. Sorry I already sold the sensors.

Brainfart on my part. I do think the DME cuts fuel to prevent over-revving, but it would of course do it using the injectors, not the fuel pump. D'ouh!

Injector checks/replacement + the "flakey" sensor.

then drive her!!!!!!!!!!!!!!!!!!!!!



best,

Doyle

as to DME relay stuff,..it's quite amazing what problems these aging relay contacts can cause. a simple approach/swap...I changed mine out a year ago and was completely convinced of a performance improvement.. I now change it out every year, as as routine deal....always keeping a couple of spares...gotta' love that fresh copper contact (I do),..

One thing at a time is good advice.

Best of luck,

Doyle


Best,

Doyle

Fuel injectors, check this thread

http://forums.pelicanparts.com/porsche-911-technical-forum/451721-buick-fuel-injectors-3-2-a.html

Barn flapper door sticking ?

Ingo, Pete over at Schneider's mentioned a gap that can be adjusted on the relay.

Try the voltage regulator in alternator, I have same problem before .

Thanks all for your suggestions: Especially dfink with his generous offer.

However, I will be traveling for a couple of weeks and in the meantime I have asked the owner of the car to get a new DME relay and one sensor. Once I am back we will replace first the DME relay and then the sensor and see where that leaves us.

If that does not cure it I might have to go down and check the injectors and see if we get some sort of high-current draw there. I myself have several tested DME's for a 3.2 on the shelf so I think I am covered there. Stay tuned - I will give you an update when I am back in town.

Ingo

failed CHT anyone?

http://forums.pelicanparts.com/porsche-911-technical-forum/455137-3-2l-head-sensor.html

Once I disconnect the CHT the engine has a hard time to even start or idle. So I assume it is working. Thanks for suggesting. Ingo

I read this thread quick and my bet is that sensor reading 300Kohms, replace it it's bad. The sensor should read 1000ohms +/- 20%. If one is reading 300K it's bad. And yes a bad sensor will and can cause the DME to stop producing both ignition coil signal as well as injector pulses.

Also, Steve is dead on. A injector coil that is starting to fail (short out internally) will draw to much current from the injector drive circuit and the circuit is designed to shutdown if this happens. Check the resistance of each injector as well they are in the 3ohm range the idea is that all 6 injectors should read the same. What injectors are you running? stock ones?

Sal, I agree 100% and that is why I asked the owner to get a new sensor right away. In my bench test fixture I can run a variety of different waveforms and amplitudes into the senor inputs and get the DME to work. So the input circuit is very tollerant and I can see where even a marginal sensor is able to run the engine up to a certain extend.

I think the marginal sensor is the culprit, too. I measured several times and triple-checked my results. It was 300k and not 1.2k like the other. Rick proposed a very interesting theory as to what could be going on. We will see when I get back from my trip.

Ingo

Sorry I jumped the gun on my post. Sucks that a second DME didn't solve the issue.

Ingo,

You can buy the BMW sensor for 1/2 the price and it's the same exact sensor. I think this is the BMW part number 12-14-1-708-619-M14 our host sells it.