mjshira

UPDATE PART TWO

FOund the issue. I need a reman distributer. Mine is adding too much timing on the secondary side... anybody know where to do get them?

Geoffrey

Umm, I don't think so. The distributor does not have any effect on the ignition timing, only the phasing of the rotor to the plug post. Please think about this before you buy a $900 reman distributor. The ignition timing is controlled entirely by the ECU which uses the 60-2 flywheel for crank position reference. The only information the distributor provides is what stroke the engine is on, and as long as the distributor is installed correctly, the rotor movement due to the weights and springs will not move the distributor signal to the other side of the missing teeth on the flywheel.

This setup is not like a distributor triggered ignition system such is found on a carburated, MFI, CIS controlled engine.

The rotor phasing occurs to ensure that the rotor is even with the distributor at higher RPM where there is more advance required and the time for the the rotor is near the plug pin is reduced.

What plugs were you running and what plugs did you go to?

I would have a strong look at the ignition timing mapping of your custom programming.

mjshira

thanks Geoffrey

Here is what has me confused at the moment. I had stock heat plugs, we went one step down in heat, this had no positive effect. This in my view rules out plus.

We then ran the car on the dyno. First with the same map we've been using and we got the same result. Next we disconnected the secondary dizzy (not the one next to the fan and connected to the crank) ran the can and no detonation (obviously we didn't do this run for power put just inspection)

All I know is that no matter how much timing we pull with the software we get the detonation in about the same spot.

We even left the timing values stock and this happened the same way.

This is what left me with the conclusion that it was a mechanical advance issue. Thanks for your post, I appreciate it even while it leaves me all the more confused frankly.

The good news is that I am learning one step at a time.

mjshira

How can I validate the distributor is installed right? I've not messed with the distributor at all.

Could this be the reference mark sensor?

dfinnegan

There are alignment notches on the distributor housingings. With the crank positioned at "z1" top dead center for cylinder 1, the rotors should point at these notches. Note that cyl 1 and 4 overlap, so be sure you have cyl 1 which, normally, is determined by the rotors pointing at these notches.

mjshira

thanks buddy, I will check that.

I am at a loss here because the following parts are new:

DME
Coils
Coil drivers (sorry if not correct term)
CHT sensor
Reference mark sensor (I had to splice new one which is a 996 part number) to the old plug, set the gap to the flywheel, etc

On the dyno we could see AFR was right on and never lean, on the piggyback ECU we could see timing from the DME. Both seemed right. We concluded the dizzy was the issue (which it seems it is not now based on Geoffrey's comment) because we saw the mechanic advance and springs on the secondary dizzy. We figured that since we had pulled timing to attempt to stop the detonation and not had an impact and then we'd disconnected the secondary dizzy and noticed that the detonation didn't happen. Regardless, that was how we came to the conclusion we did.

Now I've got to really think this over again because frankly, I am stumped at the moment. The engine really wants to run.

Geoffrey

Obviously I'm not there to check the parts, and there could still be an issue with the distributor, however, the issue is not the one you posted, I can assure you of that.

Check the simple things like

1) Are the plug wires on the second distributor correct?
2) Is the distributor rebuilt correctly with the rotors correctly oriented
3) check the timing with a timing light to ensure it matches what the computer says it is
4) the Z1 mark on the pully.

mjshira

Thanks, I will do that.

The car seems like it has too much advance in the car from the start. The fuel quality position switch on the DME has been tested at position zero and 4. The later position pulls about 2.8 degrees out across the band. All this did was move the detonation down in the RPM band.

Prior to changing plus we 'thought' we had a pre-dontation issue due to the former plugs potentially being too hot. I am going to go back to these plugs as the engine ran better with them.

The dizzy belt looked good and also looked as if it has been replaced in the recent past.

mjshira

I also should mention that the car is a 1990, the distributor has venting installed which tells me it must be a rebuild unit as to my understanding the distributors didn't come with venting in 1990.

Question: could an improper spacing on the reference mark sensor (between the senor head and flywheel) cause this?

mjshira

Check List:

I have created this check list which I want to run by you guys to validate it is complete:

1. Inspect distributor belt (done)
2. Validate that distributors are aligned correctly (using Adrian's book for this)
3. Check reference mark sensor gap - should be 1mm plus or minus .02 mm's
4. Check reference mark sensor with meter (ranges found in Porsche 964 technical manual)

Here are some questions I need help with:

1. How do I replace the knock sensors? I have two new ones on the bench and was thinking I'd swap them to validate I've got good working knock sensors (if the Hall sensor is not working the engine won't start, I am confident the Hall sensor is working)

2. Since we know that the pre-detonation was eliminated when the secondary distributor was disconnected (I disconnected the coil driver for the secondary bank, ran the car on the dyno, and while she was down on power, there was no pinging...) how can I further isolate this?

Geoffrey

The space between the sensor and the DME ring only affects the amplitude of the mag sensor wave the ECU sees. There is a minimum voltage that the DME requires and beyond that it doesn't make a difference unless the voltage exceeds what the ECU can handle (MoTeC doesn't like over about 125 volts). So, if you set the sensor at the factory specification then you will be fine. The telltale sign that the sensor is priducing too much voltage is that it acts like a rev limiter and that is not your issue.

I think the engine will still run with the hall sensor disconnected or failed, so you need an oscilloscope, hammer, or PST to check is operation. If it is defective, I believe the knock control will not function properly since the ECU does not know which cycle the engine is on.

To replace the knock sensor, you need to remove the black engine shroud - not a pleasant job with the engine in the car.

The fact that when you disconnect the secondary distributor the detonation stops is interesting. What happend when you ran the engine with the primary distributor disconnected? I still think it might be something obvious like crossed plug wires or a misaligned distributor rotor.

mjshira

I thought it was odd too. Then I looked at the secondary dizzy, noticed the springs and came to the incorrect conclusion that they'd become weak over time and where advancing those 6 plug firing timing and that leading to pinging.

Regarding inspecting wires, I can do that and will add it my list. But I don't have a high degree of confidence that will be the issue.

As we ran the car on the dyno, we put fuel where it needed to be and that didn't stop or prevent the pinging. This is what confuses the heck out of me. I think this is a mechanical, sensor or maybe a cross wire issue.

I know the DME piggyback is working right because I can read and input changes via my laptop. I've got larger injectors, lightwt flywheel and the car idles perfect. The pinging happens about 4500-5000 RPM.

I will run down my check list, including checking the reference mark sensor (just to be ****). Regarding the hall sensor, Adrian has said

"A Hall sensor in the primary distributor body monitors cylinder 1 top dead center. This allows the DME control module to synchronize the knock-sensor signal with the correct cylinder.

"The Hall sensor monitors the poistion of the rotor arm in the primary distributor. Its output is not dependent upon engine speed.

"If the Hall Sensor fails the engine will not start. The whole distributor assembly will have to be replaced."

Geoffrey

Adrian is incorrect on this one. If you disconnect the plug, the car will still start, run, and appear normally. Try it.

The engine pings at 5000rpm because that is where the engine's peak volumetric efficnency is and therefore where the cylinder pressure is highest. Throwing fuel at it won't stop or prevent detonation to any large degree. The engine is detonating because there is too much ignition advance at those points and the only way to solve that is to reduce the ignition timing. This is why steady state tuning is important for proper ignition timing.

mjshira

agreed. Here is the deal, we've pulled a good bit of timing out out of the car, fuel switch at position 4 which is about -3 degrees, plus we're using the piggyback to pull more.

the engine has been rebuilt, had the cyl head update, head work, valve work, porting, and cams. the receipt I have from the previous owner doesn't detail this well and thus I am attempting to tune the ignition to an engine that is not a stock engine anymore.

what do you think the timing should be at about that RPM and 60-80 percent load?

(BTW thanks a ton for your threads and posts here, I've learned a ton)

mjshira

here is a shot of the engine on the dyno in the range where pinging happens.

you can see where I've pulled timing and you can see where the engine timing is.

there has to be something off with my cars timing before I attempt to tune thus preventing me from optimizing.