need help with engine misfire-MSD experts
 

I installed an MSD 6AL2 programmable ignition and am having a terrible misfire after the engine has started.

The rotor is the NAPA non resistor rotor and brand new cap. Wires are stock OEM braidied wires and the plugs were gapped at 0.040.

I am able to start the car and it runs fine for the first 1 minute or two then it starts to stumble and misfire. As soon as I touch the gas peddle to rev it up the engine starts to backfire and stall. AF ratio is bouncing all over the place going lean then rich as this is happening.

My max timing was set at 32 degrees BTDC and I have programmed the retards to take out timing to exactly simulate the stock 79 turbo timing curves from the manual. Idle timing is 2 degrees BTDC at 950 rpm.

I have replaced the front fuel pump and all the relays thinking it migh be fuel related. Front pump was making some noise so I installed a Bosch 044 front pump and rewired the fuses to break up the circuit to each relay. --No change.

Fuel pressures are 3.5bar warm CP and SP is 6.4bar all within spec.

I have read on the MSD forums about rotor phasing thinking this might be the cause, but am not sure if I am barking up the wrong tree here.
HAS ANYONE HAD ISSUES WITH ROTOR PHASING WHEN USING THE MSD 6AL2 UNIT? Is this what it sounds like, as I am running out of hair to pull out of my head and need to get this thing running.

Any help is greatly appreciated from you guys this has me stumped now.

Fred
http://forums.pelicanparts.com/uploa...1368967360.jpg

Fred,

What are you using to trigger the MSD?

You can either use a crank trigger to trigger the MSD, and have it do the calculations for your advance, and leave the advance mechanism in place in the distributor to properly phase the spark; or

You can use the distributor to trigger the MSD, and zero out the advance in the MSD, relying on the mechanical advance curve to control the timing.

Leave boost retard out of it for the moment.

Which is it? You most certainly cannot do BOTH-- that is, have the distributor/advance trigger the spark and use an advance table in the box.

You didn't lock the advance, did you? I think the consensus is that leads to misfiring-- even the Factory used advance mechanisms in the 3,2 and 3,6 engines to "phase" the distributor rotor for correct orientation even though the Motronic control unit and crank trigger determined the timing.

Are you sure the wires to the MSD trigger from your distributor or crank trigger are not backwards?

Positive they are good.

The distributor was locked out at full advance with the counterweights wired tight. The vacuum pot was left on but the ends are capped so no air gets in or out. With the internal spring inside the pot I dont think it will move the plate inside the dizzy without a vacuum signal and remain stable.

All my timing retard is controlled via the MSD program as I have the distributor timed to 32 degrees BTDC and locked down.

Fred

Try a flat spark table at 15 degrees retard for all RPM. You may be requesting advance that falls off the edge of the rotor onto a neighboring cylinder... causing a misfire which will be indicated on the wideband as air to fuel flying about.

I have the same MSD set-up you have. I would check the ohms at your mag pick-up. Quote
“The published specification for the magnetic pulse generator is 600 Ohms +- 100. Our experience has shown that a variation of +- 25 Ohm indicates a need to repair defective connections or to replace the magnetic pulse generator windings.”
I bought a new Mag Pulse Gen and I’m right at 600 Ohms. Also I would check your power supply wire, and ground. BTW mine did the same thing until I figured out the battery was low.

My money is on a rotor phasing problem; especially since you effectively locked the weights. They do more than simply provide part of of centrifugal advance.

You will need a distributor machine (or access to) and a spare cap to see if this is the root of your issues or not.

So lets say that I have a rotor phasing issue. What can be done to fix it?

The only way I can think of is to somehow index the rotor tip thru some angle to attempt to center it with the distributor cap post. MSD sells an adjustable rotor for this purpose but no such thing exists for my 930 distributor. OR DOES IT?

What about grinding down the trailing edge of the rotor tip to keep it from jumping to the #5 cylinder when it should be firing on #1 for example. Sort of making a pointed tip on the rotor contact versus a broad tip like it currently has.

With the advance mechanism locked at full advance like I have it I assume the rotor wants to jump to the cylinder one clockwise from the target cylinder with a counterclockwise rotating dizzy is this correct? So for example when its supposed to fire #1 it jumps to #5, and so on. I tried to check for this with a timing light on the #1 plug wire with the engine running and it was very stable and gave no indication of a misfire. This perplexed me.

I dont want to/can't afford to pay someone with a distributor machine for their time and really want to solve this on my own. Any tips appreciated.

thanks for the input so far,
Fred

Hi Fred,
On my 930 the dizzy is locked out full advance mechanical/vacuum.
Question
1. Did you tap into your original 6 plug pigtail from the Bosch CDI for your trigger and power.
2. What’s your LED light doing on the MSD when you were able to get it to run.
On my car since I had the engine out, I rebuilt the whole engine wire harness, eliminated the AAR and thermo time switch in the L/H chain cover, my cold start valve has a manual switch I can toggle, anyway there is no more 6 pin plug. I ran the power wire to the starter +, the negative to the ground stud behind the Fuel filter ground point. I ordered 930-602-907-01 for the signal wire into the dizzy, and an extra Bosch electrical connector, so basically the MSD plugs right into the dizzy trigger wire. Nice and clean.

From MSD
http://forums.pelicanparts.com/uploa...1369160105.jpg

To Dizzy
http://forums.pelicanparts.com/uploa...1369160188.jpg

Fred,

Unlock the advance mechanism and use a crank trigger to fire the MSD box. This will allow the rotor to be in a better position relative to the cap for each ignition event.

The only thing requiring a bit of fabrication would be a bracket for the sensor, as most solutions (Clewett's, for example) call for removing the distributor.

Just for fun switch the wires on the mag pickup.

It changes the rotor phasing quite a bit, ask me how i know. My car sounded very similar to what you describe.

Any luck with this? I was educating myself on the MSD for my own application and I see they state small distributor caps should be drilled to allow air to exchange in order to prevent misfire...

Still working on it.
I replaced the front fuel pump and did tests on flow and voltage drops to the rear pump. Thinking its fuel related.
Fuel flow is above spec and pressure in the CIS system is in spec. Car will not start at all now and I smell fuel so I don't think its a fueling issue.

I found that the voltage decreases to the rear pump terminals from 11.5 volts upon initially running the pumps to a final reading of 10.2 volts after 2 minutes or more. This led me to think i had a battery dropping charge since it is 7 years old.

So I purchased a new battery and am installing it today. Is it possible the battery cannot maintain a steady 12 volts and this is effecting the MSD unit? I have read that the unit is susceptible to voltage drops or inconsistency.

Is it possible the alternator has a bad voltage reg and this is effecting the unit?
I see 13.5-14 volts on my lap top readout when its connected to the MSD and confirmed at the battery posts when the car was running.

I have not tried a flat spark curve yet like you suggested to see if reducing the maximum amount of retard dialed into the curves will fix the problem. Of course this is not the way I want to run the unit, so this is only good at diagnosing its an ignition issue. I will do this next if the new battery does not change anything.
Throwing more parts at the car is not a good way of fixing it, but the battery and front pump were a rational change due to the age of the battery and the noisy front pump.

Keep the ideas coming guys, I will keep updating this thread when I can.
This is a tough one.

Fred

I wanted to answer your questions:
1) Yes i got an adaptor from Permatune and wired into the factory harness for the CDI unit to trigger the MSD and provide a switched power source. The main power and ground ot the MSD is off the starter pos terminal and grounded to the chassis ground used by the tranny ground straps.
2) checked the red light and it blinks as the car is running off the starter motor upon cranking the engine to start it. This means its getting a proper start signal from the dizzy right?
I dont know what the red light does when its actually running, but the lap top readout shows the unit runnign properly. RPM and MAP readings are correct and the retard curves are behaving like programmed.
I ran the car at night to look for arcing in the wires or cap and saw nothing sparking or arcing over.
I dont think the OEM steel braided wires would cause a problem would they? They are compatible with MSD right?

Fred

Braided wires are correct. It's the solid core that are warned against as causing problems.

The arcing would be under the cap hidden from view if you had an electrified air field issue. Moisture, high humidity, dust can all contribute to the effect. That's more of a long shot though... plenty of simpler solutions to consider first.

Fred,

The LED light should be steady with the engine running.

One thing after reading your thread, you only locked out the mechanical advance full, not the vacuum. You may be over retarded with your curve.

I would put a timing light on it, at least check your idle settings.

With my dizzy locked out full mech-vac, I’m running about the same stock settings at idle and 4000 RPM with timing light.

You have good power and ground, the only thing left would be your trigger, check your ohms about 600, with engine cold, and after it’s warmed up.

Volts check your battery before start and after you get it running, this thing likes power. MSD says it draws .7 amps per 1000 rpm

http://forums.pelicanparts.com/911-930-turbo-super-charging-forum/610488-msd-success.html


http://forums.pelicanparts.com/uploa...1370026019.jpg

Hi Dan,
I wanted to correct you in that my post says that the vacuum canister is un connected from all hoses so it sees no input on either the vacuum or boost retard nipples. They are capped with rubber plugs so the canister should not be moving the advance plate at all. It has internal springs which keep it in position so I did not think that the plate would move on its own either so no need to mechanical lock this canister mechanism.
Anyway, I ran more tests today with the flat rpm advance curve at 14 degrees retard total. The car ran fine, so far but starting it against 14 degrees BTDC was not easy and the idle at 14 degrees BTDC is too high at 1250rpm and the throttle screw is damn near closed off.
This would indicate that the rotor is jumping spark to adjacent terminals probably due to the fact that the mechanical advance is locked at full advance and the MSD needs to take out 30 degrees total retard at idle which puts the rotor in a bad position inside the cap.
I am thinking of locking the advance counterwieghts at full retard to reindex the rotor and then going back to the stock curves i had before so i can still get vacuum retard and rpm advance curves to work together. Then I can get 30 degrees retard at idle and have 2-4 degrees BTDC instead of 14-16 like I do now.

Fred

Fred,

Locking the advance counterweights at full retard is the same as having them in their current full advance position and rotating the distributor body. A fixed position over the RPM curve is a fixed position over the RPM curve.

Good Morning Fred,

Ok you got the engine running, so you can rule out a trigger problem. Based upon the PP forum that I have searched, it’s my understanding that the Centrifugal advance will give you 16 deg. and the Vacuum advance will give you 10 deg. For a total of 26 deg.- advance-. ( I don’t have access to a SUN dizzy machine to verify this, and I’m too cheap to pay anybody).

Based upon your first post, you are telling the MSD box to retard your timing 16 deg until about 1500 RPM and letting it back in full by around 2700-2800 RPM.

Since you only advanced your centrifugal, it should only see 16 deg. hopefully the arm pressure on the pot is not affecting the rotor plate +/-

If you set your engine to TDC number 1, Z1 at pulley to case parting line, drop the dizzy into the engine with the rotor pointing to the hash mark on the dizzy body, fire the engine up, with a timing light (MSD box working correctly) you should see the flash around the Z1 mark at idle, indicating you are firing at 0 TDC. I’m not sure your pulley is marked for the 16 deg. bringing the engine to your 2800 RPM should see the flash from the timing light.

Question: where is your timing light flash in correlation with your MSD run retard curve?

My California model engine and dizzy should be 16 degrees machancial advance and 18 degrees vacuum advance according to the factory manual. These are the curves I have replicated in the image on my initial post. One of the problems diagnosing the early turbos we own is the existence of several different timing curves and distributors the factory used from 76-89. I am going with the distributor in my car which is the California model with both a vacuum and boost port on the canister.
My engine pulley is marked at 10 degrees ATDC, 5 degrees ATDC, TDC, and 29 degrees BTDC. I added marks at 10 degrees BTDC and 35 degrees BTDC to help me with timing the engine with the MSD unit.
My timing light is adjustable and works fine with the MSD and shows a strong strobe. The timing is fine at 2 dgrees BTDC at idle with the timing curves shown. This is right where it should be with 32 degrees BTDC total timing as I have the dizzy set now. At least it runs this way for a minute or two before it starts to misfire and stall, long enuff to get a reading with the light.
Bear in mind the total retard at high vaccum (idle) and low rpm my curves retard 34 degees total retard, but the MSd unit truncates to 30 degrees maximum retard. Thus 32 degrees Total advance minus 30 gives 2 degrees BTDC at idle.

Fred

Fred, any luck with this?

I have modified a Echlund EP407 rotor which is a direct replacement to the Bosch but without a resistor by cutting the base and re indexing the tip about 15-20 degrees to position it better in relation to the terminal in the cap. This eliminated any potential for rotor phasing.

This made the car easier to start then before. However my real issue is that the car starts fine when cold and runs well with throttle response and no misfire for about 1-2 minutes as it warms up. Then about 2 minutes into the cycle it starts to misfire and stall and AF ratio goes off scale rich then lean.
If I manage to keep the car running long enuff or just keep restarting it to the point where it gets hot then it will maintain an idle but occasionally a misfire can be seen.
I have replaced the cap again with a new one that I drilled a hole in the base for venting. No improvement.

I have checked the fuel pressures while this is happening and they are rock steady and no oscillations in the fuel pressure or the fuel metering plate. Cold pressure rises steadily from the intial engine start to a warm control pressure of 3.5bar and the car is misfiring the whole time this takes place.

The AF ratio starts about 11.5 when cold starting and runs thru to 12.8 to 1 at WP 3.5bar. This is no different then how the car ran before my rebuild so I know it likes these numbers.

I even adjusted the WP up to 3.7bar and down to 3.2bar to see if that changed anything and no luck.

One other thing that occurs when i manage to take the car out for a drive after it is hot. It runs fine down the road thru the gears but at approximately 0.5bar it falls flat and misfires as if it hit the rev limiter or fuel cutoff. IT is no where near the rev limit I set at 7000rpm. Its as if the spark shuts down.

The only thing I have not changed at this point is the original OEM braided spark plug wires. Plugs are NGK BP8ES.

QUESTION: would bad wires cause this issue as I described?
Car starts and runs fine bone cold after sitting all night, then 1-2 minuntes into the warm up it starts misfiring. Then it also misfires above 0.5bar but below that boost level its fine.

I am out of ideas and patience with this? I dont want to keep throwing parts at this but need t get this thing fixed.

The MSD prgram is replicating the factory timing like mentioned earlier in the post. Idle timing is 2 degrees BTDC total timing 32 degrees BTDC. Timing at 0.5bar boost is approx 26 degrees BTDC.

Thanks guys......
Fred
http://forums.pelicanparts.com/uploa...1371306288.jpg
http://forums.pelicanparts.com/uploa...1371306382.jpg

Some more pix of the rotor indexing. Note that I tried this first with a stock rotor with resistor in the photos but have since done the same with a non resistor rotor as required by the MSD.


http://forums.pelicanparts.com/uploa...1371307022.jpg

I checked the ohms and they are 585 which is within range.

Thanks,
Fred

Go get resistor plugs.

The red light on the MSD box is steady when its running.

Checked the timing and it is fine and follows the retard curve programmed. I can monitor it with the lap top while the car is running. Which is a nice feature.

I am really wondering if the age of my OEM wires is causing this. they could be original and 34 years old or been replaced at sometime over 8 years ago or more.

Thanks,
Fred

Can you recommend some?

I would like a plug number that is cool enough for the turbo....can you provide a plug number?

Fred

Just checked and NGK BP8ES are resistor plugs. Aren;t they?

Fred,
I’m using the Magnecor wires and NGK BP8ES plugs. (1) I would take a heat gun to the dizzy with the engine cold, try to heat the dizzy up and see what happens. Heat will cause the resistance to go up in an electrical component, see if there is any effect on the pickup coil. (2) On my car, I did the green wire Modification, since it was 28 years old. (3) Silly check- when you took the dizzy apart, did you get the rotor trigger wheel back in the right place on the shaft with that little roll pin?
I feel something is happening with the trigger event, why would it start cold then break up after a few min.

The BP8ES comes in both flavors so it's easy to end up with the wrong one. The 2912 version is non-resistor type of the BP8ES.

Andrew,
Good catch, I ordered mine from PP, did not even check resistor/non. By the way, if I seen a Skunk-works sticker on the back of a Porsche, I would approach with caution and high boost.

Dammit man! I have a box full of about twenty 2912 number plugs at home here.

Can you give me the correct resistor part number so I can order some? PLEASE-----

Got these from our host and never had a problem with them in the stock CD ignition.

Been running them for years with the old CD in the car.

Thanks
Fred

Yes I did and the roll pin is in place. I checked the rotor and its locked down on the shaft fine.

Of course I ge ta little play in the shaft from gear backlash and the slop in the parts but its about 4-5 degrees total roatation.....not bad.

Fred

NGK Spark Plugs#739-BP8ES

Fred,

I've had cars that I could run up to 0.050" gap on a MSD6A, and other cars that would only go 0.035" with out missing because of a breakdown somewhere in the HV system. Those braided covered stock wires could be the issue. Gap your next plugs at 0.035" and see what happens.

Here's a good chart on NGK plugs:

http://www.ngksparkplugs.com/techinfo/spark_plugs/partnumberkey.pdf

A resistor plug should be BPR8ES

I would agree with the chart on this. It is pretty definitive... if it doesn't have an (R) in the long part number it must not be a resistor plug.

NGK has been phasing their inventory records to the retailers to supercede the lettering with a 3-4 digit code. If you go to autozone for example, you may not find the b code as available but the other is. Jegs has these available with resistor. I would be surprised if the host would not also have these available if you phone in.

;). And if you spot an old grey Jetta with the same and making funny noises, good luck.

Thanks Andrew.

And thanks everyone for their input.

I measured the resistance in my wires yesterday and they ranged between 39.1 to 40.4 kilo-ohms. Dont know if this is good or not and of course this was when they were cold.
Does anyone know what the resistance should be in the OEM braided wires?

Also, took a compression test and cylinders were five at 145psi and one at 150psi....pretty good. so I know its not an issue with sticking valves or dead cylinders.

I gapped the plugs at 34mils down from 40mils when the problem first surfaced so I could try dropping the gap again, but maybe I should wait before i order a set of the proper resistor plugs.

Here is a photo of the plugs removed yesterday:

http://forums.pelicanparts.com/uploa...1371393581.jpg
http://forums.pelicanparts.com/uploa...1371393603.jpg

Just wanted to post an update on this problem.

I replaced the OEM braided wires with new Magnecore 8.5mm racing wires. This alone did not make much improvement. Also rerouted the violet and green wiring harness from the distributor and used a new coax green connector instead of tapping into the 6 pin original wiring harness for this connection which used the original green wire. This made a big improvement as i made sure to run the harness away from any other wire source and as far from any spark plug wire as possible to reduce EMF interference.
This fixed 90 percent of the misfire at idle and low load driving. However the car still misfired at high boost.
So I made one last change to replace the plugs with resistor NGK IX iridium plugs gapped at 30mils. I think the original gap I was running of 34mils was blowing out at high boost.
Ran the car with all these changes and it pulls like a train now and no more misfire at high boost.
I still have a misfire when i first start driving the car after a cold start. I cant put much load on the car and can cruise around in low rpm high vacuum until the car warms up after 5-10 miles before I can give it any throttle without a misfire.
Once the car is warmed up it runs without any hesitation. Dont really know what is still causing this but the car's ignition programming is emulating the stock distributor curves with the only modification I made was to set idle timing to 6 degrees BTDC versus 0 stock.
I can live with it for now but would like to get this last thing resolved to make it 100%.

Wanted to thank everyone who contributed info to me to solve this problem.

Hope my post might help anyone else making this switch to MSD programmable ignition.

Fred