What is with distributor rotation and cap fastening?

[Walt Fricke]

I knew that from 1966 through 1977, the 911 distributors (at least for the USA motors) turned clockwise. My 2.7 did.

Then, for the 3.0s starting in 1978, the distributor rotated counterclockwise (and it's case diameter got larger, to accommodate the electronics which displaced the points).

What I didn't realize was that the 3.2s went back to clockwise rotation. Why did Porsche make these changes?

The 3.2 distributor cap fits right on the 3.0 distributor, and vice versa, and the rotors are the same. But why did Porsche change from the convenient two clips on the earlier distributors to the harder to manipulate system used on the 3.2s?

I became acquainted with these differences while measuring cam shafts in race cars at the track, which involves using a modified cap to hold a rotary indicator. The rotation direction was (due to the foresight of the wizard who designed the electronic data system in having a direction switch on the control box) easy to deal with, but getting the 3.2 distributor caps back on securely when checking was done was a whole lot more difficult than with the 3.0 clips.

[BURN-BROS]

Who knows Walt, A few of those Mag Pulse Dizzies were clockwise rotating (3.0 Turbo). It would have been much easier had they kept one convention instead of rotation and gear diameter changes.

The cap change was probably a Bosch "improvement". Looking on insulate the cap with extra material to reduce EMI. The fastener change is due to the Germans being Sadists.

[Walt Fricke]

Aaron - I am glad I am not the only gearhead who was surprised at the relative difficulty of the 3.2 cap attachment. In time doubtless I'll get better at it. One of the guys whose cams we checked at Sebring didn't know how to remove a valve cover, but happily was fine with removing, and more importantly, putting back on, his distributor cap when we were done.

Another thing I left out: The 3.2 caps use a post, not a hole. Other than not being able to use 3.0 and earlier plug wires, I suppose it isn't an issue. Is the post a firmer attachment? The coil packs I have on my modified car with crank fire have posts like that. Did the whole, or most of the automotive industry, decide to change from negative to positive or whatnot?

It certainly was harder to drill out axially than the 3.0 cap when making an adapter for measuring cam rotation!





Apologies for the fuzzy focus on the middle picture.

[Eagledriver]

I seem to recall that the SC distributor was sourced from a Mercedes application. The rotation was due to the other application. The newer retention system is probably just progress.

[manbridge 74]

3.0 was made reverse rotation to what came previous because any wear in the distributor or problems with advance/retard would cause timing to retard thereby saving the motor compared with runaway advance.

3.2 have Motronic DME so no issues with timing advancing, so they changed the direction again.

[Walt Fricke]

The 3.2 retention system doesn't seem like progress to me. Has anyone had problems with the 3.0 and earlier clip system?

[BURN-BROS]

I believe the secondary system was changed to comply with the requirements to run Motronic.

Bosch had both CW and CCW Mag-pulse in production at the same time, so who knows why the change.

[Eagledriver]

Originally Posted by manbridge 74

3.0 was made reverse rotation to what came previous because any wear in the distributor or problems with advance/retard would cause timing to retard thereby saving the motor compared with runaway advance.

3.2 have Motronic DME so no issues with timing advancing, so they changed the direction again.

Wear will cause the timing to retard regardless of the direction the distributor turns.

[Cairo94507]

That's interesting. My 3.2 was final assembled by a very respected shop in the LA area and after about 2,500 miles I had a sudden engine stumble that was persistent. Turns out the distributor cap was not fastened/secured correctly. One of the fasteners was not done. Problem solved easily enough, but still frustrating at the time. Glad to see it is a "thing" on the 3.2 motor.

[gled49]

My 2 cents, with clockwise, counter clockwise. Change in brass gear and distributor pinion size in clockwise rotation etc, was to foul update/backdate options. Same as trying to backdate cam drive from one piece sprocket to vernier system. JMO

[Speedy Squirrel]

Like all things Porsche, it's complicated, and has to do with the presence or absence of a vacuum capsule, the location of the vacuum port the capsule is connected to, and the objective of the control system.

Early cars - no vacuum capsule, could rotate either way. Bosch was using clockwise, so that is what Porsche used.

Later cars had vacuum capsules with several objectives, but the key thing to remember is that vacuum canisters can only pull the distributor plate when vacuum is applied. If you want that to produce spark advance, you need the distributor to rotate one direction, if you want it to retard, it needs to rotate the other direction.

They did away with the clips because it did not hold the cap tight enough. There is enough play to rotate the cap several degrees. The "push and turn" eliminates that to a large extent. The screw style terminals allow wires that produce a more consistent grip on the distributor terminal.

[Walt Fricke]

Well, this is interesting.

However, my SC's distributor has vacuum ports on both sides of the vacuum diaphragm, so it can both pull and push, can't it? I'd always wondered about that until a guy here had trouble with his leaking and cut it apart and we could see why a seal around the actuating connector wasn't needed.

And a few degrees of movement of the cap wouldn't make any difference in spark timing or intensity, would it? The cap posts inside at the rotor are what - maybe 5 or more degrees wide? And the end of the rotor maybe three times that?

Although compatibility is needed, isn't the advance determined by (other than by the ECU) the rotation (centrifugal and vacuum) due to where the points or reluctor bits are? And change in position at firing of the rotor itself kind of immaterial?

In designing an in-car cam measurement system using a replacement cap, my 944 friends warned me that the 944 cap retention system was rather loose rotationally, and they said it didn't need to be tight because the ECU did the timing and the rotor tip system was broad enough to handle the entire advance/retard range there.

I had supposed the post style terminals allowed an electrical/mechanical connection which was better (more like most spark plug connections) than the older push in system.

Speaking of plugs - are they all still designed with a male screw tip, and a female screw on wasp waist piece? Why not just the wasp waist part integral with the plug? Are there motors which still use a ring connector these days? Some motorcycles? One NGK plug I favored came without the screw on piece (purchased from a motorcycle shop), so I always had to take off the old screw on parts and reuse them.

[Speedy Squirrel]

Originally Posted by Walt Fricke

And a few degrees of movement of the cap wouldn't make any difference in spark timing or intensity, would it? The cap posts inside at the rotor are what - maybe 5 or more degrees wide? And the end of the rotor maybe three times that?

Although compatibility is needed, isn't the advance determined by (other than by the ECU) the rotation (centrifugal and vacuum) due to where the points or reluctor bits are? And change in position at firing of the rotor itself kind of immaterial?

In designing an in-car cam measurement system using a replacement cap, my 944 friends warned me that the 944 cap retention system was rather loose rotationally, and they said it didn't need to be tight because the ECU did the timing and the rotor tip system was broad enough to handle the entire advance/retard range there.

The new cap was never on a variable reluctance system, only DME (3.2L and later). Two degrees at the cap is 4 degrees at the crank, which matters. DME has a wide timing range of authority at all speeds. In fact, the DME distributor has a small, very stiff centrifugal advance mechanism to keep the rotor in the range of the cap posts at full advance. Isn't that interesting?

A vacuum push system just costs more than making the distributor rotate the opposite way. It's not technically impossible to make a push type vacuum capsule and a clockwise distributor, but why do it when there are better options?

Here is a cutaway of the dual action vacuum capsule. You can see how the annular diaphram allows for a push/pull system.

[304065]

Great thread and great explanation. This is what the forum is all about.

I would only add that the greatest automotive frustration I have experienced in the last 10 years was trying to get the distributor cap back on a 944 Turbo. Finally I asked Ed Mayo, who told me to heat up a screwdriver and bend it, which worked perfectly. They should have included one in the tool kit.

[manbridge 74]

Yes, good explanation. At Porsche training you just took them at their word. Not that they wouldn’t answer questions but with a lot of material to cover they’d offer a quick answer about the why and move on right away...

[Walt Fricke]

Tell me what the deal/trick/problem is with getting the 944 distributor cap back on, please. I'm working on a cam measuring system which will require taking the cap off, and using a dummy cap with a rotary encoder.

Or is the turbo distributor cap system significantly different than that on the NA motors?

[manbridge 74]

Turbo is different. It takes a good bit of ingenuity to get the cap back on. Standard 944 is easy....

[targa72e]

HI Walt,
I don't think your idea will work a the 944 cars. The cam rides in a cam tower with no cover. Only way to get at the cam is to remove the whole tower.



john

[304065]

Walt,

The 944NA and Turbo are the same, it's just that the Turbo has a lot more extra stuff shoehorned into the engine bay.

Behold the area of concern (images copyright Pelican Parts)



You can see that the cap is held on by two spring-loaded clips, which have a philips head feature at one end and an L-shaped protrusion (the "Protrusion") at the other.



To fasten the cap you place it in position and then rotate the clip until it engages a slot in the cam tower. In order to position the clips so the Protrusion grabs the slot, you must push DOWN on the clip to overcome the spring pressure, then rotate the clip with the pressure applied, fit the Protrusion in the slot then release pressure.

The upper one is easy. The lower one requires use of a screwdriver bent 90 degrees, and the Official Factory Swear (Himmel. . .Herrgott etc. etc.) The reason is that there's no clearance, the headlight linkage gets in the way.

See, infra.



I hope this discourages you from ever going near one, I think you're a terrific guy from the couple times I met you at Parade, and would hate to have you suffer the same VEXATIONS I did.

(I realize this is all related to your scrutineering, so it MUST be done. Perhaps welding some wing nuts to the heads of the clips, so that you can manipulate them AND index them to the slots, would be the solution for a fixture that would allow you to encode the cam rotation when you measure the lift. Getting the cap back on would be the responsibility of the hapless racer )

[Walt Fricke]

Thank you, Herr Moderator. We did meet - Traverse City, maybe?

Just the stuff I need to know ahead of time. Luckily, I think most if not all of the 944s racing have removed the headlight stuff. And there are a lot more NA 944s racing than 951s if they have plumbing in that area. The 944 system looks to be almost identical to the 3.2 911 system, though with less convenient access.

John - I've got one of those sitting in my garage, and about to be transferred to my desk where I type. Look at those 8 holes. With a cleverly fabricated bent this way and that indicator tip I should be able to get on the cam lobe, and maybe even on the lifter (which means you can use a small tip, and are timed like the engine is). It seems that guys use something like this to check cam timing? Though that may not require following the whole cam profile? There are a few places on which to attach an indicator holder, including others of those head bolt access plugs.

That's the plan, anyway.

The 911 system is coming along. Next is building a plot where there is either a band instead of a point line, reflecting the couple of thousandths of tolerance and measurement variation, or just a line which takes into account variation which can be used to see if a cam whose three main maxima are OK also stays within that profile.

Plus I need to figure a convention to measure duration on the cam lobe - the Porsche specs involve 0.1mm lift at the valve, so we may have to invent one for the opening and closing lifts on the lobe. Since this will be arbitrary, I suppose I could apply a 1:1.65 ratio backward to see what exhaust cam lift would would approximate exhaust valve lift.