More Distributor thoughts - long and nerdy

Dan Cordon cord4530 at
Wed Dec 10 02:18:55 EST 2003

Comments in the middle :o)

QSHIPQ at wrote:

> I'll start here, since I agree it's not the backend spark I'd be worried
> about.  Great report on the distributor measures Dan.  My concern in reality would
> be a few fold.  First, you have accounted for no slop in the distributor gear
> OR the cam chain (remember, the distributor fires off the intake cam which is
> connected to the exhaust cam, which is *then* connected to the crank via a
> belt.  Think slop.  Add all slop together, I propose it allows very little for a
> plastic gear.

I'd agree...slop is a potentially huge problem. It's clear now that distributor geometry
*isn't* the main cause of failures.

> My next concern would be how crossfire might manifest itself.  One might
> propose that if those contacts in the rotor center and the distributor fingers get
> corroded enough, then I believe the same problems Bosch encountered with the
> 99cent "platins" might actually manifest themselves.  At 40k distributor
> volts, I would expect the (next) spark to jump to the point of lowest resistance.
> IMO, that just might be the leading edge of a rotor to the outside (non
> contact portion - read clean/uncorroded) edge of the next finger in the distributor.
>  Similar to the "platin" problem in that as pitting developed the spark
> tended to travel up the side of the electrode (remember the platins were just a
> coating of ceramic, not a "casing" per sae.  I won't use the term "carbon
> tracking" but it sure is tempting boys...

To add to this, a high pressure air-fuel mixture seems to be extreemly hard for an arc to
cross. So, we have a corroded section of a tip facing a corroded terminal, connected to a
plug in a 'much less conductive than air' environment. While the clean leading edge of the
rotor has a (distant) view of a less corroded contact connected to a plug that's in nearly
astmopheric air. I could definitely see forward arcing happening. At least *occasionally*

> Historically, when the points ignition systems went HEI (read:  GM and Ford),
> the Distributors got huge/wide and the design of rotors became part and
> partial to reducing the propensity for crossfire isssue when doubling ignition
> voltage.  VW-audi/Bosch, on the other hand, freely changed from low energy spark
> points based ignition systems to high energy ignition systems, with NO
> appreciable change in distributor cap OR rotor design.   A 6mm trim up of the rotor in
> 1991 20vt only: a bandaid really.  A really bad one when you add in slop.

I also agree on this. I put a high power electronic setup in a normal Ford V8 cap. That was
a bad idea. Cross fire all over. But then, the spark would jump a full inch, and there was
less than an inch between could have been firing three plugs at once!

> Take out the slop, any slop, the problem of crossfire *should* be less.
> Remember, crossfire happened without a *broken* plastic gear necessarily.  That
> indicates to me that the right combination of slop factors in the right
> circumstances could very well add up summarily to disaster.  I suspect that addressing
> <all> slop in unison is the safest thing to do (physically replace and
> inspect tension on TB, Cam gear chain, and Distributor/gear).  IMO audi went away
> from distributors for good reason, they wanted to keep the slop in the rest of
> the design parameters that ultimately resulted in distributor slop, in a bad
> distributor design.

I still can't figure out one thing. (Okay, there's MANY things I can't figure out) When
putting the new distributor in my engine, the window for correct placement was very small.
I got no spark at all if I was + or - 2° off of the 'correct' location. Some of that is
probably due to slop in my TB and chain. But it seems that if the slop in an engine got
really bad, the ECU wouldn't get the signals from distributor and crank close enough to
fire the coil at all.

> I also think you will see a progression of less slop in Motronic up to the
> Camshaft position sensor = multi coil fire ignition.  If you look at the v8 for
> instance, you will see that the distributors are run from the exhaust cam,
> which is directly connected to the Timing Belt and auto TB tensioner (auto ign
> slop adjustment for the service life of the belt + no cam chain slop included).
> Then as the AAN (new 20vt motor) came about, the CPS is also connected
> directly to the timing belt  (albeit with manual TB tensioner) via exhaust cam.  I
> see these designs as progresive ways to not only reduce ignition related slop,
> but also eliminate cam chain tension slop.  Needed btw, if the distributor and
> rotor system design remains 30+years old in a HEI.
> Ah, you might say :)!  What about the 30v v6 cars and the 1.8t?  Their
> respective CPS are back to intake cam driven.  Yes, they are, but the
> electro-hydraulic cam chain tensioner all but eliminates (or should before *it* fails -
> don't ask) the problem of chain slop.  IOW the computer can compensate for it.
> And NO distributor.
> I'm no expert on any of this stuff.  I just enjoy watching how engineers at
> audi do things progressively and/or regressively.  In the case of crossfire and
> motronic, they addressed it in the v8 and not the 20vt for those first couple
> years.  I believe this is because audi already knew that direct fire ignition
> was on the way, and future designs (up to and including today's multivalve
> heads) were already addressing the problems inherent to crossfire ignition.
> Summary:  Motronic turbo finally dictated a 30 year old distributor cap and
> rotor design obsolete.  Redesign the cap and rotor or go multicoil?  Thank
> goodness *something* good came of all this.

Yes, this is good. I think I'm sufficiently inspired to work on a coil-on-plug system for
the 20V engine. Mihnea has convinced me that this may be a good idea. I'm already thinking
of some improvements for this next design. I won't have any time to work on this until
after this semester is over. And in reality, as a pet project it is likely to take me a
while. But at least it's a start. I should be able to do the design and prototyping (I have
a gearless distributor, and a 20V to try it on), but after some customer feedback and an
update, I really have no means of manufacturing this on any reasonable scale.

And since our campus got rid of Pro/E, I'll have to learn Solidworks....something I've been
meaning to do anyway :o) I'll keep the group posted.

> My .02 arbitraged thru the peso
> Scott Justusson
> QSHIPQ Performance Tuning

Dan Cordon
Mechanical Engineer - Engine Research Facility
University of Idaho

More information about the 200q20v mailing list