Imo000

I don't speak Elvish.

UpFixenDerPorsche

Hm. Surprised this pic has not been banned by the national "Citizens for Decency" movement. And I, Sam the pure and righteous Eagle, for one, will be leading the charge to bring it to their attention. Hm. Hm.

UpFixenDerPorsche

I think you'd be diving into a very deep pit with a couple of giant chimpanzees (they're more sadistic than gorillas) plus a few rattlesnakes.

As others have said, the use of high quality belts, the PK tensioner (a 'real' hydraulic tensioner), and attention to a few time based maintenance/inspection steps means the belt system should be virtually disaster free.

69gaugeman

Do this and the factory system is just fine for this application.

The Deputy

Note to self...never, ever, apply for a job at Audi...unless its to sweep the floor or make coffee.

I bet whomever engineered that contraption had more than a few sleepless nights in the process...and most likely wrenched on or engineered motorcycle engines once during their career.

Brian.

69gaugeman

This is one of the pitfalls of modern CAD engineering. It has allowed people to not come up with efficient and creative solutions to problems. Just add another chain! Pre-CAD, you though LONG and HARD before you put your pencil to the paper.

I think that is what is missing in today's newest engineers. Yes they can design things with much better function in smaller spaces and know beforehand if it will be strong enough, but they lose sight of the K.I.S.S. principle. I have found myself even falling into this trap a few times. The nice thing about solid modeling is you can erase and start over pretty quick (in conventional engineering time units, anyway)

jayc67

Even more simple than the mechaincs of it, look at the financials.

You do a TB/WP change every what, 5 years or so? $2500 each roughly, if you pay someone else to do all of it, less if you do it yourself.
The factory system is rock solid if you maintain it.


What is a conversion to chain drive going to cost? If a chain conversion has a 25-30 year ROI, what's the point?

If the chain lasts 15 years, and it's 20 years before break even.....that's actually moving backwards financially.

GregBBRD

One of the things that almost everyone misses is the fact that ANY change in the timing belt area is going to significantly change how the engine performs, at one end of the rpm range or the other. This includes the OP's thought of converting the entire thing to a chain, it includes changing the timing belt to a more stretchy or less stretchy belt, it includes changing how the slack is "absorbed" when the belt stretches.

The fact that the stock belt stretches on the pulling side (oil pump gear side) of the crankshaft drive gear significantly retards the cam timing. Remove the covers, rev one of these engines up on the dyno to 6,000 rpms and you can see the belt "grow"....and retard the cam timing. Put a timing light on the cams and you can watch the cam timing retard at higher rpms.....increasing at a fairly constant rate through the rpm range.

Yup, Porsche, whether by choice or accident, ended up with a system that allows the cam timing to retard at higher rpms.....and this occurs in significant degrees of crank movement with the stock belt and the stock tensioner.

It's a beautiful, very clever design....which is why I use it on every engine I work on or build.

I have actually have "tuned" high performance camshafts to run better at the higher rpms and still have great emission qualities by "playing" with the cam timing and then "playing" with the belt tension.

One of my very first questions to Roger (call him, ask him)...about the Gates Racing Belt, is where I was supposed to set the cam timing. (If the belt stretches less than the stock belt, the cam timing changes less and more initial retard is required to keep the same performance at higher rpms.) Roger (being a great guy trying to develop a usable product for everyone and attempt to answer all concerns), went back to Gates and asked the engineers if any change was required to the cam timing, because the belt stretched less. Whether they "missed" or didn't understand how the engine worked....or didn't want to sit down and figure it out, they responded that no change was required. This, of course, was absolute nonsense.....the difference in the amount of stretch in the belt radically changes cam timing at higher rpms. To this day, I'm not aware of anyone that has measured what is required....and I didn't think it was my job to measure this for everyone else, when the stock belt was still available and did exactly what it was supposed to do (as the Porsche engineers designed it.)

I'm not going to even start on what happens with an Audi tensioner that tightens the belt tighter on the "wrong" side of the crank pulley, when the engine is running and changing rpms....suffice it to say that the cam timing isn't going to retard nearly as much at higher rpms than the stock system.....any "slack" is gathered up on the wrong side of the crank pulley (non pulling side.)

Combine a Gates racing belt with an Audi tensioner....I don't think anyone can tell you what the difference is in cam timing at high rpms versus the stock belt and tensioner, because I do not believe anyone has actually considered or measured this.

Back to the OP's thought. A chain is very resistant to stretch....by design. Converting the 928 engine to a chain would either require a complete redesign of the camshaft, or more likely a hydraulic tensioner system that changed the chain length (and thus the cam timing at higher rpms), much like modern engine designers are now doing to get good emission qualities while still getting great performance out of current engines. Note that the 911 series engines have been actively doing this since 1998.

The message here is clear (read the above a couple of times and think about it....it is really a simple/logical concept to understand), when someone changes almost anything in an engine, careful thought about the ramifications of that change is required. And this requires someone to carefully consider/calculate what else needs to be altered to keep that engine performing like it was originally designed.

Chalkboss

What you are saying Greg makes sense and is very well articulated. Thanks for sharing your knowledge. It would be very interesting to see one of these engines run with the covers off and see this in action. Certainly gives me more to think about, that the system is built with a certain amount of "elasticity" to meet performance standards.

bureau13

Now I want clear cam covers on my engine...

UpFixenDerPorsche

What sort of timing change numbers are involved Greg?

Presumably worse for high performance cams?


Cheers

UpF

PorKen

Please tell me you are being sarcastic. Otherwise...dayum.

It should sound good. Thank you! Greg is cherry-picking 10 years of me beating these belt tension/belt management concepts into his head. Up until ~2014 Greg would not admit that the belt stretched after the initial break-in period. Oh well. At least folks, including Greg, are starting to understand what I have worked so hard to boil down into concise terms over the course of many many pages of discussions with Kibort, Greg, et. al. defending/explaining the PKT and PK32V'r.


Unfortunately, Greg has not quite connected all the dots yet, as evidenced by the following word-salad:

Let's review, shall we?

The Audi Tensioner/Damper is constantly managing the free belt length and thus requires a lower average tension on the belt at all times. When decelerating tension will be higher for a few seconds until the T/D releases the belt which has been stored on the slack side from any stretch of the belt generated by the pulling of the cams by the crank when accelerating. Running tension with the Audi T/D is lower than the factory system which requires high static tension, IE. pre-stretching of the belt in order to cover the range of belt length, as the factory tensioner itself does not have much range. Frankly, once the engine is running, the factory tensioner does jack-all to manage the belt.

When running, belt tension is set by the crank pulling the oil pump, water pump, and cams.

As Greg has seen (finally), cam timing retards with a stretchy belt because the belt gets longer as it is pulled taught by the crank. This extra belt length ends up in-between the crank and the 1-4 cam as the belt is spit out from the crank gear. With the factory tensioner at high rpm there is so much extra belt the belt flutters or flaps. This flapping wears the belt and gears. If there is enough flutter the belt can literally float over the 1-4 cam or crank gear. The extra belt also allows the 1-4 cam to change timing ± randomly.

What Greg has not seen - and what I have been saying since 2006 - is that the Audi T/D system never lets the belt flap because it extends and retracts as the belt rolls over it.

PorKen

What I am trying to currently rationlize is whether the benefits of the high strength Racing belt outweighs passive cam retard. If one is going for maximum change from low to high rpm, you'd want to run the stretchiest belt - the Continental - which I ran on my S3 which peaked out over 310 RWHP, vs my S4 with a Racing which has peaked around 305?

What's nice about the Audi T/D system is that it can absorb anything from the Conti to the Racing and it runs at a lower static tension, allowing more length at high rpm...

GregBBRD

Ken:

Instead of being such a troll and constantly personally attacking me....why don't you just address the topic?

You seem to think you can simply tell any lie you want and people will believe it.

I've been readjusting cam belts since you were in diapers.....I've been fully aware that cam belts stretch and need to be re-adjusted since 1978, when I was adjusting cam belt after the first 2,000 miles on brand new cars.

I've been taking "advantage" of the fact that the belt stretches under acceleration, retarding the cam timing, since way before you even thought about owning a 928.

Why make up bold faced lies, which you have done above, about things like this?

Address the topic with some theory or facts of your own.

Stop the personal attacks. Stop being a troll!

Since the very beginning of your tensioner system, I've asked you the same questions....over and over again.....what exactly happens to the cam belt with relationship to the cam timing when the engine is running, with this system? Where do I set the cam timing initially to get the same results as with the factory system? How do I set the initial cam timing with a tensioner that, by your own admission, gets significantly tighter when the engine is running and thus changes all of the stock cam timing specifications?

All I've ever gotten back is personal attacks, which divert attention away from the original topic/questions.

This was a clever tactic in "high school debate class" and might get some chuckles here, but it doesn't address the questions.

Apparently, 10 years later, you are getting ready to actually measure what happens in a "scientific" manner with your tensioner system.

I applaud that effort....get it done and give us some viable data.

PorKen

Seriously. Get over it.

Same thing that happens with the stock tensioner system, minus the flapping.

To the factory settings. (This is much easier done with the PK32V'r, BTW.)

You must spin the engine with the starter to settle the belt into the gears and equalize the tension between the gears, first.

Equalizing the belt tension should be done using the dial indicator too. This is why you can't set the timing accurately on an engine stand. You may max out the belt tension to try and get consistent results, but what happens when you run it at normal levels? As you have now noticed, the lower tension will affect the timing. Then the belt stretches from new to old. Then it stretches from low rpm to high rpm. Then the engine warms up...etc, etc.

Obviously you are still not listening to what I keep talking at you. It does not get significantly tighter. Under steady state running the tension level with the Audi T/D will equalize in a few seconds when warm. Tensioner. Dampener.

As I say, again and again, the cam timing will always be dynamic, never exactly what one would like it to be. Probably best to either set the cam timing by equalizing the compression on the 1-4 and 5-8 banks, or by picking a particular RPM that you want the cams to be spot on, EG. at max HP, then using a timing light to check/adjust the timing until they hit that target together. Like the cams themselves, there's always going to be a trade-off, some range where it works best.