Hey, Dave. Not my car, not my call, but I hear you. I've seen 2 types of tensioner failures in 30 years though. Both to do with the tensioner arm. First was due to a tensioner roller bearing seizure. Second by way of a sheared tensioner arm bolt that suffered excessive corrosion.

Never heard of or observed any failures caused by anything from within the insanely simple OE tensioner, and I can't count how many I've found that had been run dry for countless miles.

Then, there's this..

So, if the piston was freely moving, I'm not going to scold someone for not going back inside a non-leaking, functional tensioner to... clean.. and inspect/replace the o-ring.

 

FWIW when you do the T belt you should remove the tensioner and replace the gasket.

NOTE I use Hondbond or Dreibond 1209 on the gasket when fitting the tensioner back to the engine,
I put a load on the tensioner roller like the belt would apply then tighten the bolts.
This will reduce the amount of movement that the tensioner will move till it finds its working position after a few heat cycles.
Since its not pinned the tensioner will move some, this could be enough to make the low tension circuit get triggered.

Also replace the rubber boot and clamp,
orient the inner C clip so the open end faces up,
and the outer clamp so its clocked to about 11 O clock while looking at the boot from pin to adjuster bolt.

Make sure the piston inside moves freely in the teflon coated tensioner,
I think we are starting to see some of these units teflon coating starting to fail possibly from them being operated dry.

I suggest to use STP oil treatment oil to fill the tensioner, , I use a visene bottle with the tip drilled open .
Fill from the port closest to the right fender, watch for the oil to come out of the hole closest to the crank.
NOTE its possible to overpressure the boot with an oil hand pump device,
this can cause the boot to come unseated and leak, So the visesne bottle works great for not being able to overpressure the refilling event

 

Adrian,

What took you back inside the timing belt covers this time around anyway? You refreshed the tensioner fewer than 20k miles ago, so this doesn't scream of scheduled maintenance. Water pump? Bearing noise? Do tell.

Stan, by "break it down", I was referring to the relatively new clamp & boot - not removal of the tensioner. A bit obvious, with a new report from Adrian, that replacing the boot & clamp should be done.

 

it’s just that I did the last change in 2015 so was feeling guilty.

Good thing anyway given the bound tensioner.

@Mrmerlin I was intending to renew gasket, boot and rear o-ring but thanks for reminder. Last change I fed in STP but it was impossibly thick (maybe the temps here) so I will use a lighter oil this time round.

@worf928 I did replace idler and tensioner roller and bearings last change. I've just checked my records and it was 2015 but only 11K miles since since and they feel perfect and silent so I think I will reinstall them.

 

I suspect the best way to get STP into the tensioner is with a hypodermic syringe.
I used STP previously and it sure helps minimise leakage but I am not convinced it is optimal for the service especially in colder climes.
I think I used diff oil last service which I think is what Porsche recommends

 

The belt is the *least* important part to replace when you are doing a “timing belt service.”

The belt will last for far, far more than ten years if everything else is working as it should.

But, how long will a 10-year old:
- tensioner boot last?
- tensioner gasket not leak?
- tensioner be full of oil?
- main roller…
- idler…
- carrier bushings…
- crank idler…

If you are replacing the belt and leaving everything else, you might as well not have gone to the trouble.

Seriously.


@Kiln_Red The tensioner failure mode I have seen several times (and the only mode I have seen) is that it has been run dry for so long that the coating inside the bore is done. Thus the washer stack can’t move freely.

 

The Belleville washers won't do their thing if the tensioner is dry for any interval of time.

 

Due you mean that they become permanently damaged?

I’ve never thought about it in cases where the tensioner is mostly dry but the bore coating is undamaged.

In the cases where the coating is damaged and/or the stack won’t just push out easily the whole thing goes in the trash.

 

No, just that the tensioner isn't compensating the thermal expansion that it otherwise would if it were filled with gear oil.

 

That’s interesting. I always understood that the role of the oil was to transmit block heat faster, and so decrease the time before the belt was detensioned as engine heated up. If the tensioner has no effect at all when dry that sounds as if it could have very serious consequences, including fast cam gear wear?

 

Adrian,

As I am concerned the oil has two prime functions- to carry heat into/away from the Belleville pack and just as importantly [or so I believe] to act as a damper fluid. The cooling system of the motor takes about 3 minutes to warm up to 80C plus and that is when the Belt tension alarm turns on- coincidence? Ironically the engine oil takes about 10 minutes to get fully warmed but that is another matter.

The belt tension alarm feature does not exist on the early models that have a simple barrel that is filled with oil and non interferrence motors. I am not sure when the revised design was introduced but I have a feeling it was introduced to coincide with the change to the HTD cam belt sprocket system some time around 1984 and/or the arrival of Euro S interference cams. The later model tensioner has an inner and an outer annulus and there is a check valve in the back plate that presumably allows tensioner oil to flow down the outer annulus and then back into the central annulus via the check valve as it presumably flows along the Belleville pack. Why Porsche engineers modified it this way I could not fathom out with any certainty initially but then I discovered that the 16V Euro S models had interference higher lift cams ergo the belt tension alarm system and more critical tensioner design perhaps? I know the Belleville pack responds to temperature changes rapidly. Oh and both tensioiner designs have US patents on them- that tells one something albeit they will have expired long ago.

Now considering what the timing belt does there are 8 pulses generated from the exhaust cam and 8 from the inlet cams and that pulsing energy has to go somewhere- some of it will be dampened out in the cam belt as it streches and contracts and the tensioner pulley assembly follows the belt trying to keep a uniform tension and logic says that dampened energy goes into the tensioner piston- it has to go somewhere.

The big mistake many folks on this list have made over the years is thinking that when the Bellville pack is at operating temperature it is then a rigid assembly but that is just not the case at all. For sure when the pack heats up it flattens out to some extent to compensate for thermal expansion of the block and the calcs I did to figure out that degree of contraction coincided with the shrinkage I meaasured when dumping the pack into boiling water- tbhe pack shrank instantly. Although Belleville washers are very rigid on an individual basis, the very reason they are assembled in very specific groups is to custom tune both the degree of thermal contraction and the spring rate of the assembly needed hence why correct assembly is critical. Although most owners do not realise it, you can actually compress the spring pack in one's fingers believe it or not- I forget what the composite spring rate is- from memory it is something like 0.6 times the spring rate of a single washer and thus it is not really that stiff at all - thus why one can compress the pack slightly and thus how energy can pulse through the pack.

The Belleville pack being configured the way it is has oil saturated into it with lubrication between parallel washers and series configured washers and as the pack oscillates it absorbs energy part of which acts as a pump of sorts to move the oil around the circuit in one direction running forwards within the inner annulus and backwardss in the outer annulus thus improving energy transfer and damping over the original design - it is a very eloquent design, needs little attention each time the belt is changed and consequently is much more sophisticated than most realise. I suspect it is this motion that helps the pack warm up quicker and when up to temperature helps remove excess heat build up from the absorbed energy quicker.

If the oil is lost the system will work to a point but, whereass the Belleville pack will eventually heat up through conduction it will however take longer and thus possibly cause belt over tension for a short period - I also beleive it will also cause friction between the washers and cause them to wear out much faster due to fretting rather than lasting forever and a day and the shaft piston will likely scrape the walls of the inner chamber and screw up the tensioner completely.

As I said previously, filling the pack with STP is difficult- I tried it. However after studying the design in quite some detail I figured it was better to stick with the Porsche recommendations of using either diff oil [in a hot climate like mine ] or engine oil [20W50] in a cold climate to facilitate the pumping action on the later model tensioners. Check the oil level once a year or more frequently if covers high miles per annum.

As I said earlier I was not sure why Porsche changed the design but I have a feeling it may have been to fit in with the era of interference engine models where belt failure was not an option. Thus I suspect the Euro S models and the gen 2 tensioner design appear to have arrived somewhere together around 1983 or 1984- perhaps Dr Rob can offer some more complete data in that respect.

In short the stock tensioner does what the PKT tensioner does but in a less user friendly package that needs careful attention with a viable belt tension checking tool- no such need with the PKT that also happens to be half the price if the orignal item is screwed. If one does not have a belt tension tool there is an easy workaround for the later tensioner design that is probably as accurate as the factory tool or the Jay Kempf tool that I use and that is to back off the tension on a fully warmed up motor until the alarm comes on and then to tighten the adjuster nut two flats. That sort of gave the same results as my Kempf tool.

Finally the alarm itself, another brilliant design feature really needed on the interference models. Over the years we have seen owners have heart attacks when the alarm illuminates and have their car towed to wherever as a consequence. If the alarm comes on and stays on that is a sure sign there is an issue but most times simply accept the alarm sees it go out never to return- a flase alarm as it were- but still appropriate to check the tension. Fact is the alarm says "timing belt service" on the later digi dash annunciator and for sure it does not say "Tough luck A-hole you have 32 bent valves and 8 buggered pistons"! Setting the tension correcty very much depends on doing so with No1 cylinder on its firing stroke- tension the belt on the opposite crank revolution and the belt tension will be incorrect through the range.

 

I'm not convinced that the washers are sufficiently heated, without the oil, to provoke sufficient physical change. I'm only working from my own logic, but I don't think Porsche would've recommended gear oil for the tensioner if what they were seeking was an efficient conductor. Gear oil is not particularly good for this purpose, but it is okay as a thermal insulator. At first glance, this would seem to support the view that the tensioner does not necessarily need heat brought into it. We know better of course. The tensioner can't create its own heat.

But, in the absence of a thermal insulator, the tensioner is likely dissipating the engine heat too quickly to be effective. I might have a 928 here with a dry tensioner. I know I have one with a recently refreshed tensioner that is holding fresh gear oil. Maybe some testing with the IR gun is in order.

The obvious issue, of course, is accurately accounting for any disparity of cooling system efficiencies between the 2 cars.

 

Thanks Fred for taking time to write most complete analysis of the washers/oil system I’ve seen.

i do have Kempf tool but that’s also a smart way to set tension in its absence.

Also good to know that if my tensioner is damaged (was going to remove it and take apart today but other things intervened) then the PK tensioner is half the very high cost of a new ‘traditional’ item.

 

If tensioners bereft of oil did not heat up as needed, I suspect such examples of the later type tensioner would be flashing timing belt alarms left right and centre or so I would think. The engine block is an aluminium alloy and the coolant heating it circulates at 98C, the crankcase seems to operate at around 80C and heat will flow readily given the way alloy conducts heat. I would think a lack of oil will likely cause premature wear of the internals and seriously impact the ability of the tensioner internal assembly to act as a damper.

 

It's possible the way that we think the tensioner works is exactly opposite from one another.