Timing belt???
#18
Shameful Thread Killer
Rennlist Member
Rennlist Member
The Lamborghini V12 and the early Ferrari V12s use chains with good success. This thread is about the interference aspect of the engine, not really the drive choice. Designing a high compression NA cylinder with high efficiency combustion chamber, and including non-interference is pretty much impossible. I can't think of any performance engine these days that isn't interference, regardless of drive system. Of course, the belt drive makes any failure more apparent because it's always blamed for the damage done.
I've seen chain drive failures with the same damage, but it's not as common. If I were to redesign the 928 engine, I would change a lot of things, including the cam drive. However, I don't think there is any way to get around interference and achieve the performance goals of the engine. You need to account for full valve opening and TDC for both intake and exhaust.
I've seen chain drive failures with the same damage, but it's not as common. If I were to redesign the 928 engine, I would change a lot of things, including the cam drive. However, I don't think there is any way to get around interference and achieve the performance goals of the engine. You need to account for full valve opening and TDC for both intake and exhaust.
#20
Rennlist Member
Well, MB used chain drives on their SOHC V8s, and they are not famous for longevity, based on my reading of mercedesshop.com forums.
AFAIK the Miata/MX5 1.6 and 1.8 engines are non-interference (our 1.6 certainly is!), and have a higher specific output than a 928 engine - 1.6L=72hp/L, 928 16V= 66hp/L (interference version). An S4 at 350hp is only 70hp/L.
jp 83 Euro S AT 54k
AFAIK the Miata/MX5 1.6 and 1.8 engines are non-interference (our 1.6 certainly is!), and have a higher specific output than a 928 engine - 1.6L=72hp/L, 928 16V= 66hp/L (interference version). An S4 at 350hp is only 70hp/L.
jp 83 Euro S AT 54k
#21
Bjbpe, is the belt really a problem (assuming replaced and tensioned periodically) or rather a perceived problem? You've had your car for 25 years and I assume if in that time your belt had broken you'd have said so in your original post. I think the problem is deferred maintenance rather than the design per se.
#22
Addict
Rennlist Member
Rennlist Member
I think the main reason for chain driven cams having a perceived greater reliability than belts is because they generally give you lots of increasing audible warning before it all goes pear shaped thus giving the owner chance to sort out that funny noise from the engine. In general first noise that you get from a TB is as the valves hit the pistons.
TB is the more elegant engineering solution - less weight, low rotational mass, no timing case covers to leak. And in an engine bay like the 928s I suspect that if you factored in the cost of a Chain and tensioner change and resealing the timing case every 100,000 miles I strongly suspect that the 928 would come out cheaper.
TB is the more elegant engineering solution - less weight, low rotational mass, no timing case covers to leak. And in an engine bay like the 928s I suspect that if you factored in the cost of a Chain and tensioner change and resealing the timing case every 100,000 miles I strongly suspect that the 928 would come out cheaper.
#24
Chronic Tool Dropper
Lifetime Rennlist
Member
Lifetime Rennlist
Member
#27
Addict
Rennlist Member
Rennlist Member
Join Date: Oct 2003
Location: Gone. On the Open Road
Posts: 16,548
Received 1,679 Likes
on
1,089 Posts
The main issue with the 928s timing belt system is the guy that last serviced it.
The factory install will last 15+ years and 40k+ miles. I know this is true for at least two examples on which I've been the first guy to service the system since Hans and Frans in Zuffenhausen. In both of those cases there was less wear on the components and fewer things "wrong" than the average belt system where I've been the Nth guy to touch it regardless of how many miles since the last job(*).
These the-timing-belt-system-is-crap threads come up ... what? ... about every year or so. I can write from first hand experience that a lot of things can be wrong with what the last guy did to the system and it will still keep valves from hitting pistons. As non-simple systems go, I'd say its actually quite robust and with a few exceptions (e.g. water pump bearing seizure and immediate failure due to just plain getting it totally wrong) gives thousands of miles of "warnings" that can be observed directly or with a simple side-covers-off inspection.
If you want to find fault with the system then I'd nominate "design for service" as the main fault. There are a number of fiddly bits and some tricks - nothing, though, that simple careful observation won't get you - to doing it properly.
(*) For example: I just did the third belt on a 24k-mile 928. All the gears had to be replaced because the belt had been just tight-enough to not trigger the warning system since the last service 8-ish-k miles ago. And there was no oil in the tensioner. And the tensioner gasket was torn. And the cam timing was off. (I had a factory-trained Porsche "technician" try to convince me once that the tensioner got its oil supply from the block...)
Last: comparisons of general automotive engineering to Apollo or flight-critical system engineering is like comparing apples and basket *****.
The factory install will last 15+ years and 40k+ miles. I know this is true for at least two examples on which I've been the first guy to service the system since Hans and Frans in Zuffenhausen. In both of those cases there was less wear on the components and fewer things "wrong" than the average belt system where I've been the Nth guy to touch it regardless of how many miles since the last job(*).
These the-timing-belt-system-is-crap threads come up ... what? ... about every year or so. I can write from first hand experience that a lot of things can be wrong with what the last guy did to the system and it will still keep valves from hitting pistons. As non-simple systems go, I'd say its actually quite robust and with a few exceptions (e.g. water pump bearing seizure and immediate failure due to just plain getting it totally wrong) gives thousands of miles of "warnings" that can be observed directly or with a simple side-covers-off inspection.
If you want to find fault with the system then I'd nominate "design for service" as the main fault. There are a number of fiddly bits and some tricks - nothing, though, that simple careful observation won't get you - to doing it properly.
(*) For example: I just did the third belt on a 24k-mile 928. All the gears had to be replaced because the belt had been just tight-enough to not trigger the warning system since the last service 8-ish-k miles ago. And there was no oil in the tensioner. And the tensioner gasket was torn. And the cam timing was off. (I had a factory-trained Porsche "technician" try to convince me once that the tensioner got its oil supply from the block...)
Last: comparisons of general automotive engineering to Apollo or flight-critical system engineering is like comparing apples and basket *****.
#28
Chronic Tool Dropper
Lifetime Rennlist
Member
Lifetime Rennlist
Member
Dave hits resonant chord here. The original belt started life running on new sprockets, over new rollers, in a dry environment. Next guy looks at it and the bearing$/roller$/$procket$/$seals, and decides that it "just needs the belt". Those other part$ look OK. Unless he's related to my great aunt, Claire Vouyante, it's wishful thinking that all those parts will somehow last -another- 40-60k and seven years without any signs of wear or age. And without any useful inspection method short of another teardown. So somehow the belt itself is maligned as the "problem", when it's really the victim of those other part$ that just weren't quite as good any more. Want your belts to last a really really long time and without worry? Pop in all the fiddly bits that it touches new. And the oil seals. And don't overheat the car.
Chains are noisy, they stretch, they need serious lubrication, and they take up room inside a dedicated timing cover. A good belt is better. Modern belts have Aramid/Kevlar reinforcing that doesn't complain about repeated flexing the way steel-reinforced belts used to. The belts stretch very slightly after initial run-in, and are stable for the rest of their service life. Chains get longer throughout their wear life, and a stretched chain will undoubtedly destroy all its drive sprockets as it grows. Think cam gears for the 928 are expensive? Replace all the sprockets on any early 12 cyl prancing horse car with originals. Every 25k miles.
Chains are noisy, they stretch, they need serious lubrication, and they take up room inside a dedicated timing cover. A good belt is better. Modern belts have Aramid/Kevlar reinforcing that doesn't complain about repeated flexing the way steel-reinforced belts used to. The belts stretch very slightly after initial run-in, and are stable for the rest of their service life. Chains get longer throughout their wear life, and a stretched chain will undoubtedly destroy all its drive sprockets as it grows. Think cam gears for the 928 are expensive? Replace all the sprockets on any early 12 cyl prancing horse car with originals. Every 25k miles.
#29
Captain Obvious
Super User
Super User
And a flywheel isn't engineered to be as light as possible and still do their job? A fully filled torque converter has a lot of my, just like a flywheel and acts like one too. Just think about it for a second.
#30
Burning Brakes
Join Date: Oct 2008
Location: Palo Alto, CA
Posts: 972
Likes: 0
Received 0 Likes
on
0 Posts
Kevlar has a negative coefficient of thermal expansion, so it contracts as it heats, which should make the belt tighter as the engine warms up (unrelated to initial break-in stretch, methinks).