Let me start this note by saying that I absolutely love my ’87 S4 as much today as when I bought it new 25 years ago. It’s a great car but it certainly is nowhere near perfect as many owners who read this would probably agree.

As you will note, the topic for this comment is the timing belt, a subject that I would guess is as much discussed in these columns as any other.

As an aerospace engineer of some fifty plus years going back to the Apollo program, several factors were of utmost importance with Apollo. One was obviously reliability and this correlated to a significant level with quality control. As a matter of fact, every item including nuts and bolts (not just a sample) was inspected to assure that there was no possibility that an off-design part got into the mix. Another aspect of reliability was that the part or parts had to do what they were designed to do well beyond the intended product life. Also, there were extensive studies to assure that failures, should they occur, were isolated in such a way that the affect of the failure was limited in its impact on other parts of the whole. Obviously, such limits were difficult to achieve but the results speak for themselves. Apollo 13 is an example where someone missed something (a shorted wire) which resulted in the problems that occurred. That wasn’t supposed to happen.

So, now, on to the 928. I have great difficulty understanding what the Porsche engineers were thinking of by utilizing a steel-reinforced fabric belt to drive a set of camshafts that could, with relatively minor imperfection cause the pistons to crash into the valves. Of course this was supposed to be a high-efficiency, high-performance engine but when one considers the fact that the belt has to be near perfect while, at the same time, being dependent on the proper operation of a water pump, a tension device and God knows what else, it sure does not compare with what I would expect from a car that one paid close to six figures for in today’s money.

Of course the car is 25 years old and one can argue that I should not expect so much from a design that is that old. However, the argument that I am making was just as valid about the reliability of the design even 25 years ago.

I also own a relatively old (2002) Cadillac with the “Northstar” engine. It has a layout similar to the 928 engine but it is not an interference engine (somewhat lower efficiency) and the cams were driven by a highly reliable chain. I’ve got a lot of miles on that engine and have never had to service the chain.

So, here’s my question: What was the reason that Porsche went to such a relatively inferior design (at least from my outlook). I’d love to hear from folks who would like to defend the timing belt.

One more comment and this is about the reliability of other parts of the car. I do read a lot of the Q & A made by others and I am amazed at the number of times I have said to myself “Yep, been there, done that!” This applies to things like tail light burnouts, poor heater/AC controls, heater motor wearouts and on and on. I am surprised that a car that was so expensive had so many minor bugs, these on cars with relatively low mileage.

Thanks to those who would like to respond. Like I said, I love the car but it sure could have been better designed.

 

i am just amazed at how the teeth on the belt mesh into the slots in the cam gears many thousands of times a minute w/o jamming.

 

I think it comes down to the perennial calculation of best reliability at best cost.

The engineers at Porsche were tasked to design something from a clean sheet of paper that was ground breaking but to do it at a price point that would generate sales and some profit. To wit, while many of the fundamental design elements were new or rarely seen (all aluminum block without iron cylinder sleeves, "hidden" bumper elements, Weissach Axle) much of the hardware to execute the design was off-the-shelf parts and technology (blower motors, switches, etc.) that was never expected to last 20 years plus. Combine this with the fact that the 928 never became the mainstay of the Porsche line-up and many of the design choices that the original team would have recognized as being weak once the car hit the market and got significant real world use never got designed out (or at least only the major issues see previous comment re: not becoming the flagship model) and you have the current state. Most of us are constantly fighting entropy with these systems to the point that it seems like the reliability is kind of low.

I always go back to the fact that an estimated 65% of all the Porsche's EVER BUILT are still on the road today is a testament to just how good a job those engineers did calculating the balance between cost, reliability and making a profit.

My $0.02

Mike

 

It cntinues to amaze me how inovative and generally well designed the 928 is. Considering how small the company was at the time, and also the size of the racing program they were suporting at the same time.... considering that even the last 928s are now 18 years old, the cars are pretty reliable compared with their contemporaries ? The main problem is lack of proper maintainance once the cars left the dealer network.

Yes, some of the bought in items were of limited quality, but they have often lasted in excess of 100k miles.

 

Making a car isn't rocket science...

In rocket science, a lot of things can go wrong (as you would seem to know better than me) with disastrous consequences. Space ship gone, crew gone, billions of dollars gone. In a 928, when the belt breaks or the waterpump lets go, heads need to be replaced or fixed. Car is still there, driver lives, a few thousand dollars in cost.

 

The timing belt was new and cool for engineers when the engine was designed in the early 70's.

Saves space, quieter, fewer oil leak points, etc.

Compared to modern belts, the 928 belt is too narrow, IMO, but it works.


Fortunately, there is a modern alternative to many of the engineering missteps of the 928.

 

Just as a data point - the Northstar is a great engine - but it is an interference engine.

 

Non-interference engine belt failure = catastrophic stoppage, no serious engine damage, replace belt(and possibly rollers, gears, idler).

Interference engine belt failure = catastrophic stoppage, serious engine damage, replace above + valves, possibly guides).

For about 8 years, the 928 was a non-interference engine(depending on country code). The engine produced a certain amount of power, at a certain amount of fuel consumption, and a certain BSFC. The numbers were - in a word,, abysmal for the size of the engine. In order to achieve the economy consumption, and maintain or improve the power of the engine, something had to give. In this case, one of the things that had to give was non-interference engine design. The catastrophic failure was the same in either case, but the repair cost grew substantially.

Of course, the failure rate of well maintained 928s with interference engines was so low as to not be calculable. Then, a bad thing happened. Late model 928s started getting cheap to buy. They were found with deferred maintenance, and the owners didn't invest in the costly belt and water pump change as required by the factory. So - of course, things began to fail. Water pumps, gears, tensioners, belts, and all those things that Porsche said would last for 'X' miles ran for X(n) miles where n is some multiple of what Porsche said the bits are good for.

Then we blame - the car of course! No - let's blame the designer for setting all those cars up to fail! Yes, that's it, the engineer is a fault for building a 5 litre 330HP engine which can get amazing BSFC. It's not the nut behind the wheel, it's not the cheap *** owner, it's the engineers fault. Yup. those dirty bastards. A little rubber time bomb in each car. MMMuuuuuuuuaaaahhhhaahahahhahhhaahhaaaaa!!!!!!

<edit: I never sent anything to the moon. I'm just a lowly EE, rather pedestrian at that. However, everything you type here, everything you store off your local machine, every **** picture you download, and every bit that appears in front of you today travels on my network. Without the products I design, maintain, and install - your internet would basically be the same as a 1983 BBS with a 1200 Baud modem.>

 

To answer some of your questions the timing belt was utilized to reduce noise that a typical chain set up creates. The water pump was the latest technology for the time and lasts as long as any water pump on todays cars. Porsche continued to improve the systems as technology and money allowed. Don't forget it was never a money maker and came about because they thought, at the time, that US law was going to make the 911 obsolete. The feared laws never became a reality and the hard core 911 fans prevailed. It's still a fantastic car capable of going against the latest technology.

 

Servicing timing belt on these cars is a weekend job, beverage in hand.
Don't fight it.
Become "one" with it.
Do it often and do it correctly.

 

I imagine that with the length required the mass of the chain plus the added mass of steel cam cogs would be a big factor. That's a lot more mass to accelerate and would affect throttle response negatively. I'm talking about the rotating mass of the engine itself here.

 

+1

 

I cannot picture it, all steel gears, in place of what is now Aluminum. No back drive for rollers, so those have to be geared. Chain, about 5 pounds itself, then, having to lubricate it all?, Quietly?.Even a chain that long would need the tensioner. I think weight was the priority. Overall, for the car, and the engine was only part of the car, just like on Apollo.

 

Then you reduce the mass of the flywheel and all back to normal. Timing belt was a terrible choice by Porsche. They could have used two chains, one for each bank.

 

On an auto? You wouldn't have much flywheel left.