View Poll Results: How much would you spend on updated clear TB covers?
$100
14
41.18%
$150
11
32.35%
$200
5
14.71%
may cost more than $200, but they would be worth it.
4
11.76%
Voters: 34. You may not vote on this poll
Clear Lexan Timing Belt Covers Offered- Buy?
08-31-2004, 11:14 AM
#31
Addict
Rennlist Member
Rennlist Member
Mounting the distributers securely on the 32v cars' TB covers would be a PITA? No? I'm leaning towards buying an experimental set of TB covers and experimenting witn drilling holes and using plugs/portholes. Anyone with extras for an 85S they'd like to sell relatively cheap for experimental purposes?
H2
pioneer4570@hotmail.com
H2
pioneer4570@hotmail.com
08-31-2004, 11:35 AM
#32
Rennlist Member
A few thoughts here.
1) What's the origonal material of the belt covers? I have assumed that it is ABS, but I'm not sure. It turns out that you can do ABS in clear just fine. Also you can do an ABS/ Polycarb mixture in clear that is even better. I think an injection mold for this would be $20K or more, but we might be able to vacume form the parts out of Polycarb on a $500 tool.
2) I like the idea of having a more visible belt. I generally leave the passenger side cover off for a 1000 miles or so after a T.B. job just to keep an eye on things. However this will not help in cases like Big Dave's where the failure is sudden and unexpected.
3) I think the problem with the existing tension light is that you are expecting really simple electronics to solve a pretty complex problem so you end up (as usual) with a bunch of false positives and a bunch of false negatives and people just ignoring the light, and failures without the light going off until too late.
4) We could build some more complex electronics to handle the problem. I have a small electronics design and production shop and I would be happy to design and produce the system. The trouble is that, if it is to really work, it needs to be tested and tuned under all modes of T.B. system failues. Not only would that be WAY to expensive for me, but I don't even think we know what all the failure modes are. Lots of unexplaned failures out there. Lots of failures in which the root cause has been guessed at but without really the methodology, time and money it takes to understand the root cause of a trended failure. And I think for this to really work we will need lots of sensors. For example...I think we might need a sensor to listen to the W.P. bearing and detect when it is making a nouse it should not.
5) Here's another idea. How an analog belt tension gauge? The reason I think this might work is the following: In my experience, even with DSPs and FPGA and what not, it is almost impossible to build a machine that can analyize things as well as your eye. I don't think anybody really knows what 928 belt tension looks like and how it varies with load, temp, etc..... But I bet if we had a gauge on the dash we would soon get used to what it looks like and be able to spot anything abnormal.
6) As and engineer and an a product developer I would say the real reason why we are having this problem is the Porsche just did not not do the work required to figure out the problems and solve them. When you have designed a system and it is out in the field and mostly working, but occasionally failing it is very easy for the problems not to get solved. In the first place engineering often never hears about them. And then, since all the engineers are all off working on new projects (which are more interesting and profitable) when something does get sent back to engineering as a possible problem usually engineering pushes back and blames the customer or the service or the parts vendors. Or throws some fixes at it without completely understanding the problem. My sense is that this is what happened to our T.B. system.
7) Even if there are still problems in the T.B. system, and Porsche is not at all interested in doing any more work on car, this does not mean the problems cannot be solved. The biggest step in solving a problem like this is completely understanding the failures. I don't know if we have enough people (or T.B. failures) here on rennlist to do it, but here are some ideas of things we could do: a) build a T.B. failure data base. We come up with guidelines for analyzing failures and every time anybody has a failure we really try to follow the guidelines and get data about what the failure was into the data base. b) identify members that have access to resources to really understand failed components...so if a water pump bearing fails we know not only how many miles were on it, but who reman'd it, what brand of bearing was in it, what brand of seal, what the seal material was what the condition of the seal material etc. etc. c) once the root causes of the failures are completly understood solutions can be found. This is really the easy part. For example: there are lots of bearing seal materials out there and people who know all about selecting the best ones to solve specific problems.
Oops...I guess I'm ranting again. But I would really love for this problem to be solved in a satisfactory manner.
1) What's the origonal material of the belt covers? I have assumed that it is ABS, but I'm not sure. It turns out that you can do ABS in clear just fine. Also you can do an ABS/ Polycarb mixture in clear that is even better. I think an injection mold for this would be $20K or more, but we might be able to vacume form the parts out of Polycarb on a $500 tool.
2) I like the idea of having a more visible belt. I generally leave the passenger side cover off for a 1000 miles or so after a T.B. job just to keep an eye on things. However this will not help in cases like Big Dave's where the failure is sudden and unexpected.
3) I think the problem with the existing tension light is that you are expecting really simple electronics to solve a pretty complex problem so you end up (as usual) with a bunch of false positives and a bunch of false negatives and people just ignoring the light, and failures without the light going off until too late.
4) We could build some more complex electronics to handle the problem. I have a small electronics design and production shop and I would be happy to design and produce the system. The trouble is that, if it is to really work, it needs to be tested and tuned under all modes of T.B. system failues. Not only would that be WAY to expensive for me, but I don't even think we know what all the failure modes are. Lots of unexplaned failures out there. Lots of failures in which the root cause has been guessed at but without really the methodology, time and money it takes to understand the root cause of a trended failure. And I think for this to really work we will need lots of sensors. For example...I think we might need a sensor to listen to the W.P. bearing and detect when it is making a nouse it should not.
5) Here's another idea. How an analog belt tension gauge? The reason I think this might work is the following: In my experience, even with DSPs and FPGA and what not, it is almost impossible to build a machine that can analyize things as well as your eye. I don't think anybody really knows what 928 belt tension looks like and how it varies with load, temp, etc..... But I bet if we had a gauge on the dash we would soon get used to what it looks like and be able to spot anything abnormal.
6) As and engineer and an a product developer I would say the real reason why we are having this problem is the Porsche just did not not do the work required to figure out the problems and solve them. When you have designed a system and it is out in the field and mostly working, but occasionally failing it is very easy for the problems not to get solved. In the first place engineering often never hears about them. And then, since all the engineers are all off working on new projects (which are more interesting and profitable) when something does get sent back to engineering as a possible problem usually engineering pushes back and blames the customer or the service or the parts vendors. Or throws some fixes at it without completely understanding the problem. My sense is that this is what happened to our T.B. system.
7) Even if there are still problems in the T.B. system, and Porsche is not at all interested in doing any more work on car, this does not mean the problems cannot be solved. The biggest step in solving a problem like this is completely understanding the failures. I don't know if we have enough people (or T.B. failures) here on rennlist to do it, but here are some ideas of things we could do: a) build a T.B. failure data base. We come up with guidelines for analyzing failures and every time anybody has a failure we really try to follow the guidelines and get data about what the failure was into the data base. b) identify members that have access to resources to really understand failed components...so if a water pump bearing fails we know not only how many miles were on it, but who reman'd it, what brand of bearing was in it, what brand of seal, what the seal material was what the condition of the seal material etc. etc. c) once the root causes of the failures are completly understood solutions can be found. This is really the easy part. For example: there are lots of bearing seal materials out there and people who know all about selecting the best ones to solve specific problems.
Oops...I guess I'm ranting again. But I would really love for this problem to be solved in a satisfactory manner.
08-31-2004, 01:23 PM
#33
Burning Brakes
Thread Starter
Join Date: Jul 2003
Location: IL
Posts: 1,166
Likes: 0
Received 0 Likes
on
0 Posts
I have been following the timing belt failures and I am going to start a poll on the subject for further research.
I beleive the biggest problems associated to bent valves are:
water pump failure
tensioner not working or tensioned correctly
roller failure (siezure or wobbling)
old or over tightened belt snapped
I think the clear covers would give us a "heads up" on most of these things before dooms day happens.
Karl,
The vacuum form idea I had too, the forms are rather inexpensive, relatively speaking.
I beleive the biggest problems associated to bent valves are:
water pump failure
tensioner not working or tensioned correctly
roller failure (siezure or wobbling)
old or over tightened belt snapped
I think the clear covers would give us a "heads up" on most of these things before dooms day happens.
Karl,
The vacuum form idea I had too, the forms are rather inexpensive, relatively speaking.
08-31-2004, 01:34 PM
#34
Bumbling Neophyte
Rennlist Member
Rennlist Member
Join Date: Apr 2003
Location: IAX, FL
Posts: 269
Likes: 0
Received 0 Likes
on
0 Posts
I really like the vacum formed polycarb idea. The tool most likely could be made from wood with a layer or two of fiberglass over it. It's not to deep, has nice round corners. Think about what the air between the cover and the engine would look like if it was solid and on a large flat piece wood.
08-31-2004, 03:56 PM
#35
Shameful Thread Killer
Rennlist Member
Rennlist Member
Great idea you have Tony. As I am just about to start a TB/WP job I will incorporate the trip wire idea in my job. I plan to make a solid aluminum plate just above the right side pulley with a channel cut out for the belt width. Across the channel I will use two strands of approx 36ga wire, connected in parallel. Using two strands should minimize a false positive, as we would need both to break from a fray to set off the alarm.
Excellent analysis by Karl. I do some root cause and FA at work , and if we do get enough data points to make a valid case for sensors, then that can incorporate or assist the trip wire concept. From a partial background with vibration analysis, I would think a circuit with Fourier transform, on a pezio sensor contacting the belt tension arm would be ideal, but still in the formative stages.
Gathering the data will be complicated, as most belts removed will show no impending catastrophic information. Once a catastrophic failure has occured, it will be hard to determine the failure mode right up until the event. That's what we need though, is a preventative diagnostic to avert the event. Using similar, but clearly not exact criteria, if we could expand the failure mode to include ANY type of automotive belt, I think that would help in making a deterministic decision about sensors. I realize that the timing belts are more durable than a V belt, but I have seen an impending failure of a belt show strand fraying, and is why I will incoporate the trip wire concept.
Doc
Excellent analysis by Karl. I do some root cause and FA at work , and if we do get enough data points to make a valid case for sensors, then that can incorporate or assist the trip wire concept. From a partial background with vibration analysis, I would think a circuit with Fourier transform, on a pezio sensor contacting the belt tension arm would be ideal, but still in the formative stages.
Gathering the data will be complicated, as most belts removed will show no impending catastrophic information. Once a catastrophic failure has occured, it will be hard to determine the failure mode right up until the event. That's what we need though, is a preventative diagnostic to avert the event. Using similar, but clearly not exact criteria, if we could expand the failure mode to include ANY type of automotive belt, I think that would help in making a deterministic decision about sensors. I realize that the timing belts are more durable than a V belt, but I have seen an impending failure of a belt show strand fraying, and is why I will incoporate the trip wire concept.
Doc
08-31-2004, 04:32 PM
#36
Addict
Rennlist Member
Rennlist
Site Sponsor
Rennlist Member
Rennlist
Site Sponsor
Quite often what happens is teeth simply shear off the belt ...first one then another until enough fatigue that one of the gears slips my guess the crank which has all the load and is smaller diameter but the belt stays intact.
08-31-2004, 04:51 PM
#37
Burning Brakes
Thread Starter
Join Date: Jul 2003
Location: IL
Posts: 1,166
Likes: 0
Received 0 Likes
on
0 Posts
Can someone post a picture of the timing belt covers with dimensions? I have a source that could quote it but I need some pictures to work from. It could take a couple of weeks to quote something like this.
08-31-2004, 05:32 PM
#38
928 Collector
Rennlist Member
Rennlist Member
Sounds like MK's cam cogs 
Originally Posted by Jim bailey - 928 International
Quite often what happens is teeth simply shear off the belt ...first one then another until enough fatigue that one of the gears slips my guess the crank which has all the load and is smaller diameter but the belt stays intact.
08-31-2004, 06:52 PM
#40
Burning Brakes
Join Date: Sep 2003
Location: North Central Florida (Gainesville) USA
Posts: 957
Likes: 0
Received 0 Likes
on
0 Posts
I like Tony's idea of an improved cambelt warning system and Thaddeus' idea of partially clear covers.
Kudos to worthwhile aftermarket improvements for our Sharks!!!
Tim
Kudos to worthwhile aftermarket improvements for our Sharks!!!
Tim
08-31-2004, 07:31 PM
#41
Instructor
Join Date: Jan 2002
Posts: 179
Likes: 0
Received 0 Likes
on
0 Posts
[QUOTE=Tony]Good idea and all but, it would be a bit#$ to keep clean.
Only glitch is that there is an appx 3 minute buffer in the stock warning before it is actually displayed. So hopefully your belt has more than 3 minutes of life left before it completely goes.
[QUOTE]
Do I understand this correctly that there is a 3 minute time delay between the time the sensor sees a fault and the time it is displayed? Seems pretty long for such an important dianostic.
Only glitch is that there is an appx 3 minute buffer in the stock warning before it is actually displayed. So hopefully your belt has more than 3 minutes of life left before it completely goes.
[QUOTE]
Do I understand this correctly that there is a 3 minute time delay between the time the sensor sees a fault and the time it is displayed? Seems pretty long for such an important dianostic.
08-31-2004, 08:27 PM
#42
Deer Slayer
Lifetime Rennlist
Member
Lifetime Rennlist
Member
I spoke with Jim at 928 Intl because I was thinking about putting little windows in a cover as a test. (They were too spendy to justify the expense for an experiment). He mentioned the fact the covers are different, year to year... so based on that, I conclude if someone comes up with a plastic one, they'd have to have different models. Having to make several different molds for different model years seems to move this idea away from viability some...
08-31-2004, 08:48 PM
#43
Addict
Lifetime Rennlist
Member
Lifetime Rennlist
Member
[QUOTE=aaron3][QUOTE=Tony]Good idea and all but, it would be a bit#$ to keep clean.
Only glitch is that there is an appx 3 minute buffer in the stock warning before it is actually displayed. So hopefully your belt has more than 3 minutes of life left before it completely goes.
Do I understand this correctly that there is a 3 minute time delay between the time the sensor sees a fault and the time it is displayed? Seems pretty long for such an important dianostic.
The way i understand it, yes. I might be wrong though. I do know that if you forget to hook the wiring up to it, the light will come on in appx 3 minutes
Only glitch is that there is an appx 3 minute buffer in the stock warning before it is actually displayed. So hopefully your belt has more than 3 minutes of life left before it completely goes.
Do I understand this correctly that there is a 3 minute time delay between the time the sensor sees a fault and the time it is displayed? Seems pretty long for such an important dianostic.
The way i understand it, yes. I might be wrong though. I do know that if you forget to hook the wiring up to it, the light will come on in appx 3 minutes
09-01-2004, 08:06 AM
#45
Burning Brakes
Thread Starter
Join Date: Jul 2003
Location: IL
Posts: 1,166
Likes: 0
Received 0 Likes
on
0 Posts
According to Devek's catalog, the 85-86 TB cover units would be the same, and then the s4's would be the same even all the way to the gts. Its possible the cam gear covers are the same between those models, but I am still looking into it.