Kempf setting / @ post #96, where for Gates Racing Belt?
10-08-2010, 01:18 AM
#46
Addict
Rennlist Member
Rennlist Member
I would have thought that if a cam nose has snapped a belt tension warning would probably be the least painful symptom.
Could be just the belt wasn't bedded down properly - I normally do about 10 full engine turns checking the tension each time.
Other candidates are water pump bearing failure or tensioner pulley bearing failure or tensioner arm pivot bolt.
10-08-2010, 02:03 PM
#47
Burning Brakes
I let the car sit for 5 minutes running in the parking lot, no TB light , I then drove it for 10 minutes before the light came on. Sorry if I confused anyone on the specifics.
10-08-2010, 02:07 PM
#48
Burning Brakes
One way or another the 1-4 side cover has to come off.
I would have thought that if a cam nose has snapped a belt tension warning would probably be the least painful symptom.
Could be just the belt wasn't bedded down properly - I normally do about 10 full engine turns checking the tension each time.
Other candidates are water pump bearing failure or tensioner pulley bearing failure or tensioner arm pivot bolt.
I would have thought that if a cam nose has snapped a belt tension warning would probably be the least painful symptom.
Could be just the belt wasn't bedded down properly - I normally do about 10 full engine turns checking the tension each time.
Other candidates are water pump bearing failure or tensioner pulley bearing failure or tensioner arm pivot bolt.
10-08-2010, 03:06 PM
#49
Rennlist Member
I 'ld forgotten about that, and I was using a Loos gauge on Saturday.
But like the 9201 and the kempf tool they all bend the tensioned "thing" they are measuring by virtue of the deflection they impose. This doesn't matter as you are always comparing like with like ( the same object bending through a fixed angle) so the error from bending is always constant and fixed.
That also why you can't compare the rig settings on a boat with rod rigging to an identical boat with dieform wire rigging.
But like the 9201 and the kempf tool they all bend the tensioned "thing" they are measuring by virtue of the deflection they impose. This doesn't matter as you are always comparing like with like ( the same object bending through a fixed angle) so the error from bending is always constant and fixed.
That also why you can't compare the rig settings on a boat with rod rigging to an identical boat with dieform wire rigging.
10-08-2010, 04:59 PM
#50
Rennlist Member
This is why I wont touch a new belt, because there is no design limits for tension known. we are going to have to figure it out. someone will groove the cam gears and that will mean too tight, and someone will lose a motor. (too loose). 
the stock belt is fine, and has been used in the most demanding racing conditions. if maintained, there should be no isses.
I like the fact that the belt is more durable, but because its a spring mass system, audio frequency and twist test and even the porsche tool will be way off with the new stiffer belt.
the stock belt is fine, and has been used in the most demanding racing conditions. if maintained, there should be no isses.
I like the fact that the belt is more durable, but because its a spring mass system, audio frequency and twist test and even the porsche tool will be way off with the new stiffer belt.
The 9201 tool has a calibration tool that is used to check the values. It is never going to be very far off, if checked and calibrated. I personally check mine, every time I use it.
I agree on the audio frequency. I'd be thrilled if someone has a audio frequency value, for Roger's Racing belt. That would solve everything, for me. However, as I stated earlier, this belt is going to have a different "tune" than the factory belts.
I agree on the audio frequency. I'd be thrilled if someone has a audio frequency value, for Roger's Racing belt. That would solve everything, for me. However, as I stated earlier, this belt is going to have a different "tune" than the factory belts.
10-08-2010, 05:00 PM
#51
Nordschleife Master
When I made electric basses (back in the day) I got interested in the physics of a vibrating string. Google, dig, and it becomes clear why the preferred method of measuring belt tension these days - at least by mfrs I've looked at - is by measuring its vibrating frequency. (Really interesting stuff. Also used by many to tighten, or at least check tension of, spokes in bicycle wheels.)
Back to the 928 stuff. If one can measure the belt tension with an infrared sensor, then maybe we can get some ball-park indications of things maybe being FUBAR with it. I have it from an engineering authority -- google search of enthusiast discussion groups ;-) -- that with timing belt systems the slack-side tension should be something like 30% of the pull-side tension. Maybe one could measure the belt tension from both sides of a running and fully-warmed up engine. I don't know if this is physically possible. But if it is, then maybe this method could give on some indication of whether the tension is in the ball park in unusual combinations such as GRB and Porkensioner.
The science fiction department now points out that these two sensors could be made a permanent install and could be used with a micro-controller to control the toothed belt warning light. If someone does that, they might even get a real engineer involved to compute the actual desired tension ratio...
10-08-2010, 05:03 PM
#52
Rennlist Member
My bicycle measures the chain tension using an infrared sensor. A second, magnetic sensor measures the chain speed. Then the trip computer computes my power output as chain tension times the chain speed. Cycling at 0.5 kW for longer than a minute is nearly impossible for me. As is operating the the trip computer, it's so complex now...
Back to the 928 stuff. If one can measure the belt tension with an infrared sensor, then maybe we can get some ball-park indications of things maybe being FUBAR with it. I have it from an engineering authority -- google search of enthusiast discussion groups ;-) -- that with timing belt systems the slack-side tension should be something like 30% of the pull-side tension. Maybe one could measure the belt tension from both sides of a running and fully-warmed up engine. I don't know if this is physically possible. But if it is, then maybe this method could give on some indication of whether the tension is in the ball park in unusual combinations such as GRB and Porkensioner.
The science fiction department now points out that these two sensors could be made a permanent install and could be used with a micro-controller to control the toothed belt warning light. If someone does that, they might even get a real engineer involved to compute the actual desired tension ratio...
Back to the 928 stuff. If one can measure the belt tension with an infrared sensor, then maybe we can get some ball-park indications of things maybe being FUBAR with it. I have it from an engineering authority -- google search of enthusiast discussion groups ;-) -- that with timing belt systems the slack-side tension should be something like 30% of the pull-side tension. Maybe one could measure the belt tension from both sides of a running and fully-warmed up engine. I don't know if this is physically possible. But if it is, then maybe this method could give on some indication of whether the tension is in the ball park in unusual combinations such as GRB and Porkensioner.
The science fiction department now points out that these two sensors could be made a permanent install and could be used with a micro-controller to control the toothed belt warning light. If someone does that, they might even get a real engineer involved to compute the actual desired tension ratio...
Very good! Excellent way to start the weekend, with, of course, my usual Pretty Big Reward...Been totally enjoying your saga, BTW...
10-08-2010, 09:33 PM
#53
Nordschleife Master
Here's something copy-pasted after "a couple:"
Source URL: http://machinedesign.com/article/ten...lt-drives-0607
Published on Machine Design (http://machinedesign.com)
Tension in timing-belt drives
Timing-belt drives transmit torque and motion from a driving to a driven pulley or force to a linear actuator. They may also convey a load placed on the belt surface.
A drive under load develops a difference in belt tension between the entering (tight) and leaving (slack) sides of the driver pulley. This effective tension, Te, is the force transmitted from the driver pulley to the belt and to the driven pulley or load:
where T1 and T2 = tight and slackside tensions. The effective tension or working force generated at the driver pulley overcomes the driven pulley's resistance to motion. The driving torque, M1, is related to the driven torque (load), M2, by:
where d1 = pitch diameter of the driver pulley, P2 = power required at the driven pulley, 1 and 2 = angular speeds of the driver and driven pulleys of pitch diameters d1 and d2, respectively, and = efficiency (typically about 0.94 to 0.96).
Shaft forces
A force equilibrium at the driver or driven pulley relates tight and slack-side tensions and the shaft reaction forces Fs1 or Fs2. In powertransmission drives, forces on both shafts are equal in magnitude:
where theta1 = belt wrap angle around the driver pulley.
Belt pretension
Belt pretension (initial tension), Ti, is the tension set by an adjustable idler pulley. Pretension prevents belt slack-side sagging and ensures proper tooth meshing. In most cases, timing belts perform best when the magnitude of slackside tension is about 10 to 30% that of the effective tension.
Although generally not recommended, belt drives can work without an adjustment mechanism. This is possible because, after initial tensioning and straightening, belts tend not to elongate or creep. Overall belt length remains constant during operation regardless of loading conditions, provided belt sag and some other minor influences are neglected. However, reaction forces vary under load. And slack and tight side tensions not only depend on load and pretension, but on belt elasticity and structure stiffness as well.
A constant slack side tensioner is a better way to control belt tension in power-transmission drives and in some conveyors. Here, an adjustable, floating idler riding on the belt slack side compensates for a lengthening tight side. Slackside tension is maintained by an external tensioning force, Fe:
where theta e, = belt wrap angle about the idler pulley. This, and the expression for effective tension, combine to give tight-side tension, T1, and shaft reactions, Fs1 and Fs2.
Drives with constant, slack-side tension add an external load to the system and cannot be characterized by force analysis alone. Calculating tight and slack side tensions and shaft forces (two equations, three unknowns) for a given torque or effective tension, requires an additional relationship: belt elongation.
Total belt elongation equals that from pretension, neglecting belt sag and some factors that contribute little to elongation, such as belt bending resistance and radial shifting of the belt pitch line. Pulleys, shafts and mounting structures are assumed infinitely rigid for analysis purposes. Then, elongation is expressed by a geometric compatibility of deformation:
where DL11 and DL22 = tight and slack-side elongations due to T1 and T2, DLme = total elongation of the belt portion meshing with the driver (and driven) pulleys, and DL1i, DL2i, and DLmi = deformations from belt pretension, Ti.
In most cases, belt deformations at the pulleys during pretensioning and in operation are about equal (DLme ΔDLmi), so:
Tensile tests of properly loaded timing belts show stress is proportional to strain. Defining the stiffness of a unit long and wide belt as specific stiffness, csp, the belt stiffness coefficients on tight and slack sides, k1 and k2, are expressed by:
where L1 and L2 = unstretched lengths of the tight and slack sides, respectively, and b = belt width. Note that these expressions are similar to that for axial stiffness of a bar. Hooke's Law says that elongation equals tension divided by a stiffness coefficient provided that tension is constant over belt length:
Combining expressions for the stiffness coefficients with those for tight and slack side tensions gives:
where L = total belt length. These equations can be used to determine the shaft reactions, Fs1 and Fs2. In practice, a belt drive can be designed such that the desired slack-side tension equals about 10 to 30% of effective tension. This gives proper tooth meshing during operation. Then the expression for slack-side tension, T2, is used to calculate the correct pretension. As mentioned previously, the above relations apply only when these tensions are constant over belt length. In all other cases, elongations must be calculated according to the actual tension distribution.
Information for this article comes from Mectrol Corp., Salem, N.H., www.mectrol.com [1]
Links:
[1] http://www.mectrol.com
10-08-2010, 11:10 PM
#54
Drifting
As for the Gates Racing belt tension adjustment, no compensation is needed. Adjust tension as with any other timing belt manufacturer. Any differences in the belt stiffness of the various manufacturers, produce negligible errors in resultant belt tension.
The proper belt tension is specified by Porsche in the WSM using special tool 9201. The Kempf tool provides the same measurement as the 9201 with added simplicity and improved measurement repeatability.
The proper belt tension is specified by Porsche in the WSM using special tool 9201. The Kempf tool provides the same measurement as the 9201 with added simplicity and improved measurement repeatability.
10-08-2010, 11:26 PM
#55
Team Owner
Borland have you had the new GRbelt in your hand?
Or installed one of these new belts??
The new belt is quite a bit stiffer than the regular belt , I have yet to get my hands on one and there is concern about how to tension and what tension to use .
I am interested to find out more
Or installed one of these new belts??
The new belt is quite a bit stiffer than the regular belt , I have yet to get my hands on one and there is concern about how to tension and what tension to use .
I am interested to find out more
10-09-2010, 12:09 AM
#56
Drifting
Yes, I purchased one from Roger recently, but haven't installed it yet.
Yes, the Gates Racing belt seems stiffer, but that will not significantly affect the tension adjustment setting. The methods used for tension measurement have little to do with the stiffness, and more to do with the tension of the belt.
Yes, the Gates Racing belt seems stiffer, but that will not significantly affect the tension adjustment setting. The methods used for tension measurement have little to do with the stiffness, and more to do with the tension of the belt.
10-09-2010, 12:31 AM
#57
Team Owner
please post pictures and your thoughts when you get the new belt installed,
i think that there is lots of info to find out
i think that there is lots of info to find out
10-09-2010, 02:24 AM
#58
Former Sponsor
Originally Posted by ptuomov;
When my current engine blows up, the next one in the garage has both Porkensioner and Roger's racing belt. I agree that there's just one way to find out about that combo. Once I cross that bridge I'll report back.
I haven't personally set the cam timing on a single 928 engine yet, so the following goes with a huge caveat:
I though the different timing between banks was for thermal expansion of the block and not stretch of the belt. "My" logic was that aluminum block is going to expand a lot and that had to be accounted for with different timing between banks.
Belt stretching and the block expanding are basically the opposite effects. I am thinking this thru in my head now and I think that if it were the belt stretching, one would advance the passenger side bank relative to the driver side. To compensate for the thermal expansion of the block, one retards the passenger side bank relative to the driver side. Does that make sense? I am sure I got the directions mixed up in my head with 50% probability.
I haven't personally set the cam timing on a single 928 engine yet, so the following goes with a huge caveat:
I though the different timing between banks was for thermal expansion of the block and not stretch of the belt. "My" logic was that aluminum block is going to expand a lot and that had to be accounted for with different timing between banks.
Belt stretching and the block expanding are basically the opposite effects. I am thinking this thru in my head now and I think that if it were the belt stretching, one would advance the passenger side bank relative to the driver side. To compensate for the thermal expansion of the block, one retards the passenger side bank relative to the driver side. Does that make sense? I am sure I got the directions mixed up in my head with 50% probability.
The difference is all about the belt. The 5-8 side is set more advanced (opens sooner) than the 1-4 side. When the crankshaft pulls on the belt, the majority of the stretch takes place between the crank and the first camshaft that is turned, which means this camshaft would be more retarted (if it wasn't set up advanced) than the 2nd camshaft, which is, essentially, pulled by the first camshaft, not the crank.
Make sense?
10-09-2010, 02:38 AM
#59
Former Sponsor
Originally Posted by mark kibort;
This is why I wont touch a new belt, because there is no design limits for tension known. we are going to have to figure it out. someone will groove the cam gears and that will mean too tight, and someone will lose a motor. (too loose).
the stock belt is fine, and has been used in the most demanding racing conditions. if maintained, there should be no isses.
I like the fact that the belt is more durable, but because its a spring mass system, audio frequency and twist test and even the porsche tool will be way off with the new stiffer belt.
the stock belt is fine, and has been used in the most demanding racing conditions. if maintained, there should be no isses.
I like the fact that the belt is more durable, but because its a spring mass system, audio frequency and twist test and even the porsche tool will be way off with the new stiffer belt.
Keep in mind that Roger's new racing belt might work really well and solve some major issues. Quicker reving engines are going to (by definition) have more belt stretch than engines that pulls slower on the belt. Controlling this better will naturally result in a better performing engine.
I'm not resistant to change...it just has make some sense or have some practical potential improvement, before I jump onboard.
10-09-2010, 02:45 AM
#60
Former Sponsor
Originally Posted by borland;
Yes, I purchased one from Roger recently, but haven't installed it yet.
Yes, the Gates Racing belt seems stiffer, but that will not significantly affect the tension adjustment setting. The methods used for tension measurement have little to do with the stiffness, and more to do with the tension of the belt.
Yes, the Gates Racing belt seems stiffer, but that will not significantly affect the tension adjustment setting. The methods used for tension measurement have little to do with the stiffness, and more to do with the tension of the belt.