The Porsche manual says:
This, at least for my Staeger, is absolute bunk. After fiddling around for what seemed like hours with no success, I stuck a borescope in there. It's a 1.5mm Allen set-screw. Adjusts nicely once you know.
Maybe some of them are different? I don't know. But be forewarned....
Calibrating the 9201 Staeger belt tension gauge
-
Tom
- Site Admin
- Posts: 8922
- Joined: Fri Jun 25, 2021 2:04 pm
- Location: Silicon Valley, CA
- Has thanked: 932 times
- Been thanked: 3990 times
- Contact:
I have both kings. The older one has the straight slot set screw and the newer one has a hex key grub screw. I believe the original versions back in the 80's had the straight slot set screw, and so that's what they referenced when they wrote the manual back in the day.
I've done a lot of calibration testing as I develop the digital version, and can confirm there quite a bit of false precision (placebo precision?) built into the tool and its specs. I have three or four calibration bars and those differ by as much as .3 between the gauges. If you stack calibration tolerances, calibration bar variations, tool placement/use, and factory spec ranges, you could easily end up with a belt that is in spec with one tool and well out of spec on the other. You can even get slightly different readings even on the same calibration bar by just placing it a little deeper or shallower on the posts for example.
I'm confident that's why people get away with the twist method or water pump slip method. It's not that those methods reliably get the car within the factor tension specs (they don't) but rather that the motor can tolerate a much wider tension range than the placebo precision of the tool suggests.
I suspect, however, that the factory specs were intended to balance the overall life and reliability of the parts -- pulleys, bearings, the belt, rollers, etc. -- and the you'd need long term data across a large set of cars to see the benefits of the factory tool over the more subjective tests (and/or to confirm if that's even true). With all the material and mfg process changes over the last 40 years since the specs were published, it's hard to even know if those specs are even still optimal. That said, it's the best we have, so I'm a diehard believer in setting the belts to spec with objective measurements like the factory tool (and our 3D tool) offer. 
I've done a lot of calibration testing as I develop the digital version, and can confirm there quite a bit of false precision (placebo precision?) built into the tool and its specs. I have three or four calibration bars and those differ by as much as .3 between the gauges. If you stack calibration tolerances, calibration bar variations, tool placement/use, and factory spec ranges, you could easily end up with a belt that is in spec with one tool and well out of spec on the other. You can even get slightly different readings even on the same calibration bar by just placing it a little deeper or shallower on the posts for example.
I'm confident that's why people get away with the twist method or water pump slip method. It's not that those methods reliably get the car within the factor tension specs (they don't) but rather that the motor can tolerate a much wider tension range than the placebo precision of the tool suggests.
I suspect, however, that the factory specs were intended to balance the overall life and reliability of the parts -- pulleys, bearings, the belt, rollers, etc. -- and the you'd need long term data across a large set of cars to see the benefits of the factory tool over the more subjective tests (and/or to confirm if that's even true). With all the material and mfg process changes over the last 40 years since the specs were published, it's hard to even know if those specs are even still optimal. That said, it's the best we have, so I'm a diehard believer in setting the belts to spec with objective measurements like the factory tool (and our 3D tool) offer. While I'm mostly in agreement with you, if I'm completely honest I'm just a tool junkie. The cars are just an excuse.Tom wrote: Sat May 16, 2026 11:31 am ... so I'm a diehard believer in setting the belts to spec with objective measurements like the factory tool (and our 3D tool) offer....
