fishboat

(This is a bit long..it's mainly targeted at novice, and possibly not so novice, wheel builders and those searching for info in the future.)

I recently built up two sets of wheels. They were my first venture into building bicycle wheels. Overall, I found wheel building reasonably easy to do. It takes some time initially, though speeds up into a routine process with some experience. I did rely on tensiometers to guide the builds. (As a brief aside, for those who have lots of wheel building experience, I would agree a tensiometer may not be needed. However, without that experience, a tensiometer is very handy and simplifies the process.)

I finished the first set and things seemed to go well. When I was about 90% of the way done with the second set I started to question "where I was at". I was using a "ZTTO" (Park Tool TM-1 clone, exact clone) tension meter. I picked it up off Amazon a couple years ago and used it primarily to check spoke to spoke consistency in existing wheels. For this purpose, the cheap instrument does quite well. I also found the precision (repeatability) of the ZTTO to be very good while building my current wheels.

Where I started to stumble is when I was approaching the final tension of the second set of wheels. It seemed I was almost there and yet I seemed to be a l-o-n-g way away from the tension I measured in a half dozen wheels I've purchased in recent years. The tension in the half dozen wheels appear to be very consistent within themselves(wheel to wheel). However, all six wheels appeared to be at MUCH higher tension than my "finished" set of wheels and my nearly finished set.

What to do? Most any measuring device requires calibration. Since I had no idea of my device's calibration state, that was the next step. So..BF comes to the rescue once again with options to build an inexpensive calibration setup. I did that (pic below) and checked the ZTTO device. Yikes. It was off. Really off. The instrument was measuring considerably high and therefor yielding spoke tensions very low of where they should have been. The tool had very good precision, but very poor accuracy.

No problem, so I thought, as the ZTTO, just like the Park TM-1, has a bolt on the back side to calibrate the device. I used the bolt to increase the tension on the spring and thereby brought the ZTTO output up to match the tension I set on the calibration setup (set at 117kgf). The ZTTO felt much stiffer when I squeezed it.

Ok, back to finishing my wheels. The newly calibrated ZTTO did indicate the spokes were at a very low tension. At about 6-8 spoke checks into a wheel the number output from the ZTTO started getting very odd & inconsistent. I checked the device against the calibration setup and it was way off again. I tried to recalibrate it..no dice. The spring would no longer hold sufficient tension. I could get maybe the equivalent of 100kgf out of the spring. If I tightened the spring further it would yield and the error just increased. Junk tool. It’s nicely made and finished well and yet, at the end of the day, it’s all about the spring. It does look good though.

I tried to locate a replacement spring..nope. I tried to locate a replacement Park spring..nope(one vendor in EU had them for about $5 with $40 in shipping).

What to do, version 2. I considered picking up a Park TM-1, but pretty much lost faith in the design of this tool. Web research indicated the Park also needed frequent calibration. Park does calibrate them for a small fee, though I didn’t want to mess with it…or invest $80+ in finding out how much I’d have to futz with it. I ended up buying a Wheel Fanatyk (Jobst Brandt design) tensiometer. So far..two thumbs up. I checked the calibration when I received the instrument and it was right where it should be. I went through both sets of wheels and tightened them up. They are now consistent with the hand built wheels I’ve purchased in recent years.

Below you’ll see the data. It’s from the second set of wheels I built and was about 90-95% complete. The spokes just needed a little more tension and the wheels would be done(they were radially true, laterally true, and dished correctly). The ZTTO data columns indicate the “nearly final” tension before I started questioning the build process. The Wheel Fanatyk columns report the actual spoke tension of those same wheels(spokes) once I received the WF instrument. As you can see..the tension is way low of where it should be. I’m fortunate I didn’t ride these wheels. The first set of wheels I built had similar (low tension) numbers.

Wheel specs: HP Plus Son Archetype rims(125kgf max), Sapim DB SS spokes (2/1.8/2), Sapim brass nipples, Ultegra 6500 hubs.

Take Away Summary: I understand the allure of inexpensive tensiometers. If you use a tensiometer in wheel building, make sure you have some way of checking the instrument calibration. Without a calibration check, you’re just winging it and hoping for the best. I’m a retired chemist and well acquainted with instruments of all sorts as well as running Gage R&R evaluations to establish their capability. Just because an instrument pumps out a number doesn’t mean it means anything.




Finished rear wheel(my spreadsheet):


Finished front wheel(my spreadsheet):


As long as this is being posted..below is the Wheel Fanatyk conversion (meter reading to kgf) data regression. To get the best fit I regressed just the data around a finished tension rather than all the data. The best regression equation was of the form y = A + (B/x):

Aardwolf

A good point about needing to calibrate tension meters.
I got to exactly the same place a couple of months back: I've got the Ztto TC-1.

What I had previously spotted is that on the Park Tools spoke tensioning app it gives the kgf to 2 decimal places (https://www.parktool.com/wta).
So by putting 10, 11, 12, ... in as the spoke tension readings you can get a TM-1 conversion table accurate to 2 decimal places.
Using the statistical package R and the basicTrendline library I fitted an exponential curve to the TM-1 conversion table:


From that I guessed the Ztto TC-1 would have an exponential curve and used a crane scale to get some readings, then I adjusted the TC-1 to get the maximum sensitivity on the kgf range I wanted (40-90 kgf):

Dean51 

fishboat Good work with the DIY calibration rig! I did something similar to calibrate my homebuilt copy of the Brandt tensiometer.....see pics below. FWIW, I've rechecked the calibration a few times over the years and it has held up perfectly.

Dean

IMG_2439

IMG_2443