Friday, February 27, 2015

Musson's Professional Guide to Wheel Building: But does it work!?

My review of Roger Musson's Professional Guide to Wheel Building is very positive. But how does his advice hold up in practice? Well, I just finished building a 650B wheelset following his advice and using some of the DIY tools described in the book.

DIY Nipple driver

Musson argues that commercially available nipple drivers such as the Park Tool ND-1 suffer from a design flaw in that they lead to an initial spoke tension that is too high[1] . Musson's design, built from a cheap screwdriver or bit has a central pin 3mm long, significantly longer than that of most other nipple drivers. The idea is that the spoke moves up inside the nipple, pushes against the pin, and then disengages the driver, allowing the builder to quickly achieve the same thread engagement on all spokes. To achieve that in previous builds I had used a visual indicator: Engage all nipples to the point where the threads just disappear into the nipple. Depending on your spoke length, this may already be too much tension, and Musson's method also sounded faster so I decided to give it a try.

Crude, but it works


I started out with a drill bit, later to be used with an electric drill. I don't have a vise, and consequently the tool I produced is neither pretty nor particularly precise. Musson also advocates against using an electric drill. He argues that with a hand tool, it is easier to precisely feel get the disengagement of the tool just right. After having used an electric drill in this build, I tend to agree. I ended up with fairly good equal tension in all spokes, but it could have been better still. Another problem I encountered was that after using the tool, the spokes were still very slack, requiring several more turns with a regular spoke wrench. However, that probably was less the fault of the tool but of slightly too long spokes (see below).

Spoke lengths

I had ordered spokes before buying Musson's book and reading his strongly worded advice to not rely on manufacturer or any other measurements that you haven't done yourself. While I actually had measured at least the front hub myself, for the ERD of the rims and the rear hub I relied on the value provided by the manufacturer/the Edd online database. This did not work out so well. Assuming that my tensiometer is still accurate, in order to get adequate spoke tension, I had to engage the nipples to a point where I'm almost out of thread. One or two millimeter shorter spokes definitely would have been better—and probably would have prevented the problem with the nipple driver described above.
Spoke is already far in the nipple, despite not being at final tension yet

Lacing instructions

The lacing instructions were perfect. I got the lacing (32 spokes, three cross) right on the first try and for the first also managed to properly align the hub label with the valve hole (for purely aesthetic reasons, one is supposed to be able to read the hub label when looking through the valve hole). Between the written instructions and the helpful illustrations it is hard to go wrong.

Dishing gauge

When reading the instructions for building a dishing gauge, I was skeptical: Building a precision tool out of cardboard, double-sided sticky tape, and zip ties!? But lo and behold, this is a very solid design. Instead of using Musson's design of a bent spoke as the feeler gauge, I simply stuck a barbecue skewer through the corrugated cardboard. Because my truing stand—in contrast to Musson's design or the Park Tool TS-2.2—requires me to have the quick release skewer in the wheel for truing, the BBQ gauge was not ideal, requiring me to remove the QR skewer each time I wanted to check the dishing. This would be easily fixable, though, and with the QR removed, the dishing gauge worked great! As an alternative to cardboard, Musson suggests using thin plywood. At this point I think that for an infrequent wheelbuilder like me the (more easily manufactured) cardboard version is totally sufficient. Instead of using double-sided tape to attach the wood pieces that sit against the rim, I used superglue, which allowed me to more easily align.

I used E-6000 glue instead of double-sided tape to attach the wood pieces

Tensioning

Spoke torsion flag (and marked valve hole)
As a way of achieving spoke tension, Musson, like others, suggests using the sound a spoke makes when plucked with a fingernail or guitar pick: Pluck the spoke and compare the sound to that of the other spokes. I must say that this has never worked for me. You always get a mix of different frequencies and I find it hard to compare them to each other. It's quite possible that this is just a skill I lack (I do like death metal and grind core after all...), but I rather relied on my Wheelsmith tensiometer. Of course, as Musson correctly points out, this is more time consuming than the plucking method, but it does have the advantage of also providing you with absolute values for the spoke tension.

Musson's approach of iteratively checking lateral trueness, vertical trueness, dish, and tension works well, especially if you have the luxury of the amateur wheelbuilder: You can take as much time as you want to. For the final stages of tensioning I found his advice of using a masking tape flag to visualize spoke torsion to be especially useful. Under high tension, turning the nipple will start lead to a small amount of twisting of the spoke instead of further engaging the threads. So if you turn the spoke half a turn, it will probably only have engaged the equivalent of one third of a turn earlier in the building process. Previously I had just overtightened the spokes by about one eight of turn and then turned them back the same amount to get rid of the spoke torsion. But with the flag method it is possible to get the exact amount of required overtightening and backing out.

Conclusion

All in all, using Musson's tools and instructions allowed me to build a great set of wheels with little pain. Obviously I can't say for sure how they're going to hold up in the long term. But the tension is even, and trueness and dish are within tight tolerances so that I have a lot of confidence in them. The Professional Guide has proven itself to be an excellent manual for wheelbuilding while also saving you money by building your own tools.

[1] I believe this is not true for this VAR tool. But then that's 50 bucks...

4 comments:

  1. This was interesting to read. I've built wheels a few times now, and used only a spoke tool. Truing (in all directions) was checked in my upside-down bike, spinning the wheel and checking the clearance to the brake. I wasn't concerned with dish in the front, as equal tensioning should result in a good wheel, and on the rear I just tightened the drive side more in the beginning, and it came out well each time. I guess tools could help ensure more perfectness, but I didn't find it necessary.

    ReplyDelete
    Replies
    1. Yeah, my first one or two wheels were built in a similar way, and they also held up just fine. I used zip ties attached to fork/seat stays to check the truing. If I had to pick just one wheelbuilding tool in addition to the spoke wrench, it'd probably be the truing stand. It's just much more comfortable and precise than the in-bike method.

      Delete
  2. This is a great review and I love knowing that I can build some inexpensive tools to help with the build. Thanks so much for the thoughtful details. I haven't built a wheel yet, though I have repaired a badly damage wheel by transferring the spokes to a new rim. would love to hear from you, if you have any idea where I can get a copy of the book? All I can find is a PDF online, no hard copies, and I do so much better with pages in hand. Cheers!

    ReplyDelete
    Replies
    1. I'm happy to hear you found the review helpful. I don't think the book is available as print. Roger regularly updates the book, which I guess lends itself to an e-book. Of course you can always just print it yourself. I fully agree that for a wheelbuilding book paper is rather than precariously balancing your laptop on your workbench... :)

      Delete