Sunday, August 21, 2005

Rear Triangle Follies

For the better part of a month, I've been trying to figure out the alignment of my rear triangle. You see, it's wrong but I haven't been able to determine what or how?

Part of my problem has been the availability of concentrated periods of time. Today, Janet and the kids went up to Wisconsin for the weekend to see Grandma, then my showing appointment canceled. Suddenly I had a whole day to work in the shop!!

Without a frame jig, I've been using Suzy's approach of a straight beam off the front triangle to position/align the rear triangle. In concept this should work well - albeit with a certain amount of effort. Suzy makes it work, but I've had problems.

BTW, this once again reinforces my respect for Suzy Jackson who truly must be a goddess among garage bicycle builders.

When the chainstays were first installed, I put my Dropout (DO) tools on and they aligned perfectly. I was surprised but happy. Then, I installed a wheel. Whoa Nelly!!!!! It's off to the right by an awful amount and crooked vertically.

So I start measuring things like: Length from the back of the DO to the center of the BB (or the inside of the back edge of the BB, or the outside of the front edge of the BB), level side-to-side of the chainstays starting near the BB and working my back to the DOs, same as previous items except to the back of the axle slot in the DO, and finally in desperation I ran string DO to DO via the head tube and measured its offset from the seattube.

I did many measures over a long period of time because I didn't want to rush in with a repair unless I was sure of the problem. Unfortunately, none of these measures showed what the problem was. Whenever there seemed to be discordant measures left to right, a re-measure would refute the first. Arghhhh!!!!!!

I must be poor at measuring. At one point I was looking for a twist in the BB such that the stays could be good relative to the BB but not to the rest of the frame - I couldn't find it.

So, yesterday, I began by rechecking the front triangle w/ v-blocks on a flat surface. It checked out fine. Then, two beams were run from the downtube past either side of the seattube and over the DO tools. I didn't use shims to equalize for tube sizes, but made sure that the beams aligned to the downtube (and so had only one point of contact on the seattube). I then sighted down between the two beams which clearly indicated that the back of the rear triangle was offset to the right relative to the front triangle. There were several more tries to ensure that the result was consistent - and it was.

I looks like the difference is 1-1.5mm, which isn't much - but lets see what we can do.

I took the approach of installing my dummy axle and adding a couple of mm to its normal width (132). Then I heated the left stay socket until the flux was nicely melted. After things cooled, I screwed the dummy axle back to its normal width and heated the right socket.

After cooling and defluxing, it was back to measuring. The results of all measures were largely the same as before: All good except the beams showed an approximate 1.5mm shift to the right and the wheel was still to the right. There was one key difference: the wheel wasn't as far to the right as before. Concluding that the beams weren't giving me an accurate measure of how much to adjust, I looked at how much the rear wheel needed. It was 5mm (yeow!).

So, using the same method as before, I gave the rear triangle a 5mm shift to the left. Measure and check the wheel, it looks pretty good. Now there is still a difference vertically from left to right, and it appears that the left DO is out of phase with the right DO. So I cold set the stay to achieve the proper height (drop?).

After thinking about the alternatives, I decided to remove, clean up and replace the left DO. With stainless, I didn't want to take a chance on weakening the joint by merely heating and twisting the DO into phase. So, the DO is out and I've started cleaning it and the tube. a little careful refitting and I should have an aligned rear-end.

At that point, I'll need to finish the top end of the left seatstay, install the brake and chainstay bridges, and sand & file the frame.

Its a relief to think I finally have the rear-triangle mystery solved. That said, I think its time to acquire a jig to avoid these issues, and particularly to avoid re-heating joints.

That's it for now. Hopefully there will be more pictures soon.

No comments: