Interested in what tolerances people have achieved with the takedown sleeve system. I have just been through the process of fitting to my yew billets. Spent a lot of time trying to achieve a "perfect sliding fit", not convinced I did as well as I could have.
Basically did as follows:
1. fitted the brass sleeves to the shaped ends of the billets, being as careful as I could not to distort them.
2. Used a vice to shape the outer (stainless) sleeve to fit. One problem being that both the brass inner sleeves were now very slightly different in shape, so not possible to achieve a close fit with the stainless outer on both.
3. got round this by making a looser fit on one (mix of file on inner and vice on outer) and going for as accurate a fit as I could on the other.
4. Then glued the looser fitting sleeve into the outer sleeve, so no movement there.
The other sleeve (the one I went for as accurate) has a maximum gap of around 0.25mm at one point of the fit. This allows about 5mm of "wobble" at the end of the 36" limb (sorry about the mixed units!). Obviously there is no wobble on the other limb as it is glued into the outer sleeve.
The wobble is front to back so shouldn't affect string alignment. However, would have liked to have had less/no movement at all (however, would then have been impossible to push it into or take it out of the handle as the air wouldn't have been able to escape!).
How close a fit have others managed to achieve. What is possible using a vice to shape a piece of stainless steel piping and some emery paper/files to shape the brass inner?
As a point of comparison. A fine clearance hole on a 24mm bolt is defined as 25mm in diameter (i.e. a 1mm gap which would translate into 18mm wobble at the end of a 36" bow limb). An ISO "loose running fit" for a 25mm shaft in a machined bearing is defined as 0.26mm clearance (i.e. close to what I achieved).