I have done it the way Arne suggests and where the difference between the two diameters is small, that, as Arne says, is easier than making a wood sleeve. But if the difference is great (say between 6" and 4.5" - I think wood sleeve is better and accuracy can be achieved without a lot of effort. I offer you a sort of "article" about it:
Mast made from lapped aluminium tubes
I have made a couple of “composite” aluminium tube masts, one was done in three pieces and one was done with two. The aluminium/aluminium overlap epoxy-glued join seems to be successful, though in neither case have they been rigorously tested. I can say only that I am satisfied.
It has been suggested that epoxy glue is not ideal for aluminium as they have different mechanical characteristics, and that a more flexible glue such as Simsons is preferable. I don't disagree with this, but my gut feeling is that, in this case, provided there is sufficient overlap (bury), the glue used need do little more than fill the annular gap between the two tubes. A necessary step is to add an external fairing of thickened epoxy to smooth the transition between the two diameters, then an external fibreglass bandage applied over the join - and this alone ought to be strong enough to resist the twisting forces that the rig imposes on the top of the mast, and the downward force (which wants to make the mast “telescope” shorter). Actually, I think it would be preferable if the join could be made so that the two tubes could be disassembled with minimum damage, in the event of a breakage, but I don't know how to do it without the two parts being irrevocably stuck together. This is why I don't think it matters much what type of glue is used, though I may be corrected on this.
In the three-part tubular mast the differences between the diameters was small enough that an epoxy/glassfibre bandage was sufficient to fill the gap between the tubes at one joint, and on the other join, the gap was a little larger so I made a bandage of what seemed to be a sort of thin polyester carpet which I thoroughly soaked in epoxy before sliding the tubes together, and the composite was resilient enough to ensure that there was no floppiness while the epoxy was curing – ie that the tubes would remain co-axial. The above seems like a lot of words (it’s a bit hard to explain) but the process was simple, straight-forward and intuitive. I think there will be a photo of that mast in the next JRA magazine. I agree with Arne that is a much easier solution than making a tapered wooden topmast or trying to make an accurate, thin, wooden sleeve. It really was easy.
With the other mast, the one made from just two tubes, the difference in diameters was great enough that I was worried about how to keep the two tubes accurately aligned, and there was no way the annular gap between the two tubes could be filled with just epoxy-and-fibre bandages. At least, that didn’t seem to me to be an easy or economical way to do it. Instead, I filled the gap with wooden staves.
The staves were laid out on a length of duct tape, like this.
then wrapped around the mast, glued and held with plastic cable ties until cured.
Two of these wood/epoxy packings were made, the gaps filled with epoxy and roughly faired.
They were deliberately undersize and the intention was to make an accurate fit with the addition of thickened epoxy.
The mast was to be about 10 metres and I did not have the space to lay it out and ensure that the finished mast would be straight. To save space in the garage the two tubes had been stored, one inside the other - and from this happenstance I hit upon the idea of making the wood/epoxy packing for the join while one tube was still lying inside the other, ie of reversing the topmast section and sliding it all the way into the bottom tube – with just enough sticking out to allow me to apply the wooden packing. I then applied a layer of thickened epoxy resin, wrapped it all in two layers of baking paper
and slid the top mast further into the bottom section so that the wood/epoxy/thickened-epoxy packing could cure in place. The nice part was, perfect alignment was easily obtained by putting a rope grommet around each end of the inner tube and covering the grommet with enough layers of thin packing tape to make a close fit. This ensured perfect alignment and the result was a casting which was a near perfect fit, and with accurate alignment (ie accurately coaxial.) The temporary rope grommets were discarded later, of course.
There was only one worry – after the epoxy was cured, would the two parts come apart, to allow the topmast section to be reversed and re-assembled as the desired join?. I had previously ascertained that baking paper will stick to epoxy, but does not allow epoxy to soak through it, so I thought to use it as a parting membrane. I also used a liberal spray of an aerosol parting agent on the inside of the “mould” (the outer tube.) I then said many prayers, because if it would not come apart the result would be a very expensive failure. With a fair bit of beating it did come apart, and the result is a near perfect coaxial casting which, after the baking paper is sanded off, is a sliding fit.
A few drops of epoxy (or pretty much any sort of glue) and I think it would be impossible to separate the two parts.
After the topmast section is reversed and the join assembled and glued, an external conical fairing of thickened epoxy can smooth the transition between the two diameters and then, as described by Arne, a fibreglass bandage applied over the joint. This last should on its own provide enough strength to prevent rotation or telescoping. It is my surmise (correct me if I am wrong) that this last necessary step ought to make the glue which is inside the joint redundant except as a gap filler.
I am sure there are many better ways, but this is just what I did, with what I had to hand, and maybe will spark some more thinking and some improved ideas. I can say that it was quicker and easier than the description would imply, and accuracy was obtained without a lot of sanding and fairing. Personally, I would do it again that way if it were again necessary to fill a largish gap between the diameters of the two tubes. (But I would be very careful to ensure a minimum or two layers of baking paper and to ensure there are no deformities in the outer tube (effectively a non-tapered mould.) There's quite a lot of friction to overcome when separating the casting.
[I just re-read Stephen's proposal (much more economical with words) and infer that he is pretty much suggesting to achieve the same thing. If you can lay the mast out fully, keep it straight and do it all in one go, so much the better I suppose. Laying out the wooden sleeve in the form of staves, thickness and spacing carefully calculated, and initially stuck to duct tape, was, I thought, a good and easy way of making (assembling and gluing) the sleeve.]