I've been chatting with astrobeast1 (Dan) over on CN and I've come to a decision whether or not to ditch the external rings completely.

The surface tissue I used with the resin to cover the tubes is very lightweight at 30gsm. The ribs/baffles I've fitted on the inside do not quite butt-join together and I haven't bothered to fill the gaps between them yet, either. Both of these things represent weak spots in the current tubes and risk the tubes collapsing or splitting in some way.

The external rings were there to serve two purposes:

- provide additional rigidity to the tubes (keep them round)

- to satisfy the geometrical requirements of flat-end truss attachment to a dodecagon.

A layer of 300gsm chopped strand mat and additional resin around the outside will provide more rigidity - I should have used this the first time round. I will attend to the ribs where they butt against one another (or not quite, as it happens), probably with epoxy putty. Together, these two things will (I have convinced myself) hold the tubes in the round under any stresses to which they are likely to be subjected.

By dispensing with the rings then, the trusses attached to the top and bottom tubes will mate directly onto the surface of 8 out of the 12 sides, with pairs being attached just either side of a 'vertex' - so two each at top, bottom, left and right. Additional 'meat' will need to be fixed internally to accept the threaded inserts that will take the truss bolts. On the bottom tubes, this may need to be significant, given the weight of the primary which will be about 6 inches away.

At the bottom end of middle tubes, I will double up MDF triangles (so 36mm thick) to make up the required geometry of the end-profile and attach these to 4 of the 12 faces - top-left, top-right, bottom-left and bottom-right. These will be screwed from the inside with a strip of mat over the top to 'tie' them to the adjacent faces. The trusses then attach to these, either side of the top 'point'.

I want to have a full ring terminating the trusses from the top tube, such that I can bolt the whole lot to the top end of the middle tube - the scope will then be able to be split into two sections. I still need to give a little thought to the placement of those bolts, the attachment supports on the middle tube and the precise hardware I will use.

A picture is probably in order...

Anyway - getting rid of the rings will eliminate over 1kg of weight, to be replaced with not far short of that again with the mat/resin and triangle blocks. However, the overall appearance of the scopes afterwards should be much improved.

I think design-wise, it's probably the optimal solution in terms of a hybrid of the dodecagon tube and Serrurier truss designs. Aesthetically, it trumps the set of external rings by a large margin. I'm confident it will hold together, too - well, I have to be. A niggle I still have is whether the choice of material for the trusses (1/2"x1/2"x1/16" aluminium channel) is going to be sufficient. Royce's 8" open truss newt uses this but my tubes are likely to be about half as heavy again as his.

In the meantime, I need to get the hacksaw out to remove the external ring I've already attached and also see about fixing the belt sander.

This is what I get for not thinking it all the way through from the start.