I originally set out on the Hyperstar journey to take deep narrow band images but straight away ran into problems with mirror flop limiting my subs to about 300 sec, so time to put my thinking cap on and get the big hammer out. Actually the big hammer was in fact the edge of a 1/2" chisel as my first effort was to reduce the play between hollow shaft and the mirror/tube bore by whacking the shaft o/d, this raised some nice bruises which did in fact reduce play between the two parts but not by enough to go for more than about 400 sec. So plan B was to cut 3 slots in the main cell casting through which I ran three lengths of 2mm x 15mm aluminium bent over at one end and attached to the back of the primary with double sided tape (I tried silicone adhesive first but it flexed too much). To secure the exposed end I screwed 3 steel angle brackets to the back of the main casting and fitted M6 machine screws to them, these mirror support struts had slots filed on the exposed ends to allow them to move freely against the angle bracket/lock bolt. The struts being clamped after focusing by lock *****. Clamping these up initially moved the primary around but adding two washers with a dab of grease between them soon fixed that and permitted a nice firm clamp with only a very small shift which does not affect focus or collimation. I am now a pretty happy bunny as I can reliably go to 900 sec subs with no noticeable flop. I know my methods are crude and will make the engineers among you weep but I have no workshop and have to do everything on the kitchen table with simple hand tools. I think one of the main reasons this crude mod actually works is due to the small primary, a larger diameter mirror would need much more support to work well. The final touch to improve star shapes was to use an aperture mask to bring the effective diameter down to 120mm, I will have a fiddle with this to see if I can increase it but even at 120mm I am satisfied with the performance.