Assemble as shown (3 views):
Clean gluing surfaces with acetone:
Makes its own spacers! Let RTV cure at least 24 hours:
Stringing not shown - you'll want to make an alignment jig for sure. These are .014" plain steel electric guitar strings, and have a tensile strength of at least 30 lbs. As shown later, .010" strings work perfectly well, albeit with 1/2 the strength per string.
What you should note here is how the two closest vanes (in this picture) anchor at the far side of the hub assembly, crossing paths to do so. This is the key to what makes this spider so rigid and resistant to vibration. It completely removes the tendency of wire spiders to rotate in vibration along the optical axis of the OTA.
The wires could align along 90 degrees, or as 60 degrees, for a kind of fan effect on the spikes... The view from above:
Shown in the mockup secondary cage (ring). Improvised tensioning hardware at this point. ;) Note the spreaders on the near wires are not symmetrical above/below the ring:
View from the focuser position:
Can you say
There's considerable discussion in this thread on Cloudy Nights where I posted it first. Some additional work yielded the following:
After working with the secondary a bit I thought that if the wires that converged on the focuser side of the hub went from top-bottom and bottom-top instead of describing a trapezoid the resistance to bending along that side would improve. It does. Obviously a trapezoid is less stable than a triangle, which in turn is less stable than a crossed-over double triangle. Try it with a loop of string if you have doubts. OTOH, the effect is greater than expected for such a small crossing...
After that I thought I'd just go ahead and try stringing it with .010" wires and see how it worked. The results are most impressive - I can't measure any deflection from vertical to horizontal with a dial gauge, and even at medium tension the settling time from rapping on it is unaffected (less than a second).
The real explanation of why crossing the wires over such a small difference affects the vibration so strongly is subtle. Take a look at the picture below. The vibration mode involved (which was the largest remaining of a small set) consists of pivoting across the two back vanes, essentially as a rigid structure. This is shown by the small purple arc .
In the top geometry the lower attachment point first moves to the left, then to the right, as it pivots up through the plane. In the bottom geometry the top attachment point always moves to the right as it pivots up. The restraining string thus resists the pivoting much better.
BTW I originally built the spider with a longer span on the "short arm" but a broken 6-32 tap on the very last hole on the flat plate rendered that one so much scrap. I suspect but can't prove that lengthening the span in question would have little effect on the stability, given the geometry at work here, but it's something else to try. As it sits I'm extremely pleased with the performance so far.
In the quest for the
limits of the design I restrung the spider with .008" strings last
Friday (8/8/08). This works perfectly well, which is not too surprising
considering the progress so far - the settling time changes from about
1/4 second to about 1/3 second. All vibrational modes with this
geometry have strong damping. Next step will be seeing how that is in
practice, as I'm calling it done for now.
I have a very dicey internet connection today with a bad phone line and I'm heading off to OSP tomorrow. I'll update this afterwards and apologize for any mistakes or bad links in advance. :)MORE: In discussion how to align the spider with Greg Rhode, who's building one, it seems I've left out one essential clue - use a jig to do the rough alignment. I got this idea from Mel Bartels originally. You just build a structure that attaches to the upper cage or ring and clamps the spider in place well enough to align it physically. THEN you string the wires. After that, if you start with the flex-plate screws mid travel, you should always be able to achieve fine collimation. Sorry, no pics of this, as yes - 4+ years after the above, I still don't have the scope finished. :(