Sunday, April 28, 2013

Done With Big Bertha

Maybe in more ways than one.  I ended up using the steel rods epoxied to the tube to create a flat base for the Moonlite truss connector blocks -- and this worked out well.  Minimal material, minimal weight, and once the screws were tightened down, there is no risk of those rods working loose.

After wrestling with getting the truss tubes in place (rather like wrestling a very large insect, I suspect, without the horror film aspects), this is what we have:

My objectives for the rebuild were:

1. Reduce weight.

2. Improve stiffness of the structure.

3. Make it a bit easier to transport.

To my shock, #1 was a complete failure.  It was 61 pounds before, and it is 61 pounds after--in spite of cutting away more than half of the aluminum C-channel that was the base.

The truss tube structure is, I think, considerably stiffer than what I had before.  But it appears that now the flex isn't in the structure between the upper and lower cages, but between the lower cage and the saddle.  Remember that I no longer have a 72" C-channel between the two cages.  The flex is increased because of that, to no great surprise.  In addition, to save a pound, I replaced the 19" Losmandy dovetail plate with a 13.5" plate.  This means that there is less stiffness between the lower cage and the saddle.  Overall, I think I have turned a barely tolerable amount of flex which varied by angle into an intolerable amount of flex in wind of any sort -- although I won't know for sure until I roll it out under a dark sky.  But I managed to spend more than $300 and much of two days doing so.

The only way to salvage this disaster might be to replace the existing aluminum C-channel with carbon fiber composite C-channel like this.  But that will be almost $200 with shipping, and before I do that, I want some serious confidence that it is going to have deflection measured in thousandths of an inch from a 55 pound load.

The temptation is very strong to just give up on trying to have a serious 17.5" reflector on an equatorial mount.  The choices seem to be:

1. A solid tube, which makes it heavy enough that I need a $6000 mount for it.

2. Dobsonian mount, which greatly complicates astrophotography.  (There are ways, but they don't impress me.)

UPDATE: A little playing with it leads me to believe that I moved the flexibility from the structure to the lower cage.  The half-Suerrier truss appears to be completely rigid, nor does it appear where the lower cage meets the saddle of the mount is flexing.  It looks like the lower cage is flexing!  I'll take some video of it, and then play it back frame by frame tomorrow night.  If so, I might just need to have a thicker aluminum tube made -- perhaps .125" or .1875" wall instead of the current .090" wall.  Or perhaps even make a fiberglass tube.  (Carbon fiber composite?  Like $1600.)  I have made small fiberglass tubes before; I would certainly make a 4" ID then an 8" ID tube first as a experiment.  The big problem is appropriate molds.  I had thought of using 20" pizza pans for this, but what about 20" car wheels?  They have the advantage of being very wide, and perhaps I could borrow a couple from someone for this purpose.

Here's the video.  I haven't gone through it frame by frame, but it looks to me like the tube is flexing.

UPDATE 2: Definitely: the lower tube is flexing, not the truss tube structure or the support.


Moose said...

Why don't you build a truss out of carbon fiber components from Dragon Plate? Seems to be a low effort proposition...

Clayton said...

The truss components aren't the problem; it turns out to be the lower cage tube is too flexible.

RS said...

Pretty clear you aren't a mechanical engineer.

Know any?

Clayton said...

Afraid not. That's why I ask my readers, and they have come up with some good suggestions.