I have decided to begin an new project and after a couple of months without the ability to get out into the workshop, I have decided to build a 12" truss Dobsonian - basically a scaled-down version of the "big ones". Some of you ATMers may remember my initial 18" Dobsonian project here, which I unfortunately had to end prematurely for family and health reasons. I hope this thread will last longer than two pages! This scope will be a low maintenance, basic Dobsonian - no bells and whistles - so family and others can use it. More importantly, this scope will not be "over engineered" which can be an asset to a person's ability or a hinderence, depending on who you are. If this is successful, I may venture onto a larger Dob in the future.
On with the show...
Flotation Mirror Cell Design.
The first step was to design the mirror cell and despite my previous comment "This scope will be a low maintenance, basic Dobsonian" I did decide on an 18-point flotation cell. I spent a few hours last night calculating the dimensions and drawing up the results on AutoCAD.
In most part, I rescaled the dimensions from Kriege's book and confirmed them with some software and internet resources. The tolerances of the balance and pivot points, and the size of the triangles and bars are within a millimetre or two. For me, quite acceptable. Oh, and all my dimensions are in inches, don't ask me why, I just like imperial measurements! Here are the results:
..and my 1.6" thick 14" mirror rides on a mere 6-pointer, with the flotation points spaced at around 58% of radius.
The thing to do is download David Lewis's program PLOP and have a go at modelling the various options. I think you'll find that anything more than 6 points for a 12" mirror is overkill:
Thanks all for your advice. I actually used PLOP to do some of the calculations and I understand that an 18-point may be an overkill. By recalculating to a 9-point cell, the balance points for the triangles are only about 2mm further out from the centre of the mirror. No big deal as far as I'm concerned. I can easily make a 9-point cell with the same frame dimensions. I am making the triangles this week.
Flotation Cell Frame Welding Jig.
I knocked up a welding jig this evening from some old scrap. Normally, I wouldn't have done this, but welding can be a bugger of a job, especially when you want your work to be completely square. As anyone who welds knows that the heat can cause bowing and twisting so I wanted to minimise that as much as possible.
Flotation Cell Frame Parts.
Easy! 20mm square tube for the three horizontal rungs and 20 x 8mm flat bar for the two sideframes. All these materials were lying around in the shed. All steel cut and waiting to be welded. Money outlaid so far: nil.
So in summary of the ideas below & adding another, a 6-pointer:
- is lower profile
- gives better theoretical results when modelled in PLOP
- is easier to fabricate, because you have no triangles to cut out, just bars.
Thank you Geoff, Ian and Phil for your advice and input. I have been discussing a 6-point cell with Satchmo (Mark) at length today, and the pros and cons of various designs and construction methods. All your advice sounds quite valid and makes sense, so I will sit on it and weigh it all up over the next week or so. I have designed a 6-point and 9-point cell based on my calculations.
Flotation Mirror Cell Frame.
I welded up the frame today, using 20mm square tube for the horizontal rungs and 20x5mm flat bar for the side rails. I had to cut a slot for the welding clamp as I had nowhere to attach it! I am pretty happy with the end result. I might even get it powder coated as I know someone in the industry who may be able to help.
The design of the 6 point would worry me a bit as it appears that the bars are not straight with the balance points inside the collimation screws. This would put a twisting force on the plate and make collimation a bit sticker than if the bars were simply straight. Worth considering??
For the sake of clarification about the 6 point cell, the bars must be straight.
I've run the numbers in Plop for an optomised cell for Stevens case where he has allready drilled collimation bolt holes at the 56% zone. Including a little optimisation of pad angle, the flotation points on the bars will be 96mm apart ( which effectively works out at about 63% zone ) with the collimation points/fulcrums at the 56% zone. This yields a surface RMS error of 1/366 wave for the 35mm thick mirror, quite good enough by anyones standards I think.
Don't get too obsessed about the precision of the cell , there is quite a bit of leeway before you would ever notice anything at the eyepiece.
Okay, I think we are finally getting to the home stretch. Thanks to Mark again for those figures (I still cannot get Plop to play ball). Here is the revised drawing, and I am glad you agree when I say that 1mm out here or there is not going to matter in the scheme of things. I have made the bars straight and 1" (25.4mm) wide.
At the risk of opening a new can of worms: how are you planning to do the edge supports?
Interesting hot topic at the moment. Many builders of scopes with big,very thin mirrors are making sure to support the mirror on its centre of gravity. Usually with +/- 45 degree points or a cable sling. http://www.cruxis.com/scope/mirroredgecalculator.htm
But with this thickness of mirror any solution would probably be OK. That is, simple +/-45 points, seatbelt sling, or RTV-ing the cell directly to the mirror(if it doesn't wobble).
tnott, you answered Phil's questions regarding the mirror supports. +/- 45 points, sling and silicon the base of the cell onto some 1/2 inch rubber o-rings to create an uniform height on each of the six pivot points. And yes, I will be using a 6-point system base on Mark's calculations and advice.
I managed to get hold of a GSO 12" with perfect optics but a damaged tube for less than the cost of a new GSO mirror.
Last weekend I made good progress with the flotation mirror cell. Instead of threading the square tube for the collimation bolts, I fitted three 1 1/2 inch all-thread joiners (3/8 thread) into the square tube and silver braized them in place rather than welding. They protrude out approximately 8mm either side of the tube.
I then rummaged through the shed for some material to use to support the cooling fan. I couldn't find any perspex or plastic, as is commonly used, and I was determined not to spend any money, but I did find some 1mm mild steel plate. This may sound heavy for some to use, but after the 70mm hole was cut for the fan, there is not much weight in the scheme of things. It is rigid enough. Tomorrow, I'll make a small bracket for the cooling fan connector, buy the connector from the local muppets at Dick Smith, then off to the powder coaters. I discovered a powder coating firm about 3kms from me and they will put it in another batch and only charge me about $50-70.
Last edited by stephenb; 17-10-2008 at 10:15 PM.
Reason: corrected punctuation
tnott, you answered Phil's questions regarding the mirror supports. +/- 45 points, sling and silicon the base of the cell onto some 1/2 inch rubber o-rings to create an uniform height on each of the six pivot points.
Steven, Looking good. Id only recommend either +/-45 degree edge support pads or Silicon down to float points for this mirror , a sling is not necessary.
Today I made a small panel to mount the DC connector, power switch and LED using silver braizing. It consists of a long, low-profile panel with two gussets at either end, then I mounted it in between the middle and lower rungs of the cell, and braised it in place. It looks all discolored because of the silver, but once it is painted it will look okay.
Mark, I feel more comfortable using the silicone method and the +/-45 supports as a back up. I will not be bothering about a sling, although I have drilled the holes in the frame.
Last edited by stephenb; 14-10-2008 at 10:10 AM.
Reason: corrected punctuation
Mark, I feel more comfortable using the silicone method and the +/-45 supports as a back up. I will not be bothering about a sling, although I have drilled the holes in the frame.
By all means have some 'peace of mind' safety clips safety clips positioned wherever around the mirror , but you cant both glue the mirror to the cell and have anything else touching the edge, its not good opto-mechanical engineering. A mirror glued to its cell behaves as a sling support with its C of G at the rear edge of the mirror. There is no way of co-ordinating the forces between a glued down mirror and another form of edge support ( from the rear) and you will get some kind of mirror distortion that you can't control. I'm not a fan of silicon , I think Velcro is much better , as the mirror has more freedom to relax on its mount. All that being said you can get away with a fair bit mounting a 12" mirror so I wouldn't lose any sleep over it.
I change my decision and had the mirror cell frame powdercoated today, and it turned up a treat! I mounted the switch, LED, fan and connector, wired them and tested - all working okay!
Last edited by stephenb; 14-10-2008 at 10:09 AM.
Reason: corrected punctuation
