polaraligned
sage
Reged: 12/26/08
Posts: 226
Loc: P. R. of New Jersey
|
|
Quote:
sorry, that is not how it works...
anyway, dont get defensive,
No. Not defensive.
I would like to see the calculations that show that a 4" cellular is 8X stiffer than a 2" mono. My experience in beam calculations tells me this is a false statement.
Thanks,
Scott
|
polaraligned
sage
Reged: 12/26/08
Posts: 226
Loc: P. R. of New Jersey
|
|
Quote:
Yet another point of view is that a 2" monolithic mirror has plenty of stiffness. One can argue that there's no need for the cellular to be twice as stiff, and that the cellular could perform perfectly well with only 3" thickness instead of 4". In other words it's not truly an optimized design
Robert
The reality here Robert is that this mirror was originally designed as a replacement for a 2" mono mirror in dobs. I had Dream alter it and put the center hub in for me. The dob manufacturer wanted the cellular to be close to the weight of the mono mirrors. Dream will be perfectly happy designing a mirror to your specs. The weight is not an issue with me so I went with this design. Loosing 1" off the rear side would probably only save 10# or so.
Scott
|
Houdini
professor emeritus
Reged: 07/13/07
Posts: 523
Loc: Europe
|
|
Quote:
I would like to see the calculations that show that a 4" cellular is 8X stiffer than a 2" mono.
Dave's value of 8X would only apply to a monolithic 4" mirror. Whatever the exact value, for the normal gravity load a standard 18-point cell will be more than adequate.
More interesting is the analysis of the mirror and the mirror cell when looking at the horizon. How much does the mirror surface deform under the lateral gravity load?
Robert
--------------------
16" f/4.9 motorized alt-az, 25" f/5 Dobson, 43" f/4 alt-az under construction
Mirror Edge Support Calculator
|
mark cowan
Vendor (Obsidian Optics)
Reged: 06/03/05
Posts: 2156
Loc: salem, OR
|
|
Thanks for the info!
|
dave b
Post Laureate
Reged: 05/10/05
Posts: 3529
|
|
send me the dxf files and ill run it and tell you exactly.
you have to imagine that maybe 35% of the mirror is 4" thick, and another 40% may be functioning as a T beam (depending on how well it is designed): http://www.efunda.com/math/areas/TBeam.cfm
on the topic of cellular mirror nightmares, look at photo 7 (7th from the left) for some nice print through:
http://ngm.nationalgeographic.com/2009/07/telescopes/mcnally-photography
these are pics from this months issue.
--------------------
dave bonandrini
30" f/5.2 Dobsonian
President of GCAC
Astromart Moderator
|
Houdini
professor emeritus
Reged: 07/13/07
Posts: 523
Loc: Europe
|
|
Quote:
send me the dxf files and ill run it and tell you exactly
Maybe you should start by defining what "stiffness" means in the context of a cellular mirror. What load case would you apply, where would you measure the deformation?
Robert
--------------------
16" f/4.9 motorized alt-az, 25" f/5 Dobson, 43" f/4 alt-az under construction
Mirror Edge Support Calculator
|
Darenwh
Pooh-Bah
Reged: 05/11/06
Posts: 1219
Loc: Covington, GA
|
|
So, does anybody think this thread got a little off topic? 
--------------------
Daren
Covington, GA
|
polaraligned
sage
Reged: 12/26/08
Posts: 226
Loc: P. R. of New Jersey
|
|
The SolidWorks files are proprietary. It does cost money to have real engineers design these things.
Getting to "on the topic of cellular nightmares" is just showing the malicious and misinformed mindset of some in the amateur community. People are seeing things in pictures that are not there. I'll bet some members here can tell you the figure accuracy just by looking at the mirror surface.
I will not participate in this conversation anymore.
It has decayed and is off topic to the original posting.
|
JohnH
professor emeritus
Reged: 10/04/05
Posts: 734
Loc: vancouver near the wilds of B...
|
|
I have been working on four of the types of blanks being worked, plate, Thin-Aspect ULE (6" , 10" 0.8" Fused Silica), standard thickness Pyrex (2 6" X 1" ; 8" X 1.25" ; 12.5" X 2.125") and cellular (one of those old EmbeddedRF 16" blanks).
The 6" plate was by far easier to grind out to f/4.7 and polish. The surface appears to be rough due to too fast a polish stroke, leaving the infamous 'dog-biscuit' pattern in testing.
The 12" Pyrex was another matter. It had been started by soemone and left with an off-center depression some 1/2" to 1" from reaching the edge. Four hours of an off-center MOT stroke and 65 wets roughed it out to f/4.9 to the edge.
I have now roughed out the 6" fused silica blank to f/6, taking nearly twice as much time and number of wets as the aforementioned plate mirror. Very tough material.
The 16" cellular does possess some issues. There are a number of hemispherical bubbles in the front surface up to 3/16" across. The link to LookingGlass above has a photo, and the number of defects they give is consistent with the one I have. The one I got had a rough surface for the ribs, which I have been working to correct. It had a curve roughly parallel to the cast f/5 curve in the front. I started working on this by using a 320 grit SiC stone and water. Progress was slow, so I switched to 46/70 grit and a steel plate. This got all but the outside 1 1/2" of the glass and left a few fine chips on the edges of the depressions in the back. I put wood shaving in the bottom half and topped them up with plaster and I will resume lapping the back flat. I also cleaned up the inside of the tapered holes with a wooden tool charged with a mix of epoxy and SiC.
Its is unclear what the glass is though. In spite of the greatly reduced area of glass on the back where I have been flattening it, progress is slow, so I don't think it is plate of any kind. It had a number of small glass 'bubbles' that were attached to the insides of the ribs. I saved a few to identify the glass if possible.
|
JohnH
professor emeritus
Reged: 10/04/05
Posts: 734
Loc: vancouver near the wilds of B...
|
|
Here is a recent photo with the plaster poured in the holes and leveled. I got the idea and wondered if it was a smart one, and after reading the book about Palomar, found out they did the same thing when flattening the back of theirs.
|
Olivier Biot
Amused
Reged: 04/25/05
Posts: 16323
Loc: 51°N (Belgium)
|
|
How will you remove the plaster without damaging the blank?
--------------------
Prediction is very difficult, especially about the future. Niels Bohr
Tal-200K (#199) with JMI NGF-Mini2M focuser on GEM3 • Astro-Tech AT80ED • Orion Sirius EQ-G with wireless EQDIRECT • Astro-Tech Voyager • Celestron Regal LX 10x42 • Helios 15x70
ATM 14" f/5 (redesigning) • ATM 10" f/6 Portable Truss (polishing) • ATM 10" f/25 Dall-Kirkham (optics)
AstroForecast
|
Mike Lockwood
Vendor
Reged: 10/01/07
Posts: 59
Loc: Usually in my optical shop
|
|
Quote:
Mike Lockwood recently mentioned them but SFAIK he hasn't worked with any, a good place to ask is here on Yahoo.
Well, since you asked here I believe I can answer here if I don't name other businesses.
I have worked ~10 cellular blanks (from two suppliers) ranging from ~6" elliptical flats up to 28" diameter hyperbolics. (There is no secret about who the suppliers are, but I can't mention them here.)
There are several more cellular blanks coming up in my schedule very soon.
Thus far, strain has not been an issue. With care blanks polish out to figures of revolution and figure nicely. Inclusions have not been an issue. Bubbles sometimes appear, but are becoming much less common.
After years of watching 1.25"-2" thick hunks of glass cool before testing, I must say that the thermal performance is quite amazing.
Bottom line - I am eager to have my own personal telescope with a cellular primary and secondary.
However, I'm currently busy with a 14.5" F/2.6 for my own experimental telescope (I'm an ATM, it is my own ATM project, not a commercial offering), so will be a while before I get to it. It WILL happen, though.
Quote:
Dave, I contacted "embeddedrf" last year and they eventually told me they weren't sure if they were going to be going back into production - those prices are really old (see wayback machine for 2002) and no longer valid...
Additionally, those blanks had flat faces I believe. They would not have yielded a deeply curved mirror and would have had non-uniform face thickness if generated to non-flat.
Mike Lockwood
--------------------
Mike Lockwood
20" F/3 MX Starmaster, MX = Mike's eXperiment
14.5" F/2.55 self-built Newtonian
Nine other self-built telescopes, 4.25" to 30"
TeleVue Paracorr, eyepieces
http://www.Loptics.com/
|
JohnH
professor emeritus
Reged: 10/04/05
Posts: 734
Loc: vancouver near the wilds of B...
|
|
I had planned on using a dremel tool to drill a hole and then expand it close to the edge, then finish up with a brass bristle brush.
The EmbeddedRF blank I have, HAS a curve cast into it, allowing the blank to be used at up to an f/2.5 ration. I plan on going to f/2.75 as the first step of an 16 1/2" f/8.3 RC with a 32% CO.
Edited by JohnH (07/07/09 04:02 PM)
|
kfrederick
professor emeritus
Reged: 02/01/08
Posts: 657
|
|
WOW MIKE a f 2.6 It will take a lot of correction/any special coma corrector// kevin frederick
|
BluewaterObserva
Post Laureate
Reged: 05/18/04
Posts: 4968
Loc: Zuni Mtns, NM
|
|
F/2.6 is fairly typical of RC systems no? Sounds real nice to me.
|
Owen
professor emeritus
Reged: 06/21/07
Posts: 513
|
|
As a comment that could be close to heresey... On what are now seen as 'thick' mirrors, could we not grind the back using steel rod, grit and water to a pre-determined depth.
This presumes a well annealed blank, and would have to be done before grinding...
Just a thought...
Owen
|
dave b
Post Laureate
Reged: 05/10/05
Posts: 3529
|
|
Quote:
As a comment that could be close to heresey... On what are now seen as 'thick' mirrors, could we not grind the back using steel rod, grit and water to a pre-determined depth.
This presumes a well annealed blank, and would have to be done before grinding...
Just a thought...
Owen
the russians used to do something along those lines. those mirrors were said to be "cored" or "lightweighted"
coke can sized holes were bored into the glass to make it lighter and help cool the thick glass
sydor does a similar thing, but with a cnc router:
http://www.sydor.com/products-light-weight-substrates.asp
--------------------
dave bonandrini
30" f/5.2 Dobsonian
President of GCAC
Astromart Moderator
|
BluewaterObserva
Post Laureate
Reged: 05/18/04
Posts: 4968
Loc: Zuni Mtns, NM
|
|
Glass is decently expensive. Seems a ricky proposition to drill out material. Maybe not though, I don't really know what it takes to cut glass, or take out material in such a way at all.
|
mark cowan
Vendor (Obsidian Optics)
Reged: 06/03/05
Posts: 2156
Loc: salem, OR
|
|
Lightweighting can be done, but you need to devote the same kind of analysis to it as to the design of cellular mirrors, certainly if going to extremes (but why bother otherwise?  ). My favorite would be single-crystal silicon, but now we're talking $10k a kilo, if memory serves. 
Now just thinking out loud, as this thread has been hopelessly hijacked at this point:
Diamond boring tools cut through pyrex like butter, through quartz like, well, cold butter, and a fairly simple CNC boring machine could be built and well isolated, where the only byproduct is quartz dust mud - just the same as what comes off my generating machine. Certainly doesn't have to run at router speeds, a couple thousand RPM is totally sufficient. Not hazardous if kept wet. Greg Rhode built a diamond sectioning saw for quartz in a shower enclosure, basic idea.
With a little bit of design, that would allow one to pretty much halve the weight of a thin quartz (fused silica) blank while getting the face thickness down to 1/3", perhaps to 1/5" though it's riskier. You could lay it out to match the cell supports.
Fused silica comes extremely well annealed with no visible strain, so the material would take lightweighting very nicely, and the best part is? It's cheap, electrical costs have already been paid, and there are tons of it available! Suitable diamond tools are available from http://www.thk.hk with great service and fast shipping. I like it.
I'll build a machine to do the processing if I could get somebody to agree to do the modeling of the mechanical result, as I don't have the resources to pull that off. I was planning to build a CNC generator to for blanks larger than the 20" my current sine-table generator handles anyway, and this is an extension of that. 
Best,
Mark
Edited by mark cowan (07/09/09 02:31 AM)
|
Cary
Vendor (Optic Wave Labs)
Reged: 07/07/04
Posts: 398
Loc: Rancho Cordova, CA
|
|
Mark,
Let me know what you want and I will design and do the FEA for you.
--------------------
Cary Chleborad
http://www.opticwavelabs.com
|
Previous
Index
Next
Threaded
[Re: 
Edit
Reply
Quote
