- This topic has 38 replies, 6 voices, and was last updated 4 years, 4 months ago by
.
-
Posts
-
Hi everyone,
I open this new thread to bring, and to have your opinions, tips, impressions on the new project progress I've begun to undertake.We must say that the project concerns the construction of a parabolic primary mirror, for a Newtonian telescope, 420mm in diameter and focal length of 1900-1950mm or so. For now, the super-precision focal length does not matter to me, because the telescope structure still does not exist and I will adapt accordingly.
The important thing is that it proves to be in the neighborhood of the given length, which I find a good compromise between ease of processing, space telescope and eyepiece ground clearance.With regard to the processing I mean roughing, the fine grinding and polishing machine, while the parabolizzazione could do it by hand in a standar, or always at hand, but by exploiting the rotary plane of the machine, or totally to machine using the method of Carl Zambuto.
For qust'ultima processing phase I will be able to think, although to me today intrigues, at least try, the technique Zambuto.As for the optical tests I think that I will use the classic Ronchi and Foucault and then the 'Hartmann that I really like.
Without an overview of what is the road that I'd like to go, here's the progress I've made so far:
Blank primary:
The primary mirror I realized after numerous attempts at fusing, during which I have tried to achieve a mirror having a ribbed grilling on the precise retro, determined after numerous calculations and structural simulations, but they have not been successful for a number of process issues that I encountered, as integrity of the mold to high temperatures, temperature curves, occurrence of cracks in the glass during cooling, and a thousand other again ...
So after a series of attempts I opted not for Fusing, but for a softening process, which it consists in being able to bring the various pieces of glass that make up the blank to a temperature at which they begin to soften / melt just enough because it firmly sticking to each other. Followed then by adequate cooling curve in order to avoid the occurrence of internal tensions.
The idea was to take a 42cm diameter disc of glass and 10mm thick and place it on top of a convex shape in plaster having a radius of curvature approximately equal to that which would like to have the mirror finished, so that during the softening of the sheet is adagi above assuming the shape of a meniscus.
On the back of this meniscus then a whole series of pieces of glass have been applied, suitably cut and shaped so as to rest perfectly to the curvature of the meniscus on one side and forming a floor dal'altra, in addition to a series of perforated disks in the center, that have both the function of securely connect together the various ribs making the structure rigid, both as anchor points to attach the mirror to the support cell.
And here is the result:Here assemblavo while the various pieces:
Here just before baking, and just after turning off the oven, then the maximum temperature (In fact, it is seen that is a bit 'giallino):
and here the final result:
Just to let you appreciate the difference in temperature in play, between the softening process and the fusing, just look as it appeared the oven just after the shutdown of the same to almost 1100 ° C (cry from the yellow that you saw in the picture first
):In other developments soon…ciao
Ciao, thanks Uraniborg…
We hope then that the results are up to…
Inspection / control of the bank:
First they started with the check blank out of the oven, and it seems that the weld between the various pieces of glass has come good, the low melting interposed between the various junctions, to facilitate the process, He has done his duty to the great. Even by tapping with his knuckles in various points of the glass, He always feels dry beautiful noise, just as we would expect from a single compact body beautiful.
However, analyzing the curvature of the upper surface, I noticed that evidently the glass disk is not able to recline perfectly above the mold convex plaster, In fact, moving the spherometer above it, in different positions, is easily seen as the lancet of the comparator is continuously moves, sign that the curvature of the glass changes from point to point.
vabbeh, It will mean that the step of roughing will be a little longer than expected. In theory I should not have big problems to eliminate all these undulations…
In the following movie you can see how just described:
See you soon…ciao
hello Mirco I think the “trouble”but this is not physiological in this type of realizations.
homogeneity of materials,tensions,cooling mode ecc.portino to not having a
surface “perfect”,feat. as you've pointed out the work will be a little longer.
I follow with great interest
I see about five and a half tenths ripple. I also think just for the kind of work.
Mirco I missed something: What do you mean by the “low dark”? Perhaps the calcium-sodium structure behind the disc pyrex?
hello Giulio,
and, It may not seem like much, but if referring to the radius of curvature we can be seen how this varies between 2300 and 5600 mm. However it should not be a problem to remove the undulations, the only consequence of this is that the upper disk will not be perfectly uniform thickness, but there will be slightly thicker areas and other slightly thinner.There are pieces in the Pyrex blank, It is completely composed of calcium-sodium glass, cut from the same glass batch, to be as safe as possible that the coefficient of thermal expansion of the various pieces were as equal as possible. Pairing pyrex and calcium-sodium, It would not be possible (at least for my modest experience) because they have too different coefficients of thermal expansion and cooling phase be formed first and then the cracks will break all.
The low melting point is nothing more than a fine ground glass (powder) which is then mixed with water to create a kind of fairly thick cream (If you look closely in the picture of a previous post, where I'm placing hand the pieces, you will notice a white stuff between the various pieces of glass, here is what is the low melting point). As a low melting point is used a glass that "rests" at lower temperatures compared to the glasses that are to be welded together in such a way that would facilitate the process. In my case, I arrived, however, at a temperature so that even without the low melting glasses would be attached to the same, but I put it because tests done previously proved a better soldering (more resistant ) compared to non mettercelo.
P.S: E 'was important to choose among all the glass powders to be used as a low melting point , one that had the same coefficient of thermal expansion of the glass to be joined.
I make a small digression, compared to going work, to say a few words about the blank structure.
The final position I have assigned to the rear ribs, They are the result of a series of tests and ideas, that gradually I have refined, until you get the result today.
The simulation with a special software, which it allows me to evaluate the deflection of the upper surface due to its own weight of the glass, It gave the following results:As you can see, and the colors help in this sense, the deflection of the upper surface varies between -1.76nm of zones in giallino (the higher ones) and the -5nm of bluish areas (the lower ones), or a difference, say P-V, attributable to the coupling blank-cell of approximately 3-4nm entire surface support, therefore far more than acceptable.
I have not given the lower central area (about -7nm) because it obscured by secondary.See you soon…Hello to all…
Thanks Mirco. I understood. I believe, however, that the fusion of the small glass crumbs intuitively occurs more easily and a little early in the oven, with respect to the pieces of glass and little more massive conductive, because the small mass of crumbs and the intimate heat contour can bring perhaps slightly, but for first in softening temperature.
Absolutely Giulio, you are right, the crumbs have from them a high surface / volume ratio that tends to soften them before the rest, taking into account that besides being a low melting point, however, softens before…
Tool:
In order to work the blank, already precurved, I could use a glass disc plane, as it is usual to do when we start with the machining of two plane-parallel discs, but I had to build a tool, convex, having the same curvature as the blank.
For the type of processing that I intend to bring forward, or roughing and polishing machine, It is necessary revving, having a diameter equal to 75% the diameter of the mirror, or about 310mm.
To realize I mixed water and gypsum by dentists (kindly offered me the head of the laboratory of my trusted dentist) in proportions 35/100 (or 25% of water and 75% gypsum) I then poured into a circular shape cardboard, I had previously realized and supported above the concave surface of the blank, until obtaining a thickness of about 3cm.
Once the plaster has hardened and dried out for good, I lined the posterior and the lateral surface, with a pair of polyurethane paint from boats, to make the waterproof tool and protect it from dirt.
With regard to the part that will rub against the mirror with the interposition of the abrasive, There are several schools of thought or methods that you can follow, there are those who drowns directly into the plaster casting a whole series of stoneware tiles or hexagonal steel nuts:or as I decided to do myself, pasting of glass squares with the epoxy bicomponent glue.
As mentioned, in primis, I cut a series of squares of about 1,5 cm side by a sheet of glass from 6 mm thick, and then I placed them on top of a sticky paper (in my case scotch) in such a way that they can lift and arrange all together over the glue once spread on the tool.At this point I mixed with about 60g of epoxy sealing film, and I coated with a metal rod (P.S: the glue is very viscous and it is hard to pull) on the curved surface of the tool, taking care not to exceed the 5 min of time that you have available before the beginning of the crystallization process of the glue, I applied and above all glass squares, pressing them well to get a good bonding.
On the back of the tool, I finally drilled a hole, into which I glued a plastic insert, printed with the 3D printer, in whose interior will engage the spherical head of the machine oscillating arm.
To be continued…
Support:
The entire mirror machining the mean accomplish with the blank already mounted on its cell, even because, as for the back is made, It would not be very easy to be fixed on the work surface, not being able to bind with the usual 3 wedges lengo.
In fact, as I have thought from the beginning, inside the holes (8mm) drilled in the center of each glass diskette present in the blank back, I'm going to paste of threaded plastic inserts M6, so as to be able to fix directly the various of the blank cell holder triangles same. This allows me to keep it firmly in position above the mirror cell, preventing the vertical movements, that side (good thing if I have to work it in the car) which will then allow me to not have to predict, nor the mirror support strap, generally necessary with the inclination of the telescope vary, any of the side latches to keep it centered with the axis otic.The cell is a classic cell 18 support points, whose positions have been calculated already before realizing the blank, with the special software GuiPlop, and the triangles of supports have been realized in PLA by means of a 3D printer.
The cell I finally screwed on the back of the blank with the M6 screws, and I is interposed between the supporting triangles and the glass of a thickness of 1.5mm rubber diskettes. Despite this rubber is rigid enough, It will always be the most compliant element in the chain and will he eventually to deform, to the glass leaving the freedom to expand or contract the little that serves as the temperature varies.The support elements were finally mounted on an aluminum plate of thickness 10mm, which will in turn be bound on the rotating surface of the machine. Whenever I want to wash, to clean, make test, on the mirror, I'll just disengage the latches that hold in position the aluminum plate on the rotating floor, and I can xport in blocking all the assembly blank cell +.
And here all positioned on the rotating surface of the machine, during the phase of centering:Well, It has finally arrived to get started with the grinding grain 80…
soon and hello to all… Roughing ("grit" 80):
He eventually began roughing, I'm doing with the grain 80.
At the moment, since there are ripples on the surface of the glass to be processed, I set a relatively low number of the number of revolutions of the rotary plate of the machine, or around 12 giri / min.
Note that this is not a classic work Fixed pure post, but I admit a small right-left shift, the order of 2-3 cm, tool to avoid excessive repetition of abrasion in some typical areas of the right fixed post.
The swiveling out of the tool with respect to the blank below hovers around 15% the tool diameter, then roughly 47mm, and periodically, is the extension of the left-right tool racing, both the amount of the ooze do vary slightly, so as to avoid as much as possible the formation of zonal errors.Mirco, compliment you seem reductive, you're doing a beautiful job
an experimental study from casting to machining of blank ribbed( and knowing you imagine how many tests and checks shall come forth 
) I do not think you've ever seen outside of specific professional environments or research.
I follow everything and I read with interest and enthusiasm, you're discovering a new frontier for all Grattavetro !
- You must be logged in to reply to this topic.