Start with a 500 / F3 : it is feasible ?

[Alberto Bugoloni]

Goodmorning everyone.
I'm brand new to this blog (in general of all blogs, so if I'm wrong in something you will forgive me, I hope) and they are just as brand new in telescope construction. However, I happened to buy one, a small Dobson 150 that made me return the love for space but also a lot of bitterness. And, because after a few days, indeed night, of “WOW but then Saturn really does have rings” and of “Ganymede doesn't exist only in science fiction movies”, I said to myself: “it's not enough for me! I want a bigger one”
The greatest, Well… 3 is the perfect number then 150*3.3333 = 500mm. I took a look at the prices and came to the conclusion that the only way to have a bigger one is to build it myself.
So I called a friend of mine who gladly embarked on the company and who fortunately has a lot of more or less precision machinery available and above all has access to various suppliers of glass of any type and thickness, ma fino a 30mm.
Second thing. We have children. And I would like my son to 10-12 years could see through the eyepiece without scales, so I would like to get to a focal length 3 (which is still the perfect number). My 2 year old son is now… so I have time to scratch, make mistakes and start over.
I have dedicated a little’ of time in understanding how difficult it is to get to a similar focal length on a “monster” da 500mm, but now I'm determined to try.
But first I would like to understand a few things (then many others but for now these) and your advice, I'm sure, it will be decisive:
1) 30 millimeters of thickness are enough? From my first accounts, I know that to get to the parabola f3 I have to scratch a little more than 10mm in the center (when it comes to sagitta, the sphere is meant as a reference, the parable or both?). I know not, the problem, which? The deformations?
2) If the problem to the point 1 are the deformations, it would not be enough to work on a glued blanck? The upper disk (to dig) of 30mm and the lower one of 20mm. From what they tell me (but I have no experience) the bonding they use is phenomenal, then, I said to myself, why not just paste (with all the precautions) two discs? Eventually the disk below could also be pre-cut honeycomb style to lighten it but guarantee stiffness cmq.
3) If the 30mm were insufficient and the gluing was unsuitable, start with a 30mm or 20mm meniscus (I would have a chance to have those too, but I'm not really sure for 30mm) therefore already sphericized, could help?

For now I would say that I have already taken away too much time… but anyway let me thank everyone – especially Massimo – for this exciting blog!!

Thank you very much
A

[Massimo Marconi]

Hi Alberto and welcome again !
I moved your message by creating a new dedicated thread, I entered this title because it seems to me to be the synthesis of the speech,( but you can change it if that's not what you had in mind) and it is certainly a topic that interests most of those who are preparing to start “career” from Grattavetro.

Having said that it is difficult to give a summary answer to all your questions, on each of the points you have highlighted a treatise could be written
My opinion is that it is sconsigliatissimo start a machining on those diameters, not to mention the short focal lengths, the risk , on the contrary, the greatest probability is that you will stop shortly after and abandon the project, not because you don't have the skills or wouldn't be able to do it, but because there is no training and experience that allows you to understand the simplest things and tools to gradually apply them to more complex things. With small diameters everything becomes easier and faster, you learn the same concepts and the same techniques in a much shorter time and any error is easily and quickly solved. Once done a little “school mirror” you are able to decide independently what the next step to take is, having experienced the difficulties firsthand , problems and times that, as the diameter increases, increase exponentially.
But, more than mirrors, the human soul is much more complex which often leads us to initiate “mission impossible” for the simple taste of challenges to the limit of logic e, when we get something in our head, all the tips and advice of caution are of little use.
To give a very quick and not in-depth opinion to your questions;

1A – may suffice with an adequate support cell for the mirror.
1B – the sphere for the grattavetro, the conic that is being designed for optical design software. In any case, the difference is a few microns and there is no need for such precision for the focal length, unless telescopes are to be mass-produced.
1- – and

2 – I do not know, I have never tried and I have not read of studies done on glued glass for problems and behavior at an optical and mechanical level.

3 – I very much hope so, because that's exactly what I'm doing right now

Thanks for the appreciation to the Blog, We are very pleased to know that what we do helps to stimulate the curiosity and interest of enthusiasts.

[Giulio TiberinI]

Hi Alberto and welcome.
You say right: A parabolic mirror 20″ (508mm) F3, assumes a 10.58mm deep excavation arrow in the center of the initial spherical cap; and therefore of discreet commitment if it can be recommended by machine, with one of the known methods, such as pendulum grinding with a radius of 3048mm (equal to twice the focal length). Or with an edger on a lathe carriage or a CNC milling machine, or "in copy" (on a sheet metal template cut with that radius). Or again forming the curve in the meniscus, by softening in the oven at 500 ° C).

But the problems would become great in polishing, necessarily to be done with sub-diameter tools "healthy carriers" of astigmatism. Then going up in difficulty to very big problems in the parabolization flaring, which would differ at the edge of the parabola with respect to the edge of the sphere, was i 18,43 microns (that are 18430 nanometers high.), that is an enormity of excavation to do a “free hand” with the maximum tolerance of deviation error of that parabolic surface of only 68,75 nanometers with respect to the theoretical parabola taken as a constructive reference, to obtain final accuracy “minimum wage” lambda / 4 error between maximum peak and valley.

In conclusion, I realize that I have played the part of the devil's advocate a little. But the purpose, before embarking on a journey, is to collect all the information necessary to plan the route, evaluating the degree of difficulty to be overcome to reach the goal, putting at stake availability of time and patience “in reasonable quantities”.

[Alberto Bugoloni]

Hi Massimo, Giulio
Well, meanwhile, thanks for the immediate answers, better than any helpdesk: you can see that there is passion!

You do well to point out the difficulties, probably insurmountable, of such a realization on the first try. I will tell you, actually I was thinking of a Ritchey-Chrétien but I don't know why hyperbole scares me more than the parable .
In any case I will have to think about it a lot. My problem, like that of almost everyone, it is time and the fact that the workshop where I can work is not where I live (Me). But this is also the reason why I have thought about deadlines very far in the time… in the end between attempts, restarts and more, I would really like to finish it by when my son is old enough to appreciate it. If, on the contrary, I had the opportunity to spend a few days a month ahead (or below) to the telescope, maybe it would be different, maybe I wish I had some’ bigger than my current D150 but I would like it now. Anyhow, I'll think about it and don't rule out that while I think about 500 can also make attempts with a 200 or so. Let's say that for now I'm almost tempted to switch to an f4 instead of an f3. Is’ It is true that I would see Vega standing on a stool, but after all ... who ever looks at Vega?? And the nanometer problem would more than halve, if I'm not wrong.

For the initial excavation, I thought not so much about a horizontal pendulum (I also thought of a vertical pendulum, in height I would be there ) but to a machine that I would very gladly build myself. I am still undecided whether to put the eccentric or not, but I understand that it would suit me. Anyhow, more than a whole pendulum I thought of a “pendulum in half” rather, a 2.82°. This is why I was thinking of a machine with one arm (possibly with eccentric but for now let's consider it fixed) able to position the tool parallel to a 2.82 ° inclined plane with respect to the rotating table top. In this way, I could dig the flat face relatively quickly with a 300mm diameter rotary tool, such as diamond blade or diamond cup or something similar. I need to see what's available for purchase. By letting the arm go down vertically but maintaining the inclination of the top, the sphere arises as a result. It is then a question of perfecting or polishing it and real problems arise there, from what I understand and you tell me. All this if I don't get the meniscus (and apparently a 20mm meniscus can go) although I suppose it will need to be fixed anyway. Let's say for now I want to tackle one problem at a time and the first is the sphere.
So my next step will be to understand the testers (if I'm not mistaken Ronchi is the most suitable for a fast focal length) and design-build the machine.
As the glass or glass arrives, I'll keep you up-to-date, hoping you will assist me just to tell me “we all told you!!” but above all assuming you have the time to start the adventure!
Good night and thanks for the welcome, info and warnings.
A

[Giulio TiberinI]

Your "half pendulum" reasoning is correct, because with that system you would get (even digging maybe even just a straight line with an inclination of 2.39 ° from the center to the mirror edge) as excavation a course very close to the desired radius of curvature. The difficulty lies in “curvature” approximate, which may not help right from the start.

On that same way of roughing there are interesting works of realization on Youtube , by Gordon Waite of Waite Research, which I would advise you to watch, if you don't already know them.

For jobs of that heaviness (like making a 20″F3) I believe that the minimum is to have a revolving table, which offers many advantages to learn how to use it.

The advantages are that only one economical three-phase gearmotor is used 220 was the 4 poli, powered in three-phase converted from single-phase with an equally economical Toshiba inverter, with speed regulation from zero to 60 rpm.
The method of dragging the tool resting and ballasted on the rotating mirror, is given by the differential speed resulting from the greater or lesser decentralization of the tool rotation axis with respect to the mirror rotation axis.

For this reason, the resulting "machine" is called "Fixed Post" or "fixed pole" of the tool rotation pin that enters a “glass” center of the tool diameter 12mm.

Moving the tool towards the center of the mirror digs into the edge, making the mirror convex; while vice versa by moving the tool towards the edge, the center is excavated making the mirror concave. It goes without saying, therefore, that there is an internal media decentralization measure that realizes a plan, and that there is another that creates the desired spherical cap, always with only one engine.
The correct measurement to obtain a spherical cap is to use a tool with a diameter of 70% mirror, decentralizing it by a certain amount which is described in a dedicated tutorial movie by Waite “Rough Grinding a 20″ Quartz Telescope Mirror on the Fixed-Post Machine”; such as the Building a 20 series″ F/3.3.

It may also be useful to play around with a fixed post machine simulator, by clicking on this link; https://www.grattavetro.it/simulatore-di-non-solo-lucidatura-specchi-parabolici-software-polsim/

But also read here: https://www.grattavetro.it/forums/topic/posto-di-lavoro-piano-rotante/
where you will find more precise information on the Fixed post and on the decentralization settings.

Getting a good ball is therefore not difficult. The rest is different.

[Alberto Bugoloni]

hello Giulio
Thanks for the explanation.
In fact, data and information are still being collected. Rather, let me share the sequence of notes that I "took".
In particular I am trying to understand the correct sequence of operations at least to get to the glossy sphere. I need schematics… otherwise I get lost in options .
It seems clear to me that the use of a car will be necessary and that probably the best way (always at an amateur level), at least as more documented, is what you indicate of the "fixed post" even if many, I saw, they use a machine equipped with an eccentric. The path imposed by the eccentric on the tool is some arc of circumference, if I'm not wrong.
Let's say that because of the idea I got, without going into details, the steps in case of using a machine, will be these:
1) Mirror smoothing (I think this is what I do). In case of meniscus, regularization of edges and convex surface (and I do it myself with a full diameter concave tool).
2) Mirror edges smoothing.
3) Rough excavation of the sphere, if blank flat or even if meniscus but "quite" deformed with respect to the sphere
until . Macchina Fixed Post. With toroidal tool, extremely rigid, not consumable (for example steel disc, pulley and so on) with a diameter equal to half the diameter of the blank, positioned so that the edges are tangent, being careful not to exceed the desired sagitta. In this case, I suppose considerable pressure can be exerted, perhaps limited only by the power of the motor which may possibly not be able to rotate the blank-tool pair.
b. Macchina Fixed Post. With consumable tool with radius r equal to 70% of the blank radius R, offset by a certain amount (R-r) + 10%R[mm]<x<20%R[mm]. In this case the aim is to hollow out the concave blank and the convex tool, so that the surfaces obtained sphericize each other. I think this is the best approach in case of a deformed meniscus.
C. Machine with eccentric. With sub diameter tool, style diamond cup cutter or similar. Full diameter center-to-center pass, minimum transshipment. Again, I suppose considerable pressure can be exerted, perhaps with lesser limitations than those in point a as the contact surface is smaller. Perhaps it is the fastest way to excavate the approximate sphere. Be careful not to exceed the sagitta.
4) Refinement of the spherical excavation. In the event that a non-consumable tool was used in the previous activity, or the already built tool has worn out too much, the tool will have to be built using the spherical face of the blank as a "mold". Otherwise, the point tool will be used 3)b. During the refinement, the tool will be positioned on the mirror in the manner described in 3)b, taking care to loosen the blank stops to a minimum to allow micro-displacements (it should help prevent astigmatism).
5) Sphere polishing. Tool construction in pitch, starting from the mold of the concave face of the blank, pressure - heating - techniques to reconfigure the surface of the tool after the polishing session? Tool dimensions: full diameter of the blank or slightly less (70%)? Little pressure, reduced speeds, frequent checks.
until . In case of Fixed post, see point 3)b.
b. In case of machine with eccentric, as random as possible but without exceeding the edges too much.
This is a bit of the operation that I imagined after reading here and there, but please, if I'm wrong somewhere - or anywhere – feel free and be kind enough to correct me and above all extend each point with new content (my m is not laziness I'm afraid of writing unspeakable filth).
Last thing (unfortunately or fortunately you can find everything online): the rotating table of the machine, what I do? I have read from steel to aluminum to Corian, to marble to MDF ... If it goes well, the latter is clearly the fastest and least expensive thing, maybe properly painted etc etc ... . In the end, I don't think I'll open a parabolic mirror shop, so it must last in time to create my ...

Thanks ∞
Alberto

[Giulio TiberinI]

Hi Alberto.

As an indispensable premise to all my comments to your many questions, I must remember that each car is different from the others, and the processing on each of them always presupposes accurate and specific experimental control tests for all the activities assigned to it.

This is because the manufacture of a glass mirror, as known, occurs by abrasion, with a very high quantity of mechanical actions (for example of rotation or "to and fro" of a tool) each of which unitarily removes a very small quantity of glass.

Hence the work of building an astronomical mirror, it is governed more by statistics than mechanics.

In the mechanical actions of “and fro”, machines with arms have the defect of being “too precise” in repeating them at the same point of attack or inversion of the motion. And so if there is no immediate corrective action, in a few passes there is a deep local error that is difficult to correct.
While instead working freehand, the natural manual imprecision of the attacks and of the reversal points becomes an advantage, because statistically the errors of opposite sign can automatically cancel each other out.

You write 1): Mirror smoothing (I think this is what I do).

REPLY 1: Ok

You write 2): In case of meniscus, regularization of edges and convex surface (and I do it myself with a full diameter concave tool).

REPLY 2: Ok it's possible. But about the making of menisci, the works that produce a better approach to the spherical cap, occur by dropping the reflective face of the future mirror onto a good driving surface, spherical convex.
You can find an interesting work by searching on youtube for the movie entitled “Slumping a 16 inch diameter telescope mirror”, where a mold is concave milled to create a convex base in refractory gypsum, to be put in the oven to guide the "Slumping" of a 16 "dF3.3 in a sheet of common glass 19mm thick (3/4″).

You write 3): Mirror edges smoothing.

REPLY 3: The bevel of the mirror edge must always be done anyway, in order to avoid chipping during processing. If the 45 ° plane of the bevel will be extended over a 4mm wide surface, it is likely that it will last until the end of processing, saving the risk of chipping while you will have to redo it at an advanced stage of processing

You write 4):Rough excavation of the sphere, if blank flat or even if meniscus but "quite" deformed with respect to the sphere
until . Macchina Fixed Post. With toroidal tool, extremely rigid, not consumable (for example steel disc, pulley and so on) with a diameter equal to half the diameter of the blank, positioned so that the edges are tangent, being careful not to exceed the desired sagitta. In this case, I suppose considerable pressure can be exerted, perhaps limited only by the power of the motor which may possibly not be able to rotate the blank-tool pair.
b. Macchina Fixed Post. With consumable tool with radius r equal to 70% of the blank radius R, offset by a certain amount (R-r) + 10%R[mm]

REPLY 4: From the essay said "Do not bandage your head before having broken it" It is much better to produce a good "spherical cap" immediately, experimenting properly, to avoid the problems you describe.
Its a Fixed post, playing and checking the result of the tool decentralization, I would prefer choice b, to get closer, or even center the spherical curvature leading to the desired arrow depth.

You write 5): C. Machine with eccentric. With sub diameter tool, style diamond cup cutter or similar. Full diameter center-to-center pass, minimum transshipment. Again, I suppose considerable pressure can be exerted, perhaps with lesser limitations than those in point a as the contact surface is smaller. Perhaps it is the fastest way to excavate the approximate sphere. Be careful not to exceed the sagitta.

REPLY 5: It is the rudimentary method used by the experienced craftsmen of the sector, who have the experience that makes up for the roughness. But that method would leave newbies in the middle of the ford of sub-diameter tools…

You write 6): Refinement of the spherical excavation. In the event that a non-consumable tool was used in the previous activity, or the already built tool has worn out too much, the tool will have to be built using the spherical face of the blank as a "mold". Otherwise, the point tool will be used 3)b. During the refinement, the tool will be positioned on the mirror in the manner described in 3)b, taking care to loosen the blank stops to a minimum to allow micro-displacements (it should help prevent astigmatism).
5) Sphere polishing. Tool construction in pitch, starting from the mold of the concave face of the blank, pressure - heating - techniques to reconfigure the surface of the tool after the polishing session? Tool dimensions: full diameter of the blank or slightly less (70%)? Little pressure, reduced speeds, frequent checks.
until . In case of Fixed post, see point 3)b.
b. In case of machine with eccentric, as random as possible but without exceeding the edges too much.
This is a bit of the operation that I imagined after reading here and there, but please, if I'm wrong somewhere - or anywhere - feel free and be kind enough to correct me and above all extend each point with new content (my m is not laziness I'm afraid of writing unspeakable filth).

REPLY 6) The technical basis of the conclusions you arrive at, would require rewriting here, of the existing technique, because the correct or intuitive answers would derive from your knowledge of the manual processing of mirrors, whose actions must be "translated" into mechanical action, to get some with the machine, the same results obtained manually.
About this, and for this reason, usually it is advisable to make at least one mirror by hand “school”. So I think it would be helpful if you download and study well the manual processing described and illustrated very well by Jean Texereau in his book from the 1930s., still and always valid, downloadable in whole or in chapters in PDF format here: http://www.astrosurf.com/texereau/

You write 7): Last thing (unfortunately or fortunately you can find everything online): the rotating table of the machine, what I do? I have read from steel to aluminum to Corian, to marble to MDF ... If it goes well, the latter is clearly the fastest and least expensive thing, maybe properly painted etc etc ... . In the end, I don't think I'll open a parabolic mirror shop, so it must last in time to create my ...

REPLY 7):Chipboard made of particles is not suitable for the construction of the rotating table, nor the MDF made with sawdust (which are not cross-referenced, they both flex, but you need a rigid base of a composite material, as is plywood, even if only poplar; gluing two discs from 12 or vice versa 15 glued overlapping with vinyl, and then painted to make them waterproof.
In various projects of rotating table for mirror processing that you find online, you will see that most are then supported by three wheels of "Rollerblade" skates arranged at 120 degrees very close to the radius of the table, under it.

[Author]

Posts