- This topic has 7 replies, 4 voices, and was last updated 4 years, 2 months ago by
.
-
Posts
-
Hi everyone,
Let me introduce myself I am a longtime amateur astronomer, training electronic engineer.
Many years ago, when I was still in high school, I had tried to build myself a Cassegrain from 20 cm, I had managed to get a decent figure with handwork, using the light of Texerau, but then I had various problems and I had to abandon the project, so I would like to start again now, with the knowledge of ing.
First thing, so I was asking you about the configuration of the laboratory.
My idea is the following:
– Bank 1: Zeiss machine
I wanted to recover an old and very solid metal frame that I have available and insert 3 rotating shafts, one for the table and two for the tool arms, independently controlled with V / f controls made with Pic and inverter bridges (electronics are cheap, everything I can move in electronic rather than mechanical I move), all mounted on a tank with drain and taps for effective washing at each change of grain.
– Bank 2: Electronic line.
I have seen on aliexpress 3m linear encoders with sensitivity from 5 microns around 50$, I therefore thought of mounting a cart integral with them for the various tests, to the limit to be associated with a pc (I think a Rasberry Pi 4 is also overkill) and a webcam to simplify testing, mounting them on a welded steel bench integral with the wall.
– Accessories:
> Optical slits: I was thinking of 3D printing a rotary-linear diaphragm with a stroke around the mm to make some optical slits, for example for the source of the focault test.
> I think the spherometer costs me less to buy it.
What do you think?
Given the length of the test bench, I can not exceed a focal length from 1500 mm, what do you say, as a first exercise, a 300 mm f/5 o un 200 mm f/7?Thank you very much and nice to meet you.
Carlo MorHi Carlo and welcome !
A 300 I think F5 is one of the best choices to try your hand at self-building a mirror, because it is quite complex, enough to experiment with hand most of the constructive aspects of a reflection optics, without ever jeopardizing the success of the project, which remains within the reach of even those approaching these creations for the first time.
On the machines, on the other hand, I don't really agree, my opinion is that the machines for the construction of optics are a point of arrival and not a starting point..
I mean, there are so many variables to keep under control during processing that, without adequate manual experience ( with fewer variables) , it would not be possible to determine if any manufacturing errors are due to the technique, from the strategy conceived, or from the machine not working properly.
But I repeat, it's just my opinion and in the field of application of mirror-machines both Giulio and Mirco could be much more precise than me!Hi Massimo, thank you so much.
The point is that, mindful of the previous experience, I really wanted to try to switch to machining also because:
– I think a rotary table is comfortable anyway, so once I make the card for an engine I can also make it for 3 (as I record), although I could make them modular, prepare outputs for additional shafts and eventually re-flash the controller.
– I wanted to play a little’ with the effects of the various races, getting rid of the variability in terms of regularity.
– On the other hand, I think the desk with the electronic scale is useful in general terms, reading a distance on an absolute encoder a 5 microns (there is also the ad version 1 microns but I'm afraid it costs a kidney) it is easier and faster than with a tape measure and a vernier (that if it suits me is centesimal), moreover, it is easier to make a few dozen measurements to have a decent statistical dispersion.
– I enjoy playing with electronics and PID controllers.
Thank you.Hello Carlo
welcome to the blog from me too.
Also according to me a 300 f5 is the ideal choice, as a first realization. Adequately complicated to learn and deal with the main construction difficulties, but still manageable and with not excessive construction times.
As for machine making, a couple of notes:
1) I agree with Massimo, in saying that mastering manual processing is very important in the first place, as it is possible to understand how the various processing techniques modify the surface of the mirror in a different way.
2) Having said that, I do not deny that I too would be for machining , especially for the roughing and polishing phase, these phases are very long and in my opinion boring. The phase in which you really need to have the ability to generate an excellent mirror then, parabolization remains, this phase that I find very complicated to manage exclusively by machine. that, if the parabolization must be done by hand, however, it is necessary first to have acquired the necessary mastery of the techniques. In short, I would not start from scratch with machining, but I wouldn't be super good at hand crafting either, if the final intention is to focus on machining, which is totally different.That said, if you have the technical skills to build a car, I won't say no, also because already only a rotating table (possibly with rotation speed regulator) it can also be very useful in manual processing alone.
I see you don't lack knowledge of electronics, if you need it, I can give you some tips on mechanics. I don't know if it serves a real three completely independent trees, maybe even just enough 1 and then transmit the motion by means of pulleys.I don't think you need the electronic line at the moment, unless you want to know the focal length of the mirror with micron accuracy (useless thing I would say), if ever you just need a linear encoder of about twenty centimeters max, enough to cover the range of motion of the tester (in that range yes, you may need measurement accuracy).
Lastly, I wouldn't focus on the crack at the moment, it could be an upgrade to do in the second place, you will see that even if only with a well-made fixed slit, you'll have a great time and you won't need anything more complicated.
Hello and see you soon
MircoHello Carlo.
I see that you are well determined in your intentions on several fronts!
And it is a good thing in itself and a harbinger of your sure satisfactions.In fact, the Zeiss machine reads both that and, thanks to its three trees, it is more reliable in freeing the processing from the danger of “precise repeatability”, than in a few “rounds” engine could dig area errors much deeper than those caused by the much less efficient manual work. Errors that are much more difficult to recover than manual ones, especially when you are just starting out and you need to launch into experiments.
Certainly today with inverters powered in single-phase that convert it into truly three-phase current, with a greeting without regret to the old “willing ”but insufficient capacitors used to“ build ”a fictitious second phase, necessary to start the rotation of the motor, with a loss of almost half the nominal power, the task of mechanical construction is brilliantly outclassed by the infinite combinations of speeds and random paths to the tool that can be used in the Zeiss.
I therefore think that the car could be the preferential initial target for self-construction and experimentation on a (for example) 300F5 in 19mm thick normal glass (saw that the two mirror discs and tool, they cost a hundred euros, or even less, in industrial glass factories that produce furniture, walkable floors, etc. in glass).
Instead with regard to the rest 3 self-constructive points (measuring bench, for example of very precise focal length; Adjustable slit in different openings; sferometro) I would refer them to your future evaluation of usefulness resulting from the use of the machine, because each of them would certainly provide high precision performances, which, however, is not required by normal mirror construction technology, not just amateur (and name, to create binoscopes where the two optics of the instrument must necessarily be identical).
The reasons for the postponement to later, would be the following:
The measurement of the exact focal length, normally never needed, because in the technical practice the one deriving from the final arrow obtained / evolved in the working is accepted, which may not even be known with a rough accuracy better than a centimeter, due to the practical fact that in the final construction phase of the telescope the distances of the mirrors are finally adjusted, until a star aimed for the purpose is in focus.Ditto for the construction of a slit with variable opening, which is not strictly necessary as a slit from 20 micron is universal, and it is very easy to do by bringing two half razor blades together, the cutting edges have been smoothed out to make them straight.
("Smoothing out" its sharpness and its inevitable microscopic undulations, rubbing the cutting edges on a piece of glass, and then fixing the two blades with double-sided tape on a common base plate, with interposed spacer to the facing cutting edges, consisting of a piece of magnetic tape formerly a videotape, which has a standard thickness of approx 20 microns).Finally, as for the spherometer; It would also be quick and easy to build using a thick multilayer board (like i did) or an old pulley with three holes for three screw-feet arranged at 120 °; in whose central hole is fixed a micrometer unscrewed from the bow of a palmer purchased on market stalls, or scrapped by ISO9000 company metrological procedure.
The spherometer, however, I have found that it is not necessary to make parabolic mirrors with focal ratio equal to or greater than F5., because a full diameter mirror tool is used in machining, which by its nature produces by wet abrasion, between tool and mirror, two complementary spherical surfaces, which reach their equal spherical radius of curvature, as soon as it is seen to disappear in transparency between them, the air bubble which denounces the non-uniformity of the common curvature of the two surfaces in contact, present since the creation of the initial arrow.
On the other hand, when working with sub diameter tools (mandatory for mirrors with a focal ratio lower than F5) it is of ambiguous utility, because the Ronchi i di Foucault tests are much safer and more eloquent.
(the Foucault test records the same draft on a spherical cap both in the center and at the edge of the mirror; and Ronchi shows the perfectly straight lines of the lattice of a sphere).Because I say of ambiguous utility??
Since it is necessary to consider that with the spherometer a significant difference is measured between the plane identified by the three peripheral feet and the central micrometer, from which the radius of curvature of the hypothetical sphere represented by that measure…But with that measure, difference in height between a floor and a point, there is not the slightest certainty that the surface measured with that difference in height (which is the only sure thing) has the course and curvature of a sphere.Thanks to everyone for the valuable answers.
I summarize what I have understood so far:
– Although it is preferable to start getting experience with manual processing, considering previous experience, although dated, building an electronically controlled zeiss machine could be interesting anyway, also thanks to the possibility of removing any (and devastating) periodic errors, for example by giving prime rotational speeds to each other. To the ugly ones, only the axis of the rotary table is used, which is in any case convenient.
– Placing the turntable on a tub for easier cleaning seems like a good idea.
– The processing of a mirror 300 F / 5 as a first test is optimal in terms of times, costs, collected experience and executive difficulty
– The 3m @ 5um electronic scale is not needed. However, since I would have applied it to a general purpose bench, if truly an absolute encoder from 3 m coast 50-60$, gluing a decent meter and applying a vernier on the edge costs absolutely a little less, so at the limit it is overkill but not so devastating (that is, it doesn't matter if the focal length is 1500 or 1508, since it is absolute, I can read the focault around the focal length, and having the 5um in electronic reading is very convenient), then I can also use it for something else.
– I forget the slit and the spherometer, to check the depth of excavation just the ruler and the crescent and the template.Thanks a lot for the information, See you soon.
I agree.
The important thing is that you do not abdicate your aspirations.
In other words, I would not like my advice to be a limitation to your innovative thoughts, towards the limited scope of my specific well-traditional direct experience, but far from innovative in this area.
- You must be logged in to reply to this topic.