- This topic has 142 replies, 6 voices, and was last updated 9 years, 1 month ago by
.
-
Posts
-
Giulio, I think I did not understand… I decrease further if the diameter of the holes in the longitudinal reading becomes even more difficult
In addition, the camera is an accessory “supplied”, the readings I'm trying to make them from the eyepiece, with the webcam this type of image ( from behind the eyepiece ) It comes out less sharp and defined.Just to realize, this is the sequence of images on the area 2 starting from intra-focal , in which ( in theory ) the fire for a semi-central area, It should be just after the point of overlap of the slits.
at 45 ° to the center line of the slit is the hair of Mirco
Between one image and the next the longitudinal carriage path has two millimeters !!! tell me if you can figure out which is the image in focus, I can only see that the last three are blurry…
I still think that is escaping me something…
1) I see a fire, as average, the sixth pair (because in reality I do not distinguish which is best in focus between torque 5, 6 or vice versa 7).
For this reason I would have liked instead of Mirco hair, a “Ave Maria” …or the text of” international” written on microfilm, you can be able to read at least one word in order to be considered sufficiently focused.
2) As regards the diameter of the holes, it is my mistake to propose a smaller diameter. In fact I thought that the difficulty could come from the fact that a diameter of R / 100 could still include too large an area (and with great slope) within which search for the focus of the slit (or the hair Mirco).
But I think you're right ! the point is just that you have explained:
Also in this test, it is assumed / approximation that the hole 15 mm discovers a spherical sector, which contains a fire ( center of curvature ) univocal. In reality it is not so, the area still has a different slope from the sphere, This explains why there are many longitudinal point where the image appears in focus.In other words, in images 5,6,7 there is certainly the center of curvature of a sub-sector of the hole 15 mm, while the rest are still out of focus ( slightly ).
Moving longitudinally we only focus in sequence all the internal areas to the hole leaving a little out of focus other.(It must necessarily be so for a simple consideration on the continuity of caustic.)
This explains why the image is never perfectly defined, even when we are in the center of curvature of a portion of the hole, there is the remaining area “dirty” fire.
We should calculate how much must be large hole in a F2.6 so that the change of curvature produces negligible effects, certainly, as you said, narrowing the hole the problem is minimized, I realized just now
Apart from this, there is a test you could do starting from the only thing certain:
The only certain situation and unambiguous that so far I've seen is when the two images overlap, ie crossing the optical axis.
This event can be measured with extreme precision.But this situation is exactly what we measured during the test Foucault.
So I thought that by measuring exactly all the overlapping of the various zones, and analyzing measurements with Foucault, the results should coincide with the foucault-of couder mask previously made tests.Added after:
This verification is not just to see if the results are similar ( if the policy is correct must be ) but why It would be much simpler and much less subjective Foucault run the tests in this way .I too agree on minor subjectivity of caustic on Foucault.
It would remain personal observation type (If we want) pseudo-philosophical, the lesser simplicity of the second plays in his favor until one is satisfied its evaluation ratiocination, strengthened by the star test, and you do not need to prove with a certificate, the precise value of the lambda obtained.
Conversely, for example, There are a majority of “Yankees” which are even less demanding, always using only Ronchi, ending a mirror with the star test without even feeling the need to know the lambda.I think for focal Sopta F5 or use the Ronchi Foucault, or rather of both, sostanzialemnte is a choice which stems from their own experience that may be able to provide security in the simplification of their work, even if the optical quality is not documented it with a star test.
I would totally agree if I were to test a parabolic , but my problem is to measure a hyperbolic, the star test I can use it with the system to complete, for now it would not give indications on the fundamental parameter that interests me, which it is not the lambda but the conic constant.
What I need is to measure the value of the conic constant of the mirror with the highest precision, which should not necessarily be to project, there is a certain tolerance, the important thing is that the measurement is accurate because the secondary will be built basically on the values of the conical constant and the primary focal.
The primary + Secondary will definitely tested on the sky and most likely will be finalized in that context, I believe in fact that is the most reliable test, the eye, on the contrary certificates, never lies…
In fact I think the caustic test is adequate to test much more restrictive needs of those who like you are working on a tool that is not a simple Newton.
My statement was in fact related to the construction of newton, where the Foucault test with Ronchi tell you that your mirror “It is not less than Lambda / N”, while not indicandotene the exact N.
INSTEAD About the caustic test I found this interesting discussion Cloudy nights, (maybe you already know),
http://www.cloudynights.com/topic/478023-caustic-testers-and-other-testers-info-need/Where to post No. 4 is said that there are wrong data in the calculation procedure of Article Schroader book on ATM 1 (I had not perhaps you passed….I do not remember, and I do not know whether the data from which is inspired Lecleire in the test text translated from the French), and in this regard it is recommended to use a software called SIXTESTS: Windows version (2003-10-15, Win32, 372k) downloaded from here:
http://home.earthlink.net/~burrjaw/atm/odyframe.htmQu’ we talk about that error and improvements to the skilled Caustic tests:
http://www.prl.res.in/~library/Banerjee_DPK_1227_98_abst.pdf
Thanks Giulio
I read the article by Jeff Baldwin that the discussion makes reference but not the discussion.
Regarding the software I use for the caustic Foucault test analysis ( towards. 2006 ) that is configured to also analyze the results of this test.Among other night I tried the caustic test using the Couder mask
Apart guessed difficulty with that arrangement to cover / uncover the holes , I must say the fire detection is much easier and more precise.
It depends of course from the fact that the area turns out a larger area and thus a lower aperture ratio and, at the same time, the exterior-interior of a zone rays are more content with respect to the circular hole of 15 mm, then they discover an area with less variation of curvature.Is’ to see how it should be reset ( and it can be used ) the caustic test with Couder mask, maybe further modifying the mask in this way:
What do you think Giulio ? you think might work ? sectors are “tighter” with respect to the circular hole, in the image 3 mm against 15 mm ( it would be better 5-6 mm ), thus an area less variation gradient, while the reflective area ( measured with the cad ) It is exactly the same.
Ù ,mjknhhhc ncxn h7tg5n
cxd .-hb lp .,,,,,,,,,, -nk lst 4445rgvfThis part of the message is of little Daniel
who wanted absolutely ” tell her”, I think they are his greetings 
Ha ha ha, in my little Daniel is right…
I have a question because of the caustic test do not know much:
All the images you made, They were made from different longitudinal positions, but all he is taken on the optical axis (In fact, the 2 cracks appear symmetrical with respect to the center). You should instead move longitudinally, but also laterally in such a way that only one of the two slit images appear in the center of the eyepiece? Then once you find the focus position of a, It moves only laterally since it is the other to be in the center of the eyepiece? so you can know the lateral distance between the two positions…
As you have done instead of the image that seems to focus not have cleavage perfectly del'oculare center and this probably means that it is not in perfect focus. I say huh, that I am not about ferratissomo….It Mirco, what you said is exactly what should be done to measure the distance between the two slits.
The problem is that the distance you see between the two slits varies, in those images, from a few cents up to two to three tenths of a millimeter (in the central series), I cut out pictures, but the eyepiece field is much broader, then the slits for better or worse they are near the center.
Now, at the moment I can not seem to find the point of longitudinal fire with an accuracy less than one millimeter,
the cart of the lateral shift of a few cents may not bring appreciable variations.For this I have to be able to optimize the tester and find ways to make longitudinal accurate readings, then you can concentrate on those transversal, that they are an order of magnitude lower but much more influential on the calculation of the caustic.
I'm confused…
sorry if I insist but you know how I am, if I have a worm on my mind I get to do a lot of questions.
In the picture below I took one of the many images that you posted and I made a red circle. It is roughly on that dot that marks the center of the eyepiece?
Another thing, according to me decrease the radial extent of the windows is counterproductive. I say this because as you see in the image below a smaller window creates a very narrow cone of light in the neighborhood of red dot, which represents the point at which you must put the eyepiece we must move very longitudinally to blur the image. While in the above the cone of light is more accentuated and just move a little to blur the image.
Other thoughts: just to get an idea you could calculate what the longitudinal distance from the point of overlap of the windows should theoretically be the focus point of the caustic so you should better understand which of all pictures taken should be the fairest.
1- and, The eyepiece center falls roughly them.
2 Absolutely right, the reasoning and the graphics do not make a turn
But there is another element to consider: if we enlarge the hole, add sectors of different curvature mirror, so we add other small cones, like that of your second chart, with different focal points, very little but different, making it feel more image in focus, although there is moving longitudinally.3- I tried the simulation but there is not a defined point, there is a range of values in which the combination reading longitudinal + transversal reading occur caustic and then the conical. For each longitudinal increment corresponds in fact an increasing lateral distance from the optical axis. What occurs is the conical their combination.
In any case, the caustic is coincident with the optical axis only in the center of the mirror. for the central areas of the longitudinal displacement is a few cents up to a few tenths ( that cross in the central areas is of the order of thousandths of a mm ), while the outer areas can grow to several millimeters , Obviously in the case of a mirror as my 300/2.6.1)This is what I meant in the previous post. It does not have the lens center being at the center of the two images, but it must be located right above one of them.
So:
As if you were using one slit at a time. Make account to obscure one of the two windows and try to see if you can find the exact point of focus of one, keeping it well in the center of the eyepiece. Then dark window and reopen her twin and you move only sideways until she is at the center. How did you do instead, there was never a window exactly in the center of the eyepiece (apart at the point where the two overlap)…
Stop me if I'm telling a bullshit ...:good:
You're saying the right ! rather, rereading “sacred texts”, the one you've described is just the exact steps, especially for the central areas where the slits are almost overlapping, but also for the devices you better act like you said !
- You must be logged in to reply to this topic.