Expansion would cause convexity and your finger-heat caused a curvature that goes against the rest of the pattern, so the rest of the surface is concave. I see a wide area of constructive interference on the left of the image, so this would mean an area of less slope in reference to the flat surface. That would mean answer A. (As it is concave and a 'flatter' area relative to the flat surface is on the left) Love the video. :)
in situations where the measured curvature is so small that few or even no lines are visible. When heated enough, the interference pattern inside the convexity caused by an expanding heated spot would always create a continuous loop of a fringe which has a thicker belly facing down the slope if the surrounding surface is concave, as the surface facing up the slope would have a change in the gradient from horizontal in relation to the rest of the surface in cross section, while the surface facing down the slope would appear continuous. This would also make the loops of fringes appear to generate off centre from the heated spot towards the higher side.
Good explanation. I imagine the fringes as height lines of a sphere with a decentered vertex due to the tilt. That's how I can rule out B) and C). The extra info comes from the convex thermal expansion. Digital interferometers usually use phase shifting to determine the correct sign of the surface.
BTW a good analogy is cutting an onion far from the center which yields a similar pattern to the interference of a plane and a spherical wave because of the constant shell thickness.
Same! Honestly optics was the only section of physics I enjoyed when I took it, so this channel is my little guilty pleasure to binge watch every once in awhile😂 it’s a nice break from the other fast paced stuff I watch
I do like this channel for the reason that the content when it comes is well done and not part of the daily information flood. Quality over quantity. Another similar channel is Brainiac75, and I also like the LockPickingLawyer where you'd almost can see the quality of the lock picked featured by the length of the video. But I think that Google has considered that quantity is more important than quality. Project Farm and Mustie1 are channels with weekly content, which is a decent frequency for regular, much like the weekly magazines that used to exist. And for those that don't mind a bit colorful language then AvE is a channel with high and low parts of a lot of hands on stuff. So there are a number of channels there with pretty good quality content. I do a few videos myself at random intervals, and I try to make stuff that do show things that might interest at least someone.
What a great video! I've been working at Zygo for over two years now in our precision lens assembly division that focuses on telescopes. Love seeing these deep dives into the tools we use every day
The tiny "bye bye" at the extreme end of the video caught me off guard. Thank you for explaining all these concepts so thoroughly and clearly. I enjoy all of your videos and happily await the next one.
At my job, I used to work on a few interferometer test stands using similar instruments, so this was a bit nostalgic and it made for a nice refresher course. I appreciate the attention to detail at 16:39, where the reference flat is shown with a significant wedge. Although not explained in the voiceover, I'm sure this is intentional, not a video production glitch. The wedge is necessary because light will reflect off both sides of the optic-not just the reference surface, but the opposite surface too. The wedge puts this unwanted reflection out of alignment so it doesn't pick up in the interferogram.
Fantastic video, thank you! The temperature coefficient of Zerodur's *refractive index* is two orders of magnitude greater than its thermal expansion coefficient, so in transmission (as here) the expansion effect is in fact negligible. You're seeing entirely the effect of dn/dT. As it happens, dn/dT is positive, so given your assertion about the sign, your winner will still get the right answer; but for the wrong reason.
Very interesting, what's the reason for that, my understanding was that thermal effects on refractive index were driven mostly by density changes which is related to the thermal expansion, maybe thats not correct here.
Hi it also felt extremely unlikely that with rise of temperature of temperature 0.5 - 2 Celsius Zerodur had expanded so gravely! (Too lazy to count) So you wanted to say that it was a refraction index that affected the light transmission in Zerodur ? But it was not the thermal expansion.
I have been building many displacement and surface measuring (mainly Fizeau) interferometers over the last 25 years as wel as teaching. This presentation is the best introduction video I ever saw. Thanks for your great explanation: a must see for students!
Nice Interferometer! Amazing visualisation of thermal lensing too :) a perfect way to get a sense of how important temperature control is for ultra-high performance optical systems. Cant wait to see more!
I gotta say your practical knowledge & experience with optics and the equipment is the solid foundation needed to really understand the theory behind it. It gives you an intuitive edge over pure theoretical book learning as taught in universities. Thanks for another solid presentation
Your channel is consistently well produced, informative, deliciously technical, yet reasonably approachable and absolutely fascinating. Thank you for your hard work putting these videos together, they are truely the work of a master. I can't wait to see you jam a real ccd in that thing!
@22:00 ish It strikes me that the dust particles generating waves, is like throwing a pond at a pebble to observe the waves it make when hitting the pebble. wow! I'm guessing C - because the expansion of the glas when you put your finger on it, should mean that to get that bump, in that direction, it would have to be convex. And,,, then I guess it's C, because the interference line is fatter where the surface shifts the most. Not sure but I'll go with that. ;)
The heat from your finger warping the lens blew my mind. Great video. I had a vague understanding of how this worked before but this was so easy to understand.
The art of engineering very accurate surfaces is really fun; part insane accuracy via very hard to build high tech tools and part insanely low tech "stick your finger on it to heat it up" kinda stuff.
i never went to school, i have absolutely no idea what you are talking about 90% of the time, yet for some reason, i watch every single one of your videos without fail
I paused at 25:50 for this: Assuming your glass is BELOW body temperature (a probably reasonable, but not trivial assumption), the heat of your finger causes deviation in a material with a positive temperature coefficient indicates the surface is nominally concave. For clarity, the heat of your finger causes the glass to expand causing a deviation in the Z curve. That the deviation is opposite the general trend indicates the glass is concave (so A or C). That the curve is opposite the general trend of the curves points to C - if the curve were A, heating the middle would emphasize/exaggerate the curves. IFF (if, and only if) the lens is ABOVE your body temperature such that touching the lens cools (and shrinks) it, the B surface would be the best fit. That only the middle curve is influenced by touch is very interesting, but makes sense if the touch is not too long in time. That you touch your lenses with your fingers at all seems heretical, blasphemous, and deprecated! But thank you for sacrificing your lens(es) to educate us!
The solution is A ! warming the spot increases the thickness between the interfaces: 1. The fringe moves to the left, away from the region of higher thickness. So the thickness has to be higher at the right side (A or B). 2. The fringes incircle minima or maxima of thickness. So the left edge has to be a global maxima. Using the first conclusion it has to be a global minima, so the surface is concave (A or C).
Thanks for this fascinating video! I've been interested in interferometers, since I first heard of the Michelson-Morley Experiment. Later, I worked on ambient air quality and stack gas monitoring and came across interferometry again, this time to measure individual gases in a sample or continuous flow.
I don't work in a field anywhere adjacent to optics or interferometry. However, your videos are so thorough and well-explained, I am able to follow what you're saying with just my the undergraduate physics education I got as part of my engineering degree. I don't know why it works so well but you method of delivery makes this extremely complex topic very interesting!
Thanks for this video. 50+ years ago I worked in a calibration lab using optical flats. In our case we were often calibrating the accuracy of working gauge blocks (device under test) compared to the reference gauge block. We would place them a known distance apart and bridge the 2 blocks with an optical flat. Using a monochromatic light (or course) we looked at the interference pattern and calculated the difference in length between the 2 gauge blocks. We also reported on the surface deviation of the block under test. Of course, being so long ago, I remember nothing of the calculations, only that the procedure was quite demanding, and accurate if done well! And by "calibration", I don't mean any adjustment was performed, only that we reported the precise length and surface characteristics of the block under test.
I've been working at Zygo for almost 3 years now. Although I'm just a machine maintenance mechanic, I'm surrounded by this stuff every day, and it's very cool to see it in use "in the wild"
Im just wanted to say how much I appreciate this channel! Probably its my favorite channel on whole RUclips! Im saying it without any exaggeration! After I started studying cinematography in uni I just got obsessed with physics of light. Thank you!
I’m a carpenter and through have developed an interest in precise measurements. I really appreciate your clear descriptions of these natural phenomena and how the instruments are able to detect and display them. It’s , shall we say, unlikely that I’ll be cutting any boards with this level of precision, but I really enjoy knowing the outer limits of what’s possible with relatively common materials and relatively inexpensive instruments.
I think it's a testament to the sensitivity of interferometry that the distortion due to temperature change from touching a piece of _Zerodur_ is so easily observed.
I use an optical parallel and a monochromatic light to identify imperfections in a polished surface. Lines, some curved, indicate that the surface is not smooth and flat. It's simpler but it helps me understand the topic of your discussion. Thank you for posting.
I work in a metal workshop and we have a lot of old precision measuring tools. In some of the kits there are these glass pucks that have an arrow on the side that points to one face and 0.2mkm written next to it. Can these be test glasses that were used to measure surface flatness?
I'm pretty sure they are. The arrow generally indicates the flattest side. If you clean then and put two on top of each other under fluorescent light you will probably see straight fringes.
Thank you for showing and explaining us insides of such great and rare instrument! Most of us have no chance to work with them by ourselves. In my experiments with interferometry for evaluating the surface roughness I was limited to use cheap laser diode, DSLR and UV photo filter as reference surface to get interferograms)) Working with high quality interferometer is just totally different level. Thank you!
Very nice video--contained the core of an 'introduction to optical testing' course. In my career, I never did have the opportunity to open a Zygo, so that was fun to see. By the way, the alignment system was the subject of a rather vicious patent suit between Zygo and Wyko--unfortunate, because we used that technique in graduate school long before either company used it in their commercial devices. (I'm kind of biased, though, because I was a grad student of JC Wyant.) Looking forward to subsequent videos!
A? Because the expansion of the Zerodur should cause convex bump, and we see that the expansion counteracts the local curvature, meaning the local curvature is concave, hence A or C. And because the expansion should move the surface of the glass further away, and we see the fringes move left that suggests, further = left. At 10:28 in the video we had an animation that suggested that with negative slope, closer = left, so we must have the opposite of that, positive slope. So positively sloping concave is answer A.
Respect for such a strong passion and congratulations on your new acquisition! The right answer is A. 1) The interval between the fringes decreases from left to right, which means that the thickness increases in the same direction => A or D = true. 2) A local increase in thickness leads to a shift of the fringe to the left => A or B = true. 1),2) => A is true. Incidentally, the following question arose. Where is the focus of the CCD1 sensor lens? It looks like the optical system must transform the flat front on the input of the interferometer into a flat front at CCD1? It doesn't form the optical image of the testing surface, right?
Yes A is correct. About the focus, this is realized with the lens in front of the camera. Standard it is sufficiently focused between 0.1 and 1 m ( I think). A correct interferogram can only be recorded if the surface under test is in focus on the ccd.
@@HuygensOptics Understood, thanks. Probably, stable interference occurs in any cross section of the overlapping beams, since the coherence length is quite large. But to see the surface, we must look at the surface)
A) The lines become closer together on the right. So the magnitude of the slope is greater on the right than on the left. (The sample is sloping up everywhere, unlike the diagram A, but whatever, the diagrams aren't to scale.) Thus A or D. What's more, when it expands in the middle, the lines get straighter and move to the left. The small amount of expansion seems to have totally compensated for the slight warping, at least in the place where it is hottest. If it was already convex, warming the middle would enhance the effect. So it's concave. Or was. But the slight thermal expansion made it a lot less concave. Thus A or C Combined, answer is A.
I came to learn about optics but stayed until the very end and heard the sensual goodbye and now I'm more confused than ever. To be clear, the optics explanations were first class....
What a great apparatus and while the resolution of that old camera is nothing remarkable, it still offers a massive amount of functionality as a tool. Incredible stuff thanks for showing
I love your videos! I am working now since 11 years in the field of experimental quantum optics, and I leaned something in every video of yours. Whenever you upload a video i tell my wife that i need to watch it and she needs to take care of the baby and that I am not available for lunch/dinner or whatsoever until i watched it! Maybe I remember this incorrectly, but shouldn't there be a part 3 of the mirror lenses series? Cheers
Very interesting and enlightening video! I'm using interferometry frequently when checking the flatness of lapped/polished surfaces of metallic sealing discs for industrial safety valves. Our setup is a lot more barebones though, with just a monochromatic lamp and the prisms.
God, I'm glad that I subscribed to Your channel. Having nothing in common with this I was always fascinated by optical-kung-fu-black-magic, interferometry in particular, and now I roughly understand it. Thank You. Seeing this "camera" I instantly knew that it would be replaced with something much, much better. The rest is superb, so it was kind of obvious.
Thank you very much for yours videos there are super and full of knowledge and experience, I think that the right answer is D because the thermal expansion from the back side of the material tries to correct the center in the interferometer path. Have a nice weekend, Always the best.
23:24 One caveat is Zerodur {tm Scott AG} is not an optical glass used in refractive optics. So, it's index of refraction can vary by a certain amount.
10:00, Is it like moiré fringes used in subsampling of position encoders? So you have 2 gratings that differ slightly. (for example one has 10 lines per mm, and the other 11 lines per mm and if you look through both grated stencils, you can see 0,01 mm instead of 0,1 mm).
Congratulations for your GPI LC! "LC" in fact stands for "low cost" and that is true in your case because the TS that came with your Instrument is easily worth the money if it is in good condition. May your laser last long. I saw lifetimes between 2 and 20 Jears regardless of use.
The glass gets thicker due to the higher temperature, and this region will be more convex. The original interference pattern shows opposite curvature of the disturbance, so the orginal surface must be concave. The line movement shows that the higher side is to the right, so the correct answer should be A.
I could send you an extra zygo PZT housing but normally it plugs into the housing. You would have to build a driver to drive the piezos from 0-1000volts. @@HuygensOptics
I dream of creating a laser measuring system for machining. Something based on interference, that's able to immediately measure fine structures in real time while machining. I'm certain that one or two lasers can achieve more than we know so far.
by measuring the dimension of the interference pattern (and knowing the thickness and refractive index of surfaces involved) can we obtain the light wavelength itself?
6:40 - He calls GHz low frequency 😂😂 Things I though I would never hear (I dabble in the electronics as a hobby) But yeah low frequency is relative I guess 😄
If the beam expansion has more than about 2 waves of error, there can be retrace errors that will throw off the error you measure in your part as the rays get moved around on the camera..
In the demonstration of measurement, I see clearly 2 surfaces that reflect light if using zerodur test glass (21:09). However in case the test surface is not attached to the reference surface 19:0519:44, there are probably more than 2 surfaces and even much more than 2 surfaces if using a colimating lens system. How do you evaluate the result that case?
Good question. The way you achieve this is by making sure the reflections of these surfaces do not reach the detector. It's illustrated at 19:05 in the reference element by the slight tilt in the other surface that effectively places its reflection outside the area of the small beam splitter. In the case of the object, you can reduce reflection (of for example the last surface) temporary with black spray paint. If that surface has a radius of curvature significantly different from the surface under investigation, then generally the resulting nr of fringes from this surface will be too high to spot them in the interferogram.
Utterly fascinating as always. I have absolutely no need for an interferometer, but would still love to have one, as I have a measurement fetish: The process I’d measurement itself makes me happy (especially of something that’s invisible to our normal senses), whether or not I need the results for anything 😂
I remember (a long time ago) swimming front crawl against a slightly taller boy in a short course pool. Every tumble turn he got an advantage of twice the difference in height (assuming the difference in leg length was proportional to the difference in height). That must totally explain the difference in our speed.:-). Anyway, just thinking about the factor of two at 22.48
Generally, when precision lasers are used, you'd probably want to prevent reflections from getting back into the laser. In the block schematic you drew for the Zygo interferometer, it looks as if a big fraction (12.5%, assuming 50-50 splitters) of the laser power will make it back into the laser, after reflections. Maybe, then, something is tilted a bit? Maybe the position of the GRIN lens with respect to the beam splitter with the beam dump is such that on its way back towards the laser, the beam diverges? Or it doesn't matter if all that light makes it back into the laser? I don't think you want any significant reflection back into such a precision laser. I've played with interferometers; they can be so sensitive that local air-temperature gradients show up as noticeable fringe shifts. Such (varying) reflection could then detune the laser by injection locking. I am not an optics specialist, so I could be wrong.
Yes that is correct. at 13:12 you can see that the returning reflection on the 45 degrees steering mirror is quite a bit off from the original beam (it is on the top of the mirror). So I guess generally the reflected light does not make it back into the laser cavity.
I'd say it's C, because touching it moves the fringes back where they should be if it was flat, and the positive coefficient of thermal expansion means that it has to get thicker to be flat, so it is concave. And because it getting thicker moves the fringes to the left, thats also the direction where the rest of the glass is thicker.
Your finger made the glass thicker at that spot and kind of "corrected" the bowing of that dark fringe line back to the left. This makes me think the thickness where you put your finger was too small which leaves A or C (it is concave, too low in the center). Traveling from right to left across the middle of the glass the finger made you reach the 3rd bright line sooner. So I think the thickness is increasing from right to left. So that would leave A, final answer!
Jeroen, have you ever tried making a germanium lens for a thermal imager? All I know so far is that some of them look amazing. I wonder how different the lapping process might be compared to glass, or even whether lapping is still a suitable process in this application.
I have lapped Germanium. It's softer and requires you to work very cleanly. Also, you use Al2O3 rather than CeO as a polishing agent. I've never done a video about it though.
I concluded it was likely concave before the heat hint by noticing that the edges we're slightly rolled off (very unlikely to have a high edge). Since the fringes in the center are shifted in the same direction as the fringes at the very edge, the center must also be low, thus concave.
Knowing the the rolled edges shift the fringes to the right also let's us determine the overall wedge. Intuitively the fringe is shifted right at the edge because we don't need to go so far to the left to drop to the next fringe as we do in from the edge, thus the wedge is falling off towards the left, case A.
I know it has to be A or C, the concave options, because the fringe curve induced by the heat of your finger is in the opposite direction of the curve of the fringes of the surface, and the induced local curve from positive coefficient of expansion would be convex since it would be a “bump” in the surface. Since we know it is a concave sphere, and the fringes “point” to the right, the thicker part must be on the left, so the answer is C.
finger trick was awesome. So it was concave because expansion made it "grow" up and it became flatter, and it's a C because the lowest (right) part of the wedge seems to be the flattest
![Blox Fruits Dragon Rework Update [Full Stream]](http://i.ytimg.com/vi/EqDAp8udhm0/mqdefault.jpg)
Expansion would cause convexity and your finger-heat caused a curvature that goes against the rest of the pattern, so the rest of the surface is concave.
I see a wide area of constructive interference on the left of the image, so this would mean an area of less slope in reference to the flat surface.
That would mean answer A.
(As it is concave and a 'flatter' area relative to the flat surface is on the left)
Love the video. :)
We have a winner!
I was thinking it was D as I was thinking the opposite of you.
in situations where the measured curvature is so small that few or even no lines are visible. When heated enough, the interference pattern inside the convexity caused by an expanding heated spot would always create a continuous loop of a fringe which has a thicker belly facing down the slope if the surrounding surface is concave, as the surface facing up the slope would have a change in the gradient from horizontal in relation to the rest of the surface in cross section, while the surface facing down the slope would appear continuous. This would also make the loops of fringes appear to generate off centre from the heated spot towards the higher side.
Good explanation. I imagine the fringes as height lines of a sphere with a decentered vertex due to the tilt. That's how I can rule out B) and C). The extra info comes from the convex thermal expansion. Digital interferometers usually use phase shifting to determine the correct sign of the surface.
BTW a good analogy is cutting an onion far from the center which yields a similar pattern to the interference of a plane and a spherical wave because of the constant shell thickness.
I watch a lot (too much) of YT, but for some reason whenever this channel has new content I get most exited. It's all just so fascinating.
Same! Honestly optics was the only section of physics I enjoyed when I took it, so this channel is my little guilty pleasure to binge watch every once in awhile😂 it’s a nice break from the other fast paced stuff I watch
I do like this channel for the reason that the content when it comes is well done and not part of the daily information flood. Quality over quantity. Another similar channel is Brainiac75, and I also like the LockPickingLawyer where you'd almost can see the quality of the lock picked featured by the length of the video.
But I think that Google has considered that quantity is more important than quality.
Project Farm and Mustie1 are channels with weekly content, which is a decent frequency for regular, much like the weekly magazines that used to exist. And for those that don't mind a bit colorful language then AvE is a channel with high and low parts of a lot of hands on stuff. So there are a number of channels there with pretty good quality content.
I do a few videos myself at random intervals, and I try to make stuff that do show things that might interest at least someone.
Nailed it!
You can't be watching more than me
Excited*
What a great video! I've been working at Zygo for over two years now in our precision lens assembly division that focuses on telescopes. Love seeing these deep dives into the tools we use every day
The "elite" is exploiting you and want you docile and brainwashed 👉 The Connections (2021) [short documentary] 💖
The tiny "bye bye" at the extreme end of the video caught me off guard. Thank you for explaining all these concepts so thoroughly and clearly. I enjoy all of your videos and happily await the next one.
At my job, I used to work on a few interferometer test stands using similar instruments, so this was a bit nostalgic and it made for a nice refresher course.
I appreciate the attention to detail at 16:39, where the reference flat is shown with a significant wedge. Although not explained in the voiceover, I'm sure this is intentional, not a video production glitch. The wedge is necessary because light will reflect off both sides of the optic-not just the reference surface, but the opposite surface too. The wedge puts this unwanted reflection out of alignment so it doesn't pick up in the interferogram.
The "elite" is exploiting you and want you docile and brainwashed 👉 The Connections (2021) [short documentary] 💖
Fantastic video, thank you! The temperature coefficient of Zerodur's *refractive index* is two orders of magnitude greater than its thermal expansion coefficient, so in transmission (as here) the expansion effect is in fact negligible. You're seeing entirely the effect of dn/dT. As it happens, dn/dT is positive, so given your assertion about the sign, your winner will still get the right answer; but for the wrong reason.
Very interesting, what's the reason for that, my understanding was that thermal effects on refractive index were driven mostly by density changes which is related to the thermal expansion, maybe thats not correct here.
The "elite" is exploiting you and want you docile and brainwashed 👉 The Connections (2021) [short documentary] 💖
Hi it also felt extremely unlikely that with rise of temperature of temperature 0.5 - 2 Celsius Zerodur had expanded so gravely! (Too lazy to count)
So you wanted to say that it was a refraction index that affected the light transmission in Zerodur ? But it was not the thermal expansion.
I have been building many displacement and surface measuring (mainly Fizeau) interferometers over the last 25 years as wel as teaching. This presentation is the best introduction video I ever saw. Thanks for your great explanation: a must see for students!
Nice Interferometer! Amazing visualisation of thermal lensing too :) a perfect way to get a sense of how important temperature control is for ultra-high performance optical systems. Cant wait to see more!
I gotta say your practical knowledge & experience with optics and the equipment is the solid foundation needed to really understand the theory behind it. It gives you an intuitive edge over pure theoretical book learning as taught in universities. Thanks for another solid presentation
Your channel is consistently well produced, informative, deliciously technical, yet reasonably approachable and absolutely fascinating. Thank you for your hard work putting these videos together, they are truely the work of a master.
I can't wait to see you jam a real ccd in that thing!
Yet another clear and digestible explanation of complicated optical principles for the non-physicist. Bravo!!!
@22:00 ish
It strikes me that the dust particles generating waves, is like throwing a pond at a pebble to observe the waves it make when hitting the pebble. wow!
I'm guessing C - because the expansion of the glas when you put your finger on it, should mean that to get that bump, in that direction, it would have to be convex. And,,, then I guess it's C, because the interference line is fatter where the surface shifts the most. Not sure but I'll go with that. ;)
The heat from your finger warping the lens blew my mind. Great video. I had a vague understanding of how this worked before but this was so easy to understand.
I winced when he said, "Let's touch the surface."
The art of engineering very accurate surfaces is really fun; part insane accuracy via very hard to build high tech tools and part insanely low tech "stick your finger on it to heat it up" kinda stuff.
i never went to school, i have absolutely no idea what you are talking about 90% of the time, yet for some reason, i watch every single one of your videos without fail
Be careful, I've been doing it for a few years, and I think I'm beginning to understand some of it. I can never go back.
I paused at 25:50 for this: Assuming your glass is BELOW body temperature (a probably reasonable, but not trivial assumption), the heat of your finger causes deviation in a material with a positive temperature coefficient indicates the surface is nominally concave. For clarity, the heat of your finger causes the glass to expand causing a deviation in the Z curve. That the deviation is opposite the general trend indicates the glass is concave (so A or C). That the curve is opposite the general trend of the curves points to C - if the curve were A, heating the middle would emphasize/exaggerate the curves. IFF (if, and only if) the lens is ABOVE your body temperature such that touching the lens cools (and shrinks) it, the B surface would be the best fit.
That only the middle curve is influenced by touch is very interesting, but makes sense if the touch is not too long in time.
That you touch your lenses with your fingers at all seems heretical, blasphemous, and deprecated! But thank you for sacrificing your lens(es) to educate us!
The solution is A !
warming the spot increases the thickness between the interfaces:
1. The fringe moves to the left, away from the region of higher thickness. So the thickness has to be higher at the right side (A or B).
2. The fringes incircle minima or maxima of thickness. So the left edge has to be a global maxima. Using the first conclusion it has to be a global minima, so the surface is concave (A or C).
This type of channels and videos - one of the reasons why i'm still believe in youtube. It's just amazing information, thank you again for the video.
Thanks for this fascinating video! I've been interested in interferometers, since I first heard of the Michelson-Morley Experiment. Later, I worked on ambient air quality and stack gas monitoring and came across interferometry again, this time to measure individual gases in a sample or continuous flow.
I don't work in a field anywhere adjacent to optics or interferometry. However, your videos are so thorough and well-explained, I am able to follow what you're saying with just my the undergraduate physics education I got as part of my engineering degree. I don't know why it works so well but you method of delivery makes this extremely complex topic very interesting!
Thanks for this video. 50+ years ago I worked in a calibration lab using optical flats. In our case we were often calibrating the accuracy of working gauge blocks (device under test) compared to the reference gauge block. We would place them a known distance apart and bridge the 2 blocks with an optical flat. Using a monochromatic light (or course) we looked at the interference pattern and calculated the difference in length between the 2 gauge blocks. We also reported on the surface deviation of the block under test. Of course, being so long ago, I remember nothing of the calculations, only that the procedure was quite demanding, and accurate if done well! And by "calibration", I don't mean any adjustment was performed, only that we reported the precise length and surface characteristics of the block under test.
Ladies and gentlemen, the Oscar for the most amazing scientific video 2023 goes to... Huygens Optics!
New videos from @HuygensOptics are always the best part of my day
Aaaah, this tickled the mind in just the right way 😊
Science is awesome, and your way to explain and visualise everything is excellent.
I just started a project related to phase-measurement interferometry this week, so the timing of his video could not be any better!
one of those super high-qual channels. always a pleasure.
I've been working at Zygo for almost 3 years now. Although I'm just a machine maintenance mechanic, I'm surrounded by this stuff every day, and it's very cool to see it in use "in the wild"
that collimating lens is a work of art
Love your videos! (Can’t wait to see you finish the catadioptric lenses too)
This was the perfect video to follow your Coherence series.
Finally a pure optics video again, very nice! Really looking forward to the follow ups! It is super interesting, how precise optical metrology is.
Your videos always leave me with a powerful sense of how much there is to learn. I can't wait to see these tests you tease us with!
Im just wanted to say how much I appreciate this channel!
Probably its my favorite channel on whole RUclips! Im saying it without any exaggeration!
After I started studying cinematography in uni I just got obsessed with physics of light. Thank you!
I’m a carpenter and through have developed an interest in precise measurements. I really appreciate your clear descriptions of these natural phenomena and how the instruments are able to detect and display them. It’s , shall we say, unlikely that I’ll be cutting any boards with this level of precision, but I really enjoy knowing the outer limits of what’s possible with relatively common materials and relatively inexpensive instruments.
I think it's a testament to the sensitivity of interferometry that the distortion due to temperature change from touching a piece of _Zerodur_ is so easily observed.
I use an optical parallel and a monochromatic light to identify imperfections in a polished surface. Lines, some curved, indicate that the surface is not smooth and flat. It's simpler but it helps me understand the topic of your discussion. Thank you for posting.
I work in a metal workshop and we have a lot of old precision measuring tools. In some of the kits there are these glass pucks that have an arrow on the side that points to one face and 0.2mkm written next to it. Can these be test glasses that were used to measure surface flatness?
I'm pretty sure they are. The arrow generally indicates the flattest side. If you clean then and put two on top of each other under fluorescent light you will probably see straight fringes.
Congratz on your new interferometer , you deserve it! , The content and perspective you provide is amazing! ✨✨✨
Thank you for showing and explaining us insides of such great and rare instrument! Most of us have no chance to work with them by ourselves.
In my experiments with interferometry for evaluating the surface roughness I was limited to use cheap laser diode, DSLR and UV photo filter as reference surface to get interferograms)) Working with high quality interferometer is just totally different level. Thank you!
Very nice video--contained the core of an 'introduction to optical testing' course. In my career, I never did have the opportunity to open a Zygo, so that was fun to see. By the way, the alignment system was the subject of a rather vicious patent suit between Zygo and Wyko--unfortunate, because we used that technique in graduate school long before either company used it in their commercial devices. (I'm kind of biased, though, because I was a grad student of JC Wyant.) Looking forward to subsequent videos!
A?
Because the expansion of the Zerodur should cause convex bump, and we see that the expansion counteracts the local curvature, meaning the local curvature is concave, hence A or C.
And because the expansion should move the surface of the glass further away, and we see the fringes move left that suggests, further = left.
At 10:28 in the video we had an animation that suggested that with negative slope, closer = left, so we must have the opposite of that, positive slope.
So positively sloping concave is answer A.
This is one of my favorite channels on YT. Always happy when I see a new video come out!
Where was this channel when i was in grad school? Students have it so easy now (if they can find such good information in an ocean of garbage)!!!
Respect for such a strong passion and congratulations on your new acquisition!
The right answer is A.
1) The interval between the fringes decreases from left to right, which means that the thickness increases in the same direction => A or D = true.
2) A local increase in thickness leads to a shift of the fringe to the left => A or B = true.
1),2) => A is true.
Incidentally, the following question arose. Where is the focus of the CCD1 sensor lens? It looks like the optical system must transform the flat front on the input of the interferometer into a flat front at CCD1? It doesn't form the optical image of the testing surface, right?
Yes A is correct. About the focus, this is realized with the lens in front of the camera. Standard it is sufficiently focused between 0.1 and 1 m ( I think). A correct interferogram can only be recorded if the surface under test is in focus on the ccd.
@@HuygensOptics
Understood, thanks. Probably, stable interference occurs in any cross section of the overlapping beams, since the coherence length is quite large. But to see the surface, we must look at the surface)
New optics video, know its gunna be a good day
I like how this video also tests the YT compression very nicely 😄
A)
The lines become closer together on the right. So the magnitude of the slope is greater on the right than on the left. (The sample is sloping up everywhere, unlike the diagram A, but whatever, the diagrams aren't to scale.) Thus A or D.
What's more, when it expands in the middle, the lines get straighter and move to the left. The small amount of expansion seems to have totally compensated for the slight warping, at least in the place where it is hottest. If it was already convex, warming the middle would enhance the effect. So it's concave. Or was. But the slight thermal expansion made it a lot less concave. Thus A or C
Combined, answer is A.
Very well done video! I appreciate the effort you put into the explinations and production quality!
This video is an outstanding piece of reference material!
Thank you for creating this highly educational video lecture!
Absolutely phenomenal, thank you so much for sharing! Your presentation style and the subjects you discuss are excellent.
I came to learn about optics but stayed until the very end and heard the sensual goodbye and now I'm more confused than ever. To be clear, the optics explanations were first class....
What a great apparatus and while the resolution of that old camera is nothing remarkable, it still offers a massive amount of functionality as a tool. Incredible stuff thanks for showing
Dude...this is AWESOME! Amazing presentation
I love your videos! I am working now since 11 years in the field of experimental quantum optics, and I leaned something in every video of yours. Whenever you upload a video i tell my wife that i need to watch it and she needs to take care of the baby and that I am not available for lunch/dinner or whatsoever until i watched it!
Maybe I remember this incorrectly, but shouldn't there be a part 3 of the mirror lenses series?
Cheers
Yes, that is correct, I did not get to that. But now I have a good instrument so hopefully I 'll find tye time for it in the near future.
Cool, looking forward to the next Videos... i actually dont care abou the topic :)
Awesome piece of tech! Thanks for sharing!!
Your content is so articulate and clear! Thank you for sharing your work.
Very interesting and enlightening video! I'm using interferometry frequently when checking the flatness of lapped/polished surfaces of metallic sealing discs for industrial safety valves. Our setup is a lot more barebones though, with just a monochromatic lamp and the prisms.
Always a good day when I get a dose of optics theory from you, I barely understand some of the things but I find it extremely interesting nonetheless
Very informative video, thank you for sharing your knowledge.
I just wanted to say "thanks" for making such wonderfully informative content.
Only a few minutes in and I’m enjoying the video. Thank you 😊 ❤
God, I'm glad that I subscribed to Your channel. Having nothing in common with this I was always fascinated by optical-kung-fu-black-magic, interferometry in particular, and now I roughly understand it. Thank You. Seeing this "camera" I instantly knew that it would be replaced with something much, much better. The rest is superb, so it was kind of obvious.
Thank you very much for yours videos there are super and full of knowledge and experience,
I think that the right answer is D because the thermal expansion from the back side of the material tries to correct the center in the interferometer path.
Have a nice weekend, Always the best.
Great video, I am SO excited for the next one!
23:24 One caveat is Zerodur {tm Scott AG} is not an optical glass used in refractive optics. So, it's index of refraction can vary by a certain amount.
Need to get Tom Lipton and Robin Renzetti to see this!
And Spencer Webb!
Another awesome video about a subject that is not well explaned anywhere I know. Thanks
10:00, Is it like moiré fringes used in subsampling of position encoders? So you have 2 gratings that differ slightly. (for example one has 10 lines per mm, and the other 11 lines per mm and if you look through both grated stencils, you can see 0,01 mm instead of 0,1 mm).
This is great. Really interesting. Thanks for posting and explaining so well
i love to see your videos. thank you so mutch. there is no place to become such good information like you. thanks.
❤❤ Much Love, hope you never stop making videos ❤❤
Congratulations for your GPI LC!
"LC" in fact stands for "low cost" and that is true in your case because the TS that came with your Instrument is easily worth the money if it is in good condition.
May your laser last long. I saw lifetimes between 2 and 20 Jears regardless of use.
The glass gets thicker due to the higher temperature, and this region will be more convex. The original interference pattern shows opposite curvature of the disturbance, so the orginal surface must be concave. The line movement shows that the higher side is to the right, so the correct answer should be A.
Spectacular! You should retrofit it with some piezos to do PSI!!
That is certainly an option I would like to explore!
I could send you an extra zygo PZT housing but normally it plugs into the housing. You would have to build a driver to drive the piezos from 0-1000volts. @@HuygensOptics
I dream of creating a laser measuring system for machining. Something based on interference, that's able to immediately measure fine structures in real time while machining. I'm certain that one or two lasers can achieve more than we know so far.
by measuring the dimension of the interference pattern (and knowing the thickness and refractive index of surfaces involved) can we obtain the light wavelength itself?
Sure, that is possible
I used it a lot to aline fixturing for polishing and machining.
6:40 - He calls GHz low frequency 😂😂
Things I though I would never hear (I dabble in the electronics as a hobby)
But yeah low frequency is relative I guess 😄
15:25 why are all those special components actually needed here? What are the advantages compared to a regular beam expander?
If the beam expansion has more than about 2 waves of error, there can be retrace errors that will throw off the error you measure in your part as the rays get moved around on the camera..
@@jpurrazzella ah right, so that gradient index lens is essential. Yup, this would be hard to diy
In the demonstration of measurement, I see clearly 2 surfaces that reflect light if using zerodur test glass (21:09). However in case the test surface is not attached to the reference surface 19:05 19:44, there are probably more than 2 surfaces and even much more than 2 surfaces if using a colimating lens system. How do you evaluate the result that case?
Good question. The way you achieve this is by making sure the reflections of these surfaces do not reach the detector. It's illustrated at 19:05 in the reference element by the slight tilt in the other surface that effectively places its reflection outside the area of the small beam splitter. In the case of the object, you can reduce reflection (of for example the last surface) temporary with black spray paint. If that surface has a radius of curvature significantly different from the surface under investigation, then generally the resulting nr of fringes from this surface will be too high to spot them in the interferogram.
Utterly fascinating as always. I have absolutely no need for an interferometer, but would still love to have one, as I have a measurement fetish: The process I’d measurement itself makes me happy (especially of something that’s invisible to our normal senses), whether or not I need the results for anything 😂
excellent video. very understandable - I've learned something new!
tom lipton did a video on this years ago. very cool analog tech! (i didnt know this was basis of any digital imagery!)
I remember (a long time ago) swimming front crawl against a slightly taller boy in a short course pool. Every tumble turn he got an advantage of twice the difference in height (assuming the difference in leg length was proportional to the difference in height). That must totally explain the difference in our speed.:-). Anyway, just thinking about the factor of two at 22.48
Your maximum speed is actually 1.34 x sqrt(body length). So yes, being longer helps, but I'm afraid it's not quadratic with the difference :-).
5:49 shouldn't that be the frequency bandwidth instead?
Generally, when precision lasers are used, you'd probably want to prevent reflections from getting back into the laser. In the block schematic you drew for the Zygo interferometer, it looks as if a big fraction (12.5%, assuming 50-50 splitters) of the laser power will make it back into the laser, after reflections. Maybe, then, something is tilted a bit? Maybe the position of the GRIN lens with respect to the beam splitter with the beam dump is such that on its way back towards the laser, the beam diverges? Or it doesn't matter if all that light makes it back into the laser? I don't think you want any significant reflection back into such a precision laser. I've played with interferometers; they can be so sensitive that local air-temperature gradients show up as noticeable fringe shifts. Such (varying) reflection could then detune the laser by injection locking. I am not an optics specialist, so I could be wrong.
Yes that is correct. at 13:12 you can see that the returning reflection on the 45 degrees steering mirror is quite a bit off from the original beam (it is on the top of the mirror). So I guess generally the reflected light does not make it back into the laser cavity.
is the function of the GRIN lens and transforming the wavefront flat-spherical-flat again to enlarge the cross-section of the beam?
Yes, but basically you could any short focal length lens in combination with a collimation lens.
it kind of makes sense that the components of a measuring instrument are built to higher precision than mass consumer product
I have no idea what you are talking about honestly but this was still pretty fascinating 😅
I'd say it's C, because touching it moves the fringes back where they should be if it was flat, and the positive coefficient of thermal expansion means that it has to get thicker to be flat, so it is concave. And because it getting thicker moves the fringes to the left, thats also the direction where the rest of the glass is thicker.
I agree
Ouch ...A
Your finger made the glass thicker at that spot and kind of "corrected" the bowing of that dark fringe line back to the left. This makes me think the thickness where you put your finger was too small which leaves A or C (it is concave, too low in the center). Traveling from right to left across the middle of the glass the finger made you reach the 3rd bright line sooner. So I think the thickness is increasing from right to left. So that would leave A, final answer!
Jeroen, have you ever tried making a germanium lens for a thermal imager? All I know so far is that some of them look amazing. I wonder how different the lapping process might be compared to glass, or even whether lapping is still a suitable process in this application.
I have lapped Germanium. It's softer and requires you to work very cleanly. Also, you use Al2O3 rather than CeO as a polishing agent. I've never done a video about it though.
Very informative. Thank you!
I concluded it was likely concave before the heat hint by noticing that the edges we're slightly rolled off (very unlikely to have a high edge). Since the fringes in the center are shifted in the same direction as the fringes at the very edge, the center must also be low, thus concave.
Knowing the the rolled edges shift the fringes to the right also let's us determine the overall wedge. Intuitively the fringe is shifted right at the edge because we don't need to go so far to the left to drop to the next fringe as we do in from the edge, thus the wedge is falling off towards the left, case A.
I guess if you are into ATM where turned down edges are a common phenomenon, finding out the shape wasn't really that hard.
What a remarkable video! Thank you so much for your explanation!
I know it has to be A or C, the concave options, because the fringe curve induced by the heat of your finger is in the opposite direction of the curve of the fringes of the surface, and the induced local curve from positive coefficient of expansion would be convex since it would be a “bump” in the surface. Since we know it is a concave sphere, and the fringes “point” to the right, the thicker part must be on the left, so the answer is C.
Think ill give up the idea of making a lens haha. Nice video and nice equipment.
Love these vids.. I learn something new EVER.. SINGLE.. VIDEO :)
finger trick was awesome. So it was concave because expansion made it "grow" up and it became flatter, and it's a C because the lowest (right) part of the wedge seems to be the flattest