Probably a safe assumption. The live demo was great, you can see that the fringes are clear and stable. There are so many ways to do this wrong and get noisy low-contrast fringes or just nothing. The video shows the setup is really working.
Amazing project, would love to build it when you're finished. One tip you might find useful for your reference arms: You likely want a very low CTE material. Invar or zerodur is likely way too expensive, but there is a trick you can do with two different cte materials. Example: Say you want a distance of 25mm between point a and b. Take a piece of 50mm steel, point a is on one end, now attach a piece of 25mm aluminium to the other end facing towards point a. At the end of this aluminium piece is point B. So a and b are now 25mm apart. Due to the CTE of aluminium being twice that of steel, and the steel part being twice as long, they expand by the same amount and thus the distance between a and b doesnt change. Without a drawing its a little difficult to explain so I hope its clear. If not let me know.
Hi, thank you for your great advice. I am currently using a pulltruded CFK base, which is not too expensive and has a CTE of about -1e-6. So its currently good enough for me. But I agree using clever combination of materials can get you further towards 0 expansion.
@@DiffractionLimited you could also use a strain gauge to both update length of low CTE (updating length) material, as well as using that length change to determine the temperature. Even if it isn't direct temperature control, it should at least improve the coherence length, and reduce some variables.
@@fromscratch2654 I actually looked into that but it has a thermal expansion coefficient of about 6e-6 so its worse than CFK. But thank you for the suggestion :)
Excellent video! I used Renishaw interferometer equipment in a previous career (CNC machine setup & calibration). They gave us some basics of how an interferometer works, but you're taking it another level. Moar!
I'm really looking forward to the upcoming videos! The design is fantastic and would make an excellent project for my CNC-first for milling and later for geometrical compensation of my machine and Heidenhain scales. Greetings from Germany!
I have been thinking about a laser interferometer some time ago and even discussed this with some people better knowledgeable in this topic and they all quickly told me that I cannot do this with common "hobby grade" stuff. It looks like I just need a bit more advanced "hobby grade stuff" but I also realize this will give me a lot of sleepless nights re-running the theoretical stuff in my head. So I subscribed, hoping for lots of those sleepless nights 😁
PLT5 520B is a multimode laser diode, so in order to increase spatial coherence and visibility - it might be useful to place a spatial filter in front of the diode to reduce the number of modes. The most basic form of spatial filtering is just round diaphragm (as a bonus you will get a nicer beam to work with), no doubt, it will reduce output power but in most cases you will get better SNR.
It literally says 'Single mode semiconductor laser' in the datasheet. Also I operate it way below the nominal current which often results in a single mode being excited, even for multimode diodes.
@@DiffractionLimited PLT5 520B is a typical broad area Al/InGaN laser diode with a basic fabry perot resonator. You may try to measure M^2 for example to find out that it is indeed single mode in the fast axis but high multimode in the slow one. From my experience, such laser diodes give a typical multimode picture even at low pump currents. Also you may refer to output beam divergence in different axes.
@@enclis Hi, maybe we talk about different LDs, beacuse I could not observe any multimode beam profile with my sample, even at higher currents. Its strange that you have different experience with this diode.
@@DiffractionLimited awesome! I myself was considering a similar project, though I was going to stabilize the temperature of the diode with thermoelectric devices. However, I like your idea more!
@@tomthepom98 Thank you! From the tests I did, you would have to stabilize both the drive current and temperature to quite high tolerances, as the laser diodes quickly shift their wavelength. So I would bet on the stabilization arm to be more accurate.
Very interesting. Maybe you can share some information about your readout electronics. I currently seem to have some oscillations in my photodiode preamp setup when I go above bandwidths of around 50kHz or so.
marvelous! I was about to do the exact same thing and then I find your video. I have experience in manufacturing temperature controllers with sub mK stability and heterodyne Mach-Zehnder interferometers. My original plan was to use one interferometer to stabilize the laser frequency, and then use and second heterodyne interferometry to measure the distance, so lambda /100 accuracy can be easily achieved. I considered helium-neon laser and diode laser, and I found that stabilizing the diode laser in a certain mode was quite A challenge, but if you pull it off, it could be as good as A stablized helium-neon laser
This is great work. I have tried to do this with diode lasers with limited success, got fringes sometimes but they would unpredictably fade away, my diodes I think not stable enough. Interested now to try your recommended part. I had tried a more expensive, allegedly single mode part from a German supplier, also a DIY hologram source.
Hi, I know exactly what you mean and I struggled with this as well. There is a difference between 'single transverse mode' and 'single longitudinal mode'. The latter is what you want and is often not specified for laser diodes as it depends on the drive conditions. However if you got a Laser source specifically for holography you should definetly be fine. If you had problems with good interference there are a few important things to look out for: 1. Avoid ANY backreflection into the diode, this usually destroys coherence length (see one of my older videos). 2. Filter your LD current: Add one or more RC, filters with different capacitance (e.g. 50µF elko + 10µF ceramic + 1µF ceramic) in parallel. Make sure there is no additional driver before your laser diode as it is often the case for laser modules powered by 5V. 3. Find a drive current that produces a long coherence length. This is often at lower power levels slightly above threshold current. If coherence length is short, you periodically get regions of good and bad fringe contrast depending the length difference between you interferometer arms. This can look like it is sometimes working and sometimes not. 4. Use a reasonable sized heatsink to keep the diode at roughly ambient temperature.
Great video! I love the link with the practical setup! Last year I built a homodyne displacement interferometer myself with LEGO, aliexpress and 3D printed parts, where I had to deal with exactly the same issues as highlighted by you. Let me know if you're interested in my summary document. Looking forward to the future parts of this series!
Basically how the Renishaw laser interferometers work, as far as I know. Unlike the HP ones with funny Zeemann effect to get 2 frequency laser, to do doppler shift thing with, instead of the sin/cos quadrature for direction, in/out detection.
As i recall the precious metal Rhodium, when electroplated on a surface may have useful characteristics with regards hardness (resist dimensional change) and light absorbed vs reflected (extremely good reflectance) also impervious to normal atmospheric reaction (does not oxidise)
Very cool design. The stabilizer is interesting. It wouldn’t control the absolute laser wavelength right? Just prevent it from drifting after the stabilization is activated. A reasonably accurate calibration after power up can probably be done to account for wavelength uncertainty, by moving the measure arm a precisely known distance and dividing by the counted fringes. I think gas reference cells are common for absolute wavelength control. IIRC this is what the attocube interferometer does. Finding laser diodes for metrology is difficult, so thanks for sharing this part number.
Nice video, It would be amazing to add additional laser diodes at different frequencies making parallel interferometers, maybee one of those multi-diode Optical drive laser diode arrays for both the transmit and recieve elements, This would allow for better distance discrimination, especially when combined with time of flight measurement when combined with a modulation of the laser diodes?
Love so many things about this project, very excited to see it progress in the future. I noticed that in your optical topology drawing, there are paths the beam that can take which form loops and therefore resonances (for example, point 2->reflector->point 3->reflector-> point 2). In the ideal (perfectly aligned) case, wouldn't these cause the intensity of your beam to deviate from a sinusoidal shape as a function of the movement of the moving arm? Is this something you have observed in your setup? Or is this not an issue when there are finite misalignments?
Hi, good point. I believe those loops are only present in the 2D drawing, once you look at the 3D beam paths, all beams terminate as they should. Thank you for the great question though.
It might not matter because the intensity is atenuated for successive splits of the beam, but there seems to be a possible loop between CCR1 and CCR3 in such a way that the first detector gets not only the beams that traveled both arms but also the beams that did N loops of that circuit. Is this correct?
I will go into that in future. But essentially its the front part of those cylindrical laser module cases that you find everywhere. I wanted to use these because they are very affordable and easy to come by. Here is the first best link to one of those: odicforce.com/epages/05c54fb6-7778-4d36-adc0-0098b2af7c4e.sf/en_GB/?ObjectPath=/Shops/05c54fb6-7778-4d36-adc0-0098b2af7c4e/Products/OFL14
Do you have any good source on the glass plate? Is it okay to get cheap silver ccr from aliexpress? Sorry for asking so much I will wait for more info when the project is done anyways.
Yes, I also use cheap CCRs they seem to work very well. The glass Plate is just a cutoff from a larger 6mm float glass plate (glas scribe + diamond wetstone to clean up the edges). I will cover more details in one of the next videos...
Awesome video. One small doubt: is it possible that in your diagram for a Michelson interferometer with offset the detector should read the beam coming down from the splitter at the left side instead of the right? Not really relevant but just to make sure I got how it works right.
I wonder if combining what you're doing with the mechanisms used in the James Webb telescope could result in an inexpensive mechanism with an ability to reproducibly move 10 nm or less (see Hackaday's "Working Model Reveals Amazing Engineering Of Webb’s Mirror Actuators").
8:36 wouldn't we have kind of a loop at the arm 2CCR ? If we follow the laser path at arm 1 it reflects of diagonal glass goes into CCR in arm 1 then goes down again reflecting of a glass into X and part of it passes into D2. That part is ok. Now the part that passes the glass from the laser goes into CCR at arm 2 then returns to the glass and part of the beam goes into X this is also ok. But now the part which is reflected and goes into CCR at the arm 3 goes back to the glass and partly passes into D1 but it also reflects and again enters the CCR in the arm 2 following the same path from before??
Thank you for the feedback. This problem was also pointed out by another observant viewer and the loop is an artifact of the 2D representation of the beam path. In the 3D beam path there are no loops anymore, so no problem in practice. I should have mentioned that in the video though.
Looking at this and your recent short i do have one concern. I can see how the second leg will let you fix the frequency during operation. But given how far you can still drive it i don't see how it'll let you set exactly the frequency. Like during operation holding the spot at 0 diode current will stop the frequency from drifting. But when you turn on how do you know you are at (say) 500nm or 501nm. In both cases the spot can still be held at zero and the frequency won't drift. But the counts per mm on the driven leg will. Would perhaps a very potent diffraction grating be enough to decern close enough which frequency the laser is operating at when first turned on?
Excellent question. The absolute scale has to be given to the system somehow. A diffraction grating would work, but would have to be very fine pitched and itself be very stable over time. A different approach is to calibrate the interferometer by moving the reflector a known (preferably large) distance and count the pulses. For my applications, absoluthe scale isn't very crucial though.
Thanks, youtube changed 'http' to 'https' automatically. I changed the link and it works now, however it is not clickable anymore so you have to copy it manually.
Awesome video. very informative however I think there's a discrepancy in your interferometer signal-path animation. Your animation shows the signal changed to blue after it passes through some of the prism corner reflectors (these are the out of phase beams I believe?). This blue signal then passes through the beam splitter again before reaching the detector however there is no animation of this blue beam being split when doing so. I don't think this alters the end result of allowing this to work and maybe you are aware of this and did not include it in an effort to simplify an already complicated animation however since you do not explicitly mention this it leaves me with a question. Again, amazing project and video!!!
Very obserant! You are right, I omitted some beams for clarity that do not contribute to the working of the device and I probably should have mentioned that. I also just colored them blue so you can see how the two beams overlap.
I wonder what the chances are that "OSRAM PLT5 520B" parts on Aliexpress are genuine. They list a much lower price on that part, as compared with authorized OSRAM distributors in the US like Mouser Electronics who show $37.05 each, out of stock, 7-week backorder.
To answer my own question, I tried them and found they work. As the video suggests, you need to run at low drive currents to obtain single-mode operation. Which is fine, because the rated power (110 mW) is really more than I'm comfortable with anyway.
I think you'll find common cheap laser diodes are not reliably single-mode, and drift a lot over time. At least that is my experience. I was really impressed this guy found a part that is actually stable.
Electroplated rhodium has a hardness of 400-550 on the Vickers scale. This makes it very resistant to abrasion and wear, and helps it maintain its shiny finish over time. Surface Hardness: Rhodium plating results in the Increased surface hardness of the base metal, rendering the plated objects more resistant to scratches and wear
7:49 you also created changing beam offset by rotating it, possibly because you also moved it but regardless your words do not agree with what we can see which is never a good thing when you are explaining stuff like this because it can lead to comments like this one and also to opinions that you dont know what you are talking about or we dont understand you or we dont know what an offset is and many other not so good combinations and it distracts us from the point of the video.
I would partially disagree on this one. In the video it is only mentioned that the beam path doesnt change direction, when the CCR is rotated. There is no statement about the offset when it comes to rotation. But I agree, if you where confused by the demonstration, there is probable room for improvement of the video.
@@DiffractionLimited it clearly changes wither you rotate or slide it, yes direction did stay same. I am not sure but I think rotation should not change the offset if you are not sliding it also while rotating which is pretty hard to do manually without fixing it somehow and constraining so the only possible movement is rotation.
If we are all voting, I don't agree for niche tech video like this, unless it's a skilled speaker. For me, the value in this particular video is not about the emotional expression of a live human voice. I've tried to watch some "how-to" videos with voices and accents that were so hard to understand that I just gave up. Not everyone making excellent content can also speak in clear, well-paced and easily understood English.
Outstanding. Don't shy away from the information density, I suspect anyone interested in 100nm precision is here to geek out on the details.
+1
Intermediate level demonstrations are so valuable, and used to be impossible to find.
Probably a safe assumption. The live demo was great, you can see that the fringes are clear and stable. There are so many ways to do this wrong and get noisy low-contrast fringes or just nothing. The video shows the setup is really working.
Amazing project, would love to build it when you're finished.
One tip you might find useful for your reference arms: You likely want a very low CTE material. Invar or zerodur is likely way too expensive, but there is a trick you can do with two different cte materials.
Example: Say you want a distance of 25mm between point a and b. Take a piece of 50mm steel, point a is on one end, now attach a piece of 25mm aluminium to the other end facing towards point a. At the end of this aluminium piece is point B. So a and b are now 25mm apart. Due to the CTE of aluminium being twice that of steel, and the steel part being twice as long, they expand by the same amount and thus the distance between a and b doesnt change.
Without a drawing its a little difficult to explain so I hope its clear. If not let me know.
Hi, thank you for your great advice. I am currently using a pulltruded CFK base, which is not too expensive and has a CTE of about -1e-6. So its currently good enough for me. But I agree using clever combination of materials can get you further towards 0 expansion.
+1. this is a technique traditionally used in clockmaking, search for gridiron pendulum
@@DiffractionLimited you could also use a strain gauge to both update length of low CTE (updating length) material, as well as using that length change to determine the temperature.
Even if it isn't direct temperature control, it should at least improve the coherence length, and reduce some variables.
You could get zerodur for a scrap price if you use the cooking surface of an old broken ceran stovetop. 🤔
@@fromscratch2654 I actually looked into that but it has a thermal expansion coefficient of about 6e-6 so its worse than CFK. But thank you for the suggestion :)
Very nice video and content, I really enjoyed watching this. Thanks for all the effort that you put in!
the man himself!
Really awesome stuff. The frequency stabilization idea is awesome. Looking forward to the next part.
Why are you taking a break to watch other videos when you need to be working on my next fix from your own channel??? Back to the salt mines!
@@rodfrey I second this sentiment 😂
Liking and engaging in the hopes the algorithm continues to bless you.
Excellent work, thank you for contributing to the accessibility of precision technology
The wavelength stabilizing bit is a great idea! Im very curious to see where this goes!
Great stuff! Can't wait for more. For me, information density is just spot on
Fantastic content! I'm an electronics engineer myself, but I love optics! Very good video quality and presentations. Good job!
Brilliant!
Oh wait, pun unintended, but I'll roll with it. I'm certainly looking forward to how this unfolds.
Excellent video! I used Renishaw interferometer equipment in a previous career (CNC machine setup & calibration). They gave us some basics of how an interferometer works, but you're taking it another level. Moar!
I'm really looking forward to the upcoming videos! The design is fantastic and would make an excellent project for my CNC-first for milling and later for geometrical compensation of my machine and Heidenhain scales. Greetings from Germany!
I have been thinking about a laser interferometer some time ago and even discussed this with some people better knowledgeable in this topic and they all quickly told me that I cannot do this with common "hobby grade" stuff.
It looks like I just need a bit more advanced "hobby grade stuff" but I also realize this will give me a lot of sleepless nights re-running the theoretical stuff in my head. So I subscribed, hoping for lots of those sleepless nights 😁
Just wow! I wanted to own/built in interferometer, but it just wasnt in reach. Thank you a lot for posting! 😊
Thanks for sharing! You really made me want to build one. So cool!
Awesome design! Thanks for publishing this
That's awesome I wondered about building an interferometer for some time.
great video and channel! amazing explanation and proyect, i will keep an eye on this!
Ah what an exciting project! Can't wait to follow along, thanks for posting.
PLT5 520B is a multimode laser diode, so in order to increase spatial coherence and visibility - it might be useful to place a spatial filter in front of the diode to reduce the number of modes. The most basic form of spatial filtering is just round diaphragm (as a bonus you will get a nicer beam to work with), no doubt, it will reduce output power but in most cases you will get better SNR.
It literally says 'Single mode semiconductor laser' in the datasheet. Also I operate it way below the nominal current which often results in a single mode being excited, even for multimode diodes.
@@DiffractionLimited PLT5 520B is a typical broad area Al/InGaN laser diode with a basic fabry perot resonator. You may try to measure M^2 for example to find out that it is indeed single mode in the fast axis but high multimode in the slow one. From my experience, such laser diodes give a typical multimode picture even at low pump currents. Also you may refer to output beam divergence in different axes.
@@enclis Hi, maybe we talk about different LDs, beacuse I could not observe any multimode beam profile with my sample, even at higher currents. Its strange that you have different experience with this diode.
Very cool, great job!
Can't wait to see part 2 ❤
Damm you should have a high salary with this type of skill. Thanks so much for sharing
Excellent work looking forward to the next part!
Amazing project. I'm hyped for the next parts xD
OT: Which typeface/font do you use for the slides? I really like the modern /technical style of it
Super interesting, looking forward to more videos from you 🙌
very interesting. The 10khz measurement makes me wonder if it could be used as part of a servo positioning loop for a high precision linear axis.
This is actually the goal ;)
@@DiffractionLimited awesome! I myself was considering a similar project, though I was going to stabilize the temperature of the diode with thermoelectric devices. However, I like your idea more!
@@tomthepom98 Thank you! From the tests I did, you would have to stabilize both the drive current and temperature to quite high tolerances, as the laser diodes quickly shift their wavelength. So I would bet on the stabilization arm to be more accurate.
Dang, jaw dropping, quality content as always
Looks good! I built this four years ago, but never got around to properly publish it.
Very interesting. Maybe you can share some information about your readout electronics. I currently seem to have some oscillations in my photodiode preamp setup when I go above bandwidths of around 50kHz or so.
really looking forward to this. I'd like to build a device like this for machine building purposes.
No, thanks to you. It was nice . I hope to make one myself too.
marvelous! I was about to do the exact same thing and then I find your video. I have experience in manufacturing temperature controllers with sub mK stability and heterodyne Mach-Zehnder interferometers. My original plan was to use one interferometer to stabilize the laser frequency, and then use and second heterodyne interferometry to measure the distance, so lambda /100 accuracy can be easily achieved. I considered helium-neon laser and diode laser, and I found that stabilizing the diode laser in a certain mode was quite A challenge, but if you pull it off, it could be as good as A stablized helium-neon laser
In my original design,I need a cheap acousto-optical crystal to create a 80MHz modulated laser beam.if I cant do that,then this design will failed.
This is great work. I have tried to do this with diode lasers with limited success, got fringes sometimes but they would unpredictably fade away, my diodes I think not stable enough. Interested now to try your recommended part. I had tried a more expensive, allegedly single mode part from a German supplier, also a DIY hologram source.
Hi, I know exactly what you mean and I struggled with this as well. There is a difference between 'single transverse mode' and 'single longitudinal mode'. The latter is what you want and is often not specified for laser diodes as it depends on the drive conditions.
However if you got a Laser source specifically for holography you should definetly be fine.
If you had problems with good interference there are a few important things to look out for:
1. Avoid ANY backreflection into the diode, this usually destroys coherence length (see one of my older videos).
2. Filter your LD current: Add one or more RC, filters with different capacitance (e.g. 50µF elko + 10µF ceramic + 1µF ceramic) in parallel.
Make sure there is no additional driver before your laser diode as it is often the case for laser modules powered by 5V.
3. Find a drive current that produces a long coherence length. This is often at lower power levels slightly above threshold current.
If coherence length is short, you periodically get regions of good and bad fringe contrast depending the length difference between you interferometer arms. This can look like it is sometimes working and sometimes not.
4. Use a reasonable sized heatsink to keep the diode at roughly ambient temperature.
Great video! I love the link with the practical setup! Last year I built a homodyne displacement interferometer myself with LEGO, aliexpress and 3D printed parts, where I had to deal with exactly the same issues as highlighted by you. Let me know if you're interested in my summary document. Looking forward to the future parts of this series!
Yes, I am definitely interested in your findings !
@@DiffractionLimited mail sent to your gmail!
Absolutely awesome project! Interested to buy it in a kit form as soon as it is available )
Basically how the Renishaw laser interferometers work, as far as I know.
Unlike the HP ones with funny Zeemann effect to get 2 frequency laser, to do doppler shift thing with, instead of the sin/cos quadrature for direction, in/out detection.
As i recall the precious metal Rhodium, when electroplated on a surface may have useful characteristics with regards hardness (resist dimensional change) and light absorbed vs reflected (extremely good reflectance) also impervious to normal atmospheric reaction (does not oxidise)
😍i need to build one. The best way to cure my obsession with metrology is to nurture my obsession.☝🤓
INCREDIBLE
Good job man
Very cool design. The stabilizer is interesting. It wouldn’t control the absolute laser wavelength right? Just prevent it from drifting after the stabilization is activated. A reasonably accurate calibration after power up can probably be done to account for wavelength uncertainty, by moving the measure arm a precisely known distance and dividing by the counted fringes. I think gas reference cells are common for absolute wavelength control. IIRC this is what the attocube interferometer does.
Finding laser diodes for metrology is difficult, so thanks for sharing this part number.
Youre absoluthely correct. I didnt know about gas reference cells. I will look into that - very interesting, thanks.
Nice video, It would be amazing to add additional laser diodes at different frequencies making parallel interferometers, maybee one of those multi-diode Optical drive laser diode arrays for both the transmit and recieve elements, This would allow for better distance discrimination, especially when combined with time of flight measurement when combined with a modulation of the laser diodes?
Love so many things about this project, very excited to see it progress in the future. I noticed that in your optical topology drawing, there are paths the beam that can take which form loops and therefore resonances (for example, point 2->reflector->point 3->reflector-> point 2). In the ideal (perfectly aligned) case, wouldn't these cause the intensity of your beam to deviate from a sinusoidal shape as a function of the movement of the moving arm? Is this something you have observed in your setup? Or is this not an issue when there are finite misalignments?
Hi, good point. I believe those loops are only present in the 2D drawing, once you look at the 3D beam paths, all beams terminate as they should. Thank you for the great question though.
Homemade gravitational detector, when?
Really enjoyed your video
It might not matter because the intensity is atenuated for successive splits of the beam, but there seems to be a possible loop between CCR1 and CCR3 in such a way that the first detector gets not only the beams that traveled both arms but also the beams that did N loops of that circuit. Is this correct?
Very cool!
Where would one acquire the case and lens that turns laser diode into an actual laser?
I will go into that in future. But essentially its the front part of those cylindrical laser module cases that you find everywhere. I wanted to use these because they are very affordable and easy to come by. Here is the first best link to one of those:
odicforce.com/epages/05c54fb6-7778-4d36-adc0-0098b2af7c4e.sf/en_GB/?ObjectPath=/Shops/05c54fb6-7778-4d36-adc0-0098b2af7c4e/Products/OFL14
Do you have any good source on the glass plate? Is it okay to get cheap silver ccr from aliexpress?
Sorry for asking so much I will wait for more info when the project is done anyways.
Yes, I also use cheap CCRs they seem to work very well. The glass Plate is just a cutoff from a larger 6mm float glass plate (glas scribe + diamond wetstone to clean up the edges). I will cover more details in one of the next videos...
@@DiffractionLimitedhi great video do u have an instagram ?
Can you give more details about the text-to-speech narration? I was surprised at the quality
checkout F5-TTS
Awesome video. One small doubt: is it possible that in your diagram for a Michelson interferometer with offset the detector should read the beam coming down from the splitter at the left side instead of the right? Not really relevant but just to make sure I got how it works right.
Yes, its a mistake. Should be left.
Very cool
I wonder if combining what you're doing with the mechanisms used in the James Webb telescope could result in an inexpensive mechanism with an ability to reproducibly move 10 nm or less (see Hackaday's "Working Model Reveals Amazing Engineering Of Webb’s Mirror Actuators").
8:36 wouldn't we have kind of a loop at the arm 2CCR ?
If we follow the laser path at arm 1 it reflects of diagonal glass goes into CCR in arm 1 then goes down again reflecting of a glass into X and part of it passes into D2. That part is ok.
Now the part that passes the glass from the laser goes into CCR at arm 2 then returns to the glass and part of the beam goes into X this is also ok. But now the part which is reflected and goes into CCR at the arm 3 goes back to the glass and partly passes into D1 but it also reflects and again enters the CCR in the arm 2 following the same path from before??
Thank you for the feedback. This problem was also pointed out by another observant viewer and the loop is an artifact of the 2D representation of the beam path. In the 3D beam path there are no loops anymore, so no problem in practice. I should have mentioned that in the video though.
Looking at this and your recent short i do have one concern. I can see how the second leg will let you fix the frequency during operation. But given how far you can still drive it i don't see how it'll let you set exactly the frequency.
Like during operation holding the spot at 0 diode current will stop the frequency from drifting. But when you turn on how do you know you are at (say) 500nm or 501nm. In both cases the spot can still be held at zero and the frequency won't drift. But the counts per mm on the driven leg will.
Would perhaps a very potent diffraction grating be enough to decern close enough which frequency the laser is operating at when first turned on?
Excellent question. The absolute scale has to be given to the system somehow. A diffraction grating would work, but would have to be very fine pitched and itself be very stable over time. A different approach is to calibrate the interferometer by moving the reflector a known (preferably large) distance and count the pulses. For my applications, absoluthe scale isn't very crucial though.
@6:13 the Michelson with Offset, shouldn't the detector be below the left beam coming from the top CCR?
You are correct, its a mistake. I noted it in the video description, but I can blame you for not reading it ;)
Mode Stability of Diode Lasers link is not working (says "Auth required")
Thanks, youtube changed 'http' to 'https' automatically. I changed the link and it works now, however it is not clickable anymore so you have to copy it manually.
Cool project. Is there a associated git?
Where can one get corner-cube-mirror for only 15€? :D
Yes, its being built. I got my CCRs from ebay, you if you search around you find other popular selling platforms that have these as well.
9:42 before point 4 you forgot the point 2 again would reflect and go into CCR on the right and again come to the point 3.
Awesome video. very informative however I think there's a discrepancy in your interferometer signal-path animation. Your animation shows the signal changed to blue after it passes through some of the prism corner reflectors (these are the out of phase beams I believe?). This blue signal then passes through the beam splitter again before reaching the detector however there is no animation of this blue beam being split when doing so. I don't think this alters the end result of allowing this to work and maybe you are aware of this and did not include it in an effort to simplify an already complicated animation however since you do not explicitly mention this it leaves me with a question. Again, amazing project and video!!!
Very obserant! You are right, I omitted some beams for clarity that do not contribute to the working of the device and I probably should have mentioned that. I also just colored them blue so you can see how the two beams overlap.
🌊🌊🌊🌊🌊
6:15 you misplaced the detector
Thanks for the correction, cant change it anymore unfortunately.
I wonder what the chances are that "OSRAM PLT5 520B" parts on Aliexpress are genuine. They list a much lower price on that part, as compared with authorized OSRAM distributors in the US like Mouser Electronics who show $37.05 each, out of stock, 7-week backorder.
To answer my own question, I tried them and found they work. As the video suggests, you need to run at low drive currents to obtain single-mode operation. Which is fine, because the rated power (110 mW) is really more than I'm comfortable with anyway.
I wonder if I could do this by slightly modifying a laser pointer from Walmart!
I think you'll find common cheap laser diodes are not reliably single-mode, and drift a lot over time. At least that is my experience. I was really impressed this guy found a part that is actually stable.
The drift is taken care of, also I might have a few promising candidates, so we will see.
Electroplated rhodium has a hardness of 400-550 on the Vickers scale. This makes it very resistant to abrasion and wear, and helps it maintain its shiny finish over time.
Surface Hardness: Rhodium plating results in the Increased surface hardness of the base metal, rendering the plated objects more resistant to scratches and wear
Very well, now rotate that vertically and prove Einstein is wrong, please.
7:49 you also created changing beam offset by rotating it, possibly because you also moved it but regardless your words do not agree with what we can see which is never a good thing when you are explaining stuff like this because it can lead to comments like this one and also to opinions that you dont know what you are talking about or we dont understand you or we dont know what an offset is and many other not so good combinations and it distracts us from the point of the video.
I would partially disagree on this one. In the video it is only mentioned that the beam path doesnt change direction, when the CCR is rotated. There is no statement about the offset when it comes to rotation. But I agree, if you where confused by the demonstration, there is probable room for improvement of the video.
@@DiffractionLimited it clearly changes wither you rotate or slide it, yes direction did stay same. I am not sure but I think rotation should not change the offset if you are not sliding it also while rotating which is pretty hard to do manually without fixing it somehow and constraining so the only possible movement is rotation.
Not downloadable: won't watch.
Sorry but video with AI voice, video that gets a thumbs down and the channel blocked from my youtube feed.
cut the AI voice nonsense.
If we are all voting, I don't agree for niche tech video like this, unless it's a skilled speaker. For me, the value in this particular video is not about the emotional expression of a live human voice. I've tried to watch some "how-to" videos with voices and accents that were so hard to understand that I just gave up. Not everyone making excellent content can also speak in clear, well-paced and easily understood English.
@dancevideo2 I am much more forgiving of someone with a thick accent than an insecure coward that hides behind AI synthesizer software.
Or what?
@@VEC7ORlt or the author is a fake
@@HyenaEmpyema or maybe you are, or I am, beep boop.
So what?