Repair Attempt: Coherent Verdi Narrow Linewidth Green DPSS Laser
Vložit
- čas přidán 17. 06. 2024
- PCBWays Christmas Activity www.pcbway.com/activity/chris...
Accurate & Affordable Multimeters: www.welectron.com/Handheld
5% coupon code: marcoreps5
Mark-10 precision Torque & Force sensors mark-10.com/
R&S Next generation oscilloscope www.rohde-schwarz.com/de/prod...
More from me: / marcoreps
Resources:
Fabian / voigtvision
his OSH light sheet microscope mesospim.org/
Photo source microscopy.berkeley.edu/tlm/2...
Little Bee B1 OSH Current Probe github.com/westonb/little-bee...
Referenced CZcams Videos:
The Amazing Microscopic World • The Amazing Microscopi...
fluorescence and confocal microscopies • fluorescence and confo...
Mikroskopie jenseits optischer Grenzen • STED - Mikroskopie jen...
Introduction to Confocal Microscopy • Introduction to Confoc...
Killer T cell attacking cancer • Killer T cell attackin...
Two-Photon Microscopy from Laser Quantum • Two-Photon Microscopy ...
Nano-scale 3D-printing • Castle on a pencil tip...
Green mega-laser shines over Stuttgart • Simply supernatural: G...
Timestamps:
00:00 Intro
01:36 Fluorescence microscopy
04:22 For Reps?
05:47 Verdi overview
08:24 Power conversion
12:22 Active noise filter
14:41 PCBWay
16:55 Backup battery
18:53 Service menu
20:56 Fapping
24:38 In the head
27:59 FAP replacement
30:16 Partial success!
31:40 Fine tuning
#laser #microscopy #repair - Věda a technologie
Lemo connectors used internally where nobody will even see them - definitely a sign of something built with minimal regard to cost!
I see your Lemo and raise gold plated Invar optics mounts around the V10 cavity. Those Permalign mounts heat up to the melting point of low temp solder when being adjusted; if not made of Invar the optic tends to crack from CTE mismatch. If not plated in gold, the solder won't stick. They cost Coherent more to manufacture than retail Thorlabs KS05 mounts would.
You should see what they did to the plastic Proteus flow meter for their ion lasers ... licensed the patent to swap plastic for CNC brass.
That’s because this modules get also sold single without this huge control case. You have then one production only for that modules and another for bigger systems. To produce the module with different connectors makes economically no sense.
I work for another big company on the market and we do it the same.
I have always wondered how their 'permalign' stuff works. But yeah, low temperature solder and invar mounts is certainly something.
As a sidenote: Does the low temp solder not have a CTE that pulls the optic away? @@brentwheelock8297
Better to use MIL-DTL-38999, Series I LJT, II JT, III TV, HD connecters.
@@universeisundernoobligatio3283 for sure 👍 🤣🤣🤣
Marco, just a little remark on your cooling fluid, don't know if it applies here. I hope you follow the Coherent Manual which type of water to use. I had a machine leaking through a shutter because the de-ionized water has eaten holes in the Aluminium block after months of running. The customers were not happy at all ...
Btw. I'm very impressed by your expertise, it looks like there's nothing you can't do or know in the electronics field.
Highly support that. We broke a 20k€ xray tube because the service technician just put deionized water in the cooling system. The tube manufacturer specified conductivity (reduces static build up in the tube), pH (proctects exposed parts), mineral content (in fact not "as low as possible") and some other things.
In the end they even supplied a recipe that basically read tap water plus some chemical (I think it was KOH) until the pH was at 9 and nothing else.
^^ This ^^ -- make sure you look at both the laser manual and the cooler manual if you have separate manuals for them. In the lab I work at, we use 5% ipa in water for our thermotek chillers, but we use tisapph lasers from a different manufacturer so ymmv.
Agree with everyone on this note. Pure water is a great solvent. Learned the same thing operating a pool, the water will leach minerals and metals if its not "neutralized" with the right Ph, alkalinity, and dissolved solids. Oh, and the right recipe is temperature dependent too.
We use OptiShield+ on all of our lasers.
@reps take this tip above all others, a machine running at half power is better than a machine that leaks
Neighbors: "There it is again: VERY strange brilliant light emissions and maniacal laughter..."
Ah yes, more esoteric electronics content!
If you ever worked with lasers in precission applications, you know that theres nothing esoteric about this but every bit of accuracy seen is needed
I’m glad the original owner thought to give this a new home on your bench, from where we can all learn something, rather than consign this to the dumpster.
Scrapped out isn't the dumpster. There's a lot of precious metals in high end electronics like that.
@@bobroberts8500 Of course. Anywhere I’ve worked, ‘the dumpster’ was the metaphorical term for ‘we don’t have an interest in or liability for what happens to it next’. No way this won’t one day be picked over for parts for reuse and recycle. At least I hope so.
There's something really beautiful about the design of these hyper-specialized tools, thanks for showcasing this one
The engineering and craftsmanship contained within. A beautiful thing.
“Lap-a-FAP” deserves a Pulitzer!
I'd guess the power goes down at high pump currents because something is misaligned slightly, but I can't see a kinematic mount in the main laser cavity. If you do have some kinematic adjustments in the laser cavity, hooking the power meter to one of your super DMMs and following the power/adjustment curve to it's maximum for each adjustment 2-500 times can make a big improvement.
Fyi, a cheap digital camera with the IR filter removed makes an ideal 808nm and 1064nm viewer. It's an excellent safety tool. (know where EVERY stray beam goes!)
The speckle pattern is nice and steady. Though if you really want a quick n cheap way to see how good the spectrum is, get an Iodine vapor cell. The contrast between tuned to an absorption line and off a line is a nice simple way to measure spectral purity. I'd expect Etalon temperature and LDO temp to be your main tuning parameters. (though pump power and everything else will effect it to)
Last, be aware that a lot of chillers and TEC cooler control loops are highly sensitive to ambient temperature. If you ever see the device temperature go down while ambient temps go up, it has a faulty feedback sensor design.
I "tamed" a DPSS green laser like this in my Graduate work.
Thank you for the hints! Yes, most optics in the cavity seem to be soldered. I think I'll live with 7-8W for now rather than taking any risks in there
OMG! What a glorious score!
Don't look Les !! your jealousy will turn you green !
@@andymouse too late! Can't...look...away...from...screen...
10:32 - emitter follower from oscillator to current sense input performs what is called "slope compensation", technique used in current-mode control PWM converters to improve stability at duty cycles of >50%
Can confirm, the mode locking "bump" is sometimes needed. Good old ti saph
31:20 Nice Laser speckles you made. Im working with a green LASER of a magnitude less power (~3 mW) in my lab, doing ESPI (electronic speckle pattern interferometry). Its an interesting topic worth looking into in my mind. You can extract micrometer height differences of a 2D plane through the phase information stored in the intensity of an image
Sounds great! That goes into my TODO list
Thought this was optical coherence tomography until some online reading ... sounds more like shearography. Interesting.
Can you explain what causes the speckles simply? I am a fan of physics, but have minimal formal education on it.
Thanks!
Those interference patterns (?) at full power are beautiful and terrifying!
PLCC sockets are known to go bad over time and are often a source of very erratic problems. The spring pressure in those sockets is pretty low because otherwise the sidewalls would deform to much. Removing the PLCC chip and cleaning the socket and chip will get those going again.
There's a whole lot of multidisciplinary physics and magic going on in this video. Mesmerizing! Don't look into laser with remaining eye.
27:55 Pronunciation was spot on! You made me laugh really good Repson, thank you for that! 😊😊😊
Please get a laser curtain for your windows to protect the outside ! With this kind of laser even diffuse reflexes can be dangerous.
Flying Schrodingers could or could not bump the head, and then could or could not end up in an airplane cockpit window via a reflection from an instrument transflective LCD backing. I'm sure he knew the risks, but it is a good idea to be better safe than sorry. His shots from outside looking in via the window were spectacular. The green goblin is somewhat of a scientist himself 💚
In our trade (automation) laser curtains is what we call anything that is used to detect the "presence" of a human. :))
If the reflection is actually diffuse, it’s only under 10 watts, that’s not enough to do damage to anything at more than a centimetre. Eyes won’t be damaged by such a reflection beyond maybe half a metre, not that I’d be within 5 metres without laser goggles, far more of there was even a small possibility of specular reflection. So long as he insures there’s no chance the laser can’t shine directly out of the window I don’t really see a problem, this would involve bolting it down and ensuring nothing reflective could end up in the beam path.
@@Scrogan "Eyes won’t be damaged by such a reflection beyond maybe half a metre"
That's a very bold statement. Do you have a source for that?
@@Scroganwhile this is generally true, its of course not a guarantee that there wont happen to be a reflective surface somewhere that isnt diffuse enough. you either point the laser at a diffuse and safe beamstop or you seal the room by lowering the blinds.
Thx PCBWay for sponsoring such great content🎉
i stopped breathing when he opened the head 😂
That image @5:10 literally jawdropping
Awsome necromancy on this laser, 8W isnt too bad for something that was gonna be scrapped. Also as this is the narrow linewidth version you can use it for making holograms too.
That was an amazing video, thank you for this
It's worth noting that you can operate most Ti:Sa laser with just 5W of power just fine. As for mode-locking: There's lots and lots to say about of cavity alignment, but you'll eventually get there, don't give up.
Man, I love your videos. Great information, sense of humor, and inspiration.
Happy holidays!
This is awesome! Thanks for sharing.
8:22 Wow, the cooling water temperature is displayed with two decimal places of degrees celcius. Welcome to Switzerland!
Coherent is a US company.
I love any and all Coherent and Lumonics gear... Great acquisition!!
Fascinating! I realize I can't even grasp the level of knowledge that you have. Congrats!
Stay safe, Mr. Reps. Always impressed by your thorough attention to the details of whatever interesting subject you might explore.
Happy holidays, madman.
Love Brymen.
Very impressive!
so glad I found your channel. so interesting and funny
I'm jealous of the Mira 900! I have two Verdis, but have so far not been able to track down a Mira in good condition. They're a great laser system, used to work with one from the early 90s that had been through multiple Verdis in its lifetime and still worked perfectly. Assuming you have the pump, it'll last forever.
Yea these Verdi's are quite reliable I have a few of those in my labs, but mine are air cooled, a peltier chiller is used to keep the head at a stable temperature, there should be a base plate for that.
Yes definitely! There should be normally a 7 cm block of aluminum below it and this is cooled by the peltier chiller (together with the mira). Already small temperature increases leed to thermal expansion and therefore to a misalignement of the cavity. You can easlily have 30% less power beacause you are some degrees of.
I have a Verdi riser block sitting in my barn now, would happily donate it to Marco if shipping a 30lb block of aluminum from Colorado to Stuttgart made any sense. They sure look pretty, but any 1 or 2cm thin lapped cool plate will work better. The added thermal mass of heated aluminum block takes time each morning to thermalize, which offsets alignment from pump to Ti:Sa. @@humwuuhu8433
love that you are still getting texted about that laser driver, because i literally just got one of those modules and was tempted to text you ^^
I work in microscopy field and never really new the point of the two photon microscopy, that was an amazing explanation!
Great video.
You would make a great evil villain, my friend. You have the voice and everything! Love your videos.
used to work with these once ❤
Awesome content
Just FYI you cannot determine the health of a timing belt by looking at it. I've seen plenty of timing belts that broke and still looked like they were brand new. Age and mileage alone determine when a timing belt needs to be replaced.
Stretch is the only good measurement. The fibers will stretch before failure daily reliabily
@@ionstorm66 I've been a professional mechanic for over 20 years and I've never heard of anybody using belt stretch as a way to determine when a timing belt needs to be replaced. Replace them on the schedule specified by the manufacturer. If you buy a car and don't know when the timing belt was last replaced go ahead and replace it.
Some engines use what’s called a stretch belt as well, to negate the use of a belt tensioner
That green beaty is burning my eyes even over youtube 😎
I thought this was a huygens optics video from the title! I love seeing stuff from my world show up on this channel, and all the imaging nerds here in the comments! Green DPSSs are always dying, and it's thrilling seeing you resurrect this one. If youre ever in the US in virginia, I have an ailing DPSS for you to look at :)
For clearing the loop of gunk, a vinegar wash is great at breaking up galvanic/biologic buildup on interior surfaces. Just dont leave the vinegar in there...
i gain so much, what i like to call, 'Scale of Spectrum' (pun not intended) knowledge. finding out what limits electricity has, what limits circuit designs have, new concepts in the industry i would never otherwise find out about.. and so on. I get a better idea of the spectrum of all things involved in electronics. great vids man, i cant turn down watching ur vids when they come out.... just too juicy to miss
Yep.
Mr. Marco, I wonder almost every day when you will come out with your next video. Your work is by a magnitude better than other “experts” on CZcams. I often re-watch your videos in the middle of the night when my tiny infant wakes me up. Your sense of humor is great and I laugh loud. My wife of course says WTF is wrong with you? Then I hide my phone with your video and pretend I have no idea what she’s talking about. I’ve been following you for years but now that I’m out of parents’ basement, I can actually afford tools and set up a complete JBC lab. Can’t wait for your next project and try to replicate it. You seem to live in practical world.
I spent one year with the Verdi, the clicky dial causes me flashbacks :D
oh wow! I made many parts for that laser!
Awesome! Electronics, optics...?
Thank you for showing this to us. Ideas for the power loss: the FAPs could be close to end of life and are drawing more power to deliver the needed output. At the upper end they may be generating more heat than the system can remove, throwing their wavelength off. Or there could be a small alignment error. I have found if the pump mode and cavity mode overlap are a little off, thermal lensing in the vanadate can cause the power to drop. I've been building a green DPSS for a while now and I see things like this when the power gets over about four watts of green.
Thank you for the suggestions! I think I will try some more cheap FAPs in both positions, but I don't feel sufficiently equipped to tweak things in the cavity (yet) Gonna have to live with 7-8W if there's something wrong in there
The lifetimes on the original pump diodes look ok-ish ~10k was the limit for the earlier ones. I agree with @BiranKPepin on the possible issue but think the pump wavelength stability of the replacement is the more likely issue. If you test in constant current mode then you can plot out power vs current graphs (532 nm output vs pump diode) that should be linear till you hit a roll off (some nonlinearity from the SHG). Pushing too far past saturation point of the system typically breaks something, try adjusting the pump diode temps up near the saturation point. Alternatively look at the output wavelength of some scattered light with a spectrometer, if you have one.
I usually end up just running these older systems in CC mode continuously, pretty much every one of them has the service mode enabled and its typically far more stable than the "constant power" mode. often lets you push a few extra thousands of hours out of the pumps too. The newer systems have the pump diodes in the head, far nicer. I would offer you more of the same as I have scrapped a number of these lasers over the years but the drive to Australia is a bit longer and my regulator would probably complain.
You are a wizard
You can use ordinary automotive water/antifreeze mix. It cools the laser diode mounting block. Those big gold boxes have TEC cooler modules inside that cool the pump diodes. The TEC modules in these are failure prone, but the FAP submidule inside is usually OK. The fix is to replace them and then use a simple radiator and a fan, they neglected thermal expansion and it causes the cooling crystals to crush at the torque used.
The weird transistor at 10:14 is used for slope compensation, it's an emitter follower that adds some of the RC oscillator ramp to the current sensor output to make it stable.
I used to do it support for a two photon confocal microscopy lab, that laser sure looks familiar 👍
what fun!
The technology is way beyond my education but I did a double-take at your FAP driver visual at 6:39
I don’t know if everyone else is more mature than me so didn’t mention it, or if they were too busy looking at the box.
Great video, loved the humor, and may have learned something!
Worked with a V12 before to pump a Coherent TiSa for exitation of Rubidium D line at finally 2.5 Watts output. Still more effecient than Argon Ion lasers as pump.
Very excited to see whats going under the hood of the Verdi 😊
Plating with the pump diode fiber optics changes the modes and was a trick to make the laser behave better. On some designs used to zip tie them in place.
Nice, finally! :D
Thank you for the assistance 🫡
very interesting!
yoooo I'm a neuroscientist and when I saw "verdi laser" I was like "*that* verdi laser?" and well it was. Also modelock hype
awesome!
Damn!!! I live there!!!!
As someone who knows just a little about electronics, this sounds like the turboencabultor video the whole time.
Well since my hobby is analog power supplies for valve ( vacuum tube ) stereo circuits . I found arcane for this a LOL moment . Fine video.
Oooow, shiny new toys
bringign back lapping old cpus and coolers, thsoe were the days
Nice !
I love Switzerland, such a good idea. I’m glad it came up with it.
The high output power is probably due to IR to green conversion in the crystal. If the crystal starts overheating, its efficiency drops. Thus, the green power drops. If you can fine tune the crystal temperature, try lowering it? And maybe you reach a higher power. Tho, it is usually set to a sweat spot based on a beam size.
Correct; SHG conversion drops which adds to intracavity IR light and thermal lensing in the vanadate which further increases mode size at the SHG waist. Since the servo operates on green power, it's a cascade that rolls over quickly. V10 was the IR lensing limit. On the V18, a thermal lens is factored into the resonator design and you thus need to establish a strong lens to get green conversion.
Check FAP wavelength is 808nm and/or adjust diode temp to try finding a different order of the DFB grating. Otherwise 8W of Verdi TEM00 is plenty to modelock the Mira, it's the DPSS mode that really improved Ti:Sa pumping over ion lasers, 532nm is further from Ti:Sa absorption peak than 514 or 488nm.
Oh, hell. Ok. You get a thumbs-up for the F.A.P. joke. I feel so dirty. And so 12.
Fascinating how your voice never changes... Maby think about changing modulation at some point 😁
Nooo! I know very little about electronics but listen to Marcoreps videos for the soething voice.
I think if his voice changed, it would be due to some serious outcome, like inadvertent black hole creation that will begin to eat the earth. Everything else is just chill.
Looks like you awakened a Necron tomb world.
Marco vs Darth Vader lets go
27:18 😂 you’re the best
Big Scary Lazor!
crazy !!!!!
blocks of gold:)
My laser Jedi master used to have 40w green laser mounted to cutting head. He rigged it up to cut metal sheets on a mill. Talk out about bright 😮. Even the reflections off the wall was to much. 😂
Reps looks straight-up super-villainous a 33:06. You've already got the accent, now you've got costume.
goooooooo video brooooooo
nice
29:45 for a second I got scared that that thing was a new version of the blue brushless brutality beast
"DON'T USE KEY" 👀
I snorted out loud at "lap a FAP"
Wtf new marcoreps intro?
I have never bothered messing with cavity alignment on a Verdi. They work till they break and at that point they are hardly worth fixing in the research world with the current gen Millennia lasers. That said its pretty rare the cavity breaks, usually the pumps, LBO or something electronic goes in them, as we have seen.
The Mira I have no experience with, much more use to the Tsunamis shown at 4:40. If you don't have a spectrometer, you can use the speckle pattern to check if you have it mode locked. The speckle disappears when the wavelength broadens. Get a decent isolator for the output if it doesn't have one built in otherwise it will never stay mode locked, they all hate external seeding. After that I expect nothing less than a homebrew autocorrelator to measure pulse profiles!
Looking for the speckle to disappear and avoiding back reflections are great advice, is there another source for GaP photodiodes now that Thor doesn't overnight them with Lab Snacks? Or how do you generate/isolate SHG for the autocorrelation?
The Verdi power rollover at 8W might also be fixed by adjusting the FAP temperature (I think there was a service menu for this); these are DFB pumps and need to be well centered at 808nm to minimize thermal lensing in vanadate.
I have found my kind of people! For sure, this is top notch advice.
@marco reps, The mode locking tip and the one about the isolation (esp that) are super valuable advice btw.
@@brentwheelock8297 A delay line and a thin sliver of BBO make a decent autocorrelator. A slow Si photodiode is enough, you can filter the pumps from the SHG with angle, focus the two inputs with a lens at a large angle into the BBO, SHG vector comes out straight (sum of k vectors, collinear operation). With very thin BBO and sub picosecond pulses it is very easy to be phase matched, bit of angle tuning but nothing like the critical response you have at longer pulse durations.
More! Powhahhhhh!!!!!
when i'm drunk that's the only channel that i search for, right now, since i was 17', you are the best marco !
oh wow that's at least class 3 laser
crazy
Again you surpassed my expectations. Outstanding content and video!
PS. This smells like a design from CERN. Maybe someone who worked there.
Oh? Because of the current transformers? I also heard it has some signature design techniques of a certain former HP engineer & pioneer in scope current probes
@@reps Worked at CERN (PS195 Experiment) and Uni ZH, it seems to be a definite physics design and related components.
(e.g. no illusive sillicon, standard available components and deep knowhow...).
Most PHDoc's/engineers/physicists work there for a 4 year contract and then spread around the world.
Dear Marco, how about a heads-up on the actual (working) condition of your laser? I (and others for sure :) ) share the same passion for otherworldly laser-power ;)
Listening to explanations about laser stuff makes me consider RF engineering rather simple in comparison. Even if an RF circulator is black magic in itself...
I have personally observed this kind of runaway behaviour on two different 532nm pump lasers. Both were rather old and could not deliver their rated power any more (the spectra physics one even died when I was using it) so I'm not sure if it's indicative of a common issue or if it's just how they behave. One was a Verdi V8 or V10 (can't remember) and the other one was a Spectra-physics Millenia eV.
Other guess maybe it could be the ir pump that you haven't replaced that can't produce it's max rated power anymore and goes into runaway ?
Possible! I have read somewhere that modern FAPs can last 20k hours and more, whereas mine have supposedly only worked for 6k hours. Maybe it's because they are some of the oldest
the mode-lock hammer lol
You should colab with " the signal path " for this one
I wasn't expecting to see a sticker
Where in switzerland!!!!
from star wars to lord of the rings ... creating our own little Angmar, are we ? Will the evil wizardry never end, sigh ^^
don't even want to know how much cost this :/ my old hobby 20years ago still love dpss lasers but after all of those years they are still expensive even low watage, real 100mw cost 200£ with tec coling i think this is still expensive after all those years