ไม่สามารถเล่นวิดีโอนี้
ขออภัยในความไม่สะดวก
30W Laser Burning Stone + New CC Driver + Fiber Termination Lapping
ฝัง
- เผยแพร่เมื่อ 15 ส.ค. 2024
- Microscope: www.banggood.c...
Lapping paste: www.banggood.c...
Behind the scenes: / marcoreps
github.com/mar...
In this episode of the laser series I am showing some updates and ideas for the constant current laser, that were inspired by you! One particularly interesting idea is to control the efficient DC-DC converter with an op-amp, eliminating inefficient linear regulation completely. But before the drivers can be tested I have to fix the termination of the optical fibers. Those were burnt and unusable. Even when tediously lapping the fibers manually, some special equipment was neccessary:
Lapping film: bit.ly/2t72scr
Vacuum pump: bit.ly/2JBq4kH
I think it's not a problem of how fine your sand paper is. The problem is not careful polishing. First I would try to make simple JIG to keep grinding angle constant. Some holder for optic fibers, and flat surface which you already have. For example: take thick piece of metal (even cube 2x2x2cm), drill a hole just slightly wider than diameter of fibers end, put fiber end inside, and move whole thing against sandpaper, polishing both metal and fibers, this should keep grinding angle constant and will result in nice flat surface instead of rounded one. Second thing: you have a lot of scratches left by coarser sandpaper, it means you were changing sand paper too quickly. Easiest way to do it is to change direction of move during polishing, for example: paper 1200 you move your setup against sandpaper back and forth, on one straight line, without turning, all scratches are made in one direction. Next you clean everything very well (also gap between holes walls and fibers, just take it apart and clean), cleaning is important to remove coarse particles, which may damage polished surface later. Next you go to paper 2000 ,polish the same way, nice linear moves, but now you turn fibers 90 degrees, now your new finer scratches are made perpendicularly to the ones made with 1200 grit. In this way you can easily see when you should move to next finer grit - polish as long as you can see old scratches. I am not saying that fine polishing diamond paste is not necessary, but first of all I would polish it correctly :)
Thank you for some valuable hints!!
Marco Reps Also try to use Philips contact spray(as lubricant) to wet the sand papers instead of water.
Or find someone to do some atomic polishing (argon-ion would do)
I think a figure eight movement is also recommended when polishing.
Would the jig not introduce small aluminum particles that would scratch the fiber?
That was just pre... substance
Marco this may be my favorite dry German delivery of a joke you have ever offered. Not premature in the slightest.
I've noticed Germans on youtube have a great, dry sense of humour, which goes against all the stereo types.
Yep, the sense of humour coupled with that seriuos voice makes me have a lot of fun watching Markos videos!
@Nicolas Gerald its a scam.
When I did cleaving of optical fibers it was usually done by removing the cladding and then scoring the fiber with a ruby blade, after that you put a drop of distilled water onto the fiber. The water will aid in separating the fiber along one plane of the crystal when you pull it apart. The resulting surface quality is already really good
nestoriusify could you explain a bit more? Water?? I know that a good cleave can be of optical quality already but you speak of pulling... I've read that it should be bent on a mandrel with specific angles and such. Would be a great help!!
Daniele Coccioli AFAIK you can use bending or pulling. But cleaving/splicing of fibers is one of thr fields where every lab has it's own procedure "because it worked last time". The pulling can be done in some kind of jig where you fix the fiber left and right of where you marred it and use a screw to pull. I guess that's also where the method with adding a drop of water comes from. Someone proposed it weakend the bonds between the layers and when we tried it the result was good.
9:10 Holy crap! i thought you were joking about the "Lets burn a hole in the stone" part, Damn! thats amazing!
He should put that under the microscope. I wonder if he actually burned a hole in it or if this is just on the surface
So do I!
When I did lapping for high magnification microscopy our best results came from moving in one direction then lifting. Then rotate 90 degrees. That way you aren't rubbing debris into the polish surface. After getting to ~2000 grit or so you would then move to a lubricated wheel and continue the one direction alternate 90 degree. Final finish was with a micron abrasive same method. If you didn't get a perfect surface your microscopy results would really suck because the slight hairline scratch is the size of Everest at a couple hundred thousand times magnification.
I'm rambling here, that was mainly for visible wavelengths. Lasers can be more or less forgiving. Longer than visible wavelengths can tolerate some small scratching, shorter than visible wavelength become increasingly intolerant of any surface imperfections.
For the lapping, you're doing it well, but I recommend you rinse the paper every 10 strokes or use a new spot. The tiny particles are the same hardness (or harder if it heated up at all) and will end up causing scratches. That being said, for a one off polish, I don't think you did anything bad. When you have thicker glass, it's more of a problem.
Btw i pretty much always mount mosfets or transistors "underneath" (that's relative) the board. Or otherwise on the side. So much better space saving, better board layout, better dissipation and less EMI.
Does that microscope have noticeable latency ? That't a problem I've seen on cheap video microscopes.
Built-in monitor just barely noticeable, HDMI output none at all. My HDMI recorder (left screen 6:35 ) adds a severe delay.
I would love to see a in depth look at that microscope.
You mean something like a latency measurement on HDMI signal with a DIY active differential probe supplied by TekVPI? I think something like that could be arranged ...
That sounds excellent.
I've had one of these for a while, I can recommend fitting an LED ring light as it really improves the view. flic.kr/p/25kBFm9
This one:
www.ebay.co.uk/itm/60LED-Adjustable-Ring-Light-Illuminator-Lamp-for-Stereo-Zoom-Microscope-/282715024608?var=&hash=item41d320ece0
As a workbench microscope i use: an "USB IMX 335" and a "100X Zoom c-mount Lens" on a camera stand. It works perfect, and on the computer it runs fine on windows 10 with the "camera" app that comes with windows.
Greetings,
Jeff
That was cool, looking into the fibres through the microscope was interesting - it was like spider eyeballs. The lapping worked really well
Never tried it on fibre optics but 3M fibre optic lapping film works wonders on kitchen knifes and a lot of wood working shops stock it too
You had me at "the whole dogs breakfast" TOP alter
Having the proper tools for filming makes a big difference in production quality for sure.
However your production techniques are impeccable and I loved the comparative optical clarity on eachpof the progressively finer laps throughout your process.
So much so that if I were you, I would make a few photo scenes of each and release it as an instructional pamphlet. Phenomenal work in both theory and application.
This is awesome! I went through the same thing with my laser, metal ferrule, scorched fibers, burnt epoxy, 'feeling' the polishing process. I ended up squeezing my fibers tightly into a syringe needle so I didn't have to use the epoxy. I did buy some high-temp fiber optic epoxy though, I just never used it. I purchased all of the parts to build the first edition of your driver and I look forward to building the second edition.
Ah, a syringe seems like a great idea indeed!
I see missed opportunity for clickspring collab. :D
Thanks for the Elektor Subscription in the other video (Tesla Coil). Bravo!
Uuuhhh, new intro sound, _fancy_
Grüß Gott, I might be two years too late, but I work with hollow fused silica capillaries a lot. These are similar to your glass fibres. In my opinion, the best way to cleave them is a ceramic cleaving stone. You can use it to give the capillary a nick and break it clean off. With a little bit of practice, it comes out perfect and a cleaving stone is about 5€.
Thanks for your videos!
From my experiments with polishing things to a mirror finish and examining the results on a microscope, I found the metal polish Brasso to be a fantastic substance. I usually let it dry on some cardboard, and use that as a polishing surface. I was sharpening a knife in this instance. I'm not sure what grit rating it would have, but it did get much better results. I will decline to say how long I spent achieving those results.
great content. good video showing why the fiber needs to have such a superb finish 👍
To get a tracking DC-DC converter i just use a BJT and a 2.7V zener.
Zener anode to DC-DC converter output, zener cathode to BJT emitter, BJT base to linear regulator output, BJT collector to the feedback input of the DC-DC converter and to a pullup resistor to DC-DC ref voltage.
"I got this 40kg granit surface plate" why do you have so much stuff?
Everybody likes to buy tools. You do, too, don't you?
The 40Kg granite surface plate store was having a clearance sale. Don't judge.
Á:/ 😄
Running the fiber that close to the smoking stone may deposit soot on the fiber, which can result in a burned fiber again. Get a lens inbetween, it focuses smaller and the lens is cheaper/easier to replace.
I have no criticism mate, but I thing a good ol' BRAVO would be constructive enough! Keep it up! I love your videos man, they are right up my alley, not to technical, not too simple! Just on or over my skills enough to motivate and teach me new things! Thanks!
the problem with lapping is the angle you grind at more than 12000+ grain. you should try to find a way to lap the fiber at 90, with hands you will create some micro "not 90 angles"
That would be by utilizing a polishing pad and polishing puck. The pad is a square or circular hard plastic substance that you attach lapping film to. The puck is a device that allows you to mount a connector at a perfect 90° angle and negate uneven grinding completely.
If you REALLY want to remove the camera icon there is always the option to remove it in post by taking a frame, using Photoshop 2021 CC object remover (now with AI support!), then using the cleaned up region to hide the icon. It is so much more comfortable than any other conceivable option.
This is awesome, cant wait for the next episode! When can we expect it?
be warned, i have the same vaccuum pump. It releases very fine silicon oil mist, very unhealthy indoors.
I tried to build a multistage filter, that mist goes trough anything (2xkitchenoilfilter+2xccarbonfilter. My temporary workaroud is attachting a hose, and put it outside the window. I want to try a cyclon-type filter when i have time for it.
Well even on the Telecommunications Fibre optics you have to remove the protective coating first before cutting.
The fiber is about 10 times thicker than regular telco fiber. That, and there's two layers - one is plastic sheath, which is made to be removable, and there's fused cladding, which isn't(usually).
I have the same Andonstar microscope for nearly 6 months, absolutely FANTASTIC in every single way!!! I was about to burn a LOT of cash on a high end fancy one (+/-$5000) when I saw some reviews showing this Andonstar, looked waaay superior than all the other cheap ones and perfect for electronics repair. I took the risk and wooowww... best decision ever!!! Latency using the original LCD screen is minimal, I would say not noticeable. I was always planning to connect to a bigger screen but I found the original screen is just perfect for my use. I was also worried about the LCD image quality but it is actually nice and sharp!! I really recommend it!!
Andonstar POWEEEEEERRR ohhh yeaaahhh =]
Hey when are we going to see more from you? The world is ready for some unreasonably high powered personal transportation :)
Marco Reps hahaha thaaanks!! I have been working too hard at Batrium.com and building/repairing some powerful electric cars this year but I’m coming back in style. Preparing my life to handle one video a week schedule =]
Just use superglue for mounting the fibres. The fully liquid type cyanacrylate not the pre polymerised gels.
I can't help but to express my love for all of your videos!
Have you thought about building your own sls?
th-cam.com/video/sbPpFHZL_cU/w-d-xo.html
Hi ,I think you have to look for finer diamont powder or paste to get it mirror polished.
I can't wait to see you running full power!
Dear Macro you need to first strip off the fiber coating on the fiber with a Stripper then you can use the cleaver please go through some videos on the internet about fiber splicing for reference
I was getting flashbacks to my PhD when you were polishing the termination. I was preparing optical fires for electron microscopy by setting the in epoxy in a brass tube then polishing metallographically. I think that some sort of jig to keep the bundle normal to the surface would help. There might be something useful in Journal of Materials Science Letters 12(1):1-2 but I can't find a free full-text version. But if you were to drill a hole the same size as the ferrule in a block of metal at least 25mm dia then push the ferrule through that you should end up with a more plane surface.
.... And I just noticed that I am at least 2 years late to the party.
Cleavers are only designed to work on bare fiber. The cladding needs to be removed beforehand.
Otherwise they make a squeaky toy noise when they fail (not gonna lie, I'd love for that to be a feature in tools).
you never remove the cladding, that is fused onto the core...the fiber would never work without cladding ;-)
If you're talking about standard fibers for telco', then the final layer you remove before cleaving is not cladding, but some acrylic resin and something soluble in isopropanol...
But still the fiber here is pretty much 10x thicker then the 125um fiber the cleaver was designed for. Thicker fiber=stronger fiber.
Personally I'd try a hand scoring tool and pulling the fiber, but that is going to require a lot of practice :D
AKAtheA oops. Yeah I knew that. Been a long time since I've worked with fiber. I mean coating:)
I think this is my new favorite channel!
The water is for lubrication and helps clear away the debris I would look into liquid cutting compounds for a real final polish to the fibre end thing, I'm sure you could get a mirror finish on it by hand.
do the lapping on the 3d printer ,
S1:Attach the termination to the print head
S2: weigh it down somehow paper binder clip on the fiber bundle and a spring pushing down on it , light pressure is enough
S3: write gcode to move the head in x and y in random motion
S4: gcode to do a figure 8 motion for final lapping
what do u all think is that possible ?
This fiber cleaver is a cheap chinese copy of original Japan (Sumitomo IIRC) and it does terrible work even for telecom fibers. Not sure if this have possibility to adjust height of cleave but my fujikura cleaver allows that. You may want to consider looking for Fujikura cleaver then. Probably Fitel could do the job too. :) Great job with lapping it anyway. Your channel is one of the best I found so far.
I don't know if you have fixed your fiber terminating setup yet but...my dude...ouch. Watching that was painful. I'm a professional fiber optic technician and engineer, and I can tell you that there are hundreds of products online that can help you out for relatively cheap. First some fiber strippers...that will remove the acrylite coating around the fiber. That is actually what caused the cleaver you used to fail, not the diameter of the fiber. Next that cleaver you have is perfect for fusion splicing, but as you aren't splicing I would recommend a Ruby pen scribe or replace the disc inside with a Ruby wheel for more accuracy (removal of the acrylite coating would be enough to make it work though). Next, there are actual epoxies designed for making "ribbon" fiber. In other words putting a bunch of individual fiber strands together permanently. Utilizing a hotmelt kit that can be found for super cheaply online would allow you to mount a termination properly (if you use this, a Ruby pen scribe will be necessary). After that a polishing pad, lapping film, and a polishing puck (there are pucks for every kind of connector/termination) will make your fiber polishing mirror smooth and exactly the same on every fiber. If you follow these steps (I can provide more detailed instructions in an email if you wish) your fiber will become far cleaner and your end result will be as close to perfect as is possible with currently available methods.
You could use diamond polishing paste for a true mirror finish on the fiber optic.
You sound like Vigil from Mass Effect but with a German accent. I love it.
This was the funniest and interesting scientific video that I ever seen in the last 15 minutes.. :D and btw it amazes me how powerful is that microscope!
while polishing make 8 pattern not circle also using distiled water and glass work plate
awesome project and video! Looking forward to the next one!
Hahaha... 30w laser, ohhh yeaaahhh!!!!
*I don’t think people even realise how POWERFUL a 30w laser is... so much POWEEEEEERRR =]
I worked with a 50w back in the days in University (Physical Chemistry / Quantum Chemistry). And your comment in terms of POWEEEEEERRR is waaaay to underrated ;)
DatSwissGuy I’ve played with some supa POWAAAHH ones too but having a 30w at home is just endless fun haha ohhh yeaaahhh =]
When they say "30W laser" are they talking about power output or power usage? Either way there's not going to be 100% conversion from electrical energy to light. So how efficient is the conversion?
supah powaahhh \o/
30W out of 19 multimode fibers is not that great actually.
I enjoyed your regurgitation metaphor.
what text to speech program do you use? it sounds fantastic!
:^), jokes aside, enjoy all your vids! you are the this old tony of electronics. danke schon!
I’ve used a rubbing compound to finish optical surfaces, 3m auto paint rubbing compound does a good job, and likely has better particle size consistency than that banggood stuff.
Marco could you open source that constant current driver?
It is. Search marcoreps on GitHub.
Collimator and multiple emitter elements going into a fiber optic line? That might be interesting. Also a thermal probe next to the emitter end of the line, although you probably wouldn't have very long if it goes wrong.
The firmware on those Andonstars is easily hackable; you can solder on a 3v3 uart and "poke" the firmware in ram via the built-in shell (simple RTOS not linux) to do nondestructive changes. Examples include fixing stupidnesses in the UI, removing onscreen icons, etc.
Oh? Got a link? I would really like it to display stuff and send to HDMI recorder at the same time ... Not sure if that is a firmware thing tho, might need an external splitter in the end
I'm not certain how fine diamond polishing paste comes in, but I'd imagine that you could visit some jeweller websites and find some ridiculously fine polishing paste. You did an excellent job on the polish though.
I have the version before this and OnStar latency is like 350 milliseconds at my guess you notice it only just. It's almost instant and OnStar looks like this but round slight woobly mechanical adjusting via HDMI -dvi input on my repaired CMV monitor. And OnStar is a good product the one who look A like in this video. I paid 200$ NZ purchase from someone local auction site it was new only thing had wrong USB charging cable with it .
wow, thanks for the video! I might just build a SLM printer now! Add a heated chamber and a mirrored galvanometer system and you could make ultra high resolution prints out of a plethora of plastics and maybe even a couple of metals!
That's what they do - except metal sintering is done under an inert atmosphere. Better start with nylon, easier to do.
Hi Marco dont know if you noticed, but banggood also has diamond lapping paste down to 0.25 micron. very affordable.
TH-cam channel Pierre's Garage made a video about the banggood lapping paste that show use and result.
7:09
*DEEP SCRATCHES
My heart: NOOOO :O
für best mögliche politur ergebnisse empfehle ich diamantspray. das zeug gibts bis abartige körnungen. ich benutze 60K grid um meine messer zu schärfen.
glass polishing compound is what you need.
2:38 killed me XD XD
Most maniacle episode.
What of ample heat sink, hole drilled for fibers, epoxied in-place, then polished, to maximize heat build-up at the fiber ends?
Amazing video as always! Not gonna lie - I'm kinda jealous now. That laser 😍
the microscope itself can also capture video btw, just got the smaller adsm301 a week ago
Just wow! You never seize to amaze!
He's the best jewel thief.
I don't think anybody seizes to amaze. it's involuntary,
If you don't mind sharing, where did you find the SMT lasorb package? The manufacturer only sells in units of 100.
i think i found my new favorite youtube channel lol
The heat sensitivity of the epoxy seems problematic, especially with proximity to the substance being burned. Maybe optics to increase focal length, and cooling (passive or a small fan) for the output end. Polishing is tough free hand. Make a jig to keep the fibers dead flat, or make the fibers stationary and move the paper akin to a belt sander.
I also own that same microscope. I love it.
Am I the only one here who has no clue on what he's doing but it seems intersting so I watch it anyways.
you should use jewelers rouge on news paper for a mirror finish.
That would work, silica has a knoop hardness of 1200 and fe3o2 has a knoop hardness of 2000. I would be worried about the newspaper lifting up or something, maybe spread it on something. At that point though, you might as well get a polishing liquid compound.
Another great video. You've got some great kit.
Wow, interesting project 👍😀
Thanks for sharing 😀👍
Your coatimg on the optical fibre is probably polyimide. This really needs to be stripped first before cleaving. This can be done in a number of ways; plasma beam (expensive), mechanical (difficult) or high temperature sulphuric acid ( not as dangerous as it seems as the amount of acid are tin) Still in the last case great care need to be taken with 180oC plus acid. Small amounts can be place in test tubes mounted in heated metal blocks (cartridge heater) and the fibre dipped. At high temp the coating comes off in a few seconds. Looks like you have a cheap copy of a fujikura cleaver, the original works very well, how about this one?
Sort of realise how old this video is, but for polishing fibre - you use something called a "polishing puck" which always holds the fibre at exactly 90 degrees to the surface.
You need to remove the coating before cleaving the fiber.
The whole point of the coating is to strengthen the fiber so it seems to do its job.
Since you are going to attach the termination to a CNC anyways, can't the CNC help in the lapping process as well? Total CNC newbie here.
That microscope seems to be awesome for it's price!
Nicee... i see miniSpartan6... lovely little hdmi dev board
vacuumchamber with siliconeinlay...combined with the ventilationhole which u cut in the wall... i probably know whats going on at your house. nice:)))
Would it be insane to try and put a heatsink on the termination?
I think the problem isn't the heat generated but that heat is also damaging the material (and then making it create more heat).
Also yay the AI safety guy!
Don't forget the thermal compound LMAO
This seems like a good idea.
Ideally the end shouldn't get hot but it obviously is heating up so a way to remove heat seems like a good bandaid.
I was thinking just like, drill a close fitting hole in some aluminium
This would also help with lapping the surface as it would keep it square.
Would an over polished fibre end start reflecting the light back down the strand, like an internal mirror?
Total internal reflection happen only on lower angles.
Fresnel reflection happen but i dont think is a major issue as some light propagation pattern couldnt sustain after reflect.
needed scope like that, amazon had it. thanks.
water not only helps, but is needed at those grits.
Use a paste polishing compound and the dremel.
The local HF store actually sells a very good rotary vacuum pump.
Love your work!!
Can it burn a hole in a mint to make an artificial laser polo mint?
can this laser cut metal potentially? and what thickness?
You can etch the end wit HF
Really awesome work, dude. 😊
Ever thought about making a jig on a CNC to do your fiber lapping for you?
Great little high current driver you made! I am looking at getting a 10 watt NECSEL Laser head, I think it runs about 14 amps in.
Do you have a schematic for your high current driver, are any available for purchase?
what type of de-soldering air gun do you use? or if anyone knows of a decent one thats not way over budget pls let me know.
This is so cool. I’d really like to have a metal-marking fiber laser, so when you get the DC-DC supply working, I’d very much like to order a PCB. Also, can you point me to a suitable laser diode? And optics? 😁
Is there some way to make the microscope take photos with the native sensor resolution ?
The thing is fine so far, maybe a bit noisy, but the photos are really bad - they are interpolated so much, that so much detail is in fact lost.
I'm trying to read eprom code from delidded old ICs, but it's a complete fail.
Did you manage to get the faces square ??? I noticed on mine, that I had to offset the faces from 90 degrees, just a hair off to get full performance of my 80 watt fiber laser. for some reason, (At least the crude setup I had) if I managed to get totally square (at least what I thought was) the performance wasn't as good. face reflection ???
Dunno, but a suggestion.
Nicely done anyway!
That fiber cleaver is a cheap Chinese knockoff. The real one is made by Sumitomo and is actually one of the best cleavers I have ever used for 125um fibers.
Also just a side note, it likely didn't cleave because you didn't strip iff the insulation. Only the glass portion is supposed to sit on the black rubber pads, if there is insulation there the blade won't cut deep enough.