🚨 *Embarassing correction!* 🚨 After publishing I realized the scale on my calibration grid was slightly incorrect, and so the feature sizes are wrong. Min average feature size is 10-15 microns, not 5-10 as I stated in the video. There are examples of ~5um but these are with "unreliable" settings that are prone to bridging... the crisp, consistent examples showed in the video are closer to 10-15. Apologies for this goof, it's pretty embarassing and definitely makes the results less impressive. I think the feature size could be reduced down to the 5 micron I mentioned, but will take more tuning. Wanted to mention in case anyone tries to replicate this for themselves!
Yeah tricky to get small features with such long pulses(HAZ ends up being kind of huge), unfortunately 10ps lasers are still a bit tricky to get your hands on. Also really depends on numerical aperture of your optics setup, looked like you are using a galvo f-theta setup which is nice for speed/convenience but not great for spot size. I've managed ~2um with a q switched YLF 8-10ns, but that with an objective setup w/linear motor stage and was selectively removing sputtered copper to make some transmissive gratings. For that 1thou shim material(steel, brass) though same deal as you, 5-10um if being patient, too much melt, not enough ablation with long pulses. Wavelength isn't making life easier either, probably worth the time to consider 355nm.
Agreed with all the above! Looking to get my hand on a 532nm at least (would love 355), and sorely tempted to rip the laser out of the galvo setup and just stick it above a precision motion stage. The software that comes with these lasers is absolute garbage, and I'd rather just code up my own gcode converter or something instead :) Would love to find a fs or ps laser, but as you said they are still a bit tough to find on the used or import market :( One thing I'm pondering right now is doing some kind of partial ablation (e.g. just a pass or two which removes material but doesn't fully punch through) and then etching the rest of the way chemically. If it's isotropic in theory the partially ablated regions should etch through first, maybe leading to finer features? Sorta removes the convenience advantage though... just go use photoresist at that point heh. Funny that you mentioned removing sputtered layers. I was playing around with that a while back and had good luck with patterning, but the power is waaay too high. Even at 1% and as fast as you can go while maintaining overlap, the ablation spot was around 30um. Been trying to think of a way to stick a attenuator or beam splitter somewhere in the optical path do pull the power down even more. Would be a really nice/fast way to generate photomasks.
@@BreakingTaps if you do manage to get you hands on something with a fairly short pulse but still in 1064nm ish land, you can setup a second or third harmonic generator pretty easily for these high intensity pulses. Yeah there are conversion losses, but the increase in absorption offsets it easily for most materials. Only real drawback is they are hilariously temp sensitive, and optimization is tedious. On the partial and then etching.. isotropic material sounds pretty wishful thinking to me. Grain structure gets all sorts of mucked up at the site due to local melting from long pulses. For the thin film ablation, I was working with copper on glass slides, so pretty good mirror at a micron lambda. Combined that with a decent objective, running the pulse rate on the high end and keeping the pump low to prevent high energy pulses. This was a slab style folded resonator laser though, suspect the gain was a good bit lower than your fiber laser. Assuming your beam isn't tiny(otherwise expand it) you can just use a low ND filter at 45 and use the reflected as output and beam dump the transmitted(or the reverse). Might worry about having an AR coat on the back of that filter though for these applications and beam quality considerations.
3 года назад+48
How much talent can one single person accumulate. This is awesome!
Can confirm for those (like myself) who DO NOT work in a fab all the time, even PHOTOlithography at sub 10 um becomes nontrivial - I'm really glad I've never needed to develop a recipe from scratch! On metal, and without using photo or e-beam litho to start, these are some seriously impressive results! The magnet idea makes me smack my forehead - I did one project in undergrad using shadowmasks because we were afraid photolithography developer was leaving an unclean surface behind, and I did not use magnets to hold the mask in place, just tape - I made many many fuzzy circles...
I can't take too much credit, the main inspiration came from Ingle, Frank W. "A shadow mask for sputtered films." Review of Scientific Instruments 45.11 (1974): 1460-1461. They use the magnet system to keep it clamped in the deposition chamber, and I just extended that to the patterning step too :)
First time here. The movie piece at the beginning made me think it was an ad about an actual new movie! Very well done! Sometimes a stencil mask (Shadow mask) is the only option. I have the photolithography tool, but I need to use a shadow mask on porous silicon surfaces. Porous silicon is too fragile to sustain photoresist and develop processes. and I only need one process step. So no alignment requirement. And you are doing this in your garage??? No way! a sputtering equipment and a AFM tool in a garage? 😲
I'm "only" an electronics/software tinkerer... but I'm really enjoying a lot of the content you're producing. It's shades of "Applied Science" which has got to be good. Keep up The Great Work.
Thanks! I aspire to be a tenth as knowledgeable and interesting as Ben some day. He's a national treasure! :) Can't wait for his next video, it sounds very interesting
One like for the excellent intro. Another like for the subject matter. Finally a like for the future garage fab. - I think that works for the engagement algorithm.
i use 12 micron drill bits on a daily basis at my job for making specialized valves in what i assume to be some sort of contract with IBM. im not given the details, just the blueprints and the tools
This is awesome! I'll just add a few things I've picked up from my materials science classes/the internet Mask alignment is a huge challenge even in lithography, and they do so with shapes on the edges of the mask that have to line up (for example, triangles or crosses at different points along the edge of the mask). Also, my understanding is that masks for lithography are actually made via electron beam exposure, instead of UV light. The electron beam approach is considerably more accurate than UV light based methods, but is a serial process and therefore is prohibitively slow for large scale manufacturing. Lastly, if you ramp up the voltages (and therefore ion energy) in the sputtering setup, you could actually begin to implant ions into your substrate. It could be very interesting if you purchased a small silicon wafer (the ones that are a few inches are a few hundred dollars I believe), and actually tried to make a transistor for real!
Definitely worth making an aluminum carrier for your shadow masks! .1mm (100um) offset from your features on something stiff (2ish mm) cut with a 90 chamfer mill will help a ton to keep everything flat and stable. We use a system close to that at uni for our simple sputter/evap masks we used to test a bunch of materials for thin-film transistors.
@@BreakingTaps If you're going to do your dev work on 75x25mm slides, I'd definitely make an assembly where you can drop it into a pocket (drill out the corners so you can get a tweezer in there to release) and then screw around the edges. We had dowels that would hold the shadow mask (oversized through holes to prevent tearing) around the periphery, too. Anything and everything you can do to mechanically align and park features helps a ton. We were (use in prior post should be *used*) registering multiple layers (masks) and holding 50'ish microns across multiple layers. Not that it was super critical in our workflow, but still nice.
*Super* interesting as usual! I wonder if some sort of compressed-air-jet cooling could keep things cool enough for multi-pass cutting? I’d think that more shallow cuts and cooling the material between passes would let you go a lot deeper without warping... (PS: I don’t think 10-15 microns is anything to be ashamed of. It isn’t exactly microelectronic scale, but still seems very impressive, especially since you’re punching through 25 microns of material. Here’s another thought: what if you took advantage of the closing up of the shadow mask as it’s used to deliberately pre-narrow the features by depositing on it before using it for the parts you’re wanting to make? And WOAH! An AFM??!! COOL!!!
Hmm, yeah that's a good point, some downtime in between passes and an active cooling solution might help minimize how much it warps. And would probably allow smaller features, since you could keep the "area of ablation" relatively small. Interesting idea, would be easy to test! And that's a super clever idea purposefully closing up the gap to get smaller features. I love it, definitely going to try that. If it works it might warrant a followup because it's a great hack. Cheers! (And yeah i'm super stoked for the AFM. Gonna be awesome!)
If Wiki is correct, the first 4000 series CMOS logic chips were made in a 20um process. Making a functional simple gate or a flip-flop would be very impressive as a start!
Oh interesting, I didn't realize they started out that large. Like I knew the process nodes started pretty big, but figured it wasn't until they had brought it down a little more before making chips. It's definitely on my bucket list! And I can go binge through Sam Zeloof's videos again if I get stuck on steps :)
That intro, hot damn. One of my favorite movies of all time. Looks like you had a good time editing that together in addition to doing everything else. (:
Nice video Zach! So is your sputter coater something home build or a commercial one? When I worked on oled technology (long time ago) we also used shadow masks. These were cnc milled out of aluminium, had rigid edges and had a tiny ridge to keep the device surface away from the mask to avoid damage of the devices. By the way, lithography is not that difficult and can be done fairly easily with a cheap 4k resin printer at 35um resolution. If you use "lift off" technology you can even avoid etching.
Thanks! Commercial machine I'm afraid, so not very interesting to talk about on video :) I was going to build one originally (chamber in background) but found this unit for a good deal because it needed some repairs which ended up being a simple software hack/tweak. It's a Leica SEM coater, so has lots of nice features (tilt and rotation stage, dual sputter heads, integrated shutters, turbo and internal diaphragm pump) but it since it was built for thin, even coatings it is also very slow. And DC, so can only do metals. That's interesting that the shadowmasks were milled! I suppose a sharp v-tip could make a pretty narrow cut, and you'd have the advantage of a stiffer mask to prevent warping. Interesting, might try that. Makes sense that some kind of lip or ridge would be needed to prevent scratching of the device. By time I was on my third layer you could see some damage had accumulated...doesn't take much to scratch those thin film layers :) Litho is next! I was just being super lazy haha :) That's a clever idea resuising a SLA printer as a contact litho machine!
@@BreakingTaps Good luck with the litho. Watch out, it is actually pretty addictive ;-). You can achieve much higher resolutions than 35 um by demagnified projection of the LCD of the 3d printer.
Haha yeah I know what you mean :) I really lucked out grabbing that one. But no worries, I have some vague plans to use my WIP chamber (in the background) for some laser ablation deposition which should be fun and quite an adventure I think :)
Not sure if you saw it, but Huygen's Optics recently put up a video on his home built PVD setup. (Although it is perhaps a little above what most of us can reasonably build.)
I wish you made this video a few more years back when I wasted 4 months trying to use a shadow mask to work in an evaporation process that kept shorting the top and bottom layers. :(
do i understand correctly that one of major limitations with laser cutting a shadow mask is that you can only cut shapes that don't have any dangling/disconnected bits? and that sounds like a major limitation for general-purpose use. upd: ah, you did mention that later
I think your channel quality is high enough already to have at least hundreds of thousands of subs and a decent running patreon. But finding your channel is actually non-trivial unless you're specifically searching it by name. I think what you need is a collab with one of the "big" youtubers like Thought Emporium or Appled Science to get some of their viewers because they'll no doubt like your stuff if they see it.
You can clamp the heat deformed shim stock between a couple of thicker 1/8" or thicker pieces of steel with a couple of small cheap C- clamps, put it in an oven for an hour at 450F , let it cool until it can be handled remove the clamps and your shim stock will be as flat as the pieces you clamped it in.
Have you tried using some kind of thermal adhesive tape under the metal to absorb some of the extra heat while cutting? That may help mitigate the thermal warping issue. Use a solvent to get it off the adhesive later or something.
I haven't, but that's a great idea! Will try that out next time. I also got some new parameter suggestions from another viewer that supposedly works better, less warping although takes a little longer. 🤞
Would *love* to see a tour of your sputtering and vacuum pump/chamber setup. I've tried it myself and there are definitely many, many traps novices like myself fall into, so seeing what you've got working would be amazing. Did you machine your own sputter-gun or buy it from some company, and if so how much was it? Have you sourced most of your stuff used off eBay? Great video BTW!
Thanks! Afraid the sputtering setup isn't very interesting, it's an off-the-shelf SEM sputter coater from leica. I snagged it for a very good deal because it was throwing some errors, and the repair ended up being pretty trivial. But otherwise it's all stock and therefore not very interesting :) Load sample, hit go, wait for thin film. I am going to put together a laser pulse deposition chamber later this summer I think (the big glowing chamber in the background) and plan to do a video about the chamber all the little bits I learn along the way. I have most everything I need, just need to machine up some adapters and find time to work on it!
Not that it would be recommended or even wise, but I wonder if a super critical fluid (CO²) would be a good option for heat dissipation of thinner metals. Might be enough to keep it from deforming. Also given the short distance needed cutting, I wonder if a gas or other fluid could be used to narrow/focus the beam ( like a fiber optic ) more so than expensive optical lenses could focus the same beam. Maybe two fluids in laminar flow with one in the center surrounded by the other, one hell of an engineering challenge there. Just the ramblings of a madman.
Hm, did you try using a thermally conductive backer on your thicker shim stock stencils when you cut them? Like, maybe if you used aluminum or copper (or silver) rather than glass you could reduce warpage by drawing away the heat. Might not suck away the heat fast enough to matter, but it'd probably be worth investigating. Another option might be to sputter a layer of copper or whatever on one or both sides of the shim stock to give you a very small amount of added thickness and again to help draw away heat. That might introduce new warping problems though, a la bimetallic strip.
Don't think I've mentioned it explicitly in a video (just some community posts). It's a Leica EM ACE600 machine, designed for SEM coatings. So it can pull a nice vacuum and has fun features like that tilting/rotating stage... but it's pretty slow. I was originally going to build my own but this came up for sale at a good price (needed repairs) so I couldn't resist :)
I think that my prior slight worry about the creative video intro sequences migh have been from it being the first time. I didnt feel worry about this one, I guess its because it was expected now.
Yay! Great to hear! Definitely won't happen on all the videos, this one was a ton of work for like 30 seconds :) But it was a fun diversion for me to play with. And e.g. the next one on AFM has waaaay too much interesting content to waste a few minutes up front on an intro, so it'll definitely be dependent on what's getting talked about. Thanks for watching!
Maaaaybe. I'd like to try pulsed laser deposition at some point, and depending on how well that works (or doesn't) I might convert it to thermal evap instead. Love the sputter coater but as you said, directionality isn't great. And it's very slow (and can only do metals). Something that can deposit a micron of material quickly would be very appreciated :)
Afraid it's a commercial machine (Leica EM Ace600). I got it for a screaming deal on ebay because it was inoperable, but the fix was relatively simple (some uhh, software hackery involved). Little sad that I didn't make it because it's not terribly interesting to talk about being commercial, but it works quite nicely so I'm not too sad :) Only downside is that it's a DC magnetron so can only deposit metals, and it's _very_ slow because it's designed for SEM coatings.
Yes! I would _love_ to get my hands on a picosecond or femtosecond :) Unfortunately don't seem nearly as easy to find on the surplus market, (for reasonable prices), have had saved searches running for months hoping something would turn up. I dream of the day I can do femtosecond micromachining of metal and glass 😍
@@BreakingTaps I do it every day, lasers are the best machining tool, contact free and free of consumables. Photolitho, at least for micrometer sized structures is so 19th century, althrough it is almost unbeatable at accuracy, many applications can be replaced by direct laser structuring. I have daily access to more than 30 laser beam sources... it's great fun but therefore I have no need for more and got rid of most of my private beam sources. Ultrashort pulse lasers with enough pulse energy to ablate will likely not fall much more in cost any time soon, but average power goes up and up. We're reaching 10kW+ at this point... now I leave it up to your imagination as to what we do with those :P
I have two slightly off topic questions I cannot resist asking since you seem like you might know the answers. First, aside from scraping it off, do you know of a way to remove the CFA/Bayer layer/baked on mask from a color image sensor to make it black and white but more sensitive to a broader spectrum of light? Second, when I attempt to make circuit boards by applying a UV sensitive film, the film doesn't properly wash away from where I want to etch copper and partly washes away where I want to keep the copper. Am I perhaps using bad film or is it more likely I lack skill? I've tried varying the variables but never get good results. I came for the vapor deposition. I just bought a vacuum pump and am making a vacuum chamber and eventually everything for sputtering success at sputtering.
I stumbled across this paper a while ago (www.ncbi.nlm.nih.gov/pmc/articles/PMC5087437/) which removed the filter with some photoresist stripper (Posistrip EKC830). I doubt this is easily available though, and the MSDS suggests it uses some intense organic solvents (N-methylpyrrolidone and 2-(2-Aminoethoxy)ethanol) apps.mnc.umn.edu/pub/msds/ekc830.pdf). Looks like you can get N-methylpyrrolidone on ebay at least, although not sure I'd want to use it personally :) Otherwise, scraping the filter off is the only method I've seen. Regarding the PCB film... I've actually never etched a PCB before! So no good opinions or experience from me on that topic :) I know liquid photoresist can go "off" over time, so it's possible your film might just be bad, or perhaps the PCB board isn't fully cleaned before applying? Before tossing the film, I might try some exposure grids to see if you can dial in the exposure time a little more. E.g. expose one section for a second, expose another section for two seconds, etc etc up until some large value and see which gives best contrast. Goodluck!
@@BreakingTaps Yes, EKC830 isn't available at Walmart:-) I started down the N-methylpyrrolidone route and stopped when I realized I'd be buying way more than I want laying around. It is an ingredient in some fairly common household product (herbicide?) so I might try it. Or maybe I'll keep practising with scraping, go for the nickname "Chip Slayer."
Not exactly in that setup, but I did try just regular copy paper by itself and it works quite well! I saw a paper a while back which used filter paper (www.mdpi.com/2072-666X/11/7/676/pdf) and got pretty good resolution. They used a CO2 laser, but I used my 450nm and white copy paper or brown kraft paper, both worked nicely. Think I got 30-40 micron resolution which is shockingly good for just paper in my opinion :)
Ah yeah, I wanted to mention that but totally forgot! In theory this would be perfect, assuming it doesn't "lift off" prior layers (which it probably wouldn't, metal adheres pretty well to glass). But most adhesives (or any polymer really) outgas like crazy in a vacuum and can make sputtering difficult. The kapton tape I used is "good" for vacuum and still outgasses a lot, regular spray adhesives ended up being really bad. I can almost garauntee there is a "vacuum rated" spray adhesive somewhere but I dunno how/where to find it :)
@@JuryDutySummons Hehe yeah i didn't either, figured that one out the hard way. Was pumping for ages and still not getting to desired pressure... then it dawned on me :)
Thanks! Pressure varies a little depending on the target, but I generally pump to e-5 mbar, purge once or twice with argon, then take it down to e-6 mbar pre-process (especially important for aluminum or copper due to oxides, if I'm not in a rush I'll let the aluminum runs go down to e-7). Sputtering is usually done around 2.0e-2 mbar. I'm still pretty new to sputtering though, so huge grain of salt to all the above. It's possible one or more aspects of that is quite sub-optimal :)
Can't you anneal the mask while sandwiched between flat plates? Before and/or after laser. Also, is it feasible to (electro)chemically etch back the buildup to restore the mask?
Hmm, seems like that would work. Actually now that you mention it, since this is cold rolled it probably has a ton of internal stress...pre-annealing might help a bunch too to relieve the stress before the lasering. Hmmm... would be pretty easy to test. I think in theory you could electro-etch the buildup back out, but probably have to be careful to make sure it etches evenly. Although if you design the setup correctly, you could find a chemical etchant that doesn't touch the mask. Like using ferric chloride to etch out copper buildup and leave the steel alone.
Dry etching via plasma? Hmm, I'm not sure. I suspect most metals etch at the same or faster rate than glass under an oxygen/argon or reactive plasma. You'd probably need to deposit something harder like SiN which would require a reactive DC, RF, or CVD system. I did find this paper on a quick skim that used a silicon wafer as a hard mask: www.researchgate.net/publication/223079651_Deep_plasma_etching_of_glass_with_a_silicon_shadow_mask
Thanks! Fiber laser is a random chinese 50W non-MOPA laser (if you search for "50w fiber laser" it's the variety with a galvo scanner and head that moves up/down with a crank). I believe mine has a JPT laser inside, although Raycus is common too. Sputter coater is a Leica EM ACE600. Got it used for a _very_ good price because it was non-functional, repair ended up being very easy (software fix). :)
It is sad that the video has rn only 612 likes in comparison to the 4.2k views so Guys that video was great and need more attention so use the knowledge we have about the algorithm (Spiffing Brit has shown it) and interact with the video leav a like and make a comment that's what helps this channel the most from our side
Agreed! Each time i pull an image off the machine I'm just stunned by how cool it is. And watching the probe dance around is mesmerizing. Re: semiconductors, maybe! Hopefully! I'm currently playing around with some alternative materials because I really don't enjoy working with HF. Curious to see how far (or if) you can go with alternative processes. Makes complete sense why silicon is used industrially, but for home fab there might be different routes because the goals are different. We'll see! :)
@@BreakingTaps I'm not very familiar with sputter deposition for semiconductors - you'll probably need something defect-tolerant enough to work even when polycrystalline (like the much-hyped halide perovskites in research-grade solar cells), something commonly solution-grown, or an amorphous semiconductor like amorphous Si, although I agree 100%, HF in the home shop feels like a very dangerous idea - keep the calcium gel nearby!
@@AlphaPhoenixChannel Hmm hadn't thought about that, good point! Will need to do some more reading. Right now some candidates are a few van der waal materials (WS2 and MoS2) which can be applied via abrasion (e.g. rub the powder on the substrate). Causes the platelets to exfoliate themselves like graphite and form a sorta-flat multi-layer. Most of the literature focuses on those as single monolayers, but I found a few references of it being used as a "bulk" material to some ok effect. I assume the literature ignores the bulk stuff since electrons don't like to jump platelets so it's a lot less interesting? Dunno, but I'm less picky :) The other thing I'm reading into is Cu3N and CuO/CuO2. The nitride form seems to happily self-dope itself n- or p-type depending on deposition temperature, the paper I was reading suggested it was a balance between donor and acceptor defects and the temperature changes the balance slightly. Dunno just how defect tolerant it is though, I was indeed planning to sputter that one. The copper oxide ones can be generated by selective heating/annealing so it probably helps fix itself during the process? I may be bugging you with more questions hehehe. ZnO is a definite candidate as @Daniel Heineck mentioned. I haven't done much reading on that one yet but it's on the todo list :) As you said, ye olde HF in the home shop doesn't sound fun at all if I can avoid it :)
How do you people get money for all these stuff? Around 7:55 you complain about small bridges in the material. You could try to get rid of them using something like electropolishing (it doesn't work with all metals), or just simply try to watch it a bit chemically.
I was lucky to have a well paying job (software), no children, stalking ebay for deals, and my discretionary money goes into hobby projects/equipment instead of things like cars or fancy vacations or whatever. When we do vacation we're typically camping in a tent or something cheap, because that's the kind of fun we like :) So yeah, basically just directing funds other people spend on new phones or cars or eating out all the time or whatever into my hobby projects instead. And slowly accumulating the tools that I want over the years. Good ideas about the bridging! I think something like that could work, and would help a lot
My dude, I love your videos and your production quality is fantastic. but what the fuck is this title. PLEASE for your own good rename this video. This title is like choosing to wear clogs to play in the NBA...
Haha yeah I know. It's cool, I have accepted that this video is a dud due to the intro and the title. 🙂 Viewers are confused by the intro if the title doesn't allude to it, and the title is confusing if it does mention Blade Runner somehow. I tried various combinations and it just never worked out. Lessons were learned, I'm not too fussed! Cheers for watching!
@@BreakingTaps I assume you’ve seen Veritassiums video on click bait titles? The clickbait life sounds morally challenging to navigate but the proof is in the pudding. Thanks for the interesting videos!
🚨 *Embarassing correction!* 🚨 After publishing I realized the scale on my calibration grid was slightly incorrect, and so the feature sizes are wrong. Min average feature size is 10-15 microns, not 5-10 as I stated in the video. There are examples of ~5um but these are with "unreliable" settings that are prone to bridging... the crisp, consistent examples showed in the video are closer to 10-15. Apologies for this goof, it's pretty embarassing and definitely makes the results less impressive. I think the feature size could be reduced down to the 5 micron I mentioned, but will take more tuning. Wanted to mention in case anyone tries to replicate this for themselves!
Don't worry, we all break some taps. 👍
Yeah tricky to get small features with such long pulses(HAZ ends up being kind of huge), unfortunately 10ps lasers are still a bit tricky to get your hands on. Also really depends on numerical aperture of your optics setup, looked like you are using a galvo f-theta setup which is nice for speed/convenience but not great for spot size. I've managed ~2um with a q switched YLF 8-10ns, but that with an objective setup w/linear motor stage and was selectively removing sputtered copper to make some transmissive gratings. For that 1thou shim material(steel, brass) though same deal as you, 5-10um if being patient, too much melt, not enough ablation with long pulses. Wavelength isn't making life easier either, probably worth the time to consider 355nm.
Agreed with all the above! Looking to get my hand on a 532nm at least (would love 355), and sorely tempted to rip the laser out of the galvo setup and just stick it above a precision motion stage. The software that comes with these lasers is absolute garbage, and I'd rather just code up my own gcode converter or something instead :)
Would love to find a fs or ps laser, but as you said they are still a bit tough to find on the used or import market :(
One thing I'm pondering right now is doing some kind of partial ablation (e.g. just a pass or two which removes material but doesn't fully punch through) and then etching the rest of the way chemically. If it's isotropic in theory the partially ablated regions should etch through first, maybe leading to finer features? Sorta removes the convenience advantage though... just go use photoresist at that point heh.
Funny that you mentioned removing sputtered layers. I was playing around with that a while back and had good luck with patterning, but the power is waaay too high. Even at 1% and as fast as you can go while maintaining overlap, the ablation spot was around 30um. Been trying to think of a way to stick a attenuator or beam splitter somewhere in the optical path do pull the power down even more. Would be a really nice/fast way to generate photomasks.
If you can get it down to 8 micron you can make an 8080 chip :) It'd be slightly bigger but properly within an order of magnitude.
@@BreakingTaps if you do manage to get you hands on something with a fairly short pulse but still in 1064nm ish land, you can setup a second or third harmonic generator pretty easily for these high intensity pulses. Yeah there are conversion losses, but the increase in absorption offsets it easily for most materials. Only real drawback is they are hilariously temp sensitive, and optimization is tedious.
On the partial and then etching.. isotropic material sounds pretty wishful thinking to me. Grain structure gets all sorts of mucked up at the site due to local melting from long pulses.
For the thin film ablation, I was working with copper on glass slides, so pretty good mirror at a micron lambda. Combined that with a decent objective, running the pulse rate on the high end and keeping the pump low to prevent high energy pulses. This was a slab style folded resonator laser though, suspect the gain was a good bit lower than your fiber laser. Assuming your beam isn't tiny(otherwise expand it) you can just use a low ND filter at 45 and use the reflected as output and beam dump the transmitted(or the reverse). Might worry about having an AR coat on the back of that filter though for these applications and beam quality considerations.
How much talent can one single person accumulate. This is awesome!
I'd like to think one accumulates skill not talent.
Imagine a collab between this guy and Applied Science🤯 they could probably land on mars, cure cancer, and produce the most amazing videos all at once!
The sci-eng youtuber I thought of was thought emporium.
@@Jimunu why not all three!
Nighthawk in Light too.
..what on earth will they do after lunch then?
Fusion, guys. Come on. We've been 20 years away for 60 years now.
Why does breaking taps dont have more subscribers? Im blown away by the quality content.
Can confirm for those (like myself) who DO NOT work in a fab all the time, even PHOTOlithography at sub 10 um becomes nontrivial - I'm really glad I've never needed to develop a recipe from scratch! On metal, and without using photo or e-beam litho to start, these are some seriously impressive results! The magnet idea makes me smack my forehead - I did one project in undergrad using shadowmasks because we were afraid photolithography developer was leaving an unclean surface behind, and I did not use magnets to hold the mask in place, just tape - I made many many fuzzy circles...
I can't take too much credit, the main inspiration came from Ingle, Frank W. "A shadow mask for sputtered films." Review of Scientific Instruments 45.11 (1974): 1460-1461. They use the magnet system to keep it clamped in the deposition chamber, and I just extended that to the patterning step too :)
First time here. The movie piece at the beginning made me think it was an ad about an actual new movie! Very well done! Sometimes a stencil mask (Shadow mask) is the only option. I have the photolithography tool, but I need to use a shadow mask on porous silicon surfaces. Porous silicon is too fragile to sustain photoresist and develop processes. and I only need one process step. So no alignment requirement. And you are doing this in your garage??? No way! a sputtering equipment and a AFM tool in a garage? 😲
you are freaking kidding me, this is so awesome
I'm "only" an electronics/software tinkerer... but I'm really enjoying a lot of the content you're producing. It's shades of "Applied Science" which has got to be good. Keep up The Great Work.
Thanks! I aspire to be a tenth as knowledgeable and interesting as Ben some day. He's a national treasure! :) Can't wait for his next video, it sounds very interesting
Unmistakable "Applied Science" flavour to this channel...😌
I'll definitely be coming back for the next video...
Randomly found your channel and instantly became a fan
AHH that intro is genius
I was like... did I accidentaly click on a movie preview?
Every day, I pray to the Algo Gods that this channel blows up and gets the attention it deserves!
This is far more high quality explanation about the topic I have heard compared to one taught at university.
The quality of your videos is immaculate!
One like for the excellent intro. Another like for the subject matter. Finally a like for the future garage fab. - I think that works for the engagement algorithm.
i use 12 micron drill bits on a daily basis at my job for making specialized valves in what i assume to be some sort of contract with IBM. im not given the details, just the blueprints and the tools
I'm designing Shadow mask for device fabrication and your video was educating for me. Thank you 😀
This is awesome! I'll just add a few things I've picked up from my materials science classes/the internet
Mask alignment is a huge challenge even in lithography, and they do so with shapes on the edges of the mask that have to line up (for example, triangles or crosses at different points along the edge of the mask).
Also, my understanding is that masks for lithography are actually made via electron beam exposure, instead of UV light. The electron beam approach is considerably more accurate than UV light based methods, but is a serial process and therefore is prohibitively slow for large scale manufacturing.
Lastly, if you ramp up the voltages (and therefore ion energy) in the sputtering setup, you could actually begin to implant ions into your substrate. It could be very interesting if you purchased a small silicon wafer (the ones that are a few inches are a few hundred dollars I believe), and actually tried to make a transistor for real!
Definitely worth making an aluminum carrier for your shadow masks! .1mm (100um) offset from your features on something stiff (2ish mm) cut with a 90 chamfer mill will help a ton to keep everything flat and stable. We use a system close to that at uni for our simple sputter/evap masks we used to test a bunch of materials for thin-film transistors.
Oh! I never thought of a system like that, genius! Thanks for the tip, shouldn't be hard at all (theoretically) to mill up a support like that.
@@BreakingTaps If you're going to do your dev work on 75x25mm slides, I'd definitely make an assembly where you can drop it into a pocket (drill out the corners so you can get a tweezer in there to release) and then screw around the edges. We had dowels that would hold the shadow mask (oversized through holes to prevent tearing) around the periphery, too. Anything and everything you can do to mechanically align and park features helps a ton. We were (use in prior post should be *used*) registering multiple layers (masks) and holding 50'ish microns across multiple layers. Not that it was super critical in our workflow, but still nice.
that intro is of amazing quality, but i'm honestly here for the science
What I'd really love to see is a very simple logic-gate calculator. I feel like the process you have now is nearly there.
A on Glas sputtered calculator would be cool and reminds me of almost any sci-fi computer/electronic (Stargate for example)
Going to start my Monday on the right note!
*Super* interesting as usual! I wonder if some sort of compressed-air-jet cooling could keep things cool enough for multi-pass cutting? I’d think that more shallow cuts and cooling the material between passes would let you go a lot deeper without warping... (PS: I don’t think 10-15 microns is anything to be ashamed of. It isn’t exactly microelectronic scale, but still seems very impressive, especially since you’re punching through 25 microns of material.
Here’s another thought: what if you took advantage of the closing up of the shadow mask as it’s used to deliberately pre-narrow the features by depositing on it before using it for the parts you’re wanting to make?
And WOAH! An AFM??!! COOL!!!
Hmm, yeah that's a good point, some downtime in between passes and an active cooling solution might help minimize how much it warps. And would probably allow smaller features, since you could keep the "area of ablation" relatively small. Interesting idea, would be easy to test!
And that's a super clever idea purposefully closing up the gap to get smaller features. I love it, definitely going to try that. If it works it might warrant a followup because it's a great hack. Cheers!
(And yeah i'm super stoked for the AFM. Gonna be awesome!)
If Wiki is correct, the first 4000 series CMOS logic chips were made in a 20um process. Making a functional simple gate or a flip-flop would be very impressive as a start!
Oh interesting, I didn't realize they started out that large. Like I knew the process nodes started pretty big, but figured it wasn't until they had brought it down a little more before making chips. It's definitely on my bucket list! And I can go binge through Sam Zeloof's videos again if I get stuck on steps :)
Goddamn dude you’re on fire and I absolutely love this channel. So much potential and raw talent!!!
I'm pretty hyped for the comming content
That intro, hot damn. One of my favorite movies of all time. Looks like you had a good time editing that together in addition to doing everything else. (:
Wow! Your editing skills are growing so quickly that you'll soon be the best of the science RUclipsrs.
Garage microfab sounds incredible.
I want to make a Pentium in my garage someday.
Nice video Zach! So is your sputter coater something home build or a commercial one? When I worked on oled technology (long time ago) we also used shadow masks. These were cnc milled out of aluminium, had rigid edges and had a tiny ridge to keep the device surface away from the mask to avoid damage of the devices. By the way, lithography is not that difficult and can be done fairly easily with a cheap 4k resin printer at 35um resolution. If you use "lift off" technology you can even avoid etching.
Thanks! Commercial machine I'm afraid, so not very interesting to talk about on video :) I was going to build one originally (chamber in background) but found this unit for a good deal because it needed some repairs which ended up being a simple software hack/tweak. It's a Leica SEM coater, so has lots of nice features (tilt and rotation stage, dual sputter heads, integrated shutters, turbo and internal diaphragm pump) but it since it was built for thin, even coatings it is also very slow. And DC, so can only do metals.
That's interesting that the shadowmasks were milled! I suppose a sharp v-tip could make a pretty narrow cut, and you'd have the advantage of a stiffer mask to prevent warping. Interesting, might try that. Makes sense that some kind of lip or ridge would be needed to prevent scratching of the device. By time I was on my third layer you could see some damage had accumulated...doesn't take much to scratch those thin film layers :)
Litho is next! I was just being super lazy haha :) That's a clever idea resuising a SLA printer as a contact litho machine!
@@BreakingTaps Good luck with the litho. Watch out, it is actually pretty addictive ;-). You can achieve much higher resolutions than 35 um by demagnified projection of the LCD of the 3d printer.
All your videos are very well made
Now we need a DIY PVD... That machine is way way too nice for youtube!
Haha yeah I know what you mean :) I really lucked out grabbing that one. But no worries, I have some vague plans to use my WIP chamber (in the background) for some laser ablation deposition which should be fun and quite an adventure I think :)
Not sure if you saw it, but Huygen's Optics recently put up a video on his home built PVD setup. (Although it is perhaps a little above what most of us can reasonably build.)
Highly recommend! His channel is great, he's doing some really interesting work over there. Deserves way more subs!
W O W was that intro amazing 😮
I wish you made this video a few more years back when I wasted 4 months trying to use a shadow mask to work in an evaporation process that kept shorting the top and bottom layers. :(
Well that fills a gap in my knowledge of the old CRT "shadowmasks" awesome 👍
Wow lasers are the ultimate tools... wish I had that pulsed dude
congrats on winning the battle with your hair this time
Your content is amazing!!
do i understand correctly that one of major limitations with laser cutting a shadow mask is that you can only cut shapes that don't have any dangling/disconnected bits? and that sounds like a major limitation for general-purpose use.
upd: ah, you did mention that later
Shadow Mask: coming soon to a garage microfab near you! (loved the intro)
I think your channel quality is high enough already to have at least hundreds of thousands of subs and a decent running patreon. But finding your channel is actually non-trivial unless you're specifically searching it by name. I think what you need is a collab with one of the "big" youtubers like Thought Emporium or Appled Science to get some of their viewers because they'll no doubt like your stuff if they see it.
Hitting it right out of the park!
You can clamp the heat deformed shim stock between a couple of thicker 1/8" or thicker pieces of steel with a couple of small cheap C- clamps, put it in an oven for an hour at 450F , let it cool until it can be handled remove the clamps and your shim stock will be as flat as the pieces you clamped it in.
Will give this a shot! Would certainly be an easy way to fix the issue if it works, cheers!
@@BreakingTaps I am a Bladesmith and it even works on hardened steel as long as the warpage isn't to severe. I hop it works out for you.
BT, great as always! You should get on Reddit that will deff help with the subs. R/BT !!!
Have you tried using some kind of thermal adhesive tape under the metal to absorb some of the extra heat while cutting? That may help mitigate the thermal warping issue. Use a solvent to get it off the adhesive later or something.
I haven't, but that's a great idea! Will try that out next time. I also got some new parameter suggestions from another viewer that supposedly works better, less warping although takes a little longer. 🤞
Would *love* to see a tour of your sputtering and vacuum pump/chamber setup. I've tried it myself and there are definitely many, many traps novices like myself fall into, so seeing what you've got working would be amazing. Did you machine your own sputter-gun or buy it from some company, and if so how much was it? Have you sourced most of your stuff used off eBay?
Great video BTW!
Thanks! Afraid the sputtering setup isn't very interesting, it's an off-the-shelf SEM sputter coater from leica. I snagged it for a very good deal because it was throwing some errors, and the repair ended up being pretty trivial. But otherwise it's all stock and therefore not very interesting :) Load sample, hit go, wait for thin film. I am going to put together a laser pulse deposition chamber later this summer I think (the big glowing chamber in the background) and plan to do a video about the chamber all the little bits I learn along the way. I have most everything I need, just need to machine up some adapters and find time to work on it!
DUDE!.. You are a videography/directing/creative god. Never mind the schience (Intended spelling) and make a movie!
Not that it would be recommended or even wise, but I wonder if a super critical fluid (CO²) would be a good option for heat dissipation of thinner metals. Might be enough to keep it from deforming.
Also given the short distance needed cutting, I wonder if a gas or other fluid could be used to narrow/focus the beam ( like a fiber optic ) more so than expensive optical lenses could focus the same beam. Maybe two fluids in laminar flow with one in the center surrounded by the other, one hell of an engineering challenge there.
Just the ramblings of a madman.
Heck yes to garage microfab videos.
Love your videos
Hm, did you try using a thermally conductive backer on your thicker shim stock stencils when you cut them? Like, maybe if you used aluminum or copper (or silver) rather than glass you could reduce warpage by drawing away the heat. Might not suck away the heat fast enough to matter, but it'd probably be worth investigating. Another option might be to sputter a layer of copper or whatever on one or both sides of the shim stock to give you a very small amount of added thickness and again to help draw away heat. That might introduce new warping problems though, a la bimetallic strip.
awsome sputter machine i need this machine
Have you considered submerged cutting? Some lasers and plasma cutters cut with the workpiece submerged or with fluid cooling on the workpiece
Sorry if you mentioned this already and I missed it, but what sputter machine is that?
Don't think I've mentioned it explicitly in a video (just some community posts). It's a Leica EM ACE600 machine, designed for SEM coatings. So it can pull a nice vacuum and has fun features like that tilting/rotating stage... but it's pretty slow. I was originally going to build my own but this came up for sale at a good price (needed repairs) so I couldn't resist :)
I think that my prior slight worry about the creative video intro sequences migh have been from it being the first time. I didnt feel worry about this one, I guess its because it was expected now.
Yay! Great to hear! Definitely won't happen on all the videos, this one was a ton of work for like 30 seconds :) But it was a fun diversion for me to play with. And e.g. the next one on AFM has waaaay too much interesting content to waste a few minutes up front on an intro, so it'll definitely be dependent on what's getting talked about. Thanks for watching!
Congrats on getting Harrison Ford to make a cameo in your intro
Took some convincing but he came around eventually 😇
Interesting. Did you try spring steel? And/or EDM instead of the laser (not sure of the wire diameter limits).
can someone please tell me what is the make and model of the laser cutting equipment used in the video?
Any plans to do evaporative deposition? It requires a higher vacuum, but you get more directional deposition.
Maaaaybe. I'd like to try pulsed laser deposition at some point, and depending on how well that works (or doesn't) I might convert it to thermal evap instead. Love the sputter coater but as you said, directionality isn't great. And it's very slow (and can only do metals). Something that can deposit a micron of material quickly would be very appreciated :)
What sputtermachine is that? Looks real cool did you make it your self? can't wait to see more!
Afraid it's a commercial machine (Leica EM Ace600). I got it for a screaming deal on ebay because it was inoperable, but the fix was relatively simple (some uhh, software hackery involved). Little sad that I didn't make it because it's not terribly interesting to talk about being commercial, but it works quite nicely so I'm not too sad :) Only downside is that it's a DC magnetron so can only deposit metals, and it's _very_ slow because it's designed for SEM coatings.
you need an ultrashort pulse laser and it's no problem... no heat-induced warping if you apply it right ;-)
regards - your local laser expert :P
Yes! I would _love_ to get my hands on a picosecond or femtosecond :) Unfortunately don't seem nearly as easy to find on the surplus market, (for reasonable prices), have had saved searches running for months hoping something would turn up. I dream of the day I can do femtosecond micromachining of metal and glass 😍
@@BreakingTaps I do it every day, lasers are the best machining tool, contact free and free of consumables. Photolitho, at least for micrometer sized structures is so 19th century, althrough it is almost unbeatable at accuracy, many applications can be replaced by direct laser structuring. I have daily access to more than 30 laser beam sources... it's great fun but therefore I have no need for more and got rid of most of my private beam sources.
Ultrashort pulse lasers with enough pulse energy to ablate will likely not fall much more in cost any time soon, but average power goes up and up. We're reaching 10kW+ at this point... now I leave it up to your imagination as to what we do with those :P
can P/N/un doped semiconductor be vapor deposited for actual working transistors?
Will it be on Netflix?
Soon™
I really love the video, but I wish to you had captions on it.
Are the auto-generated ones bad? I haven't looked to be honest. Or do you mean a different language?
I have two slightly off topic questions I cannot resist asking since you seem like you might know the answers. First, aside from scraping it off, do you know of a way to remove the CFA/Bayer layer/baked on mask from a color image sensor to make it black and white but more sensitive to a broader spectrum of light? Second, when I attempt to make circuit boards by applying a UV sensitive film, the film doesn't properly wash away from where I want to etch copper and partly washes away where I want to keep the copper. Am I perhaps using bad film or is it more likely I lack skill? I've tried varying the variables but never get good results.
I came for the vapor deposition. I just bought a vacuum pump and am making a vacuum chamber and eventually everything for sputtering success at sputtering.
I stumbled across this paper a while ago (www.ncbi.nlm.nih.gov/pmc/articles/PMC5087437/) which removed the filter with some photoresist stripper (Posistrip EKC830). I doubt this is easily available though, and the MSDS suggests it uses some intense organic solvents (N-methylpyrrolidone and 2-(2-Aminoethoxy)ethanol) apps.mnc.umn.edu/pub/msds/ekc830.pdf). Looks like you can get N-methylpyrrolidone on ebay at least, although not sure I'd want to use it personally :)
Otherwise, scraping the filter off is the only method I've seen.
Regarding the PCB film... I've actually never etched a PCB before! So no good opinions or experience from me on that topic :) I know liquid photoresist can go "off" over time, so it's possible your film might just be bad, or perhaps the PCB board isn't fully cleaned before applying? Before tossing the film, I might try some exposure grids to see if you can dial in the exposure time a little more. E.g. expose one section for a second, expose another section for two seconds, etc etc up until some large value and see which gives best contrast. Goodluck!
@@BreakingTaps Yes, EKC830 isn't available at Walmart:-) I started down the N-methylpyrrolidone route and stopped when I realized I'd be buying way more than I want laying around. It is an ingredient in some fairly common household product (herbicide?) so I might try it. Or maybe I'll keep practising with scraping, go for the nickname "Chip Slayer."
did you try cutting the imitation silver while on the paper and just use it on the paper?
Not exactly in that setup, but I did try just regular copy paper by itself and it works quite well! I saw a paper a while back which used filter paper (www.mdpi.com/2072-666X/11/7/676/pdf) and got pretty good resolution. They used a CO2 laser, but I used my 450nm and white copy paper or brown kraft paper, both worked nicely. Think I got 30-40 micron resolution which is shockingly good for just paper in my opinion :)
What about a light spray adhesive to the mask? I wonder if that could be done in a way that wouldn't make removing the mask again later impossible.
Ah yeah, I wanted to mention that but totally forgot! In theory this would be perfect, assuming it doesn't "lift off" prior layers (which it probably wouldn't, metal adheres pretty well to glass). But most adhesives (or any polymer really) outgas like crazy in a vacuum and can make sputtering difficult. The kapton tape I used is "good" for vacuum and still outgasses a lot, regular spray adhesives ended up being really bad. I can almost garauntee there is a "vacuum rated" spray adhesive somewhere but I dunno how/where to find it :)
@@BreakingTaps That's a really good point, I didn't even think of that.
@@JuryDutySummons Hehe yeah i didn't either, figured that one out the hard way. Was pumping for ages and still not getting to desired pressure... then it dawned on me :)
I'm curious about the pressure inside the sputter chamber. At what pressure do you run? Nice video as always :)
Thanks! Pressure varies a little depending on the target, but I generally pump to e-5 mbar, purge once or twice with argon, then take it down to e-6 mbar pre-process (especially important for aluminum or copper due to oxides, if I'm not in a rush I'll let the aluminum runs go down to e-7). Sputtering is usually done around 2.0e-2 mbar. I'm still pretty new to sputtering though, so huge grain of salt to all the above. It's possible one or more aspects of that is quite sub-optimal :)
Interlinked
Can't you anneal the mask while sandwiched between flat plates? Before and/or after laser.
Also, is it feasible to (electro)chemically etch back the buildup to restore the mask?
Hmm, seems like that would work. Actually now that you mention it, since this is cold rolled it probably has a ton of internal stress...pre-annealing might help a bunch too to relieve the stress before the lasering. Hmmm... would be pretty easy to test.
I think in theory you could electro-etch the buildup back out, but probably have to be careful to make sure it etches evenly. Although if you design the setup correctly, you could find a chemical etchant that doesn't touch the mask. Like using ferric chloride to etch out copper buildup and leave the steel alone.
Are there materials that would be suited as mask for dry etching glass?
Dry etching via plasma? Hmm, I'm not sure. I suspect most metals etch at the same or faster rate than glass under an oxygen/argon or reactive plasma. You'd probably need to deposit something harder like SiN which would require a reactive DC, RF, or CVD system. I did find this paper on a quick skim that used a silicon wafer as a hard mask: www.researchgate.net/publication/223079651_Deep_plasma_etching_of_glass_with_a_silicon_shadow_mask
@@BreakingTaps Thank you. I'll look into it.
Great video!! Make/model on fiber laser and sputter coater? Thanks!!
Thanks! Fiber laser is a random chinese 50W non-MOPA laser (if you search for "50w fiber laser" it's the variety with a galvo scanner and head that moves up/down with a crank). I believe mine has a JPT laser inside, although Raycus is common too. Sputter coater is a Leica EM ACE600. Got it used for a _very_ good price because it was non-functional, repair ended up being very easy (software fix). :)
"blah blah blah whoooooosh shish howwwl wheeee" best subtitle ever ! ( you tube video of a boat called tally ho)
It is sad that the video has rn only 612 likes in comparison to the 4.2k views so Guys that video was great and need more attention so use the knowledge we have about the algorithm (Spiffing Brit has shown it) and interact with the video leav a like and make a comment that's what helps this channel the most from our side
Oh beside Patreon etc but not everyone has the money too do so
Shadowrunner 👍 … but no photon sieves 😕
Interlinked.
Within cells interlinked
AFM super cool, so eventually you plan to make a working semiconductor ? I can't wait ...cheers.
Agreed! Each time i pull an image off the machine I'm just stunned by how cool it is. And watching the probe dance around is mesmerizing. Re: semiconductors, maybe! Hopefully! I'm currently playing around with some alternative materials because I really don't enjoy working with HF. Curious to see how far (or if) you can go with alternative processes. Makes complete sense why silicon is used industrially, but for home fab there might be different routes because the goals are different. We'll see! :)
@@BreakingTaps I have played with AFM and STM before ! this is gonna be great !
@@BreakingTaps ZnO might be accessible in a reactive sputter environment? That you can etch with HCl.
@@BreakingTaps I'm not very familiar with sputter deposition for semiconductors - you'll probably need something defect-tolerant enough to work even when polycrystalline (like the much-hyped halide perovskites in research-grade solar cells), something commonly solution-grown, or an amorphous semiconductor like amorphous Si, although I agree 100%, HF in the home shop feels like a very dangerous idea - keep the calcium gel nearby!
@@AlphaPhoenixChannel Hmm hadn't thought about that, good point! Will need to do some more reading. Right now some candidates are a few van der waal materials (WS2 and MoS2) which can be applied via abrasion (e.g. rub the powder on the substrate). Causes the platelets to exfoliate themselves like graphite and form a sorta-flat multi-layer. Most of the literature focuses on those as single monolayers, but I found a few references of it being used as a "bulk" material to some ok effect. I assume the literature ignores the bulk stuff since electrons don't like to jump platelets so it's a lot less interesting? Dunno, but I'm less picky :)
The other thing I'm reading into is Cu3N and CuO/CuO2. The nitride form seems to happily self-dope itself n- or p-type depending on deposition temperature, the paper I was reading suggested it was a balance between donor and acceptor defects and the temperature changes the balance slightly. Dunno just how defect tolerant it is though, I was indeed planning to sputter that one. The copper oxide ones can be generated by selective heating/annealing so it probably helps fix itself during the process? I may be bugging you with more questions hehehe.
ZnO is a definite candidate as @Daniel Heineck mentioned. I haven't done much reading on that one yet but it's on the todo list :) As you said, ye olde HF in the home shop doesn't sound fun at all if I can avoid it :)
10/10
A mix between applied science and Cody's lab.
interlinked.
This is not fair, I know have an online crush.
🤗
How do you people get money for all these stuff?
Around 7:55 you complain about small bridges in the material. You could try to get rid of them using something like electropolishing (it doesn't work with all metals), or just simply try to watch it a bit chemically.
I was lucky to have a well paying job (software), no children, stalking ebay for deals, and my discretionary money goes into hobby projects/equipment instead of things like cars or fancy vacations or whatever. When we do vacation we're typically camping in a tent or something cheap, because that's the kind of fun we like :) So yeah, basically just directing funds other people spend on new phones or cars or eating out all the time or whatever into my hobby projects instead. And slowly accumulating the tools that I want over the years.
Good ideas about the bridging! I think something like that could work, and would help a lot
My dude, I love your videos and your production quality is fantastic. but what the fuck is this title. PLEASE for your own good rename this video. This title is like choosing to wear clogs to play in the NBA...
Haha yeah I know. It's cool, I have accepted that this video is a dud due to the intro and the title. 🙂 Viewers are confused by the intro if the title doesn't allude to it, and the title is confusing if it does mention Blade Runner somehow. I tried various combinations and it just never worked out. Lessons were learned, I'm not too fussed! Cheers for watching!
@@BreakingTaps I assume you’ve seen Veritassiums video on click bait titles?
The clickbait life sounds morally challenging to navigate but the proof is in the pudding.
Thanks for the interesting videos!
Where is the Thought Emporium comment? Must be an error in the algorithm or something like the kids says today