It always blows me away how you can deconstruct a process to it's bare bones and get a result at a fraction of the cost and you gladly share all your knowledge. You're a true pillar of the community for sharing!
Not to be disrespectful to the guy. but he knows nothing about chemical etching. Whatever he proposed with his video is nothing but a silly game and those components will never end up in anything that you would like to use. He can maybe make some machinery which is very interesting. But he just made chemical etching look bad and he made all the mistake you can make when etching something. I guess he has his place on youtube ;) but not anywhere near what a professional chemical etcher would do.
@@Mr.Flo.Official Some salty idiot named mr flo got upset that companies are inflating costs of products for no apparent reason other then to make money. Most of the time the answers are simple and cheap.
@@Mr.Flo.Official "He knows nothing about chemical etching" he literally showed the process on how to chemically etch metal, and the issues you will have and how to solve them. You couldn't be anymore wrong on your assessment.
Another neat thing with photoetch is you can do partial depth to make fold lines,like scoring cardboard. Probably needs finer process control,though might be worth trying just putting them on one side .
Mike, Yes it is normal practice to have the fold lines etched only from one side. The Asteriod Belt shown here content.instructables.com/ORIG/FEJ/ID3W/HFSHFI10/FEJID3WHFSHFI10.jpg does that by having half etched lines. The photo tool for the back here content.instructables.com/ORIG/F8D/ELWM/HFSHFHYS/F8DELWMHFSHFHYS.png has thin lines for the folds but the front side is solid. The same for the buttons so the arrows and symbols are only etched half way through.
@@patkirk960 I didn't know youtube censored URLs. I guess if the disappear you can search for "The asteroid belt instructable" and scroll down to step 5. I used "toner transfer" there because I was told instructables readers prefered to see simpler tech than dry film photoresist.
@@TheCunningFellow Some lovely extra engineering in your instructable that's above and beyond the average. The extra yet simple hardware to clear the screen and to reduce the overhead from code was delightful.
Got to agree with everyone, it seems there is no time passing when watching, Mesmerizing videos with very informative information, Always learn something interesting
I'm thinking you could print some markings on the 4 corners of each transparent mask so that you could use a moire pattern technique for easy and clear alignment of the masks when you need to overlay them. I hear this technique is pretty good and can even allow you to quantify the degree of alignment.
Hi Ben, The specialty black inks are great, but if you want to use a regular inkjet then sometimes yellow dye inks can be more opaque to UV light because of UV inhibitors to stop fading. Also a Stouffer Gauge will help a lot working out your exposure times. Cunning Fellow (The guy that did The Etchinator)
You can see the yellow ink with your eyes no problem. A non yellow light source like a large blue LED will enhance it. But really the developed dry film is the thing you should be inspecting most closely.
Agreed. Although the uv blocking power of different coloured inks depends highly per brand and type. Epson alone already has 4 or 5 different ink lines. The best are the pigment type, the dye inks are really poor uv blockers. Your best approach is to print a step wedge with all the different ink colors and see which blocks the most. More uv blockage can really up the contrast and sharpness of the lines you want to etch
Inna - Pigment inks naturally block more light so if your printer uses pigment inks (Epson would call them DuraBright or UltraChrome) then black is the best option. If your printer uses dye ink (Epson Claria) then the black does not block the UV as well as the yellow. The yellow is most susceptible to fading from the sun, so they add UV inhibitors (sun screen) to the yellow ink. Most brands of dye yellow that claim they are light fast or don't fade in the sun will do the same.
You call that spot on from a PCM background?😂 come on man. This is an insult to the process. Good to say “playing with FeCl” but not comparing with industrial. Must admit was amazing to watch from entertaining purposes tho.
@@Mr.Flo.Official I do. For a diy at home setup it's covering all the necessary bases. When you break down the milling process you have some basic concepts that need to be implemented. He implemented all the required ones.
For registering the masks/artwork you can get an anti-parralax magnifier or collimating magnifier - they're often used for 4 color printing - look for para-mag magnifier or caprock stripmaster series. 7x or 7 power should be enough.
Never heard of those. Good tip! With appropriate fiducial marks and a microscope, you can be surprisingly accurate hand aligning. I have bonded so many 30 micron pitch flex circuits by hand.
In addition, the masks should be taped to a spacer of the same thickness as the sheets including their resist coatings (ie, a strip of the stuff) That's experience from registering and etching double sided PCBs back in the 80s ... :) I'd create some light box tubes around those 405nm lights too. This will greatly increase the luminous flux at the sheet surface whilst reducing purple light spillage where it can cause trouble - it was more important back in the days when UV sources were being used - The setup I used had a large mercury arc lamp set 5 feet from the PCB surface and you _really_ didn't want your corneas being exposed to it
It could be used to calibrate for stretching as well by printing 2 square or circle patterns and comparing them at 90° to each other. What he's worked out is incredible as is, but it's very close to being something wildly more precise.
I work with/design stamped parts and source mesh all the time… Stainless, zinc/tin plated parts and stainless/epoxy coated steel mesh. A lot of our vendors are willing to do stamping tooling on the cheap (
Hi Ben, I used to design machines used in photolithography and the photoresists coaters. We used to etch parts on sheets of glass, etching gold and titanium layers. We used to use spin coaters. The nozzles would be placed above a spinning chuck that was made of nylon. The whole thing was surrounded by a big tub with sides higher than the spin axis. The force of the acid falling, hitting the surface and then being flung off was enough at 1000 rpm to take the debris away. For the underside we had nozzles firing the acid up at the underside surface.
One method to improve the density and reduce the pinholes in laser printed output is to use color foils (also called toner reactive film). You run it through a hot laminator or back through your laser printer with the color side against your artwork, and the color film fuses the toner, but not to the areas on your artwork that have no toner. I use it when I am etching PCB's at home. (For PCBs, I print the art on release paper, laminate it to the PCB, then do a short soak in water which disolves a layer of the release paper, allowing you to peel the paper away from the PCB leaving the toner artwork on the PCB. Then I run it through the laminator with color foil which helps with density and pinholes. Then I etch the board w/ ferric choloride.) The color foils don't work with inkjet, though.
I've used a product called 'Press-n-Peel Blue' for this. You can print directly on the film then laminate that to the copper pcb. Can be tricky to get the laminating right - pressure and temperature need good control. And everything has to be kept dust free. I've used this for PCBs, and for sheet metal parts. 5mil track/space is possible (just).
@@weldandcutdotcom You can't use external links in UA-cam or they get deleted. You can also get banned, like when I used a fake comedic email name and got banned.
As per ~ 18:48 they are "Air Operated Double Diaphragm Pumps" and appear to be made by "ARO". So Basically it's almost like a drum (the instrument) with pipes over the external "drum heads". Air is pumped into the "drum", the diaphragms push out a bit, and if the "pipes" are designed all fancily and some one way valves are used one can make a pump. These not only can be really chemical resistant, but some can even pump abrasive slurries!
Couple of thoughts from when I used to home etch PCBs: Laser printing onto tracing paper works much better than transparencies,you nerd to use thick stuff,>90gsm to stop crinkling in the fuser. When aligning the artwork, stapling works well as you can align accurately with the sheets completely flat against each other,but you do need a setup where the staples overhang the edge when exposing, less of an isdue with PCBs due to the thickness of the board
That magnetic setup on the etching machine is amazing. I could see something like that used for cleaning 3D printed resin parts. This whole video is a how-to start your own photochemical machining company.
For more precise alignment (at least during the tape phase), you could use a moiré pattern , possibly with a vernier type setup (so one registration is slightly smaller on one side). Then you could align the interference pattern. Although I’m not sure such precision would be necessary, so you could probably just go with interrupted overlapping lines on the cross pattern (similar to how polymer bank notes have transparency alignment patterns)
Regarding the transparency and the special ink. I had the same problem 25 years ago making masks for exposing PCB's. Came to the conclusion that printing on ordinary paper worked well enough. The laser toner sticks really well to paper, the coverage is 100% without pinholes. And the UV used to expose the photo resist went through just fine. I had to mirror flip the design so that the printed side of the paper touched the photo resist.
Okay, that's interesting that the toner sticks fine to paper, but not transparencies. I'm more surprised to hear about the UV transparency of the paper, though!
Just went through this process and contacted all the vendors you mentioned in the intro for making a few shadow masks for a metal evap tool. Was thinking the whole time this is something I could just do in my shop and it's crazy expensive. Of course I just thought it and you actually did it! I don't have enough demand to build it but it sure would be nice if an OSH-ETCH or SendEtchSend sprang from these efforts!!
Not to be disrepectful. But what sort of tolerance you need your shadow masks to be at? The setup presented won’t offer anywhere newr +/- 1-2 mm. Also the details and peel of he gets are soooo bad to watch coming from a PCM background :)
I have been doing a variation of this. Instead of printing on transparencies and then using an UV exposure station I used a cheap resin 3d printer to expose my artwork on the resist. (The kind where a UV light is projected through a digital screen to solidify the resin. I have been using an Elegoo Saturn. I just remove the vat and put the plate on the screen.) It worked quite okay for me. I got some niece pieces. The theoretical benefit is that you have zero setup cost. The very real drawback is that if you want to etch from two sides (and you want to etch from two sides) then you need to flip the plate and place it very precisely at the same spot on the resin printer's screen. To be able to achieve that I have made a kind of jig holding a usb microscope solid relative to the printer. If you do an exposure without the plate being present you can see the invididual pixels of the resin printer with the microscope. And then you can use this to align the second side of the plate very precisely. I have made this decorative orrery using the technique: ua-cam.com/video/KK3TiUWSTbI/v-deo.html
In the past I used the head from an photo collimator for a similar UV exposure process where we needed collimated light. Your setup seemed to do the trick just fine, but we needed it to be up close in my application. Its a reasonably cheap way to get get it done as you can fine them second hand. Did have to air cool the lenses as they got a bit hot from the high powered UV light sources :) Mesh bag was a slick idea! Really great video once again
Your idea about a "really clean air source" sounds very familiar to another interest of mine, mushroom farming! They need sterile workspaces and they use large like 2 foot by 3 foot HEPA filter pointed down to create a sterile downdraft area to handle spore prints and gather samples without contamination. They don't even cost all that much, and with a custom hood could make a real nice drying area. I have thought about this for another area I am interested in, beer brewing, as an area to store a kveik ring to dry and keep from contamination.
HEPA filters really are amazing. I do research using flat sheet media and they can handle concentrations greater than 10 million particles/cm^3 at well over 5 cfm through the filter (more than a billion particles a minute!) and still capture 99.97% of the particles.
I worked in electronic manufacturing and we usually have clean booth. It is just a tent made from extruded aluminium and plastic sheet, with a filter unit on top. The booth is just large enough for a small table where a worker sit to assemble dust sensitive parts like camera or screen. The filter unit is just a fan blow into HEPA filter to create positive pressure inside the booth. There is also ion generator for ESD protection and some LED light. There is also a mini table top version, and some assembly machine have integrated filter unit on top to keep the inside clean. Pretty easy to DIY something like that.
I really appreciate how you not only show your successes but discuss your challenges in these projects. I have one suggestion if you ever wanted to revisit this. Back in my PCB etching days I found that a viable alternative to spray etching is bubble agitation of the etchant. Putting a fine mesh bubbler grid at the bottom of the tank below the part has similar performance to high pressure spray with the added benefit of not having the etchant running in the plumbing and pump. Both spray and bubbling end up with the same effect of getting fresh etchant to the metal being etched. Just putting the material in a still tank ends up with the etching site having lower concentration of fresh FeCl as the already combined products only slowly move away from the reaction plane. Bubbling seems to have a similar effect as spraying, constantly introducing fresh FeCl to the etch site increasing the overall reaction speed. My first test of this setup didn't anticipate how dramatic the difference was vs simple etchant circulation, so I etched too long and ended up undercutting the mask so much that all the copper was removed from the board! Not quite the result I intended.
The problem with bubblers is that no matter how much filtration you have, there's a constant outflow of ferric chloride saturated air from the tank and things get..... corroded... in the vicinity. We ended up swapping out the bubbler for a couple of model boat props on very long shafts run through the tank lid When etching copper, there are alternatives to ferric chloride. It's reliable but messy and makes even stainless steel rust High strength hydrogen peroxide + hydrochloric acid (we used to call this the 6 second etch - extremely dangerous in high concentrations) - this works really well and lasts virtually forever (used ferric chloride is an environmental hazard due to the copper in it) - this also eats stainless steel so use with caution Ammonium persulphate works really well too There are a few instructables around on this stuff, but etching steel or other metals will have different ideal etchants
@@miscbits6399 I don't disagree that air filtration is a necessity with bubbler systems. I used an activated charcoal fume capture with good results. I didn't mention that point since I was comparing to a spray etch system which has very similar exhaust air filtering needs, but thanks for pointing this out.
Why would you go for that instead of normal Horizontal conveyor with oacillating sprays top and bottom? It’s an oxidation reaction which will allow good circulation of fecl3 while solids deposit at the bottom of sump and also have enough flow contact with the substrate. Sounds interesting what you tried anyway.
@@Mr.Flo.Official Good questions. Answers: this was ~1981 in a lab setup for prototyping boards. I already had a etchant circulating dunk style tank which I could convert by reversing the hose connections to pump air vs liquids, and didn't need to add parts. This was long enough ago that I was prototyping by drawing circuits with resist tape, not even photo resist.
wow i'm sure that i am not alone in thinking that i'd happily wait months for a vid from Ben. they always all-around perfect. length, audio, video.. then the content.. absolutely fantastic. seeing prototyping is amazing, but seeing Bens prototyping is the epitome of excellence for video. descriptions or ups and downs and gotchas... full of information and helpful and still concise.. WHY IS NOT 1M SUBS YET???
As I'm in the process of researching etching, in my case to make data plates, you post this video. Usually your videos send me down a new rabbit hole. It's amazing how much trial and error, research, and practical experience you can compress into one of these videos.
If no one else has mentioned it, look at the clean air boxes that mushroom growers use. It might work for drying the metal after cleaning if you’re doing large batches
I've watched this a couple times over the last few months, and the ingenuity in this whole process design is just mind-blowing. I would absolutely back this business!
I make circuit boards using laser printer toner for the resist. I use a heated 'laminator' to apply it. My artwork _IS_ my resist. It took a few tries to get right combination of backing paper, heat, and pressure dialed in but it was an inexpensive and rapid process. I purchased the highest resolution BUDGET monocrome laser printer I could find and have much success. I don't know if toner would survive your mesh bagged etching intact though.
One note on using toner transfer. I don’t know if this is still the case, but at least formerly, Brother’s toner didn’t work for the process. I don’t recall the details, think it maybe had a much higher fusing temperature so it was hard to get it properly melted to stick to the PCB material.
I’m blown away. You just singlehandedly recreated and improved upon a modern manufacturing process in your garage. There might be what, 300 people in the world who could pull this off?
As others have said, great work as usual X) For the registration alignment of the two layers of transparency - what about using an overhead projector? Put both sheets on the glass, then use the projected image to align the parts. Not a huge amount of magnification, but even 3~5x with good ergonomics would help.
This whole process is very similar to offset printing plates. But they were in aluminum, and only partially etched into the plates for printing. We did quite a lot of these in our class. This was highschool graphics class, back when the 486 was being conquered by the pentiums. We used all Macs though for that class. Don't remember what they were, but early 90's style machines. The large camera is what we used to make our transparencies. Literally a giant wall with a huge lens, and a room behind it to put the transparent film onto it. We developed it right then and there. I bet film transparency would be even better than inkjet as far as detail goes. We then used this on a the plate that was pre-covered with resist, put it in the light bath thing for a few minutes, and then washed it off in a special bath and unknown chemicals (at the time), and then rinsed them, and the plate was done.
Nice that you highlighted the alignment problem many people will run into. I've seen cases where people even closed off a third side, creating a sort of envelope for the metal sheet.
It's almost creepy how you come out with videos that answer the questions that I AM CURRENTLY STRUGGLING WITH. I'm talking about... the same day. It's super helpful, but also unsettling.
@@TqSNv9R0iG5Ckxew At the time I was in the process of torturing myself trying to get the dry film photoresist to adhere good enough to copper laminate. If you look really closely you can tell Ben is actually having the same issue. The film delaminates during the etch around the edges. Sometime delaminating completely. I haven't found any technique that will work well enough to have 1mm or less resolution. But I have found something that does actually work pretty well. It is labeled as "Photoresist Anti-etching Blue Ink Paint For DIY PCB." It's a thick blue ink that needs to be cut with acetone. You can either spray it on with a "Devilbiss" style paint sprayer (which will give you a better, flatter finish" OR you can go to an auto parts store that carries "Preval" paint sprayers. They are like a mix your own spray paint kit. pretty cheap, and works well enough. You have to dry the board well (quickest I've found is with a couple fans in a closet with one of those Edenpure heaters set to 90 degrees F. Important note: You can not heat the photoresist ink past 100 degrees or so or it will cure the ink and will spoil the board. It will dry and not be tacky, or will only be very slightly tacky. Then you can expose and develop the same way you would the dry film. As long as you clean the board as per Ben's instructions you should have NO problem with lift off and the edges are crisp. Thing to note, the vapors whilst spraying will probably give you cancer... sooo.... well ventilated area, a respirator, or roll them dice.
@@jonathanseagraves8140 so way on earth would you not use wet resist? When i see amateur etchers trying to use dry resist i just turn it off. He has so many flaws in this video. Fun to watch. But not the setup You should use for better results in your garage :)
@@Mr.Flo.Official Short answer is I don't know where to buy it + sunk time fallacy + I don't know what product I should be looking for in the first place... I will take any info you have.
Wow, I'm the person the algorithm's working for right this instant - I just switched tabs from a online auction where I'd spent spent about a quarter hour studying the photos for a photochemical machining setup which for sale for about $25k. I just came to the conclusion that it's unnecessarily complicated and expensive for what I can do in my studio given a little patience and slightly more swearing. Of course it would make sense if you needed 60 of something.
I've gone through almost exactly this process in the last week so this couldn't have come at a better time! I ended up going with a spray photo resist and then removing the resist with a galvo laser since I was making a decorative brass piece but this seems like a fascinating method too! I'm not sure if adding citric acid to the ferric chloride will help with stainless but so called Edinburgh Etch is a more efficient etch used in copper plate print making. The theory being the citrate helps to keep the etched material soluble.
In case others want to take a deep dive: I've also though about using a resin 3D printer as a direct digital-to-plate technology. I've seen videos about doing this for PCBs. Since they have a UV source and the screen acts as a digital resist, you can put the sensitised plate on the screen directly. As the printers get bigger, hopefully there'll be more spare parts available to make a dedicated unit for exposures at a usable size.
Dude! Awesome. You inspire me to get out in my garage and tinker around. Thanks for letting all of us look over your shoulder while you work on projects!
I'm always impressed by your big brain projects and how well you explain them so the layman can still understand the process. I'm retired and will likely never do most of what you show but I really enjoy learning new things and find your projects interesting.
Dude, you are one of the most formidably brilliant and capable people I've ever encountered. This process alone by any right should make you wealthy. And this is just one thing you did for a random project... I hope you're rewarded for your endeavors.
Very cool! It's quite similar to the process I've used for home PCBs, an Epson inkjet is pretty much mandatory for getting that super-dense artwork-- and yeah, pretty annoying you have to use a piece of paper as a backing to get it to detect the transparency. I use a medium-format Artisan 1430 outfitted with a CISS system to make sure I never run out of ink. Now that Epson has come out with Econo-cart refillable systems it sure makes the whole process straightforward, using special high-density inks. For developer I do like using sodium metasilicate since it seems a bit more forgiving. Also using a vacuum-bag to hold the artwork against the boards is a lot less hassle than getting pairs of glass pieces to sandwich the artwork/board and making sure it has even pressure across the whole piece. For single-sided boards I just use a consumer vacuum-sealer/bag, though I have to make a bag out of heavy plastic tarp for double-sided boards since they have printed stuff on one side of the vacuum bags. To seal them I use the rubber tape which is used for carbon fiber/fiberglass resin infusion, as well as sealing around the vacuum hose. Very easy to use, strong, and virtually leak-proof.
Ben, thanks for the video, brings back memories of make PCB boards before you could buy them dirt cheap. I like your etching tank, for PCB's I has a splash tank. It was about a foot cube tank epoxied plywood. You put about 1 to 2 inches of etchant in the bottom. A motor driven dasher paddel splashed it all over the board with good force. --worked very well. The final improvement that I never finished was to make a raster direct UV laser diode imager. No need to do any lithography. I met a guy Zoltan who converted a old printer, replacing the print head with laser source and lens. He got some amazing resolution from the hack. One thought I had was to convert a laser printer poly scanning mirror and lens from the IR to a UV laser for the X axis then use a stepper driven Y axis stage to advance to the next line.
I will always admire your documenting the failures. The importance is far greater than a success. There is a process that was utilized by a straight razor company back in the 1800's for etching blades of extremely fine detail. Unequaled even today. You may find the mask and stencil techniques very useful with modification, looking past the obvious. A rabbit hole worth the effort.
plus the latent risk that the panel fails for whatever reason and having to rerun in from scracth. Plus litteral shipping and handling of parts that are very fragile. Even leaving them in the frame, you can etch artifacts that might bend.
@@sarowie However, with turnaround times of three or four times you could "outsource" quality control to the customer and it would still be faster to do a more tightly controlled rerun and ship that than the current companies' multi week turn around time.
@@sarowie fail? Come on. His results are an insult for someone who knows chemical Etching. The peel of. Losing ridgetops. Undercut he is got. Good joke. And that’s how I watched the video. Good entertaining. But let’s not say “how bad” chemical etching companies are. 😂🙏🏻
@@Kenionatus not always the case. If you project is got good forecast you are a good candidate for PCM. If you only need 1-2 components and this setup presents results to your need - why not using lasercutting? So much cheaper and faster for prototyping if quality is not needed 🤷🏼♂️
What I find fascinating is that I am watching this video of etching metal plates to precision parts through a device that used very similar processes but done on a scale about a million times smaller on a crystal of silicon than what's been done on metal sheets. It's like watching things in my undergrad textbook happen on the scale where I can see or touch, instead of on the scale only electron microscopes and 3D renderincs can barely suffice. This is so cool.
Regarding the plastic fittings at 19:40, you might want to try Fit-line PVDF or PFA Flare fittings. The PVDF fittings are cheaper and might be all you need. These fitting are more likely to handle the heat, pulsating pressure, and corrosiveness of your fluid.
This process reminds me of exposing film to plate for the Printing Industry. Its been 20 years, but I used to work in "Pre Press". You would take the film, place on top of the plate using register pins (register pins are all the same size, for the film and plate). Place the two into a Vacuum table. Hit the switch and it would create a seal between film and plate. Then close this black curtain, and press another button that would start the UV light for a specific time. Once the time expires it all shuts off. Remove the film and place the plate in a solution and the end result was is a positive etched on the plate form the negative film. Love it!!! Anytime I get a notification you've uploaded a new video, Its like Christmas Day LOL... Awesome content again! Thank you!
I know you can use Miore patterns, which are large macro patterns that naturally appear while adjusting the alignment between parts with fine regular features, to quickly alignment-tune industrial survey equipment *by hand* to a surprising degree of accuracy. (Even to the point where the expensive high accuracy measurement equipment was only necessary as a QC / fine-tuning step, improving production rates etc.) I wonder, if you put a a differently spaced dot pattern on each side of the print if that would enable you to do fine alignment of the two layers by eye ( 13:00 ).
the awesome thing about your idea is that he can still use the typical cross as rough adjust, sanity check and fallback. It adds to the precision, without taking anything away.
I used moire patterns to calibrate the flatbed printer for lenticular printing. It is annoying to get it to align perfectly, but you get some mad precision.
Aligning the tool is one of the biggest challenges when it comes to preserving misalignement top to bottom. He will never get any sort of grids good enough for any applications doing it this way. :) He never discused about how shrinking affects alignement etc etc.
The competence required to do all this is way impressive. This was cheaper than hiring a company in the same way as removing your own appendix is cheaper than private healthcare. I'm amazed with how accessible niche high-tech applications and processes are becoming thanks to people like you. Also, PVC is recommended as resistant to Ferric chloride in a SDS I found. PTFE is supposed to work great as well.
+1 on another great vid comments :) From a former pro photographer who started in the age of film, a couple of potential resources for old photo tech that could be cheaply adapted: 4x5, 5x7, 8x10 and custom sized sheet film used to be sent through the machines on these stainless steel racks that had standard and adjustable rungs and spacers that would firmly hold several sheets in each rack. The film was held firmly around the perimeter and wouldn’t come loose-I wonder if you could find some of those kicking around somewhere... dunno if Adolph Gasser is still open but they used to have huge piles of photo junk in the basement. I once scored 3 or 4 big enlarger uprights for a song. Hand cranked sturdiness... Also if anyone in your network is closer than Kevin Bacon to any of the good people who worked at the New Lab, buying them lunch in exchange for describing how their big E-6 line functioned would be invaluable, and would probably come with some great stories to boot. One last aside regarding plastic compatibility: I recently threw together an electrolyzer for water-splitting (because BANG!!!) and found the hard way that few plastics do well with strong KOH solutions, including polycarbonate and acrylic-which makes me think of all those electrolyzer kits with the clear outer walls, and how I wasn’t able to find any common clear plastic that was considered to be at all good for long-term contact with strong alkali solutions... I used polycarbonate because it doesn’t shatter but the KOH took it to pieces in days. Anyway thanks for another good episode :) Wells
We also learned, that despite the claims of many manufacturers, a variety of plastic products are not compatible with acidic products. We do research on redoxflow batteries. Ended up developing a whole process of what works and what doesn't with our chemistry. Its really nice to see your workarounds for the problems you were facing. Keep it up!
For the mask, you could perhaps use a film imager (recorder?) for offset printing. Nowadays many printshops use direct-to-plate imaging, but you may find a used film imager for cheap. The film that goes with it is high-contrast, high-resolution (we used to do 2,400 ppi) and designed for near-perfect registration. Of course, this involves its own film development workflow, which you may not want to fuss with.
Aftermarket photietch for scale models is pretty cool. Also incredibly expensive to buy pre-made stuff so there are inexpensive kits that couple everything for you. MicroMark has a small size kit with everything you need for about 110 dollars (obviously doesn't have a printer included though) Plasmo has a good video on this subject too but it's a lot less precise since it's done on the super cheap with the simplest set up possible
As a career engineer- your projects and abilities blow my mind lmao. Been following since i was in early college and the quality has been stellar for years it’s crazy
Having refined my own home dry film based PCB etching process, it was really neat to see your variations for doing just metal instead of copper on fr4 or phenolic. The laser printers work pretty well as long as you do two layers of transparencies stacked and aligned, and expose with the toner facing the dry film, at least for PCB usage. Otherwise your process is actually surprisingly similar. I just use a cheap USB aquarium air pump to agitate my (vinegar and h2o2 and nacl) etchant, it's plenty for thin copper clad PCB and a good deal safer than ferric chloride
@@AureliusR yep! Not super fast, but it will turn into brown gunk that you keep brushing off (or you let an aquarium air pump keep it turbulent enough for you), and it will etch. You still get a "liquid with copper in it" at the end to dispose of safely but it's not pure danger in a bottle like ferric chloride.
@@rpavlik1 I've worked with ferric chloride quite a few times and never viewed it as particularly dangerous. Take proper safety precautions, and don't drink it, and all should be well. That being said I might try the vinegar method sometime.
The synchronicity of this is just astounding. Yesterday I was looking for this exact thing and found no real good guides on youtube, until I found this - which you uploaded yesterday. Great video!
I can always count on this channel to introduce me to a topic or project I have never even considered before and then hold my fascinated attention for 22 minutes. Well done!
Wow, the amount of good information in this video is absolutely huge. But Ben, how are you ever going to start your own succesful etching business if you are telling us all your secrets? ;-)
The video reads a lot like "somebody please make a business out of this so I can buy cheap parts from you rather than make them myself. Look, here, I even made you a business model."
This video was kinda nostalgic to me, this is like a scale up version of the DIY circuit boards I used to make back at my electronics undergrad. Oh the memories of shorted chips due to a bad etching
what is the pumping pressure for the pneumatic pump used in this system? Did you try the technique with only one side exposed? ie etching only from one side?
It is amazing how you optimize a process that is essentially meant to be done a few times. I mean I do a prototype PCBs every now and then but my process is much more manual an tedious then your one time one. You are amazing. Also It is just a best day to have Applied Science and Ben Eater video in one day! Thank you very much for your work.
Have you tried to laser print transparencies and then just use acetone vapor to slightly melt the toner and increase the density? I had some limited success with that a few years back, what you absolutely can't do is let the print for too long on that vapor.
I’ve definitely used paper to back transparencies. I had the same problem. I’m glad you brought it up because I would have had to figure that out again. Yes that was very frustrating.
Absolutely awesome! Unbelievable that you can do it all on your own. For many people in the world America means their own things, for me America is this kind of tinkering, ingenuity and freedom to create and people who brought the joy of looking at this process. Thank you!
Thanks, Ben for sharing precious knowledge for free for everyone. I haven't had time to go through the comments yet so i might be doubling on someone's, nevertheless: back in the day when i was working in Peltier elements manufacturing we were implementing and using lots of funny techniques. 1) alcohol is pretty darn good for removing water from the substrate, multiple stages of concentrations(40\70\90\100) were employed just to prolong the higher concentrations bath lifetime (the same trick is done in machines to remove water from tissue in histology), and other solvents like acetone, or ethyl acetate can be used with great success. Subsequent hot air in laminar box was used sometimes also. 2) ultrasonic is superior for immersion washing and even sometimes etching, I've built several machines for degreasing of bismuth telluride crystals with solvents\basic wash as well as few for electro(chemical) surface etching all employing medium circulation alongside ultrasonic. It might be a problem for thin substrate sometimes as ultrasonic tends to shatter brittle crystals into dust, but nothing that cant be adjusted.
I've only ever done basic pcb etching. I thought that was wonderful at the time. This is so many orders of magnitude more wonderful. Brilliant hacky engineering.
when I worked in the PCB manufacturing industry we aligned our photo mask prints using a x 10 magnifying glass. Worked fine and was very quick. Interesting project, of course very similiar to producing your own PCB's
Thank you for uploading. I never do anything particularly useful with the information you provide but you’re always extremely informative and entertaining. I appreciate every video.
this channel is by far the best applied sciences and engineering channel I have seen. each and every video that is made is detailed to near perfection and the presentation of the work leaves you with virtually no questions about it. thank you for giving us quality content throughout the years.
I also needed to do this for a project. I tried using much simpler equipment with limited success. Thank you for a very detailed description of a successful process. I eventually decided to go with stainless steel solder masks from OshPark and Seeed Studios.
I once used a similar process at work. But, we were galvanically growing the parts we need on a soft circuit board style substrate. The guy who taught me about the process originally used the idea to make massive amounts of custom soft metal gaskets, for a particle accelerator.
I've been doing this for years with a laser printer. I just print twice on the same glossy paper and then I transfer the ink to the metal sheet with a hot iron, thus using the ink itself as a resist layer. Before etching I go over the part looking for any suspicious areas not perfectly covered with ink and just use a permanent marker to "fill in the holes". I'm getting almost as good (as in almost impossible to tell the difference) as when using UV resist, but it's faster to transfer the design via UV resist for sure, and way easier to clean of the UV resist than the ink, but if in a pinch it works really well. I'm using this for making parts for scale modelling btw. I've also tinkered with making a two step procedure for partial etching where you need to etch fold lines, or etch a step into the metal sheet for replicating wire mesh etc. I just do the process twice on the same piece. A lot of work I might add... I've got a similar setup to yours but mine is a bit more manual but it does the job - for very little money!
As always, awesome work. I've been down a similar road making stainless parts for a micro robot, my laminate was custom stainless-kapton-stainless to create flexures and I dip coated with a positive resist as it was exposed with an NC controlled UV laser so that saved time. I used a commercial cleaning product but also cleaned with pumice, your method is much easier! I can't actually remember how I did the drying, I bought parts from a dyson hand dryer but I'm not sure I used it, I also had a contraption made from too cooling fans from a Harrier jump jet (the seller claimed). I did make a laminar flow cabinet so it might have been just drip dry. It is possible to use magnetically coupled pumps for the spray and this is common in commercial machines, the trick is to buy the right nozzles. An alternative used by a UK company Mega Electronics is to have a vertical tube with holes in it that is rotated quickly by a motor and dipped into the acid source, some plastic RC boat propellers in the bottom of the tube encourage the acid inside and then centrifugal force does the rest. Cheap and simple. For precision etching it is considered good to rotate the panel in front of the etch nozzles to avoid any sort of shadowing affects. This is what I did and I have some videos of my janky belt driven system. Lastly something I read in one of the few books I could find on the subject is that rotating the panel at high speed in a tank can allow for very deep etches as the centrifugal (or lack of centripetal) force clears the dissolved material quickly. This isn't something I have tried but I'd like to some time and it could be a good budget option and rotating it horizontally in a frame would be simple mechanically.
Excellent setup for etching in volume. I used a much simpler setup to do PC board etching in my home shop, then successfully etched brass and copper sheet fine parts using the same setup and chemical etchant. To get the imaging film on the part under exposure I used a foodsaver vacuum sealer and exposed one side at a time, since foodsaver plastic is only flat on one side. For low volume it was a chep vacuum easel.
this is BRILLIANT!!!!! You are thew most HARDCORE DIY guy in existence!!!! You make the most amazing projects and you never cease blowing my mind!!! THANK YOU for EVERYTHING you do!!! :D
Thank you Sir, Well demonstrated. I used a more crude approach with iron on resist intended for PCBs, I etched aluminum and stainlessness steel to make dies for metal casting pendants. Though the relief was small, the result was satisfactory for jewelry. I appreciate you giving us not only the technical details, but also insight into business and production, innovation and as well encouraging and stimulating our minds. Adam Smith's invisible hand is ever present. Take care
This was awesome I used to be a tool and die maker at Tech Etch in Plymouth MA where they made these type of parts. I love that you just made them on your own. Brilliant
Your spray process is one of the fastest I have ever seen. While fast, it may not fit in my apartment. However, hot ferric chloride can be pumped using home aquarium fish tank pumps. The kind with no moving parts, and are air-operated, are the ones of interest. Bubbles do all the work. I've noticed that the acid only needs to flow and etching goes faster.
It always blows me away how you can deconstruct a process to it's bare bones and get a result at a fraction of the cost and you gladly share all your knowledge. You're a true pillar of the community for sharing!
it's proof that like 80% of most industries are fluff and bloated profits surely for the benefit of CEOs
Not to be disrespectful to the guy. but he knows nothing about chemical etching.
Whatever he proposed with his video is nothing but a silly game and those components will never end up in anything that you would like to use.
He can maybe make some machinery which is very interesting. But he just made chemical etching look bad and he made all the mistake you can make when etching something.
I guess he has his place on youtube ;) but not anywhere near what a professional chemical etcher would do.
@@Mr.Flo.Official Some salty idiot named mr flo got upset that companies are inflating costs of products for no apparent reason other then to make money. Most of the time the answers are simple and cheap.
@@Mr.Flo.Official insult him all you like, I will still go with this dudes methods over giving business to stingy con artists any day.
@@Mr.Flo.Official "He knows nothing about chemical etching" he literally showed the process on how to chemically etch metal, and the issues you will have and how to solve them. You couldn't be anymore wrong on your assessment.
Another neat thing with photoetch is you can do partial depth to make fold lines,like scoring cardboard. Probably needs finer process control,though might be worth trying just putting them on one side .
Mike, Yes it is normal practice to have the fold lines etched only from one side. The Asteriod Belt shown here content.instructables.com/ORIG/FEJ/ID3W/HFSHFI10/FEJID3WHFSHFI10.jpg does that by having half etched lines. The photo tool for the back here content.instructables.com/ORIG/F8D/ELWM/HFSHFHYS/F8DELWMHFSHFHYS.png has thin lines for the folds but the front side is solid. The same for the buttons so the arrows and symbols are only etched half way through.
@@TheCunningFellow Thanks.. how is it that UA-cam is happy with your links?
@@patkirk960 I didn't know youtube censored URLs. I guess if the disappear you can search for "The asteroid belt instructable" and scroll down to step 5. I used "toner transfer" there because I was told instructables readers prefered to see simpler tech than dry film photoresist.
@@TheCunningFellow Thanks. Yes they certainly remove them somehow I'm sure automatically but somehow folks circumvent the censorship!
@@TheCunningFellow Some lovely extra engineering in your instructable that's above and beyond the average. The extra yet simple hardware to clear the screen and to reduce the overhead from code was delightful.
That did not feel like a 22 minute video. Super fascinating stuff! As always, your projects get my imagination going with all sorts of new ideas.
My eyes were glued like I was watching a tv show or something. Incredibly interesting and produced.
didn't even check the timestamp: this comment just saved my lab report
It's his mellow tone. Lures you in.
Got to agree with everyone, it seems there is no time passing when watching, Mesmerizing videos with very informative information,
Always learn something interesting
Never long enough !
I'm thinking you could print some markings on the 4 corners of each transparent mask so that you could use a moire pattern technique for easy and clear alignment of the masks when you need to overlay them. I hear this technique is pretty good and can even allow you to quantify the degree of alignment.
That would be a cool methods
Hi Ben, The specialty black inks are great, but if you want to use a regular inkjet then sometimes yellow dye inks can be more opaque to UV light because of UV inhibitors to stop fading. Also a Stouffer Gauge will help a lot working out your exposure times. Cunning Fellow (The guy that did The Etchinator)
Would it be visual at the end? I mean, can you easily inspect the quality when its printed?
I'd imagine , no?
But maybe a mix of dyes?
You can see the yellow ink with your eyes no problem. A non yellow light source like a large blue LED will enhance it. But really the developed dry film is the thing you should be inspecting most closely.
Agreed. Although the uv blocking power of different coloured inks depends highly per brand and type. Epson alone already has 4 or 5 different ink lines. The best are the pigment type, the dye inks are really poor uv blockers. Your best approach is to print a step wedge with all the different ink colors and see which blocks the most. More uv blockage can really up the contrast and sharpness of the lines you want to etch
Do you know any particularly good yellow inks for this kind of thing?
Inna - Pigment inks naturally block more light so if your printer uses pigment inks (Epson would call them DuraBright or UltraChrome) then black is the best option. If your printer uses dye ink (Epson Claria) then the black does not block the UV as well as the yellow. The yellow is most susceptible to fading from the sun, so they add UV inhibitors (sun screen) to the yellow ink. Most brands of dye yellow that claim they are light fast or don't fade in the sun will do the same.
Coming from someone who has years of photochemical machining experience, your setup and process is pretty spot on. Great job!
You call that spot on from a PCM background?😂 come on man. This is an insult to the process.
Good to say “playing with FeCl” but not comparing with industrial.
Must admit was amazing to watch from entertaining purposes tho.
@@Mr.Flo.Official I do. For a diy at home setup it's covering all the necessary bases. When you break down the milling process you have some basic concepts that need to be implemented. He implemented all the required ones.
@@joshuamunson2876 agree :)
So excited every time a new Applied Science video is released!
They're literally brain candy! 🍬
It’s easy to get excited when they only release a video a few times per year.
For registering the masks/artwork you can get an anti-parralax magnifier or collimating magnifier - they're often used for 4 color printing - look for para-mag magnifier or caprock stripmaster series. 7x or 7 power should be enough.
Never heard of those. Good tip!
With appropriate fiducial marks and a microscope, you can be surprisingly accurate hand aligning. I have bonded so many 30 micron pitch flex circuits by hand.
In addition, the masks should be taped to a spacer of the same thickness as the sheets including their resist coatings (ie, a strip of the stuff)
That's experience from registering and etching double sided PCBs back in the 80s ... :)
I'd create some light box tubes around those 405nm lights too. This will greatly increase the luminous flux at the sheet surface whilst reducing purple light spillage where it can cause trouble - it was more important back in the days when UV sources were being used - The setup I used had a large mercury arc lamp set 5 feet from the PCB surface and you _really_ didn't want your corneas being exposed to it
It could be used to calibrate for stretching as well by printing 2 square or circle patterns and comparing them at 90° to each other. What he's worked out is incredible as is, but it's very close to being something wildly more precise.
Use multiple angles of diagonal lines on both bottom and top. You can see fringing very easily. Zone plate idea.
I work with/design stamped parts and source mesh all the time… Stainless, zinc/tin plated parts and stainless/epoxy coated steel mesh. A lot of our vendors are willing to do stamping tooling on the cheap (
The work you put into these projects is completely mind blowing.
I can't imagine having to sort through and store all the gadgets involved, never mind having to buy them all
Super cool project
It blows my mind that we live in an age where knowledge like this can be shared so easily
This is the true power of the information age
"the mess is going to be beyond spectacular" is a phrase that I am going to start using from now on.
Hi Ben, I used to design machines used in photolithography and the photoresists coaters. We used to etch parts on sheets of glass, etching gold and titanium layers. We used to use spin coaters. The nozzles would be placed above a spinning chuck that was made of nylon. The whole thing was surrounded by a big tub with sides higher than the spin axis. The force of the acid falling, hitting the surface and then being flung off was enough at 1000 rpm to take the debris away. For the underside we had nozzles firing the acid up at the underside surface.
Ben stuns again with his brilliance and creativity. Did you see how EASILY he defeated that printer's transparency limitation? Loved it.
One method to improve the density and reduce the pinholes in laser printed output is to use color foils (also called toner reactive film). You run it through a hot laminator or back through your laser printer with the color side against your artwork, and the color film fuses the toner, but not to the areas on your artwork that have no toner. I use it when I am etching PCB's at home. (For PCBs, I print the art on release paper, laminate it to the PCB, then do a short soak in water which disolves a layer of the release paper, allowing you to peel the paper away from the PCB leaving the toner artwork on the PCB. Then I run it through the laminator with color foil which helps with density and pinholes. Then I etch the board w/ ferric choloride.) The color foils don't work with inkjet, though.
would you mind providing a link to the release paper as well as the color foil paper that you use?
(Bookmarking for reply wait)
I've used a product called 'Press-n-Peel Blue' for this. You can print directly on the film then laminate that to the copper pcb. Can be tricky to get the laminating right - pressure and temperature need good control. And everything has to be kept dust free. I've used this for PCBs, and for sheet metal parts. 5mil track/space is possible (just).
@@weldandcutdotcom You can't use external links in UA-cam or they get deleted. You can also get banned, like when I used a fake comedic email name and got banned.
@@SidneyCritic What about 'fudging' link so that the filterbot does not recognize the link? Like mmm d0t earl d0t corn (that was typed as CORN) ??
What sort of pump did you find that can handle ferric chloride without disinregrating?
As per ~ 18:48 they are "Air Operated Double Diaphragm Pumps" and appear to be made by "ARO".
So Basically it's almost like a drum (the instrument) with pipes over the external "drum heads". Air is pumped into the "drum", the diaphragms push out a bit, and if the "pipes" are designed all fancily and some one way valves are used one can make a pump. These not only can be really chemical resistant, but some can even pump abrasive slurries!
Couple of thoughts from when I used to home etch PCBs:
Laser printing onto tracing paper works much better than transparencies,you nerd to use thick stuff,>90gsm to stop crinkling in the fuser.
When aligning the artwork, stapling works well as you can align accurately with the sheets completely flat against each other,but you do need a setup where the staples overhang the edge when exposing, less of an isdue with PCBs due to the thickness of the board
That magnetic setup on the etching machine is amazing. I could see something like that used for cleaning 3D printed resin parts.
This whole video is a how-to start your own photochemical machining company.
For more precise alignment (at least during the tape phase), you could use a moiré pattern , possibly with a vernier type setup (so one registration is slightly smaller on one side). Then you could align the interference pattern. Although I’m not sure such precision would be necessary, so you could probably just go with interrupted overlapping lines on the cross pattern (similar to how polymer bank notes have transparency alignment patterns)
Regarding the transparency and the special ink. I had the same problem 25 years ago making masks for exposing PCB's. Came to the conclusion that printing on ordinary paper worked well enough. The laser toner sticks really well to paper, the coverage is 100% without pinholes. And the UV used to expose the photo resist went through just fine.
I had to mirror flip the design so that the printed side of the paper touched the photo resist.
Okay, that's interesting that the toner sticks fine to paper, but not transparencies. I'm more surprised to hear about the UV transparency of the paper, though!
Just went through this process and contacted all the vendors you mentioned in the intro for making a few shadow masks for a metal evap tool. Was thinking the whole time this is something I could just do in my shop and it's crazy expensive. Of course I just thought it and you actually did it! I don't have enough demand to build it but it sure would be nice if an OSH-ETCH or SendEtchSend sprang from these efforts!!
What a coincidence, I've just done the same for a project too!
ETCH-PARK
@@MaltandMake Fancy meeting you here!... but not at all surprised.
@@CollectiveSoftware Yes, that's better, we'll go with that one!!
Not to be disrepectful. But what sort of tolerance you need your shadow masks to be at?
The setup presented won’t offer anywhere newr +/- 1-2 mm. Also the details and peel of he gets are soooo bad to watch coming from a PCM background :)
I have been doing a variation of this. Instead of printing on transparencies and then using an UV exposure station I used a cheap resin 3d printer to expose my artwork on the resist.
(The kind where a UV light is projected through a digital screen to solidify the resin. I have been using an Elegoo Saturn. I just remove the vat and put the plate on the screen.)
It worked quite okay for me. I got some niece pieces. The theoretical benefit is that you have zero setup cost. The very real drawback is that if you want to etch from two sides (and you want to etch from two sides) then you need to flip the plate and place it very precisely at the same spot on the resin printer's screen.
To be able to achieve that I have made a kind of jig holding a usb microscope solid relative to the printer. If you do an exposure without the plate being present you can see the invididual pixels of the resin printer with the microscope. And then you can use this to align the second side of the plate very precisely.
I have made this decorative orrery using the technique: ua-cam.com/video/KK3TiUWSTbI/v-deo.html
In the past I used the head from an photo collimator for a similar UV exposure process where we needed collimated light. Your setup seemed to do the trick just fine, but we needed it to be up close in my application. Its a reasonably cheap way to get get it done as you can fine them second hand. Did have to air cool the lenses as they got a bit hot from the high powered UV light sources :) Mesh bag was a slick idea! Really great video once again
I love this guy. He makes me happy to be alive.
Your idea about a "really clean air source" sounds very familiar to another interest of mine, mushroom farming! They need sterile workspaces and they use large like 2 foot by 3 foot HEPA filter pointed down to create a sterile downdraft area to handle spore prints and gather samples without contamination. They don't even cost all that much, and with a custom hood could make a real nice drying area. I have thought about this for another area I am interested in, beer brewing, as an area to store a kveik ring to dry and keep from contamination.
HEPA filters really are amazing. I do research using flat sheet media and they can handle concentrations greater than 10 million particles/cm^3 at well over 5 cfm through the filter (more than a billion particles a minute!) and still capture 99.97% of the particles.
I worked in electronic manufacturing and we usually have clean booth. It is just a tent made from extruded aluminium and plastic sheet, with a filter unit on top. The booth is just large enough for a small table where a worker sit to assemble dust sensitive parts like camera or screen. The filter unit is just a fan blow into HEPA filter to create positive pressure inside the booth. There is also ion generator for ESD protection and some LED light. There is also a mini table top version, and some assembly machine have integrated filter unit on top to keep the inside clean. Pretty easy to DIY something like that.
The fact he did not keep his knowledge a secret to push his business has just gained me huge respect from me. Keep up the good work
I really appreciate how you not only show your successes but discuss your challenges in these projects.
I have one suggestion if you ever wanted to revisit this. Back in my PCB etching days I found that a viable alternative to spray etching is bubble agitation of the etchant. Putting a fine mesh bubbler grid at the bottom of the tank below the part has similar performance to high pressure spray with the added benefit of not having the etchant running in the plumbing and pump.
Both spray and bubbling end up with the same effect of getting fresh etchant to the metal being etched. Just putting the material in a still tank ends up with the etching site having lower concentration of fresh FeCl as the already combined products only slowly move away from the reaction plane. Bubbling seems to have a similar effect as spraying, constantly introducing fresh FeCl to the etch site increasing the overall reaction speed.
My first test of this setup didn't anticipate how dramatic the difference was vs simple etchant circulation, so I etched too long and ended up undercutting the mask so much that all the copper was removed from the board! Not quite the result I intended.
The problem with bubblers is that no matter how much filtration you have, there's a constant outflow of ferric chloride saturated air from the tank and things get..... corroded... in the vicinity. We ended up swapping out the bubbler for a couple of model boat props on very long shafts run through the tank lid
When etching copper, there are alternatives to ferric chloride. It's reliable but messy and makes even stainless steel rust
High strength hydrogen peroxide + hydrochloric acid (we used to call this the 6 second etch - extremely dangerous in high concentrations) - this works really well and lasts virtually forever (used ferric chloride is an environmental hazard due to the copper in it) - this also eats stainless steel so use with caution
Ammonium persulphate works really well too
There are a few instructables around on this stuff, but etching steel or other metals will have different ideal etchants
@@miscbits6399 I don't disagree that air filtration is a necessity with bubbler systems. I used an activated charcoal fume capture with good results. I didn't mention that point since I was comparing to a spray etch system which has very similar exhaust air filtering needs, but thanks for pointing this out.
Why would you go for that instead of normal
Horizontal conveyor with oacillating sprays top and bottom?
It’s an oxidation reaction which will allow good circulation of fecl3 while solids deposit at the bottom of sump and also have enough flow contact with the substrate.
Sounds interesting what you tried anyway.
@@Mr.Flo.Official Good questions. Answers: this was ~1981 in a lab setup for prototyping boards. I already had a etchant circulating dunk style tank which I could convert by reversing the hose connections to pump air vs liquids, and didn't need to add parts. This was long enough ago that I was prototyping by drawing circuits with resist tape, not even photo resist.
@@IFS i see :) things changed since.
wow i'm sure that i am not alone in thinking that i'd happily wait months for a vid from Ben. they always all-around perfect. length, audio, video.. then the content.. absolutely fantastic. seeing prototyping is amazing, but seeing Bens prototyping is the epitome of excellence for video. descriptions or ups and downs and gotchas... full of information and helpful and still concise..
WHY IS NOT 1M SUBS YET???
As I'm in the process of researching etching, in my case to make data plates, you post this video. Usually your videos send me down a new rabbit hole. It's amazing how much trial and error, research, and practical experience you can compress into one of these videos.
I'm jealous of how smart and articulate this guy is. I wish more minds were like his.
If no one else has mentioned it, look at the clean air boxes that mushroom growers use. It might work for drying the metal after cleaning if you’re doing large batches
laminar flow hood, you can DIY them pretty easily
I've watched this a couple times over the last few months, and the ingenuity in this whole process design is just mind-blowing. I would absolutely back this business!
I make circuit boards using laser printer toner for the resist. I use a heated 'laminator' to apply it. My artwork _IS_ my resist. It took a few tries to get right combination of backing paper, heat, and pressure dialed in but it was an inexpensive and rapid process. I purchased the highest resolution BUDGET monocrome laser printer I could find and have much success. I don't know if toner would survive your mesh bagged etching intact though.
One note on using toner transfer. I don’t know if this is still the case, but at least formerly, Brother’s toner didn’t work for the process. I don’t recall the details, think it maybe had a much higher fusing temperature so it was hard to get it properly melted to stick to the PCB material.
I’m blown away. You just singlehandedly recreated and improved upon a modern manufacturing process in your garage. There might be what, 300 people in the world who could pull this off?
As others have said, great work as usual X)
For the registration alignment of the two layers of transparency - what about using an overhead projector? Put both sheets on the glass, then use the projected image to align the parts. Not a huge amount of magnification, but even 3~5x with good ergonomics would help.
That’s brilliant!
"arbitrary"
"extrusion"
"beyond spectacular"
this guy is on an unreachable level of DIY
This whole process is very similar to offset printing plates. But they were in aluminum, and only partially etched into the plates for printing. We did quite a lot of these in our class. This was highschool graphics class, back when the 486 was being conquered by the pentiums. We used all Macs though for that class. Don't remember what they were, but early 90's style machines. The large camera is what we used to make our transparencies. Literally a giant wall with a huge lens, and a room behind it to put the transparent film onto it. We developed it right then and there. I bet film transparency would be even better than inkjet as far as detail goes. We then used this on a the plate that was pre-covered with resist, put it in the light bath thing for a few minutes, and then washed it off in a special bath and unknown chemicals (at the time), and then rinsed them, and the plate was done.
Nice that you highlighted the alignment problem many people will run into.
I've seen cases where people even closed off a third side, creating a sort of envelope for the metal sheet.
It's almost creepy how you come out with videos that answer the questions that I AM CURRENTLY STRUGGLING WITH. I'm talking about... the same day. It's super helpful, but also unsettling.
Same thing for me, I just watched a video on model making and was wondering "how are those photo etch detail kits àctually made?"
Let's test your theory. What are you working on now?
@@TqSNv9R0iG5Ckxew At the time I was in the process of torturing myself trying to get the dry film photoresist to adhere good enough to copper laminate. If you look really closely you can tell Ben is actually having the same issue. The film delaminates during the etch around the edges. Sometime delaminating completely. I haven't found any technique that will work well enough to have 1mm or less resolution. But I have found something that does actually work pretty well. It is labeled as "Photoresist Anti-etching Blue Ink Paint For DIY PCB." It's a thick blue ink that needs to be cut with acetone. You can either spray it on with a "Devilbiss" style paint sprayer (which will give you a better, flatter finish" OR you can go to an auto parts store that carries "Preval" paint sprayers. They are like a mix your own spray paint kit. pretty cheap, and works well enough. You have to dry the board well (quickest I've found is with a couple fans in a closet with one of those Edenpure heaters set to 90 degrees F. Important note: You can not heat the photoresist ink past 100 degrees or so or it will cure the ink and will spoil the board. It will dry and not be tacky, or will only be very slightly tacky. Then you can expose and develop the same way you would the dry film. As long as you clean the board as per Ben's instructions you should have NO problem with lift off and the edges are crisp. Thing to note, the vapors whilst spraying will probably give you cancer... sooo.... well ventilated area, a respirator, or roll them dice.
@@jonathanseagraves8140 so way on earth would you not use wet resist?
When i see amateur etchers trying to use dry resist i just turn it off.
He has so many flaws in this video.
Fun to watch. But not the setup
You should use for better results in your garage :)
@@Mr.Flo.Official Short answer is I don't know where to buy it + sunk time fallacy + I don't know what product I should be looking for in the first place... I will take any info you have.
Wow, I'm the person the algorithm's working for right this instant - I just switched tabs from a online auction where I'd spent spent about a quarter hour studying the photos for a photochemical machining setup which for sale for about $25k. I just came to the conclusion that it's unnecessarily complicated and expensive for what I can do in my studio given a little patience and slightly more swearing. Of course it would make sense if you needed 60 of something.
I've gone through almost exactly this process in the last week so this couldn't have come at a better time! I ended up going with a spray photo resist and then removing the resist with a galvo laser since I was making a decorative brass piece but this seems like a fascinating method too!
I'm not sure if adding citric acid to the ferric chloride will help with stainless but so called Edinburgh Etch is a more efficient etch used in copper plate print making. The theory being the citrate helps to keep the etched material soluble.
In case others want to take a deep dive:
I've also though about using a resin 3D printer as a direct digital-to-plate technology. I've seen videos about doing this for PCBs. Since they have a UV source and the screen acts as a digital resist, you can put the sensitised plate on the screen directly. As the printers get bigger, hopefully there'll be more spare parts available to make a dedicated unit for exposures at a usable size.
Having etched circuit boards, I am impressed by your effort to do large boards and to do so with minimal human intervention.
Dude! Awesome. You inspire me to get out in my garage and tinker around. Thanks for letting all of us look over your shoulder while you work on projects!
Ben is like a single person A-TEAM that would solve issues for secret projects! I love it when a plan comes together…
I'm always impressed by your big brain projects and how well you explain them so the layman can still understand the process. I'm retired and will likely never do most of what you show but I really enjoy learning new things and find your projects interesting.
Dude, you are one of the most formidably brilliant and capable people I've ever encountered. This process alone by any right should make you wealthy. And this is just one thing you did for a random project... I hope you're rewarded for your endeavors.
Very cool! It's quite similar to the process I've used for home PCBs, an Epson inkjet is pretty much mandatory for getting that super-dense artwork-- and yeah, pretty annoying you have to use a piece of paper as a backing to get it to detect the transparency. I use a medium-format Artisan 1430 outfitted with a CISS system to make sure I never run out of ink. Now that Epson has come out with Econo-cart refillable systems it sure makes the whole process straightforward, using special high-density inks. For developer I do like using sodium metasilicate since it seems a bit more forgiving. Also using a vacuum-bag to hold the artwork against the boards is a lot less hassle than getting pairs of glass pieces to sandwich the artwork/board and making sure it has even pressure across the whole piece. For single-sided boards I just use a consumer vacuum-sealer/bag, though I have to make a bag out of heavy plastic tarp for double-sided boards since they have printed stuff on one side of the vacuum bags. To seal them I use the rubber tape which is used for carbon fiber/fiberglass resin infusion, as well as sealing around the vacuum hose. Very easy to use, strong, and virtually leak-proof.
I’m an engineer and you astound me at what you’re casually able to accomplish in your little shop. It truly blows me away every video
This is how I make my PCB prototypes :)
I use immersion oil (for microscopy), between the photomask and photoresist - the edges come out much sharper.
Me too. I use ky lube instead. easily dissolves in water
@@marcoaurelio4903 If it is suitable in terms of optical characteristics and has similar refractive properties, then it is possible.
Ben, thanks for the video, brings back memories of make PCB boards before you could buy them dirt cheap. I like your etching tank, for PCB's I has a splash tank. It was about a foot cube tank epoxied plywood. You put about 1 to 2 inches of etchant in the bottom. A motor driven dasher paddel splashed it all over the board with good force. --worked very well.
The final improvement that I never finished was to make a raster direct UV laser diode imager. No need to do any lithography. I met a guy Zoltan who converted a old printer, replacing the print head with laser source and lens. He got some amazing resolution from the hack. One thought I had was to convert a laser printer poly scanning mirror and lens from the IR to a UV laser for the X axis then use a stepper driven Y axis stage to advance to the next line.
Such elite in-house (literally) mfg capability. Desperately waiting for the applied science 1um IC PDK to drop
I will always admire your documenting the failures. The importance is far greater than a success.
There is a process that was utilized by a straight razor company back in the 1800's for etching blades of extremely fine detail.
Unequaled even today. You may find the mask and stencil techniques very useful with modification, looking past the obvious.
A rabbit hole worth the effort.
I can imagine a fully automated system doing all those steps. No wonder why the comercial option is so expensive.
Amazing video as always!
plus the latent risk that the panel fails for whatever reason and having to rerun in from scracth.
Plus litteral shipping and handling of parts that are very fragile.
Even leaving them in the frame, you can etch artifacts that might bend.
@@sarowie However, with turnaround times of three or four times you could "outsource" quality control to the customer and it would still be faster to do a more tightly controlled rerun and ship that than the current companies' multi week turn around time.
@@sarowie fail?
Come on. His results are an insult for someone who knows chemical
Etching.
The peel of. Losing ridgetops. Undercut he is got.
Good joke.
And that’s how I watched the video. Good entertaining. But let’s not say “how bad” chemical etching companies are. 😂🙏🏻
@@Kenionatus not always the case.
If you project is got good forecast you are a good candidate for PCM. If you only need 1-2 components and this setup presents results to your need - why not using lasercutting? So much cheaper and faster for prototyping if quality is not needed 🤷🏼♂️
What I find fascinating is that I am watching this video of etching metal plates to precision parts through a device that used very similar processes but done on a scale about a million times smaller on a crystal of silicon than what's been done on metal sheets. It's like watching things in my undergrad textbook happen on the scale where I can see or touch, instead of on the scale only electron microscopes and 3D renderincs can barely suffice. This is so cool.
Regarding the plastic fittings at 19:40, you might want to try Fit-line PVDF or PFA Flare fittings. The PVDF fittings are cheaper and might be all you need. These fitting are more likely to handle the heat, pulsating pressure, and corrosiveness of your fluid.
This process reminds me of exposing film to plate for the Printing Industry. Its been 20 years, but I used to work in "Pre Press". You would take the film, place on top of the plate using register pins (register pins are all the same size, for the film and plate). Place the two into a Vacuum table. Hit the switch and it would create a seal between film and plate. Then close this black curtain, and press another button that would start the UV light for a specific time. Once the time expires it all shuts off. Remove the film and place the plate in a solution and the end result was is a positive etched on the plate form the negative film.
Love it!!! Anytime I get a notification you've uploaded a new video, Its like Christmas Day LOL... Awesome content again! Thank you!
I know you can use Miore patterns, which are large macro patterns that naturally appear while adjusting the alignment between parts with fine regular features, to quickly alignment-tune industrial survey equipment *by hand* to a surprising degree of accuracy. (Even to the point where the expensive high accuracy measurement equipment was only necessary as a QC / fine-tuning step, improving production rates etc.)
I wonder, if you put a a differently spaced dot pattern on each side of the print if that would enable you to do fine alignment of the two layers by eye ( 13:00 ).
the awesome thing about your idea is that he can still use the typical cross as rough adjust, sanity check and fallback. It adds to the precision, without taking anything away.
I used moire patterns to calibrate the flatbed printer for lenticular printing. It is annoying to get it to align perfectly, but you get some mad precision.
Vernier calipers use a similar trick to let you read extremely fine measurements. It's such a clever idea, I was amazed when I learned to use them.
Great idea to use moirés! They should make very fine alignment easy without magnification or squinting.
Aligning the tool is one of the biggest challenges when it comes to preserving misalignement top to bottom.
He will never get any sort of grids good enough for any applications doing it this way. :)
He never discused about how shrinking affects alignement etc etc.
every upload is like watching a magic show.
and you still tell us how its done and im still in awe
Great to see this, thank you for your dedication and knowledge
The competence required to do all this is way impressive. This was cheaper than hiring a company in the same way as removing your own appendix is cheaper than private healthcare.
I'm amazed with how accessible niche high-tech applications and processes are becoming thanks to people like you.
Also, PVC is recommended as resistant to Ferric chloride in a SDS I found. PTFE is supposed to work great as well.
+1 on another great vid comments :) From a former pro photographer who started in the age of film, a couple of potential resources for old photo tech that could be cheaply adapted: 4x5, 5x7, 8x10 and custom sized sheet film used to be sent through the machines on these stainless steel racks that had standard and adjustable rungs and spacers that would firmly hold several sheets in each rack. The film was held firmly around the perimeter and wouldn’t come loose-I wonder if you could find some of those kicking around somewhere... dunno if Adolph Gasser is still open but they used to have huge piles of photo junk in the basement. I once scored 3 or 4 big enlarger uprights for a song. Hand cranked sturdiness... Also if anyone in your network is closer than Kevin Bacon to any of the good people who worked at the New Lab, buying them lunch in exchange for describing how their big E-6 line functioned would be invaluable, and would probably come with some great stories to boot. One last aside regarding plastic compatibility: I recently threw together an electrolyzer for water-splitting (because BANG!!!) and found the hard way that few plastics do well with strong KOH solutions, including polycarbonate and acrylic-which makes me think of all those electrolyzer kits with the clear outer walls, and how I wasn’t able to find any common clear plastic that was considered to be at all good for long-term contact with strong alkali solutions... I used polycarbonate because it doesn’t shatter but the KOH took it to pieces in days.
Anyway thanks for another good episode :)
Wells
We also learned, that despite the claims of many manufacturers, a variety of plastic products are not compatible with acidic products. We do research on redoxflow batteries. Ended up developing a whole process of what works and what doesn't with our chemistry. Its really nice to see your workarounds for the problems you were facing. Keep it up!
For the mask, you could perhaps use a film imager (recorder?) for offset printing. Nowadays many printshops use direct-to-plate imaging, but you may find a used film imager for cheap. The film that goes with it is high-contrast, high-resolution (we used to do 2,400 ppi) and designed for near-perfect registration. Of course, this involves its own film development workflow, which you may not want to fuss with.
Came here to say this. Useful ebay terms include "imagesetter" and brands "linotype" and "agfa". Beautiful output and perfect for this application.
Aftermarket photietch for scale models is pretty cool. Also incredibly expensive to buy pre-made stuff so there are inexpensive kits that couple everything for you. MicroMark has a small size kit with everything you need for about 110 dollars (obviously doesn't have a printer included though)
Plasmo has a good video on this subject too but it's a lot less precise since it's done on the super cheap with the simplest set up possible
As a career engineer- your projects and abilities blow my mind lmao. Been following since i was in early college and the quality has been stellar for years it’s crazy
Having refined my own home dry film based PCB etching process, it was really neat to see your variations for doing just metal instead of copper on fr4 or phenolic. The laser printers work pretty well as long as you do two layers of transparencies stacked and aligned, and expose with the toner facing the dry film, at least for PCB usage. Otherwise your process is actually surprisingly similar. I just use a cheap USB aquarium air pump to agitate my (vinegar and h2o2 and nacl) etchant, it's plenty for thin copper clad PCB and a good deal safer than ferric chloride
Well done. Sounds way better doing it that way.
Vinegar, hydrogen peroxide and salt can etch copper?!
@@AureliusR yep! Not super fast, but it will turn into brown gunk that you keep brushing off (or you let an aquarium air pump keep it turbulent enough for you), and it will etch. You still get a "liquid with copper in it" at the end to dispose of safely but it's not pure danger in a bottle like ferric chloride.
@@AureliusR easy. But need agitation and way of smutt removal.
@@rpavlik1 I've worked with ferric chloride quite a few times and never viewed it as particularly dangerous. Take proper safety precautions, and don't drink it, and all should be well. That being said I might try the vinegar method sometime.
The synchronicity of this is just astounding. Yesterday I was looking for this exact thing and found no real good guides on youtube, until I found this - which you uploaded yesterday. Great video!
Fascinating project as always, did you ever find a suitable material that could handle the ferric chloride solution?
I can always count on this channel to introduce me to a topic or project I have never even considered before and then hold my fascinated attention for 22 minutes. Well done!
Wow, the amount of good information in this video is absolutely huge. But Ben, how are you ever going to start your own succesful etching business if you are telling us all your secrets? ;-)
I guess he already dropped the idea 😂
The video reads a lot like "somebody please make a business out of this so I can buy cheap parts from you rather than make them myself. Look, here, I even made you a business model."
This video was kinda nostalgic to me, this is like a scale up version of the DIY circuit boards I used to make back at my electronics undergrad. Oh the memories of shorted chips due to a bad etching
what is the pumping pressure for the pneumatic pump used in this system? Did you try the technique with only one side exposed? ie etching only from one side?
It is amazing how you optimize a process that is essentially meant to be done a few times. I mean I do a prototype PCBs every now and then but my process is much more manual an tedious then your one time one. You are amazing.
Also It is just a best day to have Applied Science and Ben Eater video in one day!
Thank you very much for your work.
Have you tried to laser print transparencies and then just use acetone vapor to slightly melt the toner and increase the density? I had some limited success with that a few years back, what you absolutely can't do is let the print for too long on that vapor.
This may be the best inadvertent ad for Photofab possible. The sheer complexity, wow 🌷🤷♀️✅
"... the mess is going to be beyond spectacular... Ok lets press the start button!"
you are putting an entirely new spin on the term "Side Hustle!" Great to follow your work. and as always, thanks for sharing your time with us!
Time to brew some coffee.
JOINT420
smkwdevdy
I’ve definitely used paper to back transparencies. I had the same problem. I’m glad you brought it up because I would have had to figure that out again. Yes that was very frustrating.
Absolutely awesome! Unbelievable that you can do it all on your own. For many people in the world America means their own things, for me America is this kind of tinkering, ingenuity and freedom to create and people who brought the joy of looking at this process. Thank you!
Loved it!
And that's probably the most applied science you've shown on the channel
Thanks, Ben for sharing precious knowledge for free for everyone. I haven't had time to go through the comments yet so i might be doubling on someone's, nevertheless: back in the day when i was working in Peltier elements manufacturing we were implementing and using lots of funny techniques. 1) alcohol is pretty darn good for removing water from the substrate, multiple stages of concentrations(40\70\90\100) were employed just to prolong the higher concentrations bath lifetime (the same trick is done in machines to remove water from tissue in histology), and other solvents like acetone, or ethyl acetate can be used with great success. Subsequent hot air in laminar box was used sometimes also. 2) ultrasonic is superior for immersion washing and even sometimes etching, I've built several machines for degreasing of bismuth telluride crystals with solvents\basic wash as well as few for electro(chemical) surface etching all employing medium circulation alongside ultrasonic. It might be a problem for thin substrate sometimes as ultrasonic tends to shatter brittle crystals into dust, but nothing that cant be adjusted.
I've only ever done basic pcb etching. I thought that was wonderful at the time. This is so many orders of magnitude more wonderful. Brilliant hacky engineering.
when I worked in the PCB manufacturing industry we aligned our photo mask prints using a x 10 magnifying glass. Worked fine and was very quick.
Interesting project, of course very similiar to producing your own PCB's
Thank you for uploading. I never do anything particularly useful with the information you provide but you’re always extremely informative and entertaining. I appreciate every video.
this channel is by far the best applied sciences and engineering channel I have seen. each and every video that is made is detailed to near perfection and the presentation of the work leaves you with virtually no questions about it. thank you for giving us quality content throughout the years.
I also needed to do this for a project. I tried using much simpler equipment with limited success. Thank you for a very detailed description of a successful process. I eventually decided to go with stainless steel solder masks from OshPark and Seeed Studios.
Applied Science in five years: "Today I'll show you my home made EUV Exposure Unit for the 3 nm Masks I've developed"
Your work is always so amazing!
😂👏🏻
Wow. The level of off the cuff designs is incredible.
I once used a similar process at work. But, we were galvanically growing the parts we need on a soft circuit board style substrate. The guy who taught me about the process originally used the idea to make massive amounts of custom soft metal gaskets, for a particle accelerator.
Very inspiring. I've made several attempts at chemical etching and not been successful - this inspires me to give it another try.
I've been doing this for years with a laser printer. I just print twice on the same glossy paper and then I transfer the ink to the metal sheet with a hot iron, thus using the ink itself as a resist layer.
Before etching I go over the part looking for any suspicious areas not perfectly covered with ink and just use a permanent marker to "fill in the holes".
I'm getting almost as good (as in almost impossible to tell the difference) as when using UV resist, but it's faster to transfer the design via UV resist for sure, and way easier to clean of the UV resist than the ink, but if in a pinch it works really well. I'm using this for making parts for scale modelling btw.
I've also tinkered with making a two step procedure for partial etching where you need to etch fold lines, or etch a step into the metal sheet for replicating wire mesh etc. I just do the process twice on the same piece. A lot of work I might add...
I've got a similar setup to yours but mine is a bit more manual but it does the job - for very little money!
As always, awesome work. I've been down a similar road making stainless parts for a micro robot, my laminate was custom stainless-kapton-stainless to create flexures and I dip coated with a positive resist as it was exposed with an NC controlled UV laser so that saved time. I used a commercial cleaning product but also cleaned with pumice, your method is much easier! I can't actually remember how I did the drying, I bought parts from a dyson hand dryer but I'm not sure I used it, I also had a contraption made from too cooling fans from a Harrier jump jet (the seller claimed). I did make a laminar flow cabinet so it might have been just drip dry.
It is possible to use magnetically coupled pumps for the spray and this is common in commercial machines, the trick is to buy the right nozzles. An alternative used by a UK company Mega Electronics is to have a vertical tube with holes in it that is rotated quickly by a motor and dipped into the acid source, some plastic RC boat propellers in the bottom of the tube encourage the acid inside and then centrifugal force does the rest. Cheap and simple. For precision etching it is considered good to rotate the panel in front of the etch nozzles to avoid any sort of shadowing affects. This is what I did and I have some videos of my janky belt driven system. Lastly something I read in one of the few books I could find on the subject is that rotating the panel at high speed in a tank can allow for very deep etches as the centrifugal (or lack of centripetal) force clears the dissolved material quickly. This isn't something I have tried but I'd like to some time and it could be a good budget option and rotating it horizontally in a frame would be simple mechanically.
Excellent setup for etching in volume. I used a much simpler setup to do PC board etching in my home shop, then successfully etched brass and copper sheet fine parts using the same setup and chemical etchant.
To get the imaging film on the part under exposure I used a foodsaver vacuum sealer and exposed one side at a time, since foodsaver plastic is only flat on one side. For low volume it was a chep vacuum easel.
Awesome video that breaks down the process and demonstrates some of the quirks of each step.
You know you're my hero? You and stuff made here. If I had the means and time this is what I'd be doing.
this is BRILLIANT!!!!! You are thew most HARDCORE DIY guy in existence!!!! You make the most amazing projects and you never cease blowing my mind!!!
THANK YOU for EVERYTHING you do!!! :D
Thank you Sir,
Well demonstrated.
I used a more crude approach with iron on resist intended for PCBs,
I etched aluminum and stainlessness steel to make dies for metal casting pendants.
Though the relief was small, the result was satisfactory for jewelry.
I appreciate you giving us not only the technical details, but also insight into business and production, innovation and as well encouraging and stimulating our minds.
Adam Smith's invisible hand is ever present.
Take care
This was awesome I used to be a tool and die maker at Tech Etch in Plymouth MA where they made these type of parts. I love that you just made them on your own. Brilliant
Your spray process is one of the fastest I have ever seen. While fast, it may not fit in my apartment. However, hot ferric chloride can be pumped using home aquarium fish tank pumps. The kind with no moving parts, and are air-operated, are the ones of interest. Bubbles do all the work. I've noticed that the acid only needs to flow and etching goes faster.