The fastest way to make crisp PCBs at home!
HTML-код
- Опубликовано: 26 дек 2021
- You can make nice, crisp PCBs on any budget MSLA resin printer - here's how!
Thanks to Siraya Tech for sponsoring this video! Check out their resins on Amazon go.toms3d.org/Siraya or from Siraya Tech directly at siraya.tech/collections/all
Products shown in the video:
Etching kits on Amazon go.toms3d.org/Etch
Photosensitive PCBs on Amazon go.toms3d.org/PhotoPCB
Elegoo Mars 2 Pro on Amazon go.toms3d.org/ElegooMars2ProAMZ
Read the article to this video here:
toms3d.org/2021/12/30/the-fas...
Product links are affiliate links - I may earn a commission on qualifying purchases (at no extra cost to you)
🎥 All my video gear toms3d.org/my-gear
I use Epidemic Sound, sign up for a 30-day free trial here share.epidemicsound.com/MadeWi...
🎧 Check out the Meltzone Podcast (with CNC Kitchen)! / @themeltzone
👐 Enjoying the videos? Support my work on Patreon! / toms3dp 
Наука
![Dimension 20: Never Stop Blowing Up Trailer [Dropout Exclusive Series]](http://i.ytimg.com/vi/xYeXpbHnGEM/mqdefault.jpg)
Dry-film photoresist is the solution you need. It comes in thin blue film that you can iron on copper-cladded board (not the solder mask paste). The exposed photoresist turns darker and hardens while unexposed part dissolves in alkaline solution (sodium carbonate). It's much cheaper than presensitized UV PCB, and can be removed & reapplied if something goes wrong during exposure.
can you explain more? so once ironed and process is done you can solve in sodium carbonate and then run in etcher?
@@maevebaksa yes
Yep, and you can use vinegar plus hydrogen peroxide and table salt to etch which is much friendlier.
@@maevebaksa the dry-film photoresist can be ironed on the copper-clad board with lowest heat setting or using a laminator.
After exposure, develop the board with sodium carbonate solution, use a soft bristle toothbrush to clean the traces.
I use old-school ferric chloride for etching. The etchant is heated on Ender 3 bed and some gcode to agitate the etchant.
The cured photoresist can be clean off with acetone or let the board sit in sodium hydroxide solution. I also found that the cured photoresist peels off in IPA, while uncured photoresist dissolves in IPA.
Best thing about dry-film is if anything goes wrong during exposure you can just remove it and reapply the film. With presensitized PCB you only have one chance.
@@yonggor .
Such a simple concept with awesome results! Most of these machines are capable of displaying their logos during LCD testing so I wonder how you could just get images into it like that vs having to make a negative STL. Would greatly speed up the process, but unlikely to see with chitubox systems :(
Time to open one up and get hacking..I'm sure we could come up with some open-source replacement firmware. Too bad I don't have one or I'd be down to tear it down and experiment.
Gerber files are just a zip file of all the layers as bitmap. You can export them in eagle via the manufacturing tab.
Inability to use 3rd party slicing software is a total deal breaker for some of us getting a printer. Related, One of my favorite 3D modeling programs, POVray, is totally not designed to output meshes, but is not overly difficult for producing a series of slice images that could be processed for an SLA printer accepting an open slicer standard.
@@BloopTube Can you still? I thought chitubox said they were going to change the firmware only accept files signed by the official slicer.
I've seen that in videos but I don't think any of mine do it, they just show a solid rectangle. That rectangle is probable still an image file that could be replaced though.
OMFG.... I've been looking for a way to do this for months and the solution was in front of my face, on my desk this whole time! Thank you so much for this!!!!!!
This takes me back 50 years when I was a teenager making pcbs in the bathroom by exposing boards with a sun tanning lamp. In hindsight I had lots of failures mostly because I was spraying the Kodak liquid resist. The resulting coating thickness was not uniform enough requiring different exposure times across the board, basically clean off the resist and try again. As I recall the Kodak literature recommended spinning the board to produce a uniform film thickness. The copper surface also needed to be insanely clean before attempting resist coating. If the resin printer UV light source is columnated consider spin coating before placing the coated board on the printer. This way the flatness of the board will not be a factor.
Cheers-Peter
I did the same thing. A buddy of mine found an old home sun tanning lamp. Used to make my own and copied a commercial guitar effects pedals. I used tape and overhead plastic to make the masks.
I imagine that for prototyping and small scale fabrication, this is kind of a game changer.
Not yet.
But for some really simple shcemes - yes
Example: Using a resin printer and this technique... I wonder how many little button cells disposable flashlights you could make in an afternoon?
@@StevenIngram a lot, but less than by making them with iron and printed paper, but for prototyping is superior in terms of effort.
I've seen this done before, but I can see someone making a device that would be two LCDs that would sandwich the board to make 2 layer boards easy to make.
@@s16e559 thinking about this, you would need some sort centering clamp.
@@s16e559 yes, a jig. Used frequently in CNC milling for 2+ operations on a part.
CNC mills that do double layered PCBs pretty much always use two alignment pins that go thru the pcb board to make sure both sides perfectly align when you flip the board. The problem with that is, that you have to drill these holes at the right distance with a high enough precision into the raw PCB before you start the exposure process, and you need a way to add some alignment pins to the printer. I guess if your printer has some alignment pins to hold the resin vat in place, you could maybe use them, or build a holder with alignment features out of a resin vat.
@@ellisglyou've already got the precise 3d printer. Just print a frame that screws down with the screws that normally hold the vat and use the frame as x&y stops.
If you're adventurus enough, you can even print the stencil for the solder paste (if you use a slightly flexible resin i.e. tough resin.).
Omg this is brilliant!!! So happy someone tried this and that it actually worked
For real!!
I am ashamed I did not even consider it before now, and surprised no one else published the idea sooner!
So simple. So obvious, yet completely blinded to it until now!
Genius!
@@Inertia888 lots of people have had this idea for years I was to attempt it with another guy 5 years ago already but this UV equipment was not yet widely available.
I used to make my PCBs 30 years ago with this methode. So this is not new or some thing special! U can use a normal UV Lamp. Today, i more like to use a CNC drill to make it, about it save the chemicals. Or if you need realy professional PCB, then look at a company how make it. especial if you use more as 2 layers.
I literally thought about doing this yesterday, what is life
That result is actually pretty nice.
I've been doing prototype PCBs for our company for a long time, but we cleaned out that pcb-prototype-lab a few years ago. It just was no longer really useful compared to a professionally made PCB with plated vias/holes, multilayer, solder mask, micro detail, reliability and quality, surface finish (ENIG) and all the other benefits. Waiting a week for a pcb is really not an issue, if you plan your project accordingly, and doing proper simulation and reviews of a design to have fewer errors or failures is worth more than trying to crank out 3 hardware-iterations in a week before you get a working design.
I guess this technique here is still useful for some, and probably quite cost effective when you compare it to getting a good resolution 2D printer that can print highly opaque films and an UV lamp for pcb exposure with a vacuum pump.
I agree with what you say though I'd suggest that the cheapness and availability of rapid prototyping actually pushes the home stuff even further into least-effort territory. So laser printer-transfer paper or stripboards.
I think I am one of the 'some' that it is useful for as 1. I wait way more than a week average maybe 4 weeks, 2. postage costs are several times more than the cost of the boards 3. if i optimise for single side SMT then no holes 4. hobbiest use so low volume and my time is mine but for the rest I agree.
In addition you might have to wait for the actual components to arrive if you are building something non-basic stuff. The fab can source and assemble them which cuts down the time difference.
I've been wanting something to cheaply prototype boards for a while. Very cool!
There are a few other options to check into as well, like tried and true Laser toner (I've done that and it worked well for me). Also someone modified a laser printer that apply toner directly on boards and then used acetone vapor to fix it. Laser printers can make larger boards. The method shown in this video of using a budget MSLA printer is very cool though!
You can't CHEAPLY do this.....
This is so underrated. Compared to any other DIY process, this is clean and relatively easy. Just WOW!
First saw this method on Huygens Optics channel and was shocked that this wasn't more common! Such a clever use of MSLA printers!
Very good video except I can't subtract the traces from "box"? I say a New Video showing the detailed steps would be in order and much appreciated.
What a nice accomplishment! Thank you for the tip, now the resin 3D printer sales will increase ❤️
One tip on sodium hydroxide: I would generally suggest sourcing food-grade sodium hydroxide/lye rather than lab grade sodium hydroxide.
Food-grade NaOH is pelletized to minimize dust, because breathing in NaOH is very bad (people can debate the need and efficacy of respirators for acrylate fumes, but no one debates that breathing in lye dust is bad). The larger granule pellets make it relatively safe to handle for kitchen uses, like making pretzels because the bigger pellet size makes it much harder to dissolve if you accidentally touch it (lye isn't caustic until it is mixed with water) and has minimal caustic dust if you handle it gently.
Lab-grade NaOH might come in a different granule sizes which might not be apparent on the labelling and may be potentially more hazardous to handle, especially with respect to eye and breathing protection.
The NaOH shown in the video has large grains like food-grade NaOH. No need to be super scared of this stuff, as it is commonly used in cooking for things like pretzels or noodles. Just exercise some caution.
This is very useful information, thanks for sharing!
Would you agree that neutralizing NaOH with Acetic Acid CH3CCOH is not the best Idea?
you don't need to neutralize lye, you just pour it down the drain. It's literally drain cleaner.
@@tommihommi1 In the video Tom tells that he have Acetic acid on hand for emergencies with the NaOH and I think that’s not the best idea but I am not 100% shure about that.
@@Druidus98 I'd suggest having baking soda nearby to neutralize any spills. While lye is used as a drain cleaner, you should not just pour concentrated lye down the drain unless you need to clean the drain. If you don't need to clean the drain, just dilute the lye with a lot of water.
FWIW I'm a chemist.
Edit: I buy baking soda in bulk and keep a bag of it in my lab.
I never saw the need for a resin printer personally, but this is awesome
I feel like I've just unlocked a new skill! Very smart idea. A lot easier than printing on acetate and using a UV lamp!
I strongly recommend using ground planes on your PCB designs. It's way easier to design a board with a ground plane. It also means you remove less copper, which in turn makes the chemical last longer.
It’s harder to solder without the mask if you have ground plains. This is for prototyping and fast iteration.
@@xferme Every board I've made has a ground plane, and I didn't have any problems soldering to any of them without a solder mask. Maybe it's not the same case for SMD components; I've only ventured as far as THT, so far.
If you're doing it professionally then 4 layer with 2x signal and ground+power planes is what I'd suggest as a minimum for anything that isn't completely basic, which obviously is not possible using this method. Not just for making routing easier but for EMC compliance it's much easier as well. You still have to know what you're doing though, but not as much. When I was first playing around with 2 layer opamp boards I remember introducing error where the high gain opamps amplified the poor return path impedance where I was weaving top-bottom layers. So for serious prototyping, no I would never do this. It's just a hobby thing.
Interesting approach. The results were quite good.
This is a great idea! I've done many PCBs with the toner transfer method, but it does struggle a bit on smaller features. One suggestion - when using a small etch batch, you get a faster and better etch by removing less copper. Just use flood fills on most/all of the board. You can also make your fill a ground plane, which can help simplify routing and improve signal integrity or thermal performance in some cases as well.
I saw the thumbnail and the title and it hit me... that is brilliant. I always used toner transfer method which isn't very precise and no matter how often you do it, always remains a hit or miss.
3D print a drill guide as well so the holes will go easier. Great video!
The divot in the copper acts as a pretty good drill guide
I was going to go through the process of ordering PCB's for a thing at work that I wanted to play with, but this is much more in line with trying stuff out, perfectly timed video!
In the past, when I was still dealing with single-sided PCBs, this would have been a great technique. I then used sheets of overhead paper and a good laser printer to make transparencies and lay them over my PCB boards so that I could expose them.
Meanwhile, with the current CPUs that we use, even double-sided PCBs are often not good enough anymore, and I use multilayer PCBs. For this, you cannot go with hobby equipment (unfortunately) and you have to go to PCB factories that can usually make a PCB within 2 days.
But for anyone who needs a single-sided print, this is a great solution. Thank you for showing.
If you're able to line them up perfectly, you could do double-sided boards as well. But unless you want to go through the process of plating through-holes, you'll have to settle for making vias by soldering a short section of wire between the two sides, and either use rivets for the actual through-holes, or add the aforementioned vias next to them so it doesn't matter which side the components are soldered to.
3dprint a jig , and you can do doublesided with no alignment problems. multilayer is ofcourse out of reach at home.
@@chriskwakernaat2328 if you're printing a jig and have standard sized boards, you can easily layer boards, use thinner single sided boards, with a double sided base board and drill alignment holes, and you can easily stack them with a layer of epoxy, though you need to drill bigger holes to solder the vias, which would be the main pain point.
If you don't mind a thicker board you can use rivets for the through holes, and 3d print a spacer to hold the rivet heads.
I would love to see a follow-up video where you used a jig or fixture to align the PCB as well as made a 2 sided board.
That was part of the original plan, but I ran out of time for testing it.
Did this half a year ago, it's so easy and quick. also works for etching logos on different other metals or even mobile phone back covers
This is one of those "why didn't I think of that" revelations. Great job!
Many thanks for this. As an old hat at doing the stencil PCB Maker style boards it's refreshing to find a new way that removes some of the steps
This may be the coolest and most practical RUclips video I have ever seen.
Thanks for this! I had the same idea, but had no idea what exposure time should be. I'll still do some tests to dial in the perfect time for my specific printer and film, but this gave me a much-needed point of reference to start from!
I made my firsts PCBs 30 years ago drawing them with a sharpee whishing for something better. The time have come. This video is a great out the the box idea that will keep me thinking for some time. GREAT, GREAT, GREAT!!!!
Beautiful and simple. Looking forward to trying this myself at some point.
You could use photoresist film and apply it to the plane copper. Would be a great way for model makers to make brass etching for models or dioramas.
We used photo resist spray back when I was in the school electronics club. It was kind of an acquired art to get it even though.
How do we remove air bubbles when we apply dry film to copper clad ?
That's brilliant. Seems to provide nicer results than other transfer methods.
Happy holidays, Thomas
I love your ideas Tom. That why I started to watch your channel way way back long time ago.
Today is one of the days that I thought, thank God that I subbed to your channel! This is just brilliant!!
Had an idea about doing this with my original Anycubic resin printer some years back and never quite got round to testing it. Thanks for putting the hard work in, and getting this out there :) Its evidently way easier & more reliable than the pain of Toner transfer from laser printed gloss paper ;)
Jkdekw
Thanks for this! this is really brilliant, another way to help when drilling the holes is creating a jig using the negative plate as a guide
This brings back memories of university and using a Lazer printer and trasparencies for making masks and a UV tube with a glass bed to expose them. I remember fixing mistakes with a black marker on the transparency or a sharp knife to scratch off the printing. This is a cool method.
Oh man, that's a great implementation of a resin printer. I wish I'd thought of that for one of my clients.
Wow! Never think about this way of use lcd printer. Thank you, this is very useful!
Nice! Very promising. So much better than what I had to do in high school in the 80's.
This is absolutely phenomenal.
This is so cool! I haven't made any PCBs before as it seemed to me like making the masks and the chemical process seemed quite complicated, but this method with pre-coated photoresist boards looks like a really great way to do it; I'm looking forward to trying it out! Thanks for sharing this Tom! 💜💙
This is amazing! This is something ive wanted for so long
Aw man, that little wheelchair figure is awesome! As a chair user myself I geeked out a little.. Wish I had a 3D printer to try this out with. Awesome video!
FINALLY! I told people 2 years ago to use their resin printer with photoresist film. Great demo!
I've made small silkscreen masks this way too! You can make the circuitboard and some artwork for the enclosure onthe same machine. That in itself is a boon for any small maker/tinker.
I've been expecting someone to put a Chitu board in a flatbed scanner body tbh.
What a great video. I love your experiments
I'm thrilled to see this! Thank you for this Thomas!
Interesting technique. Last time I etched a board at home, I used the toner transfer method and etched in a muratic acid/hydrogen peroxide solution. It worked pretty well. Used the same solution to etch some brass sheets to make custom playing cards for a steampunk Gambit cosplay too.
Final video of the year sounds like a big one!
No PCB from JLCPCB was harmed during the making of this video!
I'm glad that somebody did this! I think that is one of the best methods to do pcb etching.
I'd like to build a device that does the same thing, without the 3d printer stuff.
Thank you for this very great idea. Why didn’t I think of this before, it is so obvious
This is a nice idea. I use positive 20 because it's cheaper than than pre-coated boards (just needs a bit of spray-painting skill).
To make big boards you would need a big resin printer, though.
But it's real quick. I currently start with a PDF of my tracks that I have printed on tracing paper. Then I sandwich the paper between the coated PCB and a thick piece of perspex. That goes in the sun (after a couple of test exposures to get the time right as sunlight is always changing). I get pretty good results. Suitable for prototyping.
this was a genius idea. I love it. Thanks.
This plus a small CNC mill and you could make some great detailed DIY pcbs.
That is a quite genius method. Getting good quality films is always a bit of a hassle, even when using a laser printer. Think I finally found an excuse to buy a resin printer.
Brilliant Thomas. I have the Mars 2 Pro, and an older i3 Wanhao clone that I also setup with multiple toolheads (FDM, CNC Drill to 1.5mm, CNC Engrave - sharpened 8mm stainless rod drag style). I have a bag of single sided 0.8mm PCB blanks. Was going to try a simple project like making a multi-layered Arduino or something to prove the 3D printers can do multilayer PCB's. I think the key will be layer alignment, and this can be done with first using the dremel and those small drill bits like 0.6mm. FlatCAM was made for making PCB's, has many options like drill depths for CNC drilling gcode a 3D printer with Marlin can understand.
Brilliant in it's simplicity - got to give it a go and make some etched metal parts for my models...
That's very interesting! To what level of detail could you get? Do you think this process is applicable for small SMD parts, i.e. 0402 and smaller?
I have no need for making PCBs but I did found it interesting. It's great to find other uses for stuff you already have.
I'm still fiddling about in FDM land but I've been wondering about this ever since a few friends got the Mars 1 & none of them have been willing to try it yet. Nice to know it actually does work! I figured there'd be no real reason why it wouldn't.
Such a great idea.
Thanks Tom
This is great, going to give it a go, just bought the Saturn
Top notch info Tom Thanks for making this video ! stay safe !
Always add "copper pour" to your boards that you are making with this method because it dramatically reduces the amount of copper needed to remove during etching (and with it the amount of chemicals used and fumes released). It is very easy to do, and can also serve as a ground plane so it makes board design easier.
But if you get industrial opto coupling interface large copper pour areas on the board is danger for the insulation especially if your project contains igbt or scr connected at power sources fairly different voltages of the logics supply .
@@satipsrl5207 Those boards should be designed as two separate boards and you can add copper pour only on one side or on both, but with different amount of clearance. There should also be a zone that goes under optocouplers that has no copper at all, and also it is common to put cuts in the PCB along that zone
Wow, fantastic idea! Even an entry-level resin printer would be adequate for a lot of hobby-level boards, and a mid-sized one would be plenty big enough for most anything a typical maker would need to do on a quick-turn basis. (I also second the tip below about using iron-on photoresist film.)
Another thought: A 3D-printed jig could help you align the boards accurately enough to do double-sides ones repeatably.
I'm waiting for a new preordered FDM printer to arrive and this has me looking at resin printers for this purpose. Could come in really handy to finish a car lighting project. I've had a couple orders of boards so far, but I'm going to need a few more one off boards to finish it up.
Nice idea that has real possibilities, even for double-sided boards. Great for those of us that are at the end of the world (NZ :) ) and waiting for JLCPCB / PcbWay takes over a week at the best of times.
OMG! the usual thing that has been done for 30 years done with UV lamps! INCREDIBLE
Thanks a million. This answers my questions about all I wanted to try. You just saved me a bunch of money! THANKSKSKKSKSKSKK!!!
Thanx for this very useful trick : for iterative design, that's can be incredible !!
For smt prototyping of single sided boards that looks good👍🏻 And great to have a multipurpose machine!
This is so good! Thanks so much for putting this out! A F360 question, @10:55, how did you do the subtract? I extruded the rectangle and then one by one my copper pieces (so a lot), extruded those and then did a number of subtracts and it worked but Im hoping someone can point me a tutorial or how they do that without so much fiddling. I couldnt/cant figure out how to group the copper into one piece to be the cutting tool...???
Really great project!
just love your content and how you bring always new methods, richtig innovativ!! LG
If I’m not mistaken, I think you can also use UVTools to import/replace a layer with an image. So potentially you could turn the tracks into an image with the same pixel dimensions of the printers LCD and drop that into an existing sliced file (and tweak the exposure times)?
That would also allow processing with graphics tools to adjust trace width, which might help in some cases.
If you want to try the 3d goo as a resist you could try diluting and thinning with a compatible solvent and spin coating it on a plain pcb. It looks like it's a 10x - 100x too thick. Also if the lye is too finicky with timing sensitivity maybe try sodium metasilicate pentahydrate it may be more controllable. Also pretty sure you can get positive or negative uv resists to avoid a step for the image.
I stopped trying to make my own PCBs at home a while back but this was interesting anyway. It makes me wonder if I could try again some day.
really usefull, specially when having to do smaller pcbs.
You got me super excited because I have a ton of negative resist boards. Where UV light hardens the resist, your method would be perfect as I no longer have to make exposure masks. I like negative resists because I can use sodium carbonate to develop the boards a less caustic solution.
Very interesting. I will have to try it sometime. Thank you for the video.
We need a series please!!!
I don't do 3D printing... but you've got me interested in the idea of making board masks using an LCD... that sounds really interesting.
This is very clever, especially the resin on copper pcb. I bet someone comes out with some safe purpose built resins for diy pcb’ing soon.
Very interesting. The results are impressive. Better than laser printing the traces on foil and using a SON-T lamp or the ironing method. Thanks a lot!
Awesome video, would love to see more homemade PCBs!
really nice, always wanted to try this. Maybe a good use for my now obsolete photon
Interesting idea.... I love this. I wonder if you could use one of the old LCD panels that they made for overhead projectors and mount a UV light under it. It would greatly increase your PCB size, If it was capable of blocking enough UV.
any LCD could work, just need to pull the backlight off of the panel. assuming it works at all....
This is so cool! This almost make me wanna buy a cheapish MSLA printer :D
It’s so cool 😎 Your new idea always surprise me! 👍🏻
So obvious yet so genius. I've mostly gone to getting them made and just waiting now and was using the laser printing before since the photo etching is a bit of a PITA but good to know about. I wonder how this would do for making screens for screen-printing.
This is incredible.
i m doing the same since a while on my Photon Mono X, i find it difficult sometimes to not overetch very thin traces. so 0.1 or 0.2 is difficult on bigger PCBs, there will always some interrupted traces. But if it works it works
Very nice. I have a laser engraver and I did some PCBs by painting it with permanent marker and laser the tracks away. then acid and it's done! Like what you were trying to do with the resin directly on the PCB. Also works very well and gets very fine results... depending on the laser's resolution. But 90 seconds for a "big" board is faster than the 4 minutes of my highest resolution small board... but anyway, is also fast that order it from China!
The best solution, in my case was to "laser" the circuit first to have a reference for the holes, drill the holes, clean the paint and redo the laser without holes in the drawing, to make sure tiny misalignments in the drill press get the copper to the hole, and only then, acid the extra copper away! Also, with the holes already done, it's very easy to alight a two layer design.
May I recommend Sodium Metasilicate over Sodium Hydroxide for PCB work. It’s far safer to work with and more controllable, albeit a smidge slower.
You beat me to trying this ha! Glad they turned out so well! I'm wondering how small of trace width you can push it.
This is quite the revelation. I can already think of about 3 boards I can make with this.
Was thinking about this long time ago. Would love to see more attempts on the wet method.
This is a really slick idea. I see lots of comments on toner transfer being better. For large or medium boards, if you already have a laser printer, that's probably true. But for little tiny boards, for small surface mount projects? This looks great. I'm betting the resolution of this direct process is high, and it would work great for 2 or 3 cm boards, that would just be cumbersome toner transfer. Thanks!
Back in the 90s we were taught how to make PCBs at school.
You draw the traces on the computer (Acorn Archimedes!) And then print them on the laser printer.
You put this and the special photo resist PCB blank into the tiny little UV PCB sunbed and expose.
Then you put the PCB blank into the etch bath with its hot vibrating acid.
Wash it off, buff it with wire wool and drill the pads with the 1mm bit in the drill press.
Good Times.