The EPROM eraser uses shortwave UV light that is almost completely blocked by ordinary glass. That's why the EPROMs have quartz windows. A blacklight style longwave UV lamp may work better. Have you tried acetone nail varnish remover to clean the residual glue off?
+CNLohr maybe the apliedscience's evaporative method would work? It's fast so I'm thinking maybe it would not have the time to get under the copper traces.
I really need to make more glass PCBs some time. There's so many things I want to try and new ideas so many people have had.. Right now, I'm so busy with other stuff.
I tried using an EPROM eraser to cheat making photo-resist pcb's in the 1980's - It took forever but eventually worked. The problem is that the wavelength of the UV light is wrong. It's like the difference between UV-A and UV-B.
Dude this is fucking awesome, I am a mechanical engineering student but I'm learning so much about electronics manufacturing from people like you and other open-source projects. Thanks a lot!
8 лет назад+1
+H5ck Well, I am a student of electrical engineering and I learnt everything from RUclips videos and Wikipedia. Thanks, school!
If you're doing regular circuit boards, the laminator (note this is a modified GBC H-220) is spectacular. You can usually get everything to stick really well in 2-3 times through. No ironing needed.
Your etching solution is actually etching the glass giving it a frosted look. If you use pyrex or borosilicate glass slides you will not have this problem. You will have totally clear boards! The glass you are using must be a "soft" glass which has a COE (Coefficient of Expansion) of around 130, Pyrex or Boro glass has a COE of around 30. The low COE also makes it far less susceptible to thermal shock. Clear and no cracks! Win! Also this is one of the coolest things I've ever seen!
This is so incredibly cool! I can think of so many "just for looks" applications for this to make some really cool, and actually functional, stuff. Thank you for sharing this.
Wow, this is really ambitious and a very cool idea. We have expensive Liquid Flow Meters that just use a plate of glass over some heat detecting LEDs to act as simple push buttons but yet have the circuit isolated from the externals. This...could be very handy...great job by the way!
Very good work I must say.one thing I see a lot is solder iron so big for the smallest work.And all so in a right mess .i have 7-soldering irons all different and shiny clean.one iv just bought is 12-volt and I Added a tiny micro switch so when I hold it I just press with my finger and all added a white bright Led the Led stays on all the time when I'm using this micro soldering iron.your work is good and would be easier for you to use a small soldering iron for when you re-solder some of you micro work.great video keep up the good work
I dont do this stuff, but am interested in learning how. So for what it's worth from an inexperienced observer, perhaps using a hot plate or inverted iron to heat the glass evenly when melting your solder paste would limit the possibility of cracking it. place the slides in a thin frame (thumb tacks point up through cardboard) to prevent them from sliding around as much when applying the poly. Split a small dowel in half and pinch steel wool between the halves. Insert one end into a drill.
Neat method for single sided surface mount designs but a few tips: 1. As Java Lu (and maybe others) said, TRIPOD - PLEASE!! 2. Tin board with tinning solution before assembly - gives superior soldering 3. Use several coats of auto-spray lacquer on final assy for vastly superior finish. 4. When spraying remove all connectors & protect the header pins with sleeving or low tack masking tape for larger blocks 5. CLEAN your soldering iron tip PLEASE & use a much finer tip for such work
Have you ever tried selectively gluing the copper on? I'm thinking: -UV glue the copper sheet to the slide -tape a negative of your circuit on the non-copper side -Expose the glue through the negative (so that only the glue under where the traces are meant to be gets exposed) -Peel off the reset of the copper
@CNLohr I think orgameron means leaving the copper traces in place while brushing a solvent over the areas where you don't want/need the UV glue and just using the actual traces to protect the glue like the toner protects the copper during etching. It seems like that's more or less what you're doing with the polyurethane though as it cuts through the UV glue like a solvent but is also acting as a new method to adhere the traces at the same time.
I've had the notion to try and make some PCBs out of unusual material like glass for a little while now, but I didn't really know where to begin (I've done the toner transfer method, but I had no idea if I could even buy copper sheets), thanks for the inspiration!
I doubt the slide has a UV coating - it's just that glass is not very transparent at the short wavelength used in UV erasers (253nm) Eprom windows and eraser tubes are made of quartz for this reason. It may also be the case that the glue is not very res[ponsive at that wavelength. You will probably do better with a longer-wave UV source like an insect-killer tube (365nm).
There have been a lot of suggestions for this. I've switch to a spray polyurethane, but a number of products would probably work. I really don't have the time or resources to try them all! I recommend making some glass PCBs yourself and trying it out.
This is an awesome idea. I think I'm going to try something like this, and see if I can pull off a through hole board. It's got to be possible if you drill real carefully, and have lots of extras on hand in the event of a slip of the hand.
Schematic is in ExpressPCB. Full version of eagle is not free. I don't use anything that's for pay unless I absolutely cannot avoid it. It's simply discharges the circuit, then waits for it to charge via 1M-ohm resistor. Depending on how long it takes to charge, that's how it knows how much capacitance you are adding.
@CNLohr basically the same thing you do to eat away the copper but for the UV glue you have. Use a solvent instead of ferric chloride(not sure if that's supposed to be one word or not) and use the etched pcb traces instead of toner. That way the glue is removed from the glass and the traces protect the glue under it. Like I said though you're already doing something similar with the polyurethane but with the added benefit of replacing the glue with something else more permanent.
I do not agree that you don't need PTH. In most of the cases you can find alternatives, but as soon as you need a D-Sub mounted on the PCB, you're in trouble, because even the SMT version need holes. But one thing you should consider. Since you already have the UV eraser, why not use it as an exposure box ? Personally I've used the toner transfer method a number of times, but I prefer using the UV exposure method when I can, as it's more precise and gives less errors/problems.
That was a nice spark when you plugged in your programming board ~18:10. Your last resistor looks like it had a short under it. From our view, the top one to our right of the IC chip. The first time you place the copper on the glass slide, there is a better way to prevent bubbles to begin with so you don't need to squeeze them out. Basic bio-lab slide making 101.
Hey there, I dig the video man! I'm a Artist/Glassblower/Musician I'm working on a cigar box guitar amplifier at the moment I was wondering if you had ever tried making PCBs on round/ed glass or other shapes for that matter? I would love to do a collaboration circuit or something........? I'm just barely getting into the diy pcb/circuit making thing so I've be watching a lot of vids on it and so far you are the best I've seen, you seem to have high standards when it comes to high quality pcb crafting the video is a little rough around the edges but whatever I have a few vids that purdy rough myself! lol! Anyhoo I've been blowing glass for just about 13 years now so I got the glass down and you seem to be quite versed with your electronics/pcbs/circuits etc, I'm sure we could come up with some pretty amazing stuff. Anyway if your interested hit me back and let me know what you think,Ideas?,glass shapes/designs/colors? Maybe send you some glass to play with? Just a thought.Cool vid! ttysh Zac ;-)
I have made a PVD and i made a completely clear and very thin PCB with that some years ago, looked pretty cool. Have also made copper, silver and aluminum mirrors that looked awesome.
Thanks for the fast reply. I have ordered some copper that is normally used for guitar pick-up shielding to experiment with. I have the exact same printer in this video to use I can test with and see what I can get it to do :). If I find a reliable method I will let you know.
Nice job. If you want to take some advice, here's one. Learn to use knives for the precision cuts, they are better if you master them. If you wear gloves and use some bath to degrease the component before the copper plating there will be no risk of contamination so the PCB will last much longer
I'm laying one up with clear epoxy resin at this very moment. I'll let you know how it works out, I think it'll be a little less 'touchy' than the UV glue as it's what is used in conventional copper clad laminates
Hi, I've done quite a lot of thru hole PCBs and I saw this video and was amazed, I've been trying to Fe into smd/smt pcb fabrication, and I was wondering what kind of solder paste are you using? Do you have another video with in-depth description on smd soldering? That'd be awesome! Thanks a lot and keep up the awesome work!
I have not tried it, but I would imagine yes. I recently found out that UV lamp is not ideal for curing the glue because the wavelength is wrong. You would have to try.
If making a clear circuit was done in a mass scale we would have a clear phone not entirely clear but the first company to build it would the first step for the other companies. It would change the entire world.
Consider using an optical grade UV curing glue and gluing the shiny (polished) side of the copper foil to the glass slide. This should minimize the translucency you see after etching and the traces will look shiny through the glass.
btw: what about using a transparent casting epoxy like those used for embedding carbon fiber or fibreglass you could just pour a thin layer of epoxy directly on the copper - so you won't need any glass at all, get seamlessly bond material (transparent of course) and you don't have to worry about thermal issues while soldering technically it would be an FR4 or 5 PCB without the strengthening fibreglass sheets
Also a syringe works well for depositing solder paste in tight areas, and automotive clear coat finish is highly resilient when it comes time to use solvents such as acetone, and is somewhat flexible until cured.
I use Liquid Tin quote often on most of my projects. The boards look 'cooler' with bare copper rather than tinned. I have still never successfully done a dual layer (not that I haven't tried). I would imagine the static properties are almost identical to regular FR4. I've been using glass circuit boards for a number of touch projects in my house which is carpeted, and even noticed some serious static, but haven't had a failure yet.
Not sure if this would help at all, but for finer wood building applications, marine spar varnish works better (ie. more clear and less translucent) than polyurethane.
Compressive forces on the surface of the pane make it much harder to break. However, high tensile strengths in the glass pane make it shatter if cut, chipped or even sometimes scratched. This is why car windows, for example, will shatter into small, barely sharp parts instead of large, razor-sharp pieces.
It works on multiple substances, including film, however, you would not be able to toaster-oven film. You can do this on kapton, and it does toaster oven. I have never tried to do two layer.
Yes. I got some glass that was angel gilded, chemically plated on. I attempted to electroplate it further, but could not get it thick enough to electroplate effectively... The areas that I did were lumpy and did not go on even. Keep in mind, you need about 1 mil of metal to make a good circuit board.
excess solder can also be removed from bridged terminals on a surface mount components by simply stripping a piece of stranded copper wire using it like a sponge. i also prefer to take a grinder and chuck my tip in a drill to grind it to an extra fine point, then use liquid tin to prime the tip. Moving fast with a solder iron and solder paste you can avoid heating the glass by not exposing the glass to long periods of heat like you would using a messy soldering tip..
This is totally awesome. I've been wanting to make my own pcb for some radio stuff I've been doing ugly style. May as well go big for the first one so if I fail I fail big. A question, how do you think this would work for larger PCBs? It looks like maybe using some UV glass like the glass used in picture framing would work like your slide? If that makes sense...
No ditching the connector at my place, then it's better not to have a glass PCB. You'd also get in trouble if you needed a BGA100 (or larger) chip mounted - they require VIAs. Surely you can bend the legs and mount a LM3914 component if you need those. I don't think the solution is to ditch a component, but instead finding a high-speed drill that drills well through glass (diamond drill would do), and perhaps use a very strong glass. Anyway, you could try mounting some LEDs on your mug now. ;)
Have you tried removing the UV glue in small sections, then painting on a thin layer of thin acrylic? if you do it in small sections, you'll probably not be prone to large amounts of slippage, and will be able to replace the UV glue with something more aesthetically pleasing.... perhaps even embed the circuit in acrylic, or a 2-part clear setting polymer. depending on how you are able to make one-off molding, you could possible embed the entire device in such a material, leaving the connection pins exposed for programming / testing etc.
+ArcAiN6 Regarding pulling the traces off... that sounds much more manual than what I can reasonably do. But, regarding potting stuff, I agree that would likely be cool, but I rework my boards soooo much that it would make me feel bad not to be able to work on my PCBs.
I believe the quantity is somewhere in the order of one part per 30,000 dilution. If you pour the FeCl down the drain, bad things happen, if it's just a few drops, and you run plenty of water, it's not that bad.
Great presentation! Glass PCBs are probably better for building electrometers and such (low surface resistance?). May I suggest you try HCl / H2O2 for etching ? A lot cleaner and controllable, and you can get rid of that nasty FeCl3 forever!.
That's really cool - I tried angel gilding and electroplating it thicker and a few other things, but nothing could stand up to what I was after like this.
Yes, it would. But unless you are considering a Class D amplifier I'm afraid the conventional A and B designs may create unwanted and isolated hotspots in the circuit. If the amplifier is enclosed, a fan would be used as forced cooling. Glass is not a good conductor of heat and when heat is applied at one end while cooling is applied at the other, even tempered glass might product stress fractures.
You put the carborundum paste onto the glass, chuck copper then spin it in the carborundum. some grit gets trapped by copper and a hole is drilled. Usually it is how larger holes are drilled but I don't see why smaller ones wouldn't work. Drilling glass is slow going though. I can't imagine drilling a whole circuit out in glass.
I use the attiny44. You don't need any special IC, just a 1 megaohm resistor and an GPIO. You can do this on most of their processors, not things like the attiny2313 though. Also, you don't need qtouch.
The reason the dried glue is matte is because it has picked up the texture of the copper. The polyurethane fills in the grooves but is a slightly different refractive index. Have you tried coating with more UV glue or gluing on the shiny side of the copper?
I've discovered that the UV glue is tacky because it is very sensitive to moisture while it cures. If you coat your boards with UV Glue then place a sheet of clear plastic on top that will peel off later after it is cured. I tried a polythene bag and this worked well. :-)
After I am done, I typically dry and warm my boards, the UV glue is quite hard. Even with copper that's shiny on both sides, the texture is still quite pronounced. I am slowly transitioning away from the UV glue, because I am trying to use Riston for my resist mask. The same chemicals that dissolve the Riston dissolve the glue.
It's an interesting idea. I have often thought about using an acrylic, but that would have problems at solder temperatures... I was always under the impression epoxies without a strengthener were very brittle, though.
To those asking about thru-holes, why not just have a ribbon cable on one edge that goes front to back? You're still going to be restricted to SMD, but if you could just get say 10 or 20 around that would be all that most projects would ever need.
This is awesome. I'd love to see a controller like that for say, changing the radio or some sort of game input device. Thanks for showing us how you did it.
That seems to be what most people are using. I've read that they've changed the design recently (see the PCB "Fab in a box" page) for the worse. I was able to find this one on amazon that I'm going to try out called the Apache AL13P. Hopefully it won't need modification.
I may try this... I don't have a laminator or uv oven though. It's been years and years since I fooled with circuit design and/or breadboarding. I've never made pcb's. My line of work is BGA reball/reflow and general computer repair. Love anything to do with electronics!
Absolute legend! I was just thinking about this the other day, love the fact you can use it capacitive touch! Fair play to you! Currently working on a few things.. Maybe hit me up for a chat some time? Cheers!
I have tried, with many different epoxies, glues, etc. to no avail. Also, on glass, you can't use a clothes iron, but on regular PCBs, it's not that bad.
perhaps I missed it, but why do you want to use glass? Granted, it's quite neat. But do you have a specific need for it? By the way, you could use bismuth solder, which melts at ~140C to reduce breakage.
The glue is not conductive, while the glue is translucent, you'd still need a material that can conduct. I do use this process for touch sensors, I put designs into the copper.
When using the matte back copper, it doesn't really rub off.... ever it seems. But, I've been trying some shiny-backed copper and that is a major concern when removing the toner. Once the polyurethane is on, it's all on there for good. There's a lot of equipment, and not much space, I'll admit.
As it turns out it appears the wavelength used in the eeprom eraser is not optimal. The glass is transparent to UV, but you need to use a source like a mercury hallide grow lamp to get the right wavelengths.
Nice! I can see this working well using visible LEDs UV glued emitter down to the board. Just a thought, but would this work if you used copper leaf from an arts and crafts shop? Its a lot thinner but you could maybe overplate it with copper to the correct thickness?
Some safety notes: 1. When working with UV eraser, please don't open the drawer when the light is on. The UV-C wavelength used is 253.7nm; can cause skin/eye cancer or cataracts; 2. When working with chemicals, such as FeCl, use eye protection or a face mask rated for use with corrosive or caustic chemicals; 3. When using volatile chemicals or chemicals producing gaseous emissions, please work with a carbon based respirator in a well ventilated room, or in a fume cupboard using forced convection
I found that tweezers for medical use work very well: precise, large variety of forms, stainless steel (anti magnetic) and can be cheaply bought on Amazon.
Perfect thanks for the additional documentation! How's the video going regarding the SMD soldering? did you have any chance to edit the video? sorry I just got my chips (from china) so I'm ready to learn! lol!
I have this listed under my "to buy" section. I doubt the ferric chloride will be able to eat through the glue, but it may be possible to silkscreen the stuff.
Thank you both. With respect to "1pcfred", I do not know what you mean by drilling with copper. Do you mean a copper-coated drillbit of some sort? I looked up the carborundum paste and I'm with you there, but you lost me at the first statement. Thanks for your time.
Awesome video! Two things though: First, you should definitely invest in a tripod, I feel like you could seriously benefit from it. Second, the tip of your soldering iron is nasty, you should seriously start to tin/clean it. Other than that though, great video!
CNLohr i hope one of this lot of thinhs is your soldering iron XD anyway, loved your video man, but this looks like a real pain to do because of the risk of breaking the glass... do you have or will do any tutorial on how to do normal diy circuit boards?
I actually use Riston now instead of toner transfer and I use a grow lamp to cure the glue. BUT! Your problem isn't that you need two sides, it's that you are using D-sub connectors. Ditch the connector.
Thanks for posting this video, all aspects a great info, especially the bit around 16m where you reveal the SMT process, we've been wondering how this worked and your video has clearly answered this. I wonder if you'd care to respond with key process points about this method? Also, I note you've not applied any tags to the video which I suspect is why I haven't come across it previously when searching for "SMT" or similar. I found this video via the sidebar suggestions for PCB production...
Awesome! I completely overlooked that file, thanks again. I was starting to pull my hair out because express pc doesn't like to run under my setup in crossover.
I made something similar, but first step was to vapor-deposite copper on glass, then a bit of galvanic process to enhance layer, then mask + etching and OALA circuit on glass was done :)
Excellent video! I really enjoyed your various comments throughout the process on why you should or shouldn't do certain things, relating to the UV glue and the toner, etc. Do you think that you after gluing the copper you could coat it in a photoresist solution, and then transfer the circuit using UV exposure?
That is 100% dependent on the size of what you want. Also, I'm doing ones with very shiny undersides now, too, not just the matte brown. I wouldn't charge extra for them. I don't know... a 1"x3" probably $15... A 5"x5" (the biggest I can reliably do at present) probably $60? I can make the board decent outlines with glass cutting. This is a first for selling.. I've printed hundreds but they just get tested heavily.
Yeah, I have a 5N now. I do tin boards, many times, but it ruins the look of the board. And, I haven't had any trouble with solder-ability when I use paste. I actually have switch to a spray polyurethane now, too, doing just that. And, I totally am good with that hulk tip.
Ni ! Very nice work, you make a careful job. Did you try to print directly on the copper sheet, before gluing it ? If it works it will save you the transfert step.
It seems like the UV glue is a weak point in the whole thing - Why did you pick that, exactly? Have you tried anything else? Excellent vid, by the way. Looks beautiful. What are you making? I found my way here because I'm thinking through ideas for something coffee-table sized, so your vid isn't *too* helpful, but it's fascinating to watch, and you're way way way ahead of me technically...
The EPROM eraser uses shortwave UV light that is almost completely blocked by ordinary glass. That's why the EPROMs have quartz windows. A blacklight style longwave UV lamp may work better. Have you tried acetone nail varnish remover to clean the residual glue off?
bigclivedotcom Acetone causes the traces to slup off. I think it gets between the traces and the back of the PCB.
+CNLohr maybe the apliedscience's evaporative method would work? It's fast so I'm thinking maybe it would not have the time to get under the copper traces.
I really need to make more glass PCBs some time. There's so many things I want to try and new ideas so many people have had.. Right now, I'm so busy with other stuff.
I tried using an EPROM eraser to cheat making photo-resist pcb's in the 1980's - It took forever but eventually worked. The problem is that the wavelength of the UV light is wrong. It's like the difference between UV-A and UV-B.
Dude this is fucking awesome, I am a mechanical engineering student but I'm learning so much about electronics manufacturing from people like you and other open-source projects. Thanks a lot!
+H5ck Well, I am a student of electrical engineering and I learnt everything from RUclips videos and Wikipedia. Thanks, school!
If you're doing regular circuit boards, the laminator (note this is a modified GBC H-220) is spectacular. You can usually get everything to stick really well in 2-3 times through. No ironing needed.
Wow. This is the first RUclips user I've ever encountered who actually knows the proper use of RUclips annotations. Well done sir!
Your etching solution is actually etching the glass giving it a frosted look. If you use pyrex or borosilicate glass slides you will not have this problem. You will have totally clear boards! The glass you are using must be a "soft" glass which has a COE (Coefficient of Expansion) of around 130, Pyrex or Boro glass has a COE of around 30. The low COE also makes it far less susceptible to thermal shock. Clear and no cracks! Win! Also this is one of the coolest things I've ever seen!
It's so cool the way you can see the etching happening through the glass.
This is so incredibly cool! I can think of so many "just for looks" applications for this to make some really cool, and actually functional, stuff. Thank you for sharing this.
Wow, this is really ambitious and a very cool idea. We have expensive Liquid Flow Meters that just use a plate of glass over some heat detecting LEDs to act as simple push buttons but yet have the circuit isolated from the externals. This...could be very handy...great job by the way!
This is a flawless process and you did it with one hand, while filming and while talking.
You are my hero.
:)
Great video!
Very good work I must say.one thing I see a lot is solder iron so big for the smallest work.And all so in a right mess .i have 7-soldering irons all different and shiny clean.one iv just bought is 12-volt and I Added a tiny micro switch so when I hold it I just press with my finger and all added a white bright Led the Led stays on all the time when I'm using this micro soldering iron.your work is good and would be easier for you to use a small soldering iron for when you re-solder some of you micro work.great video keep up the good work
its a relief to see a very well done instructional video on youtube! Cheers to you my brother! Greetings from germany
I dont do this stuff, but am interested in learning how. So for what it's worth from an inexperienced observer, perhaps using a hot plate or inverted iron to heat the glass evenly when melting your solder paste would limit the possibility of cracking it. place the slides in a thin frame (thumb tacks point up through cardboard) to prevent them from sliding around as much when applying the poly. Split a small dowel in half and pinch steel wool between the halves. Insert one end into a drill.
Neat method for single sided surface mount designs but a few tips:
1. As Java Lu (and maybe others) said, TRIPOD - PLEASE!!
2. Tin board with tinning solution before assembly - gives superior soldering
3. Use several coats of auto-spray lacquer on final assy for vastly superior finish.
4. When spraying remove all connectors & protect the header pins with sleeving or low tack masking tape for larger blocks
5. CLEAN your soldering iron tip PLEASE & use a much finer tip for such work
Have you ever tried selectively gluing the copper on? I'm thinking:
-UV glue the copper sheet to the slide
-tape a negative of your circuit on the non-copper side
-Expose the glue through the negative (so that only the glue under where the traces are meant to be gets exposed)
-Peel off the reset of the copper
@CNLohr I think orgameron means leaving the copper traces in place while brushing a solvent over the areas where you don't want/need the UV glue and just using the actual traces to protect the glue like the toner protects the copper during etching. It seems like that's more or less what you're doing with the polyurethane though as it cuts through the UV glue like a solvent but is also acting as a new method to adhere the traces at the same time.
I've had the notion to try and make some PCBs out of unusual material like glass for a little while now, but I didn't really know where to begin (I've done the toner transfer method, but I had no idea if I could even buy copper sheets), thanks for the inspiration!
I doubt the slide has a UV coating - it's just that glass is not very transparent at the short wavelength used in UV erasers (253nm) Eprom windows and eraser tubes are made of quartz for this reason. It may also be the case that the glue is not very res[ponsive at that wavelength. You will probably do better with a longer-wave UV source like an insect-killer tube (365nm).
There have been a lot of suggestions for this. I've switch to a spray polyurethane, but a number of products would probably work. I really don't have the time or resources to try them all! I recommend making some glass PCBs yourself and trying it out.
This is an awesome idea. I think I'm going to try something like this, and see if I can pull off a through hole board. It's got to be possible if you drill real carefully, and have lots of extras on hand in the event of a slip of the hand.
Schematic is in ExpressPCB. Full version of eagle is not free. I don't use anything that's for pay unless I absolutely cannot avoid it.
It's simply discharges the circuit, then waits for it to charge via 1M-ohm resistor. Depending on how long it takes to charge, that's how it knows how much capacitance you are adding.
That is absolutely amazing!.. two thumbs for you my friend because that is technology and art in one piece.. not to mention it is functional as well.
Pretty cool project. The camera focus bit was funny. A tip, just put the camera in macro mode. No more focus problems.
@CNLohr basically the same thing you do to eat away the copper but for the UV glue you have. Use a solvent instead of ferric chloride(not sure if that's supposed to be one word or not) and use the etched pcb traces instead of toner. That way the glue is removed from the glass and the traces protect the glue under it. Like I said though you're already doing something similar with the polyurethane but with the added benefit of replacing the glue with something else more permanent.
I do not agree that you don't need PTH. In most of the cases you can find alternatives, but as soon as you need a D-Sub mounted on the PCB, you're in trouble, because even the SMT version need holes.
But one thing you should consider. Since you already have the UV eraser, why not use it as an exposure box ?
Personally I've used the toner transfer method a number of times, but I prefer using the UV exposure method when I can, as it's more precise and gives less errors/problems.
That was a nice spark when you plugged in your programming board ~18:10.
Your last resistor looks like it had a short under it. From our view, the top one to our right of the IC chip.
The first time you place the copper on the glass slide, there is a better way to prevent bubbles to begin with so you don't need to squeeze them out. Basic bio-lab slide making 101.
Hey there, I dig the video man! I'm a Artist/Glassblower/Musician I'm working on a cigar box guitar amplifier at the moment I was wondering if you had ever tried making PCBs on round/ed glass or other shapes for that matter? I would love to do a collaboration circuit or something........? I'm just barely getting into the diy pcb/circuit making thing so I've be watching a lot of vids on it and so far you are the best I've seen, you seem to have high standards when it comes to high quality pcb crafting the video is a little rough around the edges but whatever I have a few vids that purdy rough myself! lol! Anyhoo I've been blowing glass for just about 13 years now so I got the glass down and you seem to be quite versed with your electronics/pcbs/circuits etc, I'm sure we could come up with some pretty amazing stuff. Anyway if your interested hit me back and let me know what you think,Ideas?,glass shapes/designs/colors? Maybe send you some glass to play with? Just a thought.Cool vid! ttysh
Zac ;-)
I have made a PVD and i made a completely clear and very thin PCB with that some years ago, looked pretty cool. Have also made copper, silver and aluminum mirrors that looked awesome.
Really cool. I wrap a paper clip around my soldering tip to get in really tight spots.
Thanks for the fast reply. I have ordered some copper that is normally used for guitar pick-up shielding to experiment with. I have the exact same printer in this video to use I can test with and see what I can get it to do :). If I find a reliable method I will let you know.
Nice job. If you want to take some advice, here's one. Learn to use knives for the precision cuts, they are better if you master them. If you wear gloves and use some bath to degrease the component before the copper plating there will be no risk of contamination so the PCB will last much longer
I'm laying one up with clear epoxy resin at this very moment. I'll let you know how it works out, I think it'll be a little less 'touchy' than the UV glue as it's what is used in conventional copper clad laminates
Hi, I've done quite a lot of thru hole PCBs and I saw this video and was amazed, I've been trying to Fe into smd/smt pcb fabrication, and I was wondering what kind of solder paste are you using? Do you have another video with in-depth description on smd soldering? That'd be awesome! Thanks a lot and keep up the awesome work!
Even though i have no clue what any of this video meant, I enjoy listening to your voice, its very instructive :P
I have not tried it, but I would imagine yes. I recently found out that UV lamp is not ideal for curing the glue because the wavelength is wrong. You would have to try.
If making a clear circuit was done in a mass scale we would have a clear phone not entirely clear but the first company to build it would the first step for the other companies. It would change the entire world.
It's got so many drawbacks though. This was more for an aesthetic than any practical reason.
Consider using an optical grade UV curing glue and gluing the shiny (polished) side of the copper foil to the glass slide. This should minimize the translucency you see after etching and the traces will look shiny through the glass.
I actually use copper leaf from nimrod-hall now, so both sides are shiny :)
btw: what about using a transparent casting epoxy like those used for embedding carbon fiber or fibreglass
you could just pour a thin layer of epoxy directly on the copper - so you won't need any glass at all, get seamlessly bond material (transparent of course) and you don't have to worry about thermal issues while soldering
technically it would be an FR4 or 5 PCB without the strengthening fibreglass sheets
I have no clue what you did there or why. But it sure was interesting to watch.
Also a syringe works well for depositing solder paste in tight areas, and automotive clear coat finish is highly resilient when it comes time to use solvents such as acetone, and is somewhat flexible until cured.
I use Liquid Tin quote often on most of my projects. The boards look 'cooler' with bare copper rather than tinned.
I have still never successfully done a dual layer (not that I haven't tried).
I would imagine the static properties are almost identical to regular FR4. I've been using glass circuit boards for a number of touch projects in my house which is carpeted, and even noticed some serious static, but haven't had a failure yet.
Nice work!
Unfortunately I've not patience to do all this stuff, and never got a good results...
Your work is very good, have quality!
Not sure if this would help at all, but for finer wood building applications, marine spar varnish works better (ie. more clear and less translucent) than polyurethane.
Compressive forces on the surface of the pane make it much harder to break. However, high tensile strengths in the glass pane make it shatter if cut, chipped or even sometimes scratched. This is why car windows, for example, will shatter into small, barely sharp parts instead of large, razor-sharp pieces.
It works on multiple substances, including film, however, you would not be able to toaster-oven film. You can do this on kapton, and it does toaster oven.
I have never tried to do two layer.
Yes. I got some glass that was angel gilded, chemically plated on. I attempted to electroplate it further, but could not get it thick enough to electroplate effectively... The areas that I did were lumpy and did not go on even. Keep in mind, you need about 1 mil of metal to make a good circuit board.
excess solder can also be removed from bridged terminals on a surface mount components by simply stripping a piece of stranded copper wire using it like a sponge. i also prefer to take a grinder and chuck my tip in a drill to grind it to an extra fine point, then use liquid tin to prime the tip. Moving fast with a solder iron and solder paste you can avoid heating the glass by not exposing the glass to long periods of heat like you would using a messy soldering tip..
This is totally awesome. I've been wanting to make my own pcb for some radio stuff I've been doing ugly style. May as well go big for the first one so if I fail I fail big. A question, how do you think this would work for larger PCBs? It looks like maybe using some UV glass like the glass used in picture framing would work like your slide? If that makes sense...
Wow, that is amazing sir well done. Have you come up with any uses for it ? maybe touch remote or something.
FANTASTIC WELL DONE!
1. have you try this on transparent film?
2. have you think about how to make a double layer one? so you can make led matrix?
No ditching the connector at my place, then it's better not to have a glass PCB.
You'd also get in trouble if you needed a BGA100 (or larger) chip mounted - they require VIAs.
Surely you can bend the legs and mount a LM3914 component if you need those.
I don't think the solution is to ditch a component, but instead finding a high-speed drill that drills well through glass (diamond drill would do), and perhaps use a very strong glass.
Anyway, you could try mounting some LEDs on your mug now. ;)
Have you tried removing the UV glue in small sections, then painting on a thin layer of thin acrylic? if you do it in small sections, you'll probably not be prone to large amounts of slippage, and will be able to replace the UV glue with something more aesthetically pleasing.... perhaps even embed the circuit in acrylic, or a 2-part clear setting polymer. depending on how you are able to make one-off molding, you could possible embed the entire device in such a material, leaving the connection pins exposed for programming / testing etc.
+ArcAiN6 Regarding pulling the traces off... that sounds much more manual than what I can reasonably do. But, regarding potting stuff, I agree that would likely be cool, but I rework my boards soooo much that it would make me feel bad not to be able to work on my PCBs.
Completely understand that :D
I believe the quantity is somewhere in the order of one part per 30,000 dilution. If you pour the FeCl down the drain, bad things happen, if it's just a few drops, and you run plenty of water, it's not that bad.
Great presentation! Glass PCBs are probably better for building electrometers and such (low surface resistance?).
May I suggest you try HCl / H2O2 for etching ? A lot cleaner and controllable, and you can get rid of that nasty FeCl3 forever!.
Maybe someday, but Fecl3 is soooo nice.
only thing nasty about ferric chloride is they don't send it by air freight and the fumes could kill you, just don't breath them.
I just don't believe any of these other fumes are any safer.
CNLohr I got to get me a 3D printer no fumes
I do use a drop from the middle nowadays, but there are usually very, very small bubbles on the surface that do not splay out.
That's really cool - I tried angel gilding and electroplating it thicker and a few other things, but nothing could stand up to what I was after like this.
Yes, it would. But unless you are considering a Class D amplifier I'm afraid the conventional A and B designs may create unwanted and isolated hotspots in the circuit. If the amplifier is enclosed, a fan would be used as forced cooling. Glass is not a good conductor of heat and when heat is applied at one end while cooling is applied at the other, even tempered glass might product stress fractures.
You put the carborundum paste onto the glass, chuck copper then spin it in the carborundum. some grit gets trapped by copper and a hole is drilled. Usually it is how larger holes are drilled but I don't see why smaller ones wouldn't work. Drilling glass is slow going though. I can't imagine drilling a whole circuit out in glass.
You want to try Loxeal UV 30-20, its resistant to some of cleaning chemicals. Only problem is, its a bit expensive. Great video btw.
Wow this was the coolest thing I saw the past few months
I use the attiny44. You don't need any special IC, just a 1 megaohm resistor and an GPIO. You can do this on most of their processors, not things like the attiny2313 though. Also, you don't need qtouch.
its basically liquid acrylic - although in heavy coats it tends to contract quite a bit, but many light coats should do the trick !
The reason the dried glue is matte is because it has picked up the texture of the copper. The polyurethane fills in the grooves but is a slightly different refractive index. Have you tried coating with more UV glue or gluing on the shiny side of the copper?
not yet. polyurethane is just so convenient.
I've discovered that the UV glue is tacky because it is very sensitive to moisture while it cures. If you coat your boards with UV Glue then place a sheet of clear plastic on top that will peel off later after it is cured. I tried a polythene bag and this worked well. :-)
After I am done, I typically dry and warm my boards, the UV glue is quite hard. Even with copper that's shiny on both sides, the texture is still quite pronounced. I am slowly transitioning away from the UV glue, because I am trying to use Riston for my resist mask. The same chemicals that dissolve the Riston dissolve the glue.
It's an interesting idea. I have often thought about using an acrylic, but that would have problems at solder temperatures... I was always under the impression epoxies without a strengthener were very brittle, though.
To those asking about thru-holes, why not just have a ribbon cable on one edge that goes front to back? You're still going to be restricted to SMD, but if you could just get say 10 or 20 around that would be all that most projects would ever need.
you'd still need to print on both sides which is a pain.
CNLohr no pain no gain the Romans used to say
It's just a touch sensor - depending on the area you contact, it changes how strongly it believe it is pressed.
This is awesome. I'd love to see a controller like that for say, changing the radio or some sort of game input device. Thanks for showing us how you did it.
That seems to be what most people are using. I've read that they've changed the design recently (see the PCB "Fab in a box" page) for the worse. I was able to find this one on amazon that I'm going to try out called the Apache AL13P. Hopefully it won't need modification.
I may try this... I don't have a laminator or uv oven though. It's been years and years since I fooled with circuit design and/or breadboarding. I've never made pcb's. My line of work is BGA reball/reflow and general computer repair. Love anything to do with electronics!
Absolute legend!
I was just thinking about this the other day, love the fact you can use it capacitive touch!
Fair play to you!
Currently working on a few things.. Maybe hit me up for a chat some time?
Cheers!
I have tried, with many different epoxies, glues, etc. to no avail. Also, on glass, you can't use a clothes iron, but on regular PCBs, it's not that bad.
Dude, this is incredible! I have never seen anything like this. After seeing this vid, I have a few cool ideas to try soon.
perhaps I missed it, but why do you want to use glass? Granted, it's quite neat. But do you have a specific need for it? By the way, you could use bismuth solder, which melts at ~140C to reduce breakage.
Its been about a year but I finally made a board that plugs into an Atmel board. Can't wait to get them in.
The glue is not conductive, while the glue is translucent, you'd still need a material that can conduct. I do use this process for touch sensors, I put designs into the copper.
Awesome Work.....Great ! Love the sensitivity of the Glass.
Can it be used as a touch number pad...?
USE CELLULOSE PAINT THINNER to effortlessly remove pcb print after burning copper clad.
When using the matte back copper, it doesn't really rub off.... ever it seems. But, I've been trying some shiny-backed copper and that is a major concern when removing the toner. Once the polyurethane is on, it's all on there for good. There's a lot of equipment, and not much space, I'll admit.
oh but note: the glass and actual construction of the board itself was quite sufficient in this video and I want to say awesome vid ^_^
As it turns out it appears the wavelength used in the eeprom eraser is not optimal. The glass is transparent to UV, but you need to use a source like a mercury hallide grow lamp to get the right wavelengths.
Nice! I can see this working well using visible LEDs UV glued emitter down to the board.
Just a thought, but would this work if you used copper leaf from an arts and crafts shop? Its a lot thinner but you could maybe overplate it with copper to the correct thickness?
I've tried a few, but nothing ended up working out. Even now, my process has changed some but I am still using the same glue.
Some safety notes:
1. When working with UV eraser, please don't open the drawer when the light is on. The UV-C wavelength used is 253.7nm; can cause skin/eye cancer or cataracts;
2. When working with chemicals, such as FeCl, use eye protection or a face mask rated for use with corrosive or caustic chemicals;
3. When using volatile chemicals or chemicals producing gaseous emissions, please work with a carbon based respirator in a well ventilated room, or in a fume cupboard using forced convection
I found that tweezers for medical use work very well: precise, large variety of forms, stainless steel (anti magnetic) and can be cheaply bought on Amazon.
Perfect thanks for the additional documentation! How's the video going regarding the SMD soldering? did you have any chance to edit the video? sorry I just got my chips (from china) so I'm ready to learn! lol!
I have this listed under my "to buy" section. I doubt the ferric chloride will be able to eat through the glue, but it may be possible to silkscreen the stuff.
Thank you both. With respect to "1pcfred", I do not know what you mean by drilling with copper. Do you mean a copper-coated drillbit of some sort? I looked up the carborundum paste and I'm with you there, but you lost me at the first statement.
Thanks for your time.
It totally can - I hope to get my new project completed, but I will tell you now, you can put a lot of touch sensors on a board of this stuff.
Awesome video! Two things though: First, you should definitely invest in a tripod, I feel like you could seriously benefit from it. Second, the tip of your soldering iron is nasty, you should seriously start to tin/clean it. Other than that though, great video!
A lot has changed since I posted this video.
Yeah, figured as much.
CNLohr i hope one of this lot of thinhs is your soldering iron XD anyway, loved your video man, but this looks like a real pain to do because of the risk of breaking the glass... do you have or will do any tutorial on how to do normal diy circuit boards?
I'm pretty sure it's a matter of glass /really/ wanting to shatter when you drill it. I've never tried though.
I actually use Riston now instead of toner transfer and I use a grow lamp to cure the glue. BUT! Your problem isn't that you need two sides, it's that you are using D-sub connectors. Ditch the connector.
Thanks for posting this video, all aspects a great info, especially the bit around 16m where you reveal the SMT process, we've been wondering how this worked and your video has clearly answered this. I wonder if you'd care to respond with key process points about this method?
Also, I note you've not applied any tags to the video which I suspect is why I haven't come across it previously when searching for "SMT" or similar. I found this video via the sidebar suggestions for PCB production...
Awesome! I completely overlooked that file, thanks again. I was starting to pull my hair out because express pc doesn't like to run under my setup in crossover.
I made something similar, but first step was to vapor-deposite copper on glass, then a bit of galvanic process to enhance layer, then mask + etching and OALA circuit on glass was done :)
Excellent video! I really enjoyed your various comments throughout the process on why you should or shouldn't do certain things, relating to the UV glue and the toner, etc.
Do you think that you after gluing the copper you could coat it in a photoresist solution, and then transfer the circuit using UV exposure?
I do now! But, you can't etch until it's attached to the glass otherwise you get a pile of traces.
I've tried that a lot, and it almost never works out well. but you can apply the paper to the copper before it's attached to the glass.
That is 100% dependent on the size of what you want. Also, I'm doing ones with very shiny undersides now, too, not just the matte brown. I wouldn't charge extra for them.
I don't know... a 1"x3" probably $15... A 5"x5" (the biggest I can reliably do at present) probably $60? I can make the board decent outlines with glass cutting.
This is a first for selling.. I've printed hundreds but they just get tested heavily.
Yeah, I have a 5N now. I do tin boards, many times, but it ruins the look of the board. And, I haven't had any trouble with solder-ability when I use paste. I actually have switch to a spray polyurethane now, too, doing just that. And, I totally am good with that hulk tip.
Ni ! Very nice work, you make a careful job. Did you try to print directly on the copper sheet, before gluing it ? If it works it will save you the transfert step.
It seems like the UV glue is a weak point in the whole thing - Why did you pick that, exactly? Have you tried anything else?
Excellent vid, by the way. Looks beautiful. What are you making? I found my way here because I'm thinking through ideas for something coffee-table sized, so your vid isn't *too* helpful, but it's fascinating to watch, and you're way way way ahead of me technically...