If you'd like to help support the channel and help me buy more nonsense - I have a Patreon page: www.patreon.com/atomic14 We've also got a shop now! shop.atomic14.com And if you're shopping on AliExpress or Amazon then if you click on these links I will get a little bit of affiliate money: Aliexpress - s.click.aliexpress.com/e/_Ac05mF Amazon - www.amazon.com/?tag=atomic14-20 You can also join us on Discord here: discord.gg/h2DDnDdjD2 There's also the Atomic14 newsletter here: atomic14.substack.com/ And don't forget about the MakerNews newsletter - makernews.substack.com/

Interesting bit of kit, but $5200 is a little much for the average hobbyist budget. I'm going to wait a couple of months until Creality knocks it off and sells it for $200.

As someone who works in a professional manufacturing environment with sintered silver inks, surface prep is hugely important to get good bonding to the substrate. Sanding the substrate at ~200 grit then cleaning with isopropyl before printing should help a lot. There is also some benefit to doing a 2-step sinter process where you bake once as normal then do a second bake under mechanical pressure (clamp PCB between 2 pieces of jig plate metal or borosilicate glass)

You can also do this with laser printed transparencies and then standard resist and etch process. Substantially cheaper and more mechanically stable.

My local university had a few of those when they came out in 2018. I tested it but I could not get it to work consistently and after much complaints they are now only used to solder surface mounts. We got the Wegster-CNC machine to cut traces in stead.

Ah yes, back in the 80"s. Back then there existed PAY TV. Of course every engineer and technician where I worked wanted to create their own pay TV converter. I, as a technician, was self educated, and thus my approach was different. The pay TV system worked by raising the horizontal sync pulse up into the video area such that the TV could not detect it, and thus this picture was scrambled. However, the horizontal sync pulse was not touched during the vertical sync pulse. So I made a simple circuit that would check to see if the next horizontal sync pulse was in the right position or not, if not, it would make the correction. So It became a line by line correction system. Total cost for a single home made PCB and its components. was about $6.00. It required only 15 mA current at 12 Vdc. So you simply placed it within a TV converter that had volume control. Meanwhile, the engineers used the PLL method to fix the sync problem. This required lots of components and thus a complete power supply and a box to put it all within. My boss tried to then sell my design to the cable company, but they turned it down due to my design being so simple and easy to copy. Oh, and my design gave a 6 dB signal to noise ratio better performance than did the cable company decoder. They had shifted the horizontal sync pulse by dropping the RF signal strength during the horizontal pulse, by 6 dB, and then dropping the RF signal during the video section strength by 6dB at the decoder to match up again with the 6dB suppressed horizontal sync pulse. Of course my method did no RF suppressing at all, and thus the picture quality was better. . Ah, the good old days.

I would like to see if scuffing up the substrate would make the line look crisper and the header stronger.

If you mark/drill alignment dowel holes, and then mirroring the bottom traces, you can print the top, flip the board and mount on the alignment dowels, then drill and solder a bit of stranded wire between the pads as via's for double sided boards. I also wonder if you squeegeed a bit of liquid solder (the SMD stuff) over the traces then re-baked it if it would pre-tin the pads/traces Very cool and awesome development! Keep em coming!!!!

*A low cost and highly functional and reliable pick and place robot arm with OCR and machine vision (for picking loose SMT components from a box with a shaker to flip the components upright) should be a MAIN FOCUS of the international Maker community this calendar year!!!*

Printing the traces directly on a blank board is a surprising approach. I would have thought you'd do it by printing resist onto a copper-coated board and then etching off everything else.

We used to make circuit boards at school. we would print the diagram 3 times on see through plastic and layer them together to create a black diagram light could not penetrate. Than we put it ontop of a copper bread board and showered it with UV light (i believe), followed by a strong acid bath to remove the copper that was not in the diagram.

The solder paste dispensing. That's the part that got me. Man that's so nice. Really awesome demo video. It would be cool if there was a PCB (from copper plated fiber glass) milling machine that deposited solder paste as well. I can imagine that fabricating much better prototypes.

You could make the whole back of the PCB a ground and use thru-hole capacitors as bypass with one leg formed to work on a SMD pad. The idea is that the ground leg of the capacitor jumps the connection to the top of the PCB. This way there is no alignment needed. You just have to handsolder the capacitors after the reflow is done. You will need the super low temperature solder for that. Also: If you use a zero Ohm 1206 resistor you can run a wire under it to get your layout to work.

For components which will have mechanical loads applied, how about pre-milling the pads slightly to remove the gloss from the board and give a better key for the conductive paint?

I've used it at my University and I shudder every time I see one of these machines. Milling is a thousand times better than using a Voltera.

Just get a K40 laser or some other cheap laser, make some safety mods, and use that! Black spray paint goes onto Copper Clad Board, then the outline of your traces gets engraved, theb etched, and then you clean the spray paint off. Surprisingly accurate THT and SMD footprints if you are careful with the dremel when you drill it! I have been doing this for like 8 years now

Friendship ended with Pcbway. Voltera is new best friend now.

A pretty neat solder past applier: I can imagine some people buying it primarily for that. No wonder it's expensive though: It has to be a wet plotter and drill in one.

Interesting machine but the finished boards seem like the traces would be very fragile. I'm surprised we don't have a group of "do it yourself" guys developing a print head that could be adapted to a normal 3D printer. We would still be at the mercy of the ink manufacturer for ink prices. If they come up with a powdered iron mixture they could print some interesting transformer cores and electric motor hubs. These could also then be magnetized for motors, generators and guitar pickups for example. With a two headed printer such as a modified 3D printer we could use one head to lay down a layer of iron core for a transformer and then print the winding layer with the trace head. Cure it with a built in heater head and then print the next layer. Keep going to print complete motors for example.

Thks quite interesting & thinking out of the old box; Consider a 3D printer with 2 print-heads. One head prints hot insulative plastic ; the other head prints hot conductor ; both cool semi-rigid ; & can be baked rigid/hard. Thus any 3D conductor network can be printed embeded within an insulative substratum & component mount-points on/around its surface. Let me know what you think.

I would love this machine purely for applying solder paste to existing boards so that I didn't have order stencils for every iteration of a pcb design.

seems like you could use the drawing as a mask for a uv sensitized board and then etch normally. of course you could do that with a pen and a cnc machine probably cheaper. or the uv screen of a resin 3d printer

i tried really hard to use the laserjet + glossy inkjet paper + silver ink stuff technique but i just could not get it to bond right. the silver ink absorbed into the paper. i tried transparencies and even inkjet ink to try and bond / heat that but it either wiped off or the transparency just melted. i am determined to have some kinda quick positive-only PCB making method that doesnt require etching. i really like the look of this machine. it seems so much more diable than FDM printing to be honest. it has a less difficult task to do

I'm still skeptical of these things, the ink is just too sloppy for anything tighter than what's shown here, and baking seemingly adds to that, plus fully exposed traces, and the soldering situation. For home kit stuff, I think using an MSLA printer works better on PCB blanks, you get a digital mask and UV lights in one step. Not to mention the cost differential, especially when this process has consumables that're (last time I looked) proprietary to the company itself. This machine just does not make sense anywhere, other than being a rapid prototyper for a large company, yet there's also better prototyping alternatives.

Would be really cool if the voltera had a pick and place option to it as well.

I've learned to view to surface mount connectors with suspicion, and that's on normal PCBs! You should get some neutral cure silicon and stake everything down. At work, we use 3145 RTV, but that stuff is pretty expensive. You can get much cheaper neutral cure RTV at the hardware store, but I have not used it on PCBs. When I was looking for a PCB printer, the Botfactory SV2 looked like it was the closest to what I wanted. That one also prints an non-conductive resin, so you can have multilayer boards.

We used to do UV etching many years ago with pretty high accuracy. Does require using some acids though.

Seems good for retro enthusiasts.🥰

*Printing solder paste onto a board IS THE WAY that should be done!*

Been doing it for 7 months now, savings me a fortune remaking cracked boards. Just need better access to PCB routes online

Can you make a survey on print head construction? How to change ink/solder paste? How to change nozzles? Which solder paste and ink compatible with this print head?

A prototyping-only thing, but with the presented limitations, I can't help thinking that milling away minimal copper on a copper-clad board, would produce a much sturdier result.

5000€ cost!!!! Crazy

why not add texture to the PBC surface? it might increase adhesion, just like print bed for 3d printer.

Just a thought, but have u considered adding another step? Run a a drill over it cutting out the design b 4 laying the conductive ink, u might get stronger part adherence to the board.

You probably can use two vias inside pads to "nail" down that usb connetor, they dont have to go anywhere, you will prob have to add real solder as a final step

IMO, they really missed out on adding an attachment for a fine tipped paint marker to do silkscreen markings - it shouldn't be difficult to do considering it's the same process as their conductive ink prints except it's a different ink or using just a mechanical jig to hold a marker. They could probably even provide a dispenser to do UV cured solder mask (not necessarily for their printed boards) since the software already has a 'homing' function so in theory an etched board with the appropriate gerber files would allow for this. The price of the machine though... Way too steep for a hobbyist especially for those in regions where etchants are readily available (unfortunately not for me). I can see how this might be useful for a small business specializing in producing prototypes just purely as a solder paste dispenser since they'd not need to make/ procure stencils.

Perhaps some sort of adhesive could be use to bond the components that need a robust mechanical connection

Might be good for basic prototyping, but I wouldn't try doing production runs with it.

This is fantastic. I wonder how the mechanics would be if you used UV-set epoxy to reinforce the header and the pushbuttons. Since the conductors are silver, I'd add conformal coating to protect from environmental damage and oxidation.

Impressive!... How does it feel to buy such an incredible, expensive - but oh-so-beautiful - machine? I am tempted, but am experiencing guilt trips just thinking about it :-) Or did Voltera gave you one just for this review?

Very interesting, by applying 2 component clear epoxy it becomes indestructible

Interesting that the actual x-y movement is quite quick while printing, but the z movement is quite slow and there's a pause at the end of each movement before a positive z movement. I assume the pause is to reduce the sizes of holes and bumps in the printed ink. I wonder if the software has tuning for the z speed. It also looks like there could be some pretty significant efficiency optimisations that could be done, such as printing complete nets in one go rather than doing several segments and pads separately. I'm guessing that, with the way the solder doesn't wet the pads quite the same way as traditional copper pads does, solder resist isn't as important, but I don't see why they couldn't have you swap out a tube of lacquer to give you that option. Same for a "silk screen" layer which could just be a pen attachment. Aligning a double-sided board probably wouldn't be too hard if you also had the drill attachment since you'd just need to provide two or three drill sites similar to the alignment process done for the solder paste step. Also, though I see the official video shows them inserting the via and through-hole rivets after printing, I think it'd probably be easier to do it before printing the traces and pads. It would certainly be cool if this technology could be integrated into a pick-n-place machine for an all-in-one process where you'd stick a blank sheet of fibreglass and get a completed circuit board at the end (that'd probably have to only be single-sided, though unless you _really_ complicated matters by adding the ability for it to flip the board itself).

*I want one of those printers ASAP! That printer is pretty awesome! That pricetag needs to be $50 instead though. Sigh…*

Being able to fast prototype with SMD, the solder paste printing, and fast turnaround would be huge pluses, but the drawbacks of being single sided with no vias would rule this out for me. I’d rather pay 20e and wait, and get a professional board with cutouts, stencils, and holes. For prototyping I’ll keep using veroboard with breakout boards for SMD chips.

Heck just the dispensing the solder paste on an existing board might make that worth it.

I REALLY wish that you had elaboarted on the "reflow" process.....how did you heat the soldier? bake the board in an oven? heat gun? crack pipe? barbeque?

I'm not yet convinced that this is better than milling

use allignment holes for multi layer

wow man, this is amazing!

To be honest, I would like to try this additive printing process myself. Agfa has some nice ink for laserjet printers (eg. soldermask ink or photoresist ink, which is applied and cured in one printing step.). (And I think, the best printer for this could be the LP50 made by Suss.) So I would'nt need to make films with a photoplotter, as this first developement process has lots of potential for failure and i haven't got the hang for it yet.

It dispenses from a nozzle then bakes it - sounds like it would be better just to stencil an image on it with some powdered flux and then powder coat it with conductive dust (sticks to the flux) because you already have the ability to bake boards. Then you could basically make boards in one shot processes instead of having to speed up the video 10x - faster process means less footage means less editing work too! I'm super lazy and this also looks like a lot more effort to manage the tools and materials because they get wet with materials that cure, blegh always running against a clock or something with wet/activated materials... But honestly if you're gonna do it this way then you should be able to use the same gantry to pick/place your parts too, especially since it already knows the board plots.

Have you tried roughing up the board so the silver paste sticks better?

I don't see any advantage over using a laser printer, transparency film, photoresist, and a vat of ferric chloride. Neat gizmo but I'd give up and go back to how I did it in high school pretty quickly.

How about using copper plated bare pcb; cover it with acid protective paint and then remove the paint using a low power laser engraver. Finally etch it with copper etching solution. I have tried it it is cheaper and quite fast.

New subscriber always watching your videos many learning❤😊

I wonder if a vinyl cutter could be repurposed to make pcbs

That's one of the better results I've seen using conductive ink, but the limited adhesion is a sore point, I think. Maybe there's a trick to improve it, with some treatment to the board first. I'm guessing they use some kind of epoxy when they make proper PCBs. I really wish there were better and easier ways to create low-cost circuit boards. While those Chinese manufacturers are cheap, it's not ideal.

Why not go further?! This printer need to use another interchangable head with UV-resin, and install UV-lamps into case. It solves many problems - 1) It actually gives this printer a ultra purpose - to print multi-layers schemes. 2) It would allow to create more force-resistance solder joints - just by having them bigger and by covering joint perimeter with UV-resin.

maybe if you sand the pcb beforehand, the ink will stick better to it

there's headers with screw holes at the end, maybe some of those are available.

So.... could I use my diode laser to etch glass and then use the silver ink on top to make a glass pcb?

Can't see the niche for this, especially with the price tag. Seems you can get better and faster results with toner transfer at home

This is really cool! I'd love to have one

It seems more like a glorified breadboarding solution, rather than a PCB prototyping solution. I mean, with its limitations you can’t really use it for prototyping your final PCB layout design. So, once you’ve “prototyped” your “breadboarded” circuit design, you still have to go down the final PCB design prototyping path (with the usual delays waiting for your chosen PCB manufacturer to deliver).

German: Ist es möglich ein PCB zu erstellen mit einem Resinprinter? Man macht ein layer als rückseite, dann kommt daraum das design des PCBs in negativ und füllt alles mit flüssigem lötzinn auf und heizt es auf (resin ist ja ziemlich wärmeresistent)

I could learn PCB design and more advanced electronics engineering SO MUCH FASTER with that (and a pick and place machine)!

I'm wondering if using an abrasive on the board before you print the traces and pads on the board would make them adhere better. The surface of that board looks a lot like you're printing on glass, which would have very little binding strength.

I thought it was an interesting machine until I saw it was then able to cure the ink and dispense solder paste! That's fucking brilliant?

i had problems when i had an led on pin 2 or 4 (in not sure wich didnt work). when i wantet to upload an arduino programm in my esp32 dev board, the esp worked fine to upload on my laptop but on pc i always needet to press the boot button, but this didnt work... so the programming always timed out. it took me a while to find the problem, because i made 2 of the same boards but i used my pcb cnc. one for my friend, and one for me. my one did work without any problem,but the other board did not work and looked like it was write protected... all parts where the same only the ir recivers did came from old pcbs i salvaged. but the problem was that i had the resistors for the leds a little too small, and the tolerance in the resistors itselve did cause the problem. just used ones with higher resistance and it worked just fine.

would it be worthwhile to swap this pen out for an etch resist pen and cad onto copper..?

Is there a working projects of doing the same with 3d printer or regular cnc router?

With the USB port issue, couldn't you drill through the board and secure the connector to the PCB with some jumper wire or something, so there's no force exterted on the bond when inserting/removing the cable? While lining up a double-sided print would be difficult, doing some selective drilling to secure things like headers, switches etc would be easy enough?

what about an ender3 addition for cheap?

I wish there was a conductive filament as well as insulation one so you could print PCBs on an ordinary, dual material, 3d printer, lol.

Cool in concept, nice to watch on youtube - but in practice its a non starter - too limited, too expensive. Isolation milling can be had for 250eu with 3018, for 5k you can probably get a LPKF machine that will outdo this one by a mile and half.

nice idea, but I think its mostly for one off prototyping.

Loctite the surface mounts in place after soldering

You could get the ink and use it in a vintage plotter 😀

Dang why couldn’t they add the pnp to it also? Then it’s fully done.

great video, very interesting, thanks! only feedback is the volume seems quite low?

Five thousand cost? Do you realize how many professionally made boards I can buy for that! Also, in the USA, if you want to do at-home PCBs photo sensitized board is cheaply available.

If this can be used a CNC milling machine, whats the point of the silver paint? You can probably make better PCBs using the milling function.

Little question from a mechanic : why do you put soldering past on all esp32 points even the one that not used ? Doesn't it increase the risk of short cut without bringing any advantage ?

Would it help to sand the board before print to make the solder to stick better

Interesting tech. The problem is just a little more complexity and you’ll need a via. What’s the solution then? I don’t see this as being any more useful then toner transfer and good old FC. I’m not even sure it can make finer traces.

So expensive 😢. It’s a glorified 3D printer, so it’s a little wild we haven’t seen some cheaper alternatives. Or… maybe there are some? The movement system isn’t that much different, and precision seems like it’s even less important since the ink and paste squish out some. 🤔. I guess when using the drill it would need more rigidity and precision.

That's a lot of limitations for the price, but it's also pretty cool that it does the solder paste all in one. Had a thought on a really strong way to mount modules like that USB board. Instead of soldering on a pin header first, I would take the headers out of the module, and try soldering the bottom of those vias onto pads on the board, BGA style. Then for strength you could screw the module directly onto the board using the holes in the module and some corresponding holes on your PCB. I think that would hold up to quite a lot of abuse.

Hi there! What budget 3d printers do you recommend?

U need higher temp or more time in top temp during reflow dude

How well would the ink bond if you gave the G10 board a quick sanding with 400 grit before printing?

I wonder if electroplating a layer of copper would improve wetting

Would this work with something like a cricut?

How does it compare to milling a copperplated board? in terms of speed of manufacturing, accuracy, electrical characteristics and price. This printing seems nice but kinda gimmicky.

WOOW 🤩🤩🤩👍

Yeah, BUTT, the price tip stop on Mars 😒, the silver ink is very expensive as well. Per total acquiring and operating such machine is extremely expensive and is better to use a CNC to cut the traces than build them on the surface.

Что за жижа которой дорожки сделаны?

Very nice :D But is it ink or is it paint? In theory, I could make a 3d printer attachment to do this :)

isn't getting a $200 resin printer for photolithography just cheaper and faster?

Oh, I forgot.... as for this silver stuff to stick better, have you tried to remove the sheen from the PCB to see if the "ink" with adhere better? When I saw this I thought, I've love to have one. But in that the ink doesn't stick well, is a total deal breaker.