For many people, creating their own PC board is a ritual. However, nowadays, you can order super cheap circuit boards and have them in a short amount of time, so it doesn't always make sense to build your own circuit boards. However, some people like challenges, and some people don't even want to wait a few days. Maybe everyone has dreamed of a machine similar to a 3D printer, which can make beautiful PCBs. Voltera V-One has not reached that level of complexity, but it is not far from it. [Great Scott] showed us how he used the system in the video below to build two different boards. Although the results are impressive, you can also see that there are some limitations, especially when you design the circuit board without considering the machine.
The obvious thing is that the machine really needs your help. In addition to aligning the holes, you also need to install tiny rivets for the through-holes and slightly smaller rivets for the through-hole components. Last time we looked at the machine, it didn't punch holes at all, but [Scott] showed a drilling attachment, which allows the machine to produce through holes and support lead components.
The system relies on interchangeable print heads and conductive ink. Mechanically, it is not much different from a 3D printer. However, the ink seems a bit temperamental. On the first test board, because [Scott] forgot a step, one side was not cured properly. Although it looks good, it ran into problems when soldering. Because the wiring is not copper, the machine also has its own recommended solder.
However, the most interesting part was when [Scott] tried to port an existing circuit board to the system. He had to make some changes to the copper fill and the size of the vias. The tight spacing requires some manual rework, so he probably should space the traces wider in the design.
Printers are not cheap, and consumables are quite expensive. It is difficult to justify the cost just to obtain a rapid prototype that does not represent the final board well. Keep in mind that conductive ink is not as conductive as copper, and you cannot get details such as solder mask or screen printing. If you have an active hacker space, a school, or a busy laboratory, then it may be worthwhile to purchase a disposable board. However, for prototype production boards, you are unlikely to want to limit the final design to the constraints required by the printer. Also, if you don’t mind waiting for the circuit board—remember, if you don’t mind paying more, you can spend money to make the circuit board and ship it quickly—the printer's break-even point is very high.
On the other hand, 3D printers used to be expensive and are now ubiquitous. The quality of 3D printing is not as good as some other production methods, but it still has a place. Therefore, if this technology improves slightly and becomes cheaper, then one day it may enter your workshop.
As early as 2015, we predicted that manufacturing our own PCBs will become increasingly difficult to justify. At that time, many people were dissatisfied with this prediction, but the cost of manufacturing circuit boards for you was further reduced, continuing to weaken the value proposition of manufacturing your own circuit boards. Of course, sometimes you just want to do things that don't care about the economy-we have all done this. But this does make the economics of PCB printers more difficult to calculate.
I saw one up close. It is a sturdy machine, but the price of consumables is going to kill you. As far as I know, the price of these machines is 4k -5k € (I guess the price is ok), but only use conductive ink, at least from my supplier, the price of 2 ink cartridges is 150 €. For 150 euros, I can make a lot of PCBs in China, with higher quality, 4 layers... I think a cool feature is dispensing, pick-and-place, and soldering functions.
Therefore, it is not a good machine for making PCBs from scratch. It is a very interesting choice for PCBA.
I am grateful that this person took the time and talked about the challenges he faced. I am interested enough to view the specifications. There are more details about it in the article about his vid... https://www.elektormagazine.com/news/pcb-printing-voltera-v-one-greatscott. I'm still thinking about it. ... 4 hours self-made pcb is not bad for the first time.
There is one in our R&D laboratory. 4 people tried to do anything useful with it. Finally, they all ordered circuit boards from the manufacturer. Now we only use it as a drill bit and heating plate for the rework industry.
The transaction here is exactly the same. I spent a whole day trying to get it to print a circuit board that I could build on a veroboard in half an hour. It can't even dispense solder paste correctly-it will under-squeeze for a long time, and there is no software setting to increase the dispense amount. I just used the acetate transparent film laser to cut the template, polished it flat and glued it by hand.
Ah yes, software. It is cloud-based...
When I used it about a year ago, there was a clear step in the calibration where you could adjust the extrusion rate to make it distribute thick enough lines. In addition, the software is browser-based, not cloud-based, which is a meaningful difference, IMO.
There are 3 such printers in our laboratory, and they are booked every day. Although expensive, they are amazing in terms of circuit board repair and reflow. If this is something you prefer to do manually, then once the machine is up and running, it requires minimal user input so that I can handle other things.
highly recommended. Consumables are also expensive, but PCB can be used for a long time (50 ).
If anyone from Voltera sees this, pick and place will be a good addition as an add-on in the future.
The price (although it may be reasonable because it is a new design) is astronomical-just like consumables. It reminds me of the current way printer manufacturers exploit their customers by supplementing prices.
The basic price is equivalent to the manufacture of 100 professionally manufactured printed circuit boards-even for experimental purposes-printed circuit boards. In addition, any "experimental" people now do this on the breadboard-why they want to make it into a PCB version on the breadboard is anyone's guess. Layouts and tracks can be "tested" in the pcb software you will ultimately rely on.
This looks like a 3D TV made by pcb...
"You can order super cheap circuit boards and own them in a short time"
Where? The cost of making a custom 10×15 board at home is less than 4 Euros. In addition, I am not interested in ordering 10 times the same board when prototyping.
Okay, but it is difficult to achieve 2 layers at home, especially if you need vias under the IC. Multiple layers are impossible, at least not economical. I still have etching equipment and have done it here and there a few times, but this is not enough for most of my prototypes. I mainly order from JLC PCB and don't care about the extra circuit boards. If you don't want to keep them, please throw them away.
I have been doing a lot of PCB soul searches recently. This looks really expensive, and it's not a good concept.
So far my heart is to use laser ablation paint and then HCL chemical treatment. 2 layers are tricky, but via rivets are one thing. I plan to focus on SMD as much as possible and use 0 ohm jumpers instead of vias.
This seems to be the cheapest and simplest solution for the work I will be doing.
Have you checked the LPKF stuff? They have a special solder paste for making vias-faster to use than rivets, and can be used under IC. Their pcb milling machines and laser etching machines are great, but quite expensive.
I think I stick to the old photosensitive process. This looks super slow, super low detail, unreliable, looks ugly, and super expensive. Basically useless. It is not even suitable for drilling, the speed is super slow, and I think it is not rigid enough for milling.
Using the pre-coated photosensitive film to make the largest A4 size PCB takes less than 45 minutes (not including drilling). If you want a cheaper method, please use the laser printer ink transfer method. It takes 1 hour to heat each side with this machine alone, which in itself is more than the complete "manual" process of the photosensitive process.
LOL is making and ordering thumbnails. For some reason, it looks more like a novel cake decoration. The welding looks really bad.
For some reason, he insisted on using a crappy soldering iron for manual soldering instead of the actual recommended process.
No, he just doesn't know how to solder.
The tip of the iron does look too big, but this may be due to the close-up nature of the subject.
My 1.5mm tip looks like a scroll bar under the magnifying glass :)
The size can be deceptive, but it is definitely a false indication of this type of work.
When I first saw it, I thought it might cost $500 or more. Really shocked by the actual price. The quality of the printed matter also looks terrible. Toner transfer etc. are much cheaper, and possibly much easier. If you don't want to use actual normal toner, the "special" blue toner flakes you can buy are also very cheap. You can buy cheap laminators for $20 or less. Only people who like to waste money will get one of them (so they may make good deals with university research departments).
I use regular toner and agitator irons, including printers (64% toner left!) and thrift store irons. My secret to getting great results is to stick a little purple glue on the paper before printing. After ironing, the water will dissolve the glue and the paper will be clean.
good idea. Is the glue the usual PVA material?
In my opinion, the "magic" lies in consumables. At their prices, I am not sure if I will interrupt the process before their prices drop. But if I decide to give it a try, I don't see any magic in their $5,000 printer.
I'm sure that anyone who builds a 3D printer from scratch and is so inclined can come up with reasonable alternatives based on their hobby budget. What do we have here, an XY platform. Maybe it's the Z axis, but it's actually just pen up/down. The fine-point paste extruder on the base can be replaced. Small drill motor that can also be replaced. It is only used to drill directly down, so even cheap port cargo rotating tools may do so. Is there a UV lamp built into the bed? Okay, I think it's cool, but a separate UV light box might be easier and useful for other projects. I think he mentioned optional reflow soldering on the printer. So it has a heated bed. Not entirely difficult! Although...I might skip that part and build an actual closed reflow oven.
In this 5,000-dollar beast, what can children not buy from a typical online store with paper money? software? I bet FlatCam can do it.
At a critical juncture, 3D printer enthusiasts who don't have time to learn FlatCam or want to build a new machine can export the PCB layout to SVG and use one of the online utilities to create a high-level STL from it. This can be sliced on one's favorite slicer and then printed on a printer with a paste extruder. Although, a typical printer may not have the ruggedness required for truly fine surface mount traces. However, it should be acceptable for through holes and larger surface mount pads.
Wow, that's terrible. 5000 US dollars... If you order 2-4 PCBs, the printing effect of the general PCB shop is much better, then the price of the PCB is about 5 US dollars, and you will have to make full use of it to return it. That ignores the real fab'd PCB will not look like garbage.
"In addition to aligning the holes, you also need to install tiny rivets for the through-holes and slightly smaller rivets for the through-hole components."
I won't say you need.
Don't get me wrong. I am not opposed to the use of modern double-sided structures. I just want to say that if you are thinking of making your own PCB, but fiddling with rivets or getting the sides to align perfectly will prevent you, you don't need that. People have been using single panels for decades! You can still build many good projects this way.
I think it’s a bit strange, there doesn’t seem to be much middle ground. I think most people now get their PCBs from the board room. But when you see a self-made PCB, it is usually single-sided, or the manufacturer has perfectly aligned the sides, mounting vias and everything. If alignment is difficult, why don't we see a design where both sides are used, but the connection between them is kept to a minimum, and it is done with jumpers, maybe through a larger hole without routing To them, so alignment is not so important?
I totally agree with your point. I have a single-sided philosophy, no holes, limit additional processing, 1206 0ohm jumper is one thing, each save 1 via hole, most simple boards need to be on a single side Less than 10.
Most people who design PCBs only use them to connect wires, connectors, and power supplies. In any case, they integrate all the complexity. The high component count and high-density design is something I rarely see.
However, this can be traced back to the point I made. If you need a board, place it so that you can do it with Sharp, toner transfer or rubbing resist, and be happy. If you are making a large number of prototypes, you most likely want to mass-produce the circuit board later. I don’t want to tell my clients “this board is twice as large as it should be. I plan to use through holes, but I don’t like the way my prototype handles it, so I didn’t use it.”
I mean, for that matter, what we used to do with "double-sided" was to solder a wire through the holes on both sides. You can even use some (but not all) through-hole components to do this. That only applies to prototypes.
Once again, you seem to have two markets: one is a costly market, and the other is a professional prototyping that does not have enough fidelity to be useful.
Maybe if you are an affluent school that needs a lot of one-offs, and you can limit the design to meet the requirements of the tool? This may be a good use case, but I don’t think most schools can afford this kind of expendable budget.
I always think that if you are a school, then etching photos is a good way anyway-students can always have this process, and it always allows you to see that you can use what you can find Do an ordinary house instead of using professional but not really professional tools.
When I discovered the laser printer and the transmission method, I was really excited.
"Perhaps if you are an affluent school that needs a lot of one-offs and can limit the design to meet the requirements of the tools?"
So, assuming you don’t want to go the chemical route, why don’t you directly own a factory? Literally, I can't imagine this kind of thing can produce a design that the mill will not greatly exceed, and compared to the paste, the "consumable" bit is very cheap.
However, you don't necessarily need to worry about the alignment of the double panel-if you only have SMD components, then just use the entire back as a ground plane and insert the rivets. Puff, it's all done.
This is a good point. Different jobs use different tools. All these PCB manufacturing methods coexist because people use them to do different things.
The one mentioned in the article seems to hardly solve the problem I can imagine...
According to records, my school has an LPKF machine. Of course, its price is about 10 times that of this machine, but it produces beautiful boards very fast and can be produced in small batches (maybe in the range of 10 to 20) A circuit board with vias on both sides without much user intervention (it is said that it can be used for chemical vias, but I always use rivets)
Oh, I definitely only talk to one-time people. I also don't think paying customers who plan to mass produce will be happy with the one-sided restrictions!
However, "happiness" is not the exact way I describe my experience with sharpness, toner transfer, and chemicals. Maybe I was out of luck when I got weak ferric chloride? Or the pile of old PCB blanks I bought at hamfest is too oxidized, or the copper is too thick. I have no idea. It always takes much longer than each book. A set of instructions in the YouTube video said it should be done. When I reduced the etched area to fiberglass, my traces began to narrow and dent. I have never been satisfied with the results.
A few years ago, I decided to cultivate the ability to produce high-quality PCBs. I got an old HP LaserJet from work, modified a laminator for toner transfer, and switched to hydrochloric acid with an aquarium air pump for etching. This gave me the best results, but still not very good. Then the developer came out of the printer. We bought a Brother Laser for daily printing, and found that Brother toner is useless for PCB transfer. If I decide to buy a new developer and try again, I still have HP. But it takes up a lot of space, and I want to get rid of it.
At the same time, I started learning 3D printing, building my printer from scratch, and being inspired by it, I started collecting parts to build CNC. I want to skip chemicals and try milling. If this works for me, then this big printer is gone!
Therefore, the one-time cost is prohibitively high. Yes, but as I wrote in another article, the magic seems to be in stickiness. I guess we can build our own machines to apply it. But yes, goo is a bit too expensive, but it's not as problem-free as buying a machine.
Remember the people who claimed to use only vinegar and salt to wipe off unwanted copper? Yes, what I managed to do was turn a wooden board into green, which required soaking for several days!
Is that the pilot of 5 Minute Crafts?
Well, if you watch the video until the end, he just sends the nonsense back to the seller. I think in a few years, I can spend very little money to buy it as scrap.
I was the original Kickstarter funder and waited patiently for a year or so until they solved the production problem. They initially promised to use 3D printed insulators to insulate overlapping vias, but they didn't realize it, so they came up with a solution for double-sided drilling and rivets. Okay, great, but then they wanted to charge me extra drills and rivets. I initially refused, but later decided that this exercise might come in handy. I have contacted the company many times to try to buy drill bits, but they let me down. They no longer even offer it as a separate purchase.
I only used this thing once. Although the circuit board worked initially, one of the vias broke.
Honestly, looking at the final product, it does look like a dodgy kickstarter idea...
Oops, I should talk about routing instead of vias on it.
This is a very good idea, but I haven't really seen the market for this thing. As already pointed out, the cost of producing circuit boards is quite low and the speed is quite fast. If you are a company that wants to iterate every day, it may not be fast enough, but it may not be fast enough for hobbyists. If you can't wait two weeks or so, I suggest you choose a second project or other hobby to pass the time. ;) So this makes this printer suitable for situations where you really need a fast PCB. It's great that everything seems to be well integrated, but other than that, I was not completely shocked. In my opinion, the biggest problem is the lack of ability to produce fine lines. There are many ways to get a circuit board with thick traces, but making thin traces at home is a pain and requires optimization of the process. PCB milling is a mess and it is difficult to produce fine traces with it. You can etch it yourself, but you need to spend time figuring out your craft so that you can get 6 to 8 million traces to work reliably. When etching at home, you can use more space on the PCB (because a large single PCB is cheap, but the panel space is not cheap) to lay out the traces in a way that does not require too many vias. The 0 ohm jumper is very convenient for cross-wiring. Of course, photo transfer and etching are a bit painful, and at these times, the school may never expose anyone to the required chemicals, but at least for amateurs, it allows you to design PCBs for fairly common components. However, this printer does not seem to be able to print 0.5 mm pitch TQFP or QFN PCBs. I personally think this is absolutely necessary for any form of serious prototyping, unless you are doing high current design. Almost all new chips will be packaged in 0.5 mm QFN packages, so in my opinion, this is something that the home prototyping process should be able to adapt to. This printer doesn't seem to do this, so I don't quite understand this. In addition, it takes a long time to cure the ink. The printer does not seem to have any protective measures to prevent drilling without the sacrificial plate or curing when the sacrificial plate is still present, without automatic removal of drilling debris, without camera assisted alignment, everything is Need constant intervention and care, etc. So there seems to be a reason for all the very low melting point solders used in the video, this solder is very expensive and does not have good mechanical properties. I guess you can print transparencies, expose the photoresist PCB and etch it in a shorter time than the printer produces it, and the start-up quality is better.
I saw this video a few days ago and thought it was too troublesome, too time-consuming, and too expensive.
Recently, I made several breakout boards (0.5"sq to 1"sq) for soic chips using toner transfer. I used Inkscape (compared to Kicad, I'm lazy), Brother laser printer, smooth white peeling and pasting cabinet paper, and my Weller iron and 1/16-inch soldering iron tip. I can go from printing to soldering board in <1 hour.
My process: Print the reverse image on standard paper, then peel off a piece of glossy cabinet paper and stick it on the image, and then print again. Cut the image to the size of the circuit board, and use masking tape to fix the paper around the bare copper plate. C. Place the wooden board under the rag and heat it with an iron for about 1 minute. d. Pour cold water on the board and gently pull the paper back. e. Use copper wool to remove toner under running water. F. Clean the circuit board with ipa and tin it. G. Application components.
Last week, I purchased a hot air station and then frantically took out SMD chips from the discarded circuit boards I collected. I managed to create a breakout board for a 24-pin 0.65 mm pitch DAC chip with 12 million traces. After tinning with hot air, I brushed the flux, then placed the chip on top of the crowning marks and used hot air to sink it.
It's just that I won't pour ferric chloride into the sink anytime soon.
One of the closest subjects in my mind. I hope that the printed PCB process will work, but that is not the case. Not far.
My vision is a device that can print resistors and can cross traces on other traces (with insulation). Double-sidedness is really unnecessary. Print a real ground plane (conductor/resistor trace on top of insulated trace) on bare copper. Pick-n-place will be integrated. Place a component with glue. Print the trace to it. Maybe it will reach BGA in time. Although a lot of improvements are needed, all technologies already exist.
In terms of its value, I have completed a large number of PCBs through toner transfer and will never go back to that state again. For $20 a week, I can get 5 copies of my own designed professionally produced double-sided panels. And my home does not have a mini super fund website. Milling is slightly less annoying (and harmful), but I did it just to quickly prove the concept.
I am really tired of product implantation of mediocre and overpriced PCB robots on HAD.
Making circuit boards and making them at home are not the same thing. If I need more than a few or want to sell products, are prototyping for larger runs or want a beautiful circuit board, then the fab is the best. If I am making a single-sided, 20 mil thickness, 20 mil gap rule, and then milling on FR1 with an asymmetric 10° tool, I can produce a functional board in two hours instead of two weeks. Turning the circuit board in the cut profile is sufficient for screen printing with a laser diode on the same machine, and then liquid tin is used to improve solderability and corrosion resistance.
I just wish we had an online manufacturer in Australia. All Chinese products are great and cheap, but a PCB worth $5 requires a shipping fee of $16, and you still have to wait up to 2 weeks to see it. I will give an Australian manufacturer an Inc shipping fee of $30 so that it can be delivered to me within 1 week... Assuming Australia Post is willing to cooperate that week
I have used this machine in my university. The only good thing about this machine is the drilling part. In my school project, I only encountered this problem. It produces a lot of noise, is difficult to calibrate, the dispensing needle is easily broken or clogged, the reflow soldering effect is not good, and in most cases, the circuit board cannot work after reflow soldering. In addition, all accessories are very expensive. The cost of the dill part alone is about US$700, not to mention that the tip and solder can also cause trouble. It's like they have micro (macro) transactions built into this thing.
Change to a PCB factory. Actually it works very well, there are some good ones on the market.
Please be kind and respectful to help make the comment section great. (Comment Policy)
This website uses Akismet to reduce spam. Learn how to handle your comment data.
By using our website and services, you explicitly agree to the placement of our performance, functionality and advertising cookies. learn more