3 ms·

Very exciting. This will revolutionize the hobby industry. No more etching. It would imagine it would be an easy exercise to build a machine with an integrated

by teilo 9y ago

Very exciting. This will revolutionize the hobby industry. No more etching. It would imagine it would be an easy exercise to build a machine with an integrated drill, so you can go straight from printing to soldering. A PCB will literally be a PCB.

dragontamer 9y ago
In my experience... no. A modern PCB has the following: 1. Copper Traces 2. Copper-plated Through Holes (usually for Vias, but useful for through-hole parts as well) 3. Solder Mask (very important for surface mount parts) 4. Silkscreen ("Helper Text", marks that say R1 or "R2 goes here"). Given that I can order 3+ custom boards from USA suppliers for $70 or so (process in 3 days, then ship it so I get it within a week), or 10+ boards for $50 if ordered from China (takes a few weeks to travel though)... I think the modern PCB market is to just order from online. "Immediate Printing" is well served by the CNC Mill industry today. But CNC-milled boards just aren't good enough for production purposes. Even if you're a hobbyist, all the good tech is in SMT parts which really need a good soldermask. ----------------- What the technology discussed in the page will be used for is additive manufacturing. That is: the creation of new structures that are cheaper to do by additive means than subtractive means. I can imagine a small circuit of size .5mm x .5mm for example that would be more efficient to "print" than to mill or etch. I think their example of 3d Antenna built into a plastic board is the real idea behind the technology. Additive manufacturing is dramatically different than subtractive... and subtractive manufacturing is standard for a reason. The idea is that "Additive" methods can create things very different than subtractive methods. Don't try to emulate the subtractive methods through additive means. But for cheap prototyping, CNC Mill + Copper Boards is fast enough, and ordering a "professional" board in the USA or from China is cheap and easy enough as it is.
- elago 9y ago
  I also am satisfied with what the pcb industry provides anyways, but couldn't this 3d printing technique produce more than 2 layer boards? That seems like at least one potential improvement beyond milling a copper board. I do wonder how well the non-conductive material handles the heat of soldering though. I think it could be cool to combine the circuit board and enclosure into one thing, even though that's terrible for serviceability.
  - tyingq 9y ago
    I believe the resistance is also high for these inks which limits the use cases somewhat.
- new299 9y ago
  The prices you quote are pretty expensive for 2 layer boards at least. I should ordered 3 Arduino sized boards from OSHPark (US fabricated) for 40USD. From China I would probably get 10 for about 20USD. CNC Mills kind of suck. The main issue is alignment of the top and bottom layers, the lack of vias, and that even with the additional pain (and cost) the quality isn't great (no solder mask is the main pain point). A nice, simple, PCB printer would actually be useful. But they've a long way to come and I think it's quite a niche product. They'd need to produce at least 2 layer boards with vias for me to consider using one. Ideally they'd also have solder mask. I think it's quite a niche play though. There are many differences between 3D printing and PCB printing, which make 3D printing more interesting/viable.
  - StavrosK 9y ago
    ~10 boards from China cost $12 with shipping, and the quality is quite good. I've ordered multiple times, and have been satisfied every time.
    - new299 9y ago
      Sounds about right, I've found shipping to my location a bit higher than that. I've also found that some suppliers are better than others. I've had suppliers not print silk on 50% of the boards in batches for example (this was when order a batch of 100).
jack_h 9y ago
Perhaps given enough time it could. 3D printing in general is still rather lacking in the hobby industry. FDM/FFF isn't a terrible technology, but it does have a lot of drawbacks. It's main benefit is it can be implemented relatively cheaply and easily. The technology linked to in the paper appears to be the same (or very similar) to Stratasys' Polyjet technology and those machines go for tens of thousands of dollars on the low end. I do see this being useful for engineering teams though. Turn times on PCBs are either long or expensive and being able to quickly iterate on designs can be very valuable for a project.

by teilo 9y ago

dragontamer 9y ago

In my experience... no. A modern PCB has the following: 1. Copper Traces 2. Copper-plated Through Holes (usually for Vias, but useful for through-hole parts as well) 3. Solder Mask (very important for surface mount parts) 4. Silkscreen ("Helper Text", marks that say R1 or "R2 goes here"). Given that I can order 3+ custom boards from USA suppliers for $70 or so (process in 3 days, then ship it so I get it within a week), or 10+ boards for $50 if ordered from China (takes a few weeks to travel though)... I think the modern PCB market is to just order from online. "Immediate Printing" is well served by the CNC Mill industry today. But CNC-milled boards just aren't good enough for production purposes. Even if you're a hobbyist, all the good tech is in SMT parts which really need a good soldermask. ----------------- What the technology discussed in the page will be used for is additive manufacturing. That is: the creation of new structures that are cheaper to do by additive means than subtractive means. I can imagine a small circuit of size .5mm x .5mm for example that would be more efficient to "print" than to mill or etch. I think their example of 3d Antenna built into a plastic board is the real idea behind the technology. Additive manufacturing is dramatically different than subtractive... and subtractive manufacturing is standard for a reason. The idea is that "Additive" methods can create things very different than subtractive methods. Don't try to emulate the subtractive methods through additive means. But for cheap prototyping, CNC Mill + Copper Boards is fast enough, and ordering a "professional" board in the USA or from China is cheap and easy enough as it is.

elago 9y ago
I also am satisfied with what the pcb industry provides anyways, but couldn't this 3d printing technique produce more than 2 layer boards? That seems like at least one potential improvement beyond milling a copper board. I do wonder how well the non-conductive material handles the heat of soldering though. I think it could be cool to combine the circuit board and enclosure into one thing, even though that's terrible for serviceability.
- tyingq 9y ago
  I believe the resistance is also high for these inks which limits the use cases somewhat.
new299 9y ago
The prices you quote are pretty expensive for 2 layer boards at least. I should ordered 3 Arduino sized boards from OSHPark (US fabricated) for 40USD. From China I would probably get 10 for about 20USD. CNC Mills kind of suck. The main issue is alignment of the top and bottom layers, the lack of vias, and that even with the additional pain (and cost) the quality isn't great (no solder mask is the main pain point). A nice, simple, PCB printer would actually be useful. But they've a long way to come and I think it's quite a niche product. They'd need to produce at least 2 layer boards with vias for me to consider using one. Ideally they'd also have solder mask. I think it's quite a niche play though. There are many differences between 3D printing and PCB printing, which make 3D printing more interesting/viable.
- StavrosK 9y ago
  ~10 boards from China cost $12 with shipping, and the quality is quite good. I've ordered multiple times, and have been satisfied every time.
  - new299 9y ago
    Sounds about right, I've found shipping to my location a bit higher than that. I've also found that some suppliers are better than others. I've had suppliers not print silk on 50% of the boards in batches for example (this was when order a batch of 100).

jack_h 9y ago

Perhaps given enough time it could. 3D printing in general is still rather lacking in the hobby industry. FDM/FFF isn't a terrible technology, but it does have a lot of drawbacks. It's main benefit is it can be implemented relatively cheaply and easily. The technology linked to in the paper appears to be the same (or very similar) to Stratasys' Polyjet technology and those machines go for tens of thousands of dollars on the low end. I do see this being useful for engineering teams though. Turn times on PCBs are either long or expensive and being able to quickly iterate on designs can be very valuable for a project.