New method developed to 3D print fully functional electronic circuits | Hacker News Reader

New method developed to 3D print fully functional electronic circuits

nottingham.ac.uk

43 pointslevlaz20 comments

Open articleView on HN

It's not clear to me how this compares to the work from Voltera [0]. Can someone with more expertise weigh in?

[0] - https://www.voltera.io/

The primary difference is that Voltera is heat cured, this is UV cured.

Differences:

Voltera: Ink comes in syringe. Printed via standard dispenser Ink is heat cured

New Technique: Ink is printed with inkjet head Ink is UV curred

Both use off the self technologies besides the ink. Voltera uses standard paste/syringe dispensing tech, this uses standard UV cure inkjet printing technology.

I know the Dragonfly 2020 printer (developed by the Israeli company Nano Dimension) has been in development for quite a few years which (seems) to use the same method as described in the link.

The method itself isn't even that novel since it's the same the as Stratasys' polyjet printers, only with specially formulated inks. There's probably prior art to that as well, but I know Stratasys has had polyjet printers for quite a while now.

While I see the advantages there, it continues the trend of increasingly unrepairable components...

But why repair something when you can print a new one?

Assuming that wasn’t sarcastic, for one increasing consumption isn’t good from a sustainability standpoint.

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.

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.

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.

I believe the resistance is also high for these inks which limits the use cases somewhat.

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.

~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.

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).

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.

The article says, "Dr Ehab Saleh and members of the team from CfAM found that silver nanoparticles in conductive inks are capable of absorbing UV light efficiently. The absorbed UV energy is converted into heat, which evaporates the solvents of the conductive ink and fuses the silver nanoparticles."

I don't think this will be economical if it uses silver as the conductor.

I think the quantities will be relatively small. Many electronics contain gold (~100x more expensive than silver), and as far as I know that hasn't really been a problem.

I think the main interesting thing here is that one can build bulk circuits .... many, 100s of, embedded layers deep as you build a 3D object .... the next step is obviously an integrated pick&place that embeds components in something as it's printed ... and of course the step after is something with enough precision to lay down say 2u FETs ...

Embedded components could definitely be a benefit. This sort of thing is already somewhat possible (e.g. PCB cavities) but has limitations.

Another interesting possibility would be designing complex surfaces that components can attach to rather than the traditional top/bottom of a PCB. Again this is already somewhat possible with techniques like 3D-MIDs (molded interconnect devices). 3D printing would make this sort of thing far more feasible though as MIDs are inherently limited to traces on the surface, meaning you're limited to a single layer.

Pick and place machines would need to get rather advanced for this to work as well. I know there are a few out there to handle MID orientations, solder paste extrusion, and laser soldering, but from what I've seen they're quite slow.

Of course heat dissipation and electromagnetic effects could become interesting design challenges (or design patterns) if the geometries are complex enough.

The article mentions a video but I can't seem to find it. If anyone finds it, please share it.

Seems like its in here

https://www.tctmagazine.com/3d-printing-news/university-nott...

Reply on news.ycombinator.com