Welcome to the Silicon Valley Neighborhood: Nano Dimension Arrives in California
I-Connect007 Technical Editor Dan Feinberg accepted an invitation recently to tour Nano Dimension’s new USA headquarters in Santa Clara’s Silicon Valley, which included a sit-down with President and Co-Founder Simon Fried.
Dan Feinberg: Simon, your new facility and office look great. You have a warehouse with inventory, a demo lab, maintenance and conference areas, and you are up and running. I just saw the machines that you have in inventory that are ready to ship to customers. How are sales going in the U.S.?
Simon Fried: In the U.S. there are now systems on their way or already delivered to customers in the East, West and other areas of the country. Some of those machines are for customers that were part of our beta program, as some beta users have already ordered commercial machines, and others we expect will follow. It is a great vote of confidence. We are now also working with excellent channel partners such as FATHOM in Oakland, California, and TriMech in Boston.
Feinberg: Ordering commercial machines instead of presently used betas is a vote of confidence. They had the beta, they know what it does. It's not a case of you sold them something that was not ready.
Fried: Yes. Many of them have spent six, 12, or more months learning how to use the technology. It’s advanced manufacturing and requires people to revisit some of their assumptions around what is possible; our technology changes that calculus. One of the key issues is for people to get their heads around is what DFAM—design for additive manufacturing—means for the electronics space. It’s revolutionary.
Using our technology for traditional PCB designs is straightforward. It's a more direct transition as you use traditional design software and formats like Gerber and Excellon. Our goal is to not introduce any difference to these traditional PCB designers at all. So, there they can stick to the design rules, and feel at home.
But additive manufacturing of electronics enables new things to be designed, manufactured and tested, and different workflows to be adopted. The type of design and the kind of potential that non-planar, non-Gerber electronics design introduces is something entirely new—things like coils or shaped electronics. There's some thinking that needs to be done around how to design it and how to optimize it for things like print speed, or print accuracy, or which direction you want to place the part on the print tray. So, on the design for additive manufacturing of non-planar parts there's more of a learning curve.
Feinberg: You are referring to those that you've sent out press releases about, but there are others that you can't talk about yet. We just know they're coming.
Fried: As you know, the system was launched officially back at productronica in November 2017. We have installed systems in places as far away as Sydney, Hong Kong, Singapore, France, Germany, and a few other locations which we've not yet specified. We have made announcements regarding some specific customers such as Safran in France who are using the system. Other customers, including some in the United States, we aren’t in a position to mention directly for non-disclosure reasons. In time, things will evolve and we’ll be able to mention other customers by name. We're also doing some interesting work together with Harris Corporation and Space Florida looking at RF applications for satellites.
Feinberg: Well, this process is ideal for that. Do you mean for type three antenna?
Fried: Yes, our technology and materials certainly have interesting applications for antennas.
Feinberg: I remember looking at the 3D printed boards that we were shown at CES four years ago, and they look like pictures of circuit boards rather than real PCBs. These units we are looking at now look like the actual devices.
Fried: Right. I think that pretty much reflects things. The journey is one of very complex multidisciplinary integration. And, in the beginning it was very much about, "Can we get these two materials into one location?" And, then, "Do these materials have the right electrical properties?" And then, as you can see, the parts start to come together. Then you must have the right material and be able to assemble it.
Feinberg: Yes, this has come a long way. Well, one of the things I wrote a year or so ago was that I can see that almost any circuit board company—not just major ones—will continue to be print, etch, plate, etc., on most of their production for the foreseeable future, but I can see where they will eventually need 3D printing capability—one production line at least.
Fried: I think there's no question that a certain proportion of low-volume, high-mix type production—and I say production, not just prototyping—is certainly going to go this way. It's nothing that's going to happen in a massive way this year or next, but it will happen. If I look at the traction we have gained it's generally in areas such as defense and military, but opportunities for rapid adoption exist in many industries.
Feinberg: That is interesting.
Fried: What most of our customers are doing is time-compression of the R&D stage. It is speeding up prototyping. And, some are looking at new applications because there are design considerations that come with constraints that you can't meet with traditional manufacturing. So, we see a divergence where at this stage some users are turning to the 3D-printed PCBs to change work flows and speed up production, and others want to push the envelope further. There are certain things that appear simple but would be very hard to make without an additive manufacturing tool. We have folks come to us and say, "Now that you can do this 3D kind of work I need something that's basically just a small bracket, a connector, or an unusually shaped antenna and I don't need anything more complex than a few plus and a minus connections. I cannot make it in any other way so that it's going to fit in the space that I need to put things in." For now, if the volumes are not extremely high and the part is a good size for the printer then that's a discussion that must be considered for additive manufacturing because you can't make it any other way.
Those who get into it with an eye on the non-planar work, they have at the back of their mind the use of this as a manufacturing approach. So, it filters things. You know, you're not going to get people who come through the door saying, "I need to make five million of these." But, if you need to make few thousand over a longer period, maybe more than that, then this approach is possibly the only game in town.
Feinberg: You have machines in the warehouse getting ready to ship right now, so the more that you have in the field, obviously, the consumable sales go up. Are you thinking about anything beyond the present consumables, besides the dielectric and the conductor?
Fried: There is certainly a roadmap on the material side. And, it's part and parcel of the release of the technology. At the moment, we're one step better than Henry Ford in that you can have any material you'd like as long as it's one of these two. It’s reasonable to expect additional materials to follow in time.
Feinberg: Not just black.
Fried: Exactly. So, now you've got the two “colors,” conductive and dielectric. In the end we want to be customer led and deliver materials that meet quite a broad range of customer needs.
Feinberg: And you’re a step ahead because you have the new Dragonfly 202 Pro which is set up with the cartridges of dielectric, cartridges of consumables. Let's say you had a third consumable for some reason, maybe a different conductor. Maybe instead of silver it’s gold or copper. Could you then just plug that new consumable into the same cartridge with the existing plumbing?
Fried: Well, it's not as quite as easy as that. You see, everything that we make now is with the consideration of ability to inkjet print. Every material that you introduce must, in the end, work with the other materials. So, when it comes to things like adhesion and cohesion and how they react to heat, they must be adapted to one another. More printheads and more material options would certainly allow for a greater set of applications. Swapping inks in inkjet systems isn’t a hot-swap procedure.
Feinberg: Correct; that could become a contaminant.
Fried: Like other 3D inkjet machines you cannot flip from one ink to the next in a matter of minutes. But, if you have a project that requires other electrical properties for example, or other mechanical properties, you may say, "Fine. I'll run it for those kinds of jobs this week and next week we'll run it for the other kind."
Feinberg: We'll flush it and run for a week, and flush again. And, you get to a point where they flush it often enough they're going to say, "Let's just get another machine."
Fried: There are two things that we're doing on that front. One is planning our materials roadmap, and there's definitely a need for materials, just like the PCB industry. In a sense we're just mimicking what is done in traditional electronics, there is an industry need for many different electrical properties. Do we need to look at materials that are better suited to flex or bend-to-fit-type applications? The ones we have can do it to an extent, but additional mechanical properties may be required for specific applications. There are mechanical variations that we're working on. And there are dielectric variations that we're also working on.
For board assembly we currently recommend that people use low-temperature solders when they're soldering components to our boards. It's easier and more beneficial for the customer to have the assembly process be similar to a standard RoHS process. Most of the companies we work with use batch-size assembly for these boards but if they happen to have an assembly line or their regular supplier does, then it’s important that downstream activities are compatible with our prints. Currently there is a need to use low-temp, but it is certainly made up for by the time savings our technology offers.
Feinberg: This is like the inconvenience of 10 years ago when people were converting from leaded to lead-free. A 10°C or so pop in temperature was required. You know, different levels of oxidation would happen with time. Now, that's kind of the standard.
Fried: It's very similar. Now you talk to people about doing anything other than RoHS and they don't want to move the dial on anything. These are issues that the industry knows how to work with.
Feinberg: That's right. There are still some advantages to leaded solder that people forget about, ductility for one.
Fried: I've had quite a lot of interaction with people who are looking at assembly and different solder providers. It seems as if the whole assembly space is splitting in two. On the one hand you have companies that are pushing for lower temperatures, because today's components are more sensitive. Other areas, notably automotive, are interested in higher temperature assembly.
Feinberg: So, go back to leaded solder?
Fried: They're going below leaded temperatures. They're moving to tin, bismuth, indium, and I believe it was Lenovo that launched a laptop which was entirely assembled using tin-bismuth solders last year or maybe a year and a half ago. I think lead-free solders have their risks around brittleness and so on. So, they worked things out to the point where a consumer product that's assembled with low-temperature solders is out there and it's performing just fine. So, that meets our current material set quite nicely. Folks are already doing this for their own reasons, so we can tag along. Then you have the other end of the spectrum, the automotive guys. And you see the value here in Silicon Valley particularly where cars are no longer just cars—they are rolling communications devices. So, you have the military and the automotive space and the aerospace guys rushing to make things that will survive any kind of thermal cycle. And, the rest of the industry wants to make things gentler for today’s sensitive components.
Feinberg: One of the things that we're covering in the coming year is what we're calling megatrends. I don't consider 3D printing to be a megatrend yet, but I do consider it to have that potential. That’s one of the reasons we've been following 3D printing. Nano Dimension is one of few companies that are focused on 3D printing for PCB fab. Others do 3D printing and they can also print some circuit boards, but it's not what they specialize in.
Fried: I was just at the CPES trade show in Toronto and there were two companies doing 3D printing of PCBs. One company called Voltera is based in Toronto—a university spinoff. They print onto FR-4 for the hobbyists at home. Great solution. Bam, your simple circuit is ready. But, it's single layer for the time being, not multilayer. And, there is another company in New York called BotFactory, who have a printer for electronics. They're both great companies and they're very focused on electronics, primarily for the maker community.
Feinberg: I've heard of them. My first introduction to 3D printing with anything was a gentleman by the name of Bert Ohlig. His company made circuit board manufacturing equipment. Back in the ‘90s, when I was at Dynachem, we worked with them because they were making circuit board printers and we were supplying photoresist. He was showing me some 3D printing and he said, "You just wait and see. Within 20 years this is going to be a major thing. It's going to be used, maybe even to make circuit boards." I think this was about 1997, a little over 20 years ago.
Fried: That's amazing. You know, I always look at what's happening in the traditional 3D space, because it's pretty much a guide for what's evolving on the electronics side. Last year was the year that that industry shifted from talking about time compression and rapid prototyping to people buying machines and delivering products made with 3D printers. Whether it is polymers or metals there are enough stories out there where if you find the right application that solves enough of the problems in a traditional manufacturing process, then it's a no brainer to go additive.
3D is not for everything and it's not with any material. It's not for every type of customer. But, there are material sets that are available now, and the speeds that these newer, standard 3D printing machines can achieve has moved the needle. And, that threshold of where, "How many of these do I have to print before it starts to make sense for me to switch to injection molding, or higher volume traditional manufacturing?" is expanding, exponentially. What used to be 100 is in some cases today 20,000 or 30,000 units.
Feinberg: Circling back, this facility looks great. You now have a real presence in the States. And, it probably helps with the U.S. customers.
Fried: I would say it's a fundamental difference. These systems have been here for just two or three weeks from when we opened. And we're already seeing a lot of people coming just to look, to kick the tires, to get a sense of how it works. And, that makes it clear that the Bay Area is without a doubt the place for us to be. From here, we can throw a stone and when you go to pick it up you're at a potential customer's site. Many companies and industries are within fifteen minutes of here.
Feinberg: Plus, you have things like trade shows. People come to this area for trade shows and you can say, "While you are here, drop by. We're just five minutes away."
Fried: Yes, we're seeing people book trade shows and then come by.
Feinberg: Well, Simon, thank you for the tour and the time to discuss Nano Dimension’s progress—which is very impressive since we first met at CES a few years ago. I appreciate the opportunity to maintain a continuing dialogue as this new way of making circuit boards becomes more mainstream over the next few years. I look forward to covering your growth here in the U.S.
Fried: You are welcome and thank you for visiting.