International Edition

  • Interview: Rudy Ghosh Ph.D., Novacentrix, Inc.

    Rudy Ghosh

    Novacentrix, an Austin, Texas based company that has been manufacturing pulsed-light curing systems to the printed and flexible electronics ecosystem for curing nanopowders, conductive inks. They have developed a pulsed light system that can reflow SAC 305 and other solders in a fraction of a second with a machine that occupies a tiny footprint compare to other reflow systems. Is this this the next big industry revolution?

    TG: Rudy, thank you for joining us today. Tell us a little bit about the background story to NovaCentrix, who you are and where you came from?

    RG: First of all, thanks Trevor for having us over here. I had been listening to a lot of the Tech Talk Thursdays, and it’s been very, very educational for me. So it’s an honor for me to be able to come and talk to you. So as long as NovaCentrix is concerned, we are based in Austin, Texas. We have been in business, as we like to say, since last millennia. We started in 1999, so pretty much last year of the millennia. But we have been in this business for 20 years. We have our headquarters in Austin, but we have both an R & D, an apps team support for our customers in Europe, and in Asia. We also have a business team that that supports our existing customers across the world.

    We started as a company in making nanoparticles using plasma synthesis. A lot of this work was done for the defense industry, but then when we started making these metal nanoparticles, we realized that there is a much larger market for this, in other areas, and one of those areas was in printed electronics. So how do you take these nanoparticles, and make them useful for the printed electronics? Well, you make conductive inks. So that’s been a major part of our business. And we have grown that business, both by growing the internal development for the inks team, but also through acquisition of other ink companies that were doing really innovative work. And that had customers who needed those inks.

    Once we make those conducted inks and we started putting them on pretty much any surface, like plastic, paper, fabric, but ultimately there is a very important step there in electronics, which is the thermal processing. And you cannot put PT or fabric in an oven, and expect it to survive. So what we did was take our experience in controlling pulsed power and made lamps out of it. And so these are very, very high-powered xenon gas filled lamps. So they pretty much emit white light. Anything that is dark in color absorbs this light and gets hot. So that’s how we were doing it.

    But the real key part of this is the pulse part. The pulses that we’re doing here are in the microsecond range, so your tracks get hot, but the substrates underneath does not. So that’s where we are on the printed electronics side and it is what we have been doing for the last decade and a half. But in general, the printed electronics industry really wants to be synonymous with conventional electronics industry. A very important point on that is that to be able to attach components onto the substrate. For decades, soldering has been the way of attaching those components. And so that’s our next step in our evolution, and that’s what we want to talk about.

    TG: Yes, and that’s a massive step. Your background has essentially been in curing using photonics curing with pulsed light, but now you’re moving into solder reflow, and in particular reflowing regular SAC305. How did you get to that point? And tell us a little bit about how that works?

    RG: Great question, because it’s not an easy… Exactly what you mentioned, it’s a massive step. And this would not have been possible if we hadn’t done some really, really fundamental changes in the architecture and design of the tools that we have. And we did that two and a half, to three years ago. And that really allows us to tap into peak powers and average energy densities that weren’t possible for the single parts. No matter how much energy you bring in with a single pulse, you’re not going to get reflow processes done in one microsecond parts. So a very important part on that was being able to stitch multiple pulses over a longer period, moving from the microsecond, to the seconds range allowed us to go away from, not only doing curing, but also solder reflow.

    TG: So it’s quite a change in your process. What are the main advantages of curing solder, using your process?

    RG: There are two main advantages. And the first one goes back to what I was just saying earlier, is the length of the process, but it’s a long process by our standards. It is still in the seconds. So one of the biggest advantages that we have is the really, really high throughput that we can get. And the second one is we are still using this broadband, wide area, light source. This allows us to have spatial selectivity, both in terms of, in the Z direction, which is what we have done for ages, but also in the XY direction. So that allows us to really address one of the key challenges that that’s there in the SMT process, is controlling the Delta T, across either the X, Y, or the Z direction. And that’s basically been the two main advantages.

    And if you think about it, now that has allowed us to come up with two versions of our tool. And one of them is really focusing on the high throughput. So more directed towards this high-volume, low-mix sort of market. And the batch process tool, which can also be used for R & D applications, which is tailored more for that very high-mix, low-volume market.

    A lot of the initial interest that we have gotten so far has been mostly from that high-volume, low-mix side. A big market space that we have been able to look into is LED manufacturing, like multiple single components in areas, or large areas, that need the same processing over and over again. And so doing it on a variety of substrates, including ceramics. But also one of the things that has really come in the last year, or so, onto the market is this roll-to-roll manufacturing of LED arrays. Previously, it was either some sort of conductive adhesive, but if you’re going to locate this large array of LEDs outside, you want the best possible reliability on your joints, and only soldering can provide that. So that’s been a key market for us.

    TG: Looking at the regular SMT process, the reflow oven has always been the determinant of the speed of the line. It’s the slowest machine, it determines the speed… Using your system, first of all, the footprint is, when you look at the actual form size of it, is much, much smaller. So it takes up a fraction of the space of a reflow oven. But, does this allow then, this process allow the line to go faster?

    RG: Absolutely. So we basically take the slowest part of the line and make it the fastest part of the line. So that, by itself, is incredible. Anecdotally speaking, for this specific 0603 LED on substrate with etched copper, with an ENIG finish, I’ve been able to do that. Solder reflow and produce a very reliable joint that looks exactly like something that has gone through a conventional oven, in a quarter of a second. A quarter of a second. I’m talking about a single LED, but since this is wide, it’s a broad area lamp, you can do anything within the panel, in the 3″ x 12″ size within that quarter for a second. It’s possible.

    Now the other important thing is going back to the footprint. Right now, because the reflow processes are so slow, you actually have only one reflow oven per line. Now, if the reflow oven becomes really, really fast, and you can use multiple lines coming into a single reflow process, which is, in this case, would be our PulseForge tools. So, it’s saving a space in terms of footprint, not only by being small for one line, but it can actually take 10, or even more, conventional ovens, and replace them with a single tool.

    TG:  That’s quite incredible. So let’s look at the actual solder joint itself that you reflow. I mean does this affect the tensile strength of the joint or any of these other characteristics?

    RG: So that’s, again, a very important question. And as long as it is a rigid substrate, like FR-4, or ceramic, or metal, the tensile strength is pretty much exactly the same that you’d expect from a conventional reflow oven. But again, going back to a lot of the work we have done is also on flexible substrates. And in all of those, what we are seeing, the failure mode is never the solder joint, but the adhesive that connects the tracks to the flex circuit. So either the etched copper on polyimide or on PET. So that’s been very, very encouraging for us.

    There are a couple of other things that have really been very encouraging for us. The fact that we are controlling these pulses, we really have very, very good control over the peak temperature, how long we keep it at that peak temperature. That really allows us to control the thickness of the intermetallic layer very, very precisely. We can control if you want to a one micron thick intermetallic layer, we can get that if you wanted. Six microns, we can get there. Anywhere in between, we can get there.

    The other important part is the cooling is very, very fast. As soon as you turn the light off, you’re taking off all energy sources. The tool does not remain hot. Your cooling rate is very fast, which means your microstructures become very, very small, and that creates a lot of tensile strength in the joint.

    TG: What sort of feedback have you had, then, from the major solder producers? Have you been working with any of them?

    RG: Yes, we have. A lot of the initial work we did, on qualifying solders, was to include solders that our end customers bought. This is a new technology and we wanted to be able to work with existing mainstream solders and avoid telling customers “this is a new technology and you have to use this completely new type of solder or paste.”

    But ultimately, when we were doing this sort of qualification, we saw the same solders come back over and over again, because there are these two or four major manufacturers of solders, and a narrow family of their solders that go to pretty much every contract manufacturer. So we went and talked to these solder manufacturers, we qualified those specific solders. By doing this, we really built up good technology partnerships with them. And we have been able to see if they can make any new sort of solders that are optimized for this process.

    That being said, we are still working with a wide variety of solder alloys. We have done it from low temperature solders to SAC 305, which we work on a regular basis, as well as high temperature solders.

    TG: You mentioned something there about the low-temp solders, so you’ve had some experience of reflowing these, no issues there?

    RG: We do a lot of the work with tin-bismuth, tin-bismuth with silver, and that work comes from our inks team. A lot of the work they do is making screen printable solders. If, it’s with the SAC alloys, at the reflow temperature, that we need to take it to, there is a lot of dissolving off of the silver. So that’s not really an ideal solution for us. So that business unit is actively working with the low temperature solders. And we’ve seen very, very good results. We’ve also seen very good results with the low silver-content SAC alloys, like SAC105. So we have been able to really look into those from a very selfish, company perspective. But now we have actually been able to qualify those and provide it to our customers.

    TG: Well, it certainly sounds fascinating technology really. I mean, now, are there any downsides to it?

    RG: Yes, of course there is. Like this is not a solution for everything. The first thing that I said is the spatial selectivity, and the spatial selectivity is usually done by choice of color. So if you bring in a board with a black soldermask, that’s going to be a problem for us, for sure. The soldermask is going to do most of the absorbing and everything else is going to get as hard or harder. But so far we have been very happy with the feedback that we’ve gotten from our customers on that. If they have a dark-colored soldermask and we think, “Well, that’s not really going to work,” they’ve been very accommodating and changed the soldermask color. For example, if you make it green, that’s a large part of it is getting reflected. So that’s been pretty much the main issue here.

    And the second one is not necessarily a technological challenge, but a challenge for getting into an industry that really values maturity of technology. So we understand we’re bringing in a new technology, so we are working with partners to educate customers and customers of customers. And we will continue to do that. One of the things that we are very proud of, as a company, in NovaCentrix, during our time and printed electronics, is we do not really consider our customers as just customers. They’re our technology partners, they’re our advocates for our technology. So it’s never been, “Sell a tool and get out of the way.” It’s always been, sell a tool, make sure the customer is successful, and then get out of the way, till they call you back.

    TG: Right. So, how do they tweak the process? Presumably they have to gauge the amount of light exposure, depending on what materials they’re trying to reflow etc. How did they operate that and does the machine actually gather that data to store it, so that you have traceability of what happened with these boards?

    RG: Again, this is very important. And let’s go about how we tweak the process itself. There are two main parameters here. It’s the average power, and how long you bring it up for. By changing the average power, you’re bringing in different amounts of power, so you’re basically changing the ramp rate of how quickly, or how slowly, your temperature increases. And for how long you raise that is going to be controlled by the amount of time. Now, once you reach your peak temperature, you can then drop that average power, and you can flatten it for a soak. The thermal profile can be made to look very similar to what you’d see in a conventional oven, except it’s going to be condensed, instead of being in the many tens of seconds to minutes, you’re doing it within a few seconds at the most. That’s how you control the processability.

    Now, since the very beginning, a really interesting part of how we build these tools, a lot of our initial tools for curing were for production and not for our R & D or lab skill pools. So a lot of the questions about traceability of these pulses, of knowing exactly what each of these components have gone through, has been very important for more than a decade for us. So we have been working on that, and that is built into the system. I work with this one tool that’s, once in a while, that’s pretty old, it’s beat-up and old, like six years old. But I can still go through the logs and find out what was done six years ago, on middle of December Wednesday, and get that data out. So it’s something that is hardwired into those tools, but we also work on the integration side, with our customers. If we need to integrate everything to their line, and be able to make sure logging from their other units also work with us.

    TG: Yes, this factory of the future stuff coming in. It’s critical that any new production equipment has that capability and traceability. So how safe is it to use?

    RG: This has been a very important factor for us. There are safety interlocks everywhere. If you take a step back, at its peak, the amount of light energy that we’re putting down in that one microsecond to the hundred microsecond timeline, is more than half a million suns shining at the same time. So it’s a very, very bright source. So we need to make sure the users are completely, completely safe, so that there’s a lot of light shields in both our batch and inline tools. There is, on both the inline and batch tools, multiple areas where things are moving, so there’s E-Stops everywhere, close to where the operator can be.

    But ultimately the tool is designed so those E-Stops are redundant rather than necessary. That’s how our tools are designed. It has all the certifications that are required across the world. We sell a lot of tools in Germany, we also sell a lot of tools in Japan and China. And we understand each of these places has different safety requirements. Because of our years and decades of experience on making these tools and selling them, we have a pretty good handle on the safety side.

    TG: Good, absolutely fascinating story. What’s next?

    RG: The mantra of our company has always been, “Inspire, Innovate, Deliver.” So that’s what we continue to do. On the soldering side, we will continue to engage with the community, try to see what the requirements are, what is the best fit for members of that community, with our technology, and try to really actively participate with those members.

    That being said, there are other things. One of the things that we are very proud of, that we’re working on, is this contactless printing. Earlier, in one of the discussions that you had, there was this talk about the necessity of being able to put down different thicknesses of solder across different areas. How do you do that? And we think we have a solution, and we’d be very happy to talk to customers who have that specific need, and be technology partners for them.

    TG: The problem is obviously getting the speed to be able to put down these things, because at the moment, you’re using stencil printing, which is a very old technology, but it’s also very fast…..

    RG: We will do it in the microsecond range. The speed part has always been very, very important for us. So, and again, without trying to sound cliche on this, we are using lights, so why not use the speed of light?

    TG: Yeah. Absolutely fascinating discussion, I’ll tell you, honestly. And a great way to finish up our year here in Tech Talk Thursday. Really, I want to thank you for joining us. I want to wish you every success with this new PulseForge product.

     

  • Subscribe today to the industry leading magazine and newsletter

    Keep up-to-date with information about worldwide trends straight to your inbox or through our printed magazine.