Flexibility Is Key to Direct Imaging Success
Miva Technologies has a unique perspective, supplying direct imagers to both the semiconductor and PCB markets. Brendan F. Hogan, managing director of MivaTek, explains that while each industry has its own set of requirements and difficulties, the lines are blurring between the two industries, demanding more flexibility from equipment suppliers.
Nolan Johnson: Brendan, let’s start with a quick overview of MivaTek.
Brendan F. Hogan: Miva Technologies is a company that started out in film plotters and chrome mask makers. This set up an interesting dichotomy: while silver film plotters were for the circuit board market, and the chrome mask makers were for semiconductors, the base technology is very similar. Over our last 20 years of business, we’ve now evolved into direct-write methodologies for those two separate market segments. We have direct imagers for circuit boards that can do 50-micron features, and for semiconductors, we direct write 1-micron line and space on complete wafer substrates. It’s a very interesting place to be as the industry evolves from what used to be separate markets to one; the lines are blurring between what’s a micro function and what’s a circuit board function.
Johnson: We certainly see that trend in dimensions being used for PCBs.
Hogan: Whether it’s feature size accuracy for high-frequency circuits—higher-resolution direct imaging produces more accurate feature sizes due to digitization—or the throughput you’re trying to achieve, the problems faced by the machine builder are pretty challenging. And that’s the broad range of customer base with which we work. If you’re dealing in high-volume markets in Asia, you’re constantly driven about throughput. So, can you move the digital imaging data through the system fast enough? The sheer volume of data movement starts to become an obstacle.
Johnson: Is that an even greater problem if you’re in a higher-mix situation?
Hogan: Yes. When you get to the high-resolution applications the expectation is less throughput when you’re doing one-micron features. We have advantages, though, because what high-resolution laser lithography systems can do in a few hours, we can do in a few minutes; however, it’s still very challenging. The throughput question is important, but it’s more about a technical capability rather than the throughput in a lot of cases.
Johnson: I see what you’re saying; it’s more about doing the imaging as fast as possible.
Hogan: Traditional semiconductor fabrication deploys steppers that reproduce a small image repeatedly on the wafer; we can image that entire 12-inch wafer as a panel, allowing a lot of design flexibility. It can be one complete circuit or an array of circuits; the direct-write technology allows for that. It’s a lot of data and capability. The circuit board side is an incredible market, but they’re all about efficiency, throughput, and high volume. By applying the principles of PCBs to microelectronics, we are positioned to handle a broad array of challenges from a different perspective.
Johnson: At this point, there’s development in direct imaging that is a turning point, especially for the PCB side. What should we know about direct imaging right now? What’s coming?
Hogan: The teachable moment for us was our relationship with the CHIPS Consortium at UCLA. This high-technology semiconductor research group is looking to the circuit board people for manufacturing methodologies because there are some unique issues; PCB methods are relevant to microelectronics and semiconductors, and the same is true the other way. The circuit board market could look at the semiconductor aspects and how they scale or the different thought processes they have; there’s a lot to glean. We’ve been in the middle of that, and we’re participating in industry groups on both the semiconductor and circuit board sides for many years.
We see commonality and a shift going on between the two markets. All of our circuit board machines have improved a great deal resulting from our work with the semiconductor applications. Cross-training among the groups could be very useful for most of our mutual clientele, in semiconductors or PCBs.
Johnson: What specifically should the PCB side be looking for in the form of new capabilities and approaches?
Hogan: Feature size control. There are some direct imaging systems that do come from imaging and are focused on the mechanical aspects, but can they control that feature size within a couple of microns everywhere on the panel? The performance of the circuit board is influenced by the precision of the imaging and the shape of the trace. It’s not enough to produce a trace that’s three mils wide. Does it have the right taper? Is it under control and repeatable?
With a lot of the technologies that we’ve introduced, like multi-wavelength control, we can control each of our wavelengths independently so that you can shape the trace. As frequencies get higher, the shape of the trace starts to influence the quality of the signal. Those are all drivers to the quality of the end product. With PCBs, too often, it’s price and throughput, such as “Can you get 100 panels out?” The focus isn’t as much on the result in circuit quality and the influence of that.
Johnson: That’s a completely different way of thinking for somebody who’s been working with PCBs. We have to start thinking about it in the way semiconductor has thought about it for years.
Hogan: The circuit board market is also tiered, and the companies with the most resources are already on the path to that. Those newly introduced to direct imaging have to become more educated. Those are the areas where the industry would benefit from a little more collaboration. Top to bottom, getting everybody to a more common standard in mentality and sharing expertise would be huge.
Johnson: While you’re creating this equipment that sits in the middle, what are the challenges for finding people who can help you?
Hogan: The staffing question is a challenge. Whether you’re talking about business development staffing or application engineering, you become a problem solver when you have new technology, and the ROI is pretty clean. We receive a lot of inquiries from radical thinking, whether it’s a chemical milling job where it wasn’t possible a year or two ago. We’re doing some interesting work in that segment. Having a staff member that is creative and has a broad enough knowledge to handle an application and think it through in an effective manner for the customer is important. The industry is aging, especially on the PCB side, so we’re trying to acquire young talent and get them the experience to think through these problems in a creative way.
Johnson: The PCB side might need more expertise than might be coming out of the semiconductor side; I’m thinking specifically about the pre-engineering phase. What is your assessment of those talent pools?
Hogan: A lot of our focus has been from other elements and processes within the industry because we’re trying to build a portfolio of talent in our organization. We’ve been pulling some people from the circuit board industry as opposed to equipment supplier because pulling in process knowledge and what the average consumer of our product goes through is a critical aspect of making your product better.
We don’t want to target our customers for talent, but in the résumé aspect, we usually look for somebody who has worked within the industry hands-on and has “green under the fingernails,” We want them to have encountered problems and learned how to reach resolutions; they should be able to anticipate what’s going to happen instead of waiting for something to happen, and then start thinking about a solution. We try to get ahead of those things.
Johnson: When you get the candidate around the smell of solder mask, and they smile, that’s when you know you have the right person.
Hogan: Right. I’ve been through the smell of tin lead-reflow, so now some readers out there know that I must be of a certain age (laughs).
Johnson: You’re bringing people in from these various disciplines to work on the machine design. What are your plans?
Hogan: A lot of the influence for us has been how we developed the product. When we first identified the problems of developing a lower-cost solution to direct imaging—direct write—we set course to convert from film and chrome mask-making into direct-write, and we collaborated and asked a lot of questions internally and externally. We were outward-looking and didn’t assume we knew everything. That’s how I got involved with Miva in the first place; they were looking for answers from a different experience set, and then I pursued a lot of other industry people to try to establish a series of options. I think it shows in the product that we have solved a lot of those industry problems.
Our product doesn’t require as much of a cleanroom as laser-based systems. Even on the micro side, we have machines running that are not in hyper-pure cleanrooms, and we still get a very high yield. When we put a new product out, we design with realistic expectations of the market to the greatest extent possible. We spent a lot of time and effort in temperature management because it influences the registration accuracy; a lot of our customer base is not going to be able to spend that money on cleanroom and controlling to a tenth of a degree and those kinds of conditions that are commonplace in the microelectronics segment. We needed to design with flexibility to extend the investment time horizon so that down the road a few years, as your market changes and your customers’ needs change, the product can evolve as well.
The problem in our industry, especially in say the North American market, is that they need a universal tool because, in a few months, they’re going to be building something that nobody has ever built before. If you don’t build flexibility into the product nor have the ability to do many things, then you’re selling both your customer and your market short; you’re not serving each very well. The capital expenditure is too large to have a narrow, application-specific tool.
Johnson: That seems to be the one takeaway we need to know. With direct imaging, you need to be flexible.
Hogan: Exactly. You can’t think about what you’re trying to do today and your moment-to-moment problem, you need to have a bit of a roadmap because the product has to last five to 10 years. It’s very expensive capital equipment. Most of the people that buy our machines are not going to have the opportunity to spend this much capital too many times in their careers. You need to have the vision to see what you will need in a few years. And what is the market going to need? They need that from the supplier, but they also need to do it in their equipment selection choice.
Johnson: We’re having this conversation at productronica; what equipment do you have here?
Hogan: It is Miva’s newest model—the 2400 Dual Tray System—six light engines operating with four wavelengths simultaneously and very high throughput. This particular system will be integrated with a fully automatic third-party solution for automation. We support many different forms of automation, but the real push in the industry—whether it’s micro or circuit board—has a lot to do with product handling and data collection, and our system supports all of the newest approaches to barcoding, etc. When you integrate with automation, the system then becomes a data collection tool, whether it’s the sub-part on the panel or every panel. Who touched this panel? What did the panel experience? It’s all about process parameters through process results. What are the influencers so that downstream when there’s a problem, they can look at data and have that analytical ability?
Johnson: The Industry 4.0 building blocks are becoming a cost of doing business.
Hogan: Yes. People are still trying to figure it out, including us. We’re communicating a lot with our customer base, trying to see what they want and need. Too often, perfect becomes the enemy of good, so we’ve built our system to be upgradable both in software and the tools; as it changes, we can move with it. When you get the eddy currents of change, it’s not always clear where they’re coming out on the other side. You can’t take in a piece of capital equipment like this and say, “If we only knew.” That’s not how we operate.
Johnson: Fantastic. Thanks so much.
Hogan: Thank you.