ESI’s Chris Ryder, director of Product Management, and Shane Noel, Flex Systems product manager, discuss laser vias for flex users and the increasing necessity for companies to collaborate early on and become more and more involved, whether that be in the product design, or with the process or base material manufacturers.
Barry Matties: First, can you start by telling us a little bit about ESI?
Chris Ryder: ESI is a company with a long tradition in electronics. The company has been around more than 70 years. In the last several decades, we’ve focused on laser-based processes. In particular, we are market leaders in using lasers for flex via drilling and other laser-based processes such as cutting, for example. While we’ve focused primarily on the UV laser regime, we’ve also recently made moves into other laser-based markets like HDI and ICP packaging. Of course, ESI has their foot in semiconductor manufacturing as well. We’re based in Portland, Oregon, but we have a global footprint. Most of our business stems from Asia; primarily China, Korea, Japan, and Taiwan.
Matties: When people design and integrate flex into their devices, what considerations do you think they should have with regard to manufacturing as it comes back into your area?
Shane Noel: The main thing that flex designers typically think about are traces, spaces, and via sizes. The incorporation of flex generally means you’re trying to conserve space, and conserving space requires you to design smaller vias, traces, and spaces. That drives the selection of materials. It’s typically based off polyimide and copper, although we see exotic material sets come into play more and more, especially with the advent of 5G technology. In general, you think of flexibility, of course. That’s important as well as reliability over the long term.
Matties: Do OEMs come to a process team like yours to talk about their flex designs, or do they just leave it up to the fabricator?
Ryder: That’s a good question. What we see is the necessity at this stage of the process for companies like ESI to be more and more involved in the product design. We also see a tighter requirement in terms of design rules between the material manufacturers, board designers, and process manufacturers. There’s a growing need for us to be involved in that discussion.
When you look at the link between the chipset, substrate, interposer, motherboard, and flex elements that go into that total construct, it becomes so dense that you have to imagine a room full of people packed so tightly that everybody has to talk about their intention to move around at every given point in time. That’s where we are in terms of the necessity to coordinate what we’re doing for the success of the actual device.
Matties: We’re talking also rigid-flex here, and you also mentioned the HDI. Are we starting to see HDI flex come to life?
Noel: I would say that’s a true statement. Rigid-flex had explosive growth last year and the year before. Traditionally, a lot of flex has been a connector between either two HDI boards or an HDI board and an antenna. Again, to reduce space consumption and allow for more functionality in a smaller device, you see this incorporation of the rigid and flex together.
Ryder: Chip-on-flex, for example. Instead of using a substrate-based fan-out for your chip, there are efforts to put the chip directly onto the flex in certain applications.
Matties: Where would an OEM go if they are looking to become a user of flex for the first time? What sort of advice would we give to them?
Noel: When we see customers enter the flex regime from either HDI or never having used flex, what they come to ESI for is the expertise on how to process the vias. Through-hole vias are simple; anybody can punch through a piece of material. Meanwhile, blind-hole vias are a little more of an art form. To maintain the reliability of your production floor and have low yield loss, you need to understand the process and understand the laser/material interaction. That’s really where ESI excels. We’ve been doing this for over 20 years and taken pains to understand what is going on when the laser meets the material.
Matties: From a designer’s point of view, you have to think differently.
Ryder: Yes. One of the challenges that we see is that you have material manufacturers with some materials that they’re bringing out for very specific applications such as high-frequency applications. The specifications or capabilities of the material, whether it’s electrical or structural reliability, can match the OEM’s expectation on paper and the data sheet. If you’re designing an antenna and need a certain dielectric constant, it all looks good on a data sheet. The reality is that manufacturing that part could bring some challenges.
Especially with the way we’re going right now towards 5G, we have a host of composite materials. When you add that element to the material (i.e., extra layers of materials with separate a Tg, Dk, filler content, etc.), it adds complication. It brings with it a challenge that you need to address if you really want to have high-yield, high-volume manufacturing. By getting involved with process manufacturers like ESI that do laser-based processes, you help yourself and guarantee the success of high yields, which everybody wants.
We’re a part of the puzzle. The material performances are key to the chipset of the antenna of any other function you have. Again, if you’re working with a subpar yield, it’s not going to help your product be successful. That’s the challenge we have, and that’s where I think the OEM would benefit by including us in the discussion.
Matties: Let’s shift a little to the actual flex fabricator who runs the machines. What challenges are they faced with in processing flex boards? Are there any considerations they should think about?
Noel: As we’ve said, yield is always a consideration. You have to be able to produce parts at a profit. We’ve had experiences even in the past few years where customers, OEMs, and producers have had to come to ESI for a solution where they thought they had another route to go, but were experiencing low yields in production. Again, with ESI’s laser material expertise, we’ve been able to help solve a lot of those problems.
Matties: When those types of customers come to you, what challenges do they face that they couldn’t solve? What was the problem?
Noel: It was the introduction of some of these 5G material sets. When utilizing traditional flex manufacturing processes that require high-temperature lamination, these new materials have a different glass transition temperature and also react differently to the other processing steps. That has led to yield loss. We’ve seen vias move out of pads and things like that. Being able to provide a solution to a customer that we can ensure them is an easier or more reliable process to do has been able to help solve those problems.
Matties: It goes right back to what you said about starting with collaboration up front. This is a great example of where that problem could have been avoided completely.
Noel: Absolutely. To be frank, we had already done the prework with some of these customers beforehand. It was an alternative process step that was readily available.
Ryder: I want to mention the base material manufacturers in this picture as well. ESI has always been close to the base material manufacturing world, and we continue to do so. We have constant discussions with the main material manufacturers so that when they introduce new materials, the process and the needs for the processability of the material is discussed. A perfect solution may not always be possible. The materials fit a certain performance that the OEMs need, but we do make it a point to be a part of that discussion with the material manufacturers.
Matties: Is there anything that we haven’t discussed that you feel we should include for the system designers?
Noel: One thing that ESI has been seeing is a continual press for increasingly smaller via sizes. When I started in flex 15 years ago, 100 microns was pretty standard. Now, we have customers pushing well below 40 microns. That’s a challenge for us and the customer not only for laser drilling but also for downstream and upstream processing. As we explore these smaller and smaller feature sizes, we need to evolve with the industry and provide solutions to that. That’s a big one.
Plate-ability is always a question. Again, compensating for upstream and downstream processes by utilizing a laser process that gives you the desired results is vital. ESI’s understanding of those processes is again a great part of our success and our customers’ success as well.
Ryder: I think that as the cost per photon decreases, we see laser-based manufacturing as a great opportunity for some of the new products coming on the market. We’d love to work closely with the industry and customers’ supply base to make that happen. We’re proud to be a part of it.
Matties: Thank you both for being here to share your thoughts, gentlemen.
Ryder: Thank you, Barry.
Noel: Thank you.