American Standard Circuits is an early adopter of Averatek’s A-SAP process for its ultra high density interconnect (UHDI) products. In this final part of my interview with industry veteran John Johnson, vice president of business development at American Standard Circuits, we use photos, slides, and materials to discuss what he learned from his previous role at Averatek.

Steve Williams: John, let’s talk about the design benefits of UHDI technology.

John Johnson: It’s really amazing what you can do with UDHI technology. There are some things you have to watch out for in a design and because you can do the ultra-fine lines and spaces, a designer has a lot of things in their toolkit. You have via-in-pad plated over microvias, stacked and staggered microvias, and other different structures. But when designing with A-SAP, you first need to focus on using that ultra-fine line and then going to the other design aspects. Maybe use staggered microvias or several levels of staggered microvias and then use a stacked microvia if you really have to. That removes some reliability traps we've had to deal with because we couldn't reach those ultra-fine lines.

Figure 1 shows what happens when you can achieve ultra-fine features with this technology. Current subtractive technology is around 75 microns, or 3-mil line and space. Some folks can get down to 2 and 2, but it isn’t without some challenge and yield penalty. This technology allows us to get below 25 microns, a present semi-additive capability in Asia. From what we hear, they tend to struggle at 25 microns; it’s certainly not a slam dunk. This technology allows us to take it even further into the 15- to 12 ½-micron range. Ultimately, Averatek is working on the next generation of products in the 5-micron range—the semiconductor range. With packaged substrates, they would all love to have that capability today.

So, 25 microns is 1-mil?

Then sub-1-mil is 15 microns down to 5. It’s almost hard to believe that we can get there. How long until this process comes out? I’ve already run down to 15 microns in a fab facility. It’s doable and certainly with yield. That’s the amazing part. We’re actually headed that way ourselves in the not-too-distant future.

Is Figure 2 a visual reference of what you just talked about? Right. The first spiral is a representation of the 3-mil line and space and that’s done in a subtractive format. With the 25-micron, you now get a 9x increase in density. You can see that line is pretty uniform and then you go to the 12.5-micron—a half-mil line and space. You look at that under a scope, and it is beautiful.

I’ve seen some of your samples and you’re right that you have to use some high magnification to even see this definition. Absolutely. It is very hard to see. You can’t use a 10-power eyepiece and expect to see it very easily.

How has the design community accepted and used this, putting out new designs that fabricators like American Standard have to build? That’s where things are still largely in the development stage. There are a lot of designers who are looking at it. The DoD already has designs out there, but it’s about bringing this to the marketplace with designs geared for the fine line and space features. There are many cases where people are trying to get over that hurdle in routing out the fine-pitch devices. As soon as they start utilizing this technology, they will find out how easy it really is.

Interesting. John, tell me about Figure 3. What are we looking at? This is a sample on liquid crystal polymer material: a 4-mil (100-micron) thick substrate, and how far away these traces can be from a plane layer. They can actually be very tight and close to each other. In this case, it’s 17-micron, 11-micron wide traces, and 13 microns high. They can have a lot of electrical effect with each other with differential pairing because they’re so tight. That ground plane is so far away that it has very little effect at that point.

Earlier you talked about biocompatibility. That’s the beauty of using noble metals that are stable inside the body. If it’s on an LCP or polyimide material, folks are using it in neural probes, glucose monitors, and other types of implantables. Glucose monitors are out there, but when you get something without copper or nickel, it makes it easier to stay longer in the body.

We both work with FreedomCAD and we talked about UHDI at the last webinar we did with them. They’re always looking to reduce footprints or improve performance. How do you get the word out to everyone who needs to know this is a viable design option? We’ve had several seminars with our customers—discussions like we’re having here. That’s part of the key, but we’re also putting together design guidelines, recommendations, and other things you need to watch out for when you’re building parts. ENIG is a great finish, but when you have a 1-mil line and space, you put two-tenths of nickel down, and all of a sudden, it’s six-tenths of a mil space, and it gets a little more interesting.

I found it incredible that from a design standpoint, you can eliminate virtually 70% of the ball or solder connections. Right, and when you start looking at it, you can get rid of that redistribution layer and solder right down to the chip. It’s pretty amazing.

John, this has been a fantastic discussion. This technology is truly amazing, and it’s something that even 10 years ago nobody could have even dreamed about. You guys are making it happen. It’s very exciting, and that’s the fun part. I enjoy what I’m doing. In this part of my career, the ability to bring this to the point of reality is just fabulous.

I appreciate your time and your expertise on this, and we’ll be talking to you soon. Thanks Steve, I welcome every opportunity to talk about this technology.