Q&A: Semiconductors, the supply chain and what’s next

Peter Suliga is responsible for HEIDENHAIN’s electronic industry business development in North America. Considering how uncertain things were in the semiconductor world when we talked with him a few months ago, we wanted to check back in with Peter.

Just a few months ago it seemed the popularity of things like self-driving cars and graphics processing for VR and computer games wasn’t stopping any time soon. The same went for the industrial demand for chips. The market was asking for both—the more-commoditized chips and the most densely powerful—at the same time. But then, things started to stabilize, and pretty quickly.

Here are some excerpts from the discussion with Peter on the state of the semiconductor industry and the work HEIDENHAIN does to support it.

Can you describe the unique situation the semiconductor sector has been in?

What we’ve experienced over the last couple of years is really, unprecedented because of how aggressive the ramp-up was. Earlier this year, it seemed components would be purchased as quickly as we could make them. The supply chain was stressed and demand was the highest we’ve seen in our lifetime, easily.

The semiconductor capital equipment builders were faced with a tremendous amount of incoming orders. So much so that folks who typically worked in research and development, those typically tasked with developing next-generation products were actively helping manufacturing teams solve supply chain issues.

We have to be mindful that the semiconductor industry is very cyclical. With prices rising of late—and interest rates too—many car lots are full and people are spending less on expensive tech devices that run on chips. Those demand pressures are sinking.

Where is demand right now?

We’ve gone from not just great inventory numbers, but epic inventory numbers to more conversation about how to manage  chip stock. Up until recently, the problem was, “how can I get my hands on enough stock to sleep well at night.” Now, it’s transitioning to, “do I have too much stock?”

Things are still hot, but they have cooled off to more sustainable levels. There’s a strong outlook from our customers. Historically speaking, all the guidance that we have is very robust. It’s just not ‘crazy’, which is kind of reassuring; you want a stabilization that’s reasonable, versus drastic ups and downs. We’ve, basically, seen a return to normal levels—with the exception of just a few industries where there is still scarcity.

Where does HEIDENHAIN fit in the semiconductor supply chain?

I’ll walk you through a common scenario. Say Samsung or Intel is planning a new fab. Once it’s determined what node it’s going to be, or what sizes are going to be on the chip, they’ll determine which processes they’re going to run and what kind of equipment is needed. Next, they’ll consider inspection and metrology. We’re a supplier to the small group of capital equipment suppliers in that inspection and metrology space.

The customers we have are Tier 1, Tier 2 and Tier 3 semiconductor equipment manufacturers. When those companies get machine orders it trickles down to us. Our revenue is a function of their demand. That’s one way we have such a good pulse for the sector.

Can you speak more about some of the specific roles HEIDENHAIN products play in semiconductor production?

When you try to make any kind of chip, it’s a series of processes that are repeated over and over again—etching for example. They have to be the right width, the right height and the right profile. All of that is measured by metrology tools. We provide the components and subsystems to companies that build inspection and metrology machines.

These are critical to the process…if you don’t create the pattern correctly, the chip’s not going to function. The sooner you know, the better. If you have a lot of defects or your yield is going to be low, fab profitability is going to be low. It means there’s going to be a lot of waste, and you can’t have that. Fabs operate on fairly thin margins.

What do you think really drives innovation and growth in the semiconductor industry?

There has always been a semiconductor roadmap that the industry established—not all that different from Moore’s Law, but beyond it. It’s not necessarily followed as closely as it was in the past, but the guideline is still there, and most kind of orient themselves towards that.

That’s part of it. But there’s also a big market pull. Companies that sell to consumers need to have a roadmap of their own to sell more products. In other words, iPhone 13 better be better than iPhone 12 and 11 in order to generate more sales. I think that’s ultimately the reason because devices with chips sell and sell. Because you and I are buying it.

What are your typical goals when working with a customer in the semiconductor industry?

Very simply, we’re usually tasked with improving process fidelity and performance. The second thing is to reduce the cost of ownership; if we can help a customer’s machine process more chips an hour, we’re helping them look good by lowering the cost of ownership for the end user.

What makes HEIDENHAIN unique among suppliers to the semiconductor industry?

I would say the privileged position that HEIDENHAIN is in is that we aren’t just an encoder company, a motor company or an electronics company. Sure, a lot of customers are interested in individual components, an apparatus from a catalog. We definitely help machine builders who prioritize basic performance, price and delivery.

Others, though, will come to us with much broader questions. That’s our wheelhouse. HEIDENHAIN is very technology oriented. We have a strong fundamental research backbone. We’re able to discuss not just the here-and-now, basic product features, but also future challenges and how to address them now…how to actually drive the industry forward.

We have technologies and know how to talk about things that can really revolutionize how we do things and we’re actively engaged in those projects, as well as addressing, you know, machine builders where maybe basic performance price and delivery are the key points.

Our portfolio of technology and competencies is very broad and folks we engage with very closely are starting to recognize it more and more. We’re not just talking about one dimension anymore. We’re located in all six degrees of freedom.

What has you excited about the future of technology?

I’m excited that we’re always moving forward. An extraordinary accomplishment 20 years ago, or 10 years ago is just kind of nice today, nothing too crazy. The new technologies we talked about being really out there 15 or 20 years ago? People are actually looking at them today and saying, “Hmm, I could really use that.”

Even though there are challenges scaling, I don’t see the pull for better technology stopping anytime soon. People may have declared Moore’s law dead, and yet we have devices that are multiple times quicker and we’re not paying multiple times more. I always joke about buying a fancy computer back in 1996 for $2,000. That’s about $4,000 in today’s money. I would have done a lot better if I invested in stocks.

Point is, the performance of that computer compared to right now is night and day, yet the price I would pay for that performance today is minuscule. Actually, I heard in a presentation recently that if you wanted to build a computer as powerful as an iPhone in the 80s, it would be the size of an entire house and cost a billion dollars.

I don’t think it’s going to stop and I think that’s what’s so amazing.