Automation and Traceability Critical for High-Throughput Probe Testing
Peter Brandt, atg’s director of sales for Europe, India and Japan, joined Barry Matties at productronica to share his views on regional variations in electrical testing requirements, automation, traceability, and how to anticipate and satisfy the future demands of the market. Group general manager Dr. Jochen Kleinertz also comments on the need for close cooperation with customers to guide the R&D effort.
Barry Matties: Peter, thanks for taking time to talk with us. This is the final afternoon of the show and it’s just beginning to quiet down. What have been your general impressions?
Peter Brandt: This show was an unexpectedly crowded and good show this year. We have a very high number of customers here, as well as potential customers. We were very satisfied with the show this year.
Matties: Please tell us about the new technology you’re showing here.
Brandt: This year, we are exhibiting our new automatic flying probe machine. The name of the machine is A8a. It's an automatic probe machine for final boards. This machine is dedicated to high throughput, flexibility and it approaches step by step the throughput of the grid testers.
Matties: Automation is the key though, right?
Brandt: Automation is the key. Automation in combination with traceability. Automation makes sense to save labor costs, but the most sense of automation is to have traceability and to avoid having bad boards on the good board stacker at the end.
Matties: You're responsible for Europe, India and Japan. Your viewpoint of the world must be quite interesting. Tell me the differences that you see between regions.
Brandt: Oh, there are big differences within these territories. When we look at the European companies, these companies have only the need of flexibility. They make business in the automotive, aerospace, industrial markets. India’s need is mainly higher quantities, not very complex boards and simpler products, but they are increasing their technology very fast. Also, now, their intent is to buy new machines and new equipment. Japan is certainly the most difficult market for the Western companies. Here, we can sell only high-technology machines. High-technology with automation and high performance in pitch and accuracy.
Matties: There's also the nationalistic attitude there, and so competing in Japan as a German company must be difficult. How do you overcome those barriers?
Brandt: First, you need a strong agent or representative there. You need somebody who can convert the technical knowledge from the English language into the Japanese language, and you must convince the operators by technical arguments. That's very important in Japan. The most important thing is that you have a unique product that is not available in the Japanese market. That makes a big difference.
Matties: What sort of questions or challenges do potential customers come to you with at a show like this?
Brandt: Their biggest challenge in the European market is the throughput—the speed of the machine. They want to avoid building fixtures. A fixture test is a very high-cost process. You must produce a fixture. Then you debug the fixture. At the end of the day, you have problems with the reliability of a fixture test machine. The flying probe machine has the flexibility. You can change net lists and you can change electrical parameters. With a fixture, you are much more rigid. When you’ve already prepared the fixture, you can change nothing anymore.
Matties: But in a volume shop, flying probe just isn't going to be the solution, right?
Brandt: No. It's a combination between complexity and volume. For high-volume and less complexity, the fixture test can do this. It still has a need on the market, and from an economic point of view, there’s a big market share for the fixture tester. But especially for complex boards, substrate boards or embedded component technologies, there you will find automatic flying probe test systems for higher quantities like 5,000, or 10,000 PCBs per lot.
Matties: That's pretty good, and the throughput is satisfactory for that?
Brandt: Yes. Customers always want to have the fastest machines in the world. From year to year, we try to improve. We improve the performance and the speed of the machine. This is not a big single jump, but step by step, we improve our machines.
Matties: What is your background, and what is your favorite part of the job?
Brandt: I'm an electronics engineer. The favorite part of my job is talking to customers about technical demands, new technical projects in the market, and then bringing this information back to atg to develop the right equipment for the worldwide market.
Matties: You mentioned R&D. When your R&D team is working, how much customer input do they take to drive their next generation? Or are they looking ahead of their customers and saying, "Here's what you are going to need?"
Brandt: This is one of the most critical issues. In the R&D department, with what they are doing, they have very little contact with the final customers. The sales team are the people who catch this information from customers and communicate it to our R&D team. We look at different worldwide markets, because their segments of market demands are sometimes completely different. We work out what the global demand of a technology can be, and if this kind of technology is an improvement for us. Then, we generalize and give this information to our R&D people. Then, internally, in the big meeting, we decide our future developments.
Matties: That must be a tough decision though, because it's a big investment to develop new equipment.
Brandt: Yes, you need to start at least two years before and you need a decision at least three years before the machine will be available on the market. It's a long process of specification, pre-development, assembly and then the final test.
Matties: Not to mention all the competitors chasing the same opportunities, right?
Brandt: Exactly. You always need to be one step in front of the rest of the competitors.
Matties: As I look around, there are more and more competitors, especially in the Chinese market. With densities getting smaller and tighter, what are the testing challenges there for you?
Brandt: The testing challenges are for the HDI products, or the regular mobile phone and workstation products in HDI. The biggest challenge is the 4-wire Kelvin measurement. For special applications, it’s the latest test technology, the 4-wire Kelvin test. In that case, you want to test neckdowns on tracks or weak microvia connections. At the end of the day, also a big challenge is the high-frequency measurements. This is a big challenge, because under a certain level, when the tracks have a little bit of tolerances, their high frequency condition can be completely different.
Matties: I was recently at the Altium Live Event in Munich, and Lee Ritchie, a top designer in the world, was telling the 300 designers in the room, "If you're not doing HDI now, you're about to be doing HDI." Have you seen that in your customers? Is that what you're experiencing out there?
Brandt: We see that the HDI technology has also entered the market of server board applications, of military and space applications. We see a trend from the conventional mechanical drills and multilayer technology to the HDI technology driven by higher clock frequency. The boards also, in the server board technology, will become smaller and smaller. They try to reduce the number of layers, and the requirement to do this is to introduce the HDI technology for this kind of business.
Matties: And really, if you don't get it right in the testing, it's too late.
Brandt: Yes. You must also prepare them for testing. You need a tester that can do the density and has the performance and the electrical measurement parameters to do this.
Matties: Now that we've finished the board, we've tested it, and we've collected a lot of data. How are your customers using that data?
Brandt: It really depends. It depends how much data they want to collect. In general, they can collect every single measurement. They can store it on a big server. Then at the end of the day, they can analyze lot numbers, dates, tolerances, measurement parameters, etc., but also system operating parameters. There is a huge variety of data mining that can be done.
Matties: What is the biggest challenge when developing new products?
Brandt: To develop new products, we need strong relationships and partnerships with our customers. They open their future roadmaps and push us for the latest technology. Knowing their schedules guarantees that the development will be on time. We know the people. We know our technology and our customers’ technology. Without these key customers, we would not be able to introduce a new products into the market. We need this strong partnership between our key customers and our R&D team.
Matties: How does that process work?
Brandt: For certain cases, we force our R&D people to go directly with the customer team together, to share information, so we don’t get some information filtered or lose any. We always try to have direct contact. So our R&D people are involved with them and can have a big influence in order to move the development.
Dr. Jochen Kleinertz: There are also key projects which drive our development and, of course, also the development of our customers.
Matties: You have a vested interest.
Kleinertz: Yes. In Asia and the U.S., we're talking about 5G or about substrates. There is a lot going on and new requirements coming up. In the end, it's those projects which we do together with the customer that help us to develop our capabilities.
Brandt: We need this customer. When we develop a technology or equipment, this isn't the first theoretical idea, but an idea which is working on prototypes and how this will work with different products and different customers. At the end, you need your customer in order to improve this kind of development.
Matties: What advice would you give a fabricator looking at a testing solution?
Brandt: In my point of view, when you are looking around in the worldwide market, please take care about good technical support. Please do not make a short decision to any equipment supplier where they only want to sell a machine, and do not give you good, long-term support. This means valuable technical support, with good telephone support and fast support. Our machines have very fast setup times because we are at the end of the production process, which means if our machines are not working our customer cannot generate any revenue.
Matties: Is there anything that we haven't talked about that you feel like you should share with the industry?
Brandt: No. At the moment, my brain is completely empty. (Laughs)
Matties: Peter, thank you so much. I appreciate your time.
Brandt: Thank you very much.