Microwave Journal

Executive Interview: Brian Dupell, President, Dielectric Laboratories Inc. (DLI)

January 3, 2008

Brian DuPell, President of DLI, spoke with Microwave Journal (MWJ) editor David Vye about today’s spectrum management challenges, the RADAR market and the role of High-Q ceramic-based RF/mW components play in achieving the system performance required by today’s military and commercial applications. Dielectric Laboratories Inc.

(DLI) is a global supplier of specialty, high performance components, including high Q Single-Layer (SLC) and high Q Multi-Layer Capacitors (MLC), broadband DC blocks, precision Variable Capacitors (VC) through its acquisition of Voltronics Corp., and a new line of custom thin film components including RF/mW/mmW ceramic filters and resonators.

Mr. DuPell has served as President of DLI since 2001, after joining the company in 2000. Prior executive management roles include Executive VP of TAM Ceramics and Director of Manufacturing for a division of BASF Corp. He has earned Bachelor of Industrial Engineering and Master of Business Administration degrees. He is married, has three adult children and lives in Manlius, NY.

MWJ - According to the “Value-added & Application Specific Ceramic Capacitors: World Market” report that was released by Paumanok Publications1 in 2005, the global application-specific ceramic capacitor market at that time was $470 M with high Q components for applications over 1 GHz making up 23 percent of the entire market ($110 M). Of that segment, base stations represented 65 percent and defense communications represented 15 percent of the total high frequency ceramics end-use markets. Are these numbers more or less in line with your own end-user’s applications?

DuPell – It is a matter of definition, but we do not think of capacitors for base stations as application-specific components, a market where there are several suppliers with a similar group of products competing for the same relatively higher volume business. In any event, we see base station revenues being lower than the 65% Paumanok reported.

MWJ - Has recent activity in WiMAX, homeland security, or other defense applications shifted the profile of the end-use markets?

DuPell –In the cases of homeland security and defense related applications, yes.

MWJ - High-Q ceramics are clearly critical to high frequency performance. What is the approximate frequency whereby filters and/or resonators based on other technologies such as SAWs can no longer compete? Can you give us some comparative performance metrics such as temperature stability?

DuPell – Customer requirements and trade-offs define where one technology is or is not competitive. Size, weight, temperature stability, phase and amplitude tracking, and vibration insensitivity are the primary benefits of our products when compared to competing technologies. Many of our customers’ applications are in sub-gigahertz frequencies, although our technology’s relative benefits become much more powerful at frequencies above 2.5 GHz where SAW performance tends to fall off.

MWJ - What are some of the emerging applications and markets at or above this frequency?

DuPell – We have introduced a family of filters at GPS frequencies that can be used either in commercial applications or in high-reliability environments. You mentioned WiMAX earlier, a higher frequency commercial application that is beginning to take root. Automotive RADARs, satellite communications, precision targeting and counter-measures, battlefield broadband communication systems, and a variety of other military and homeland security applications utilize frequencies above 2.5 GHz – some at much higher frequencies. Point-to-point radio systems operate at frequencies up to and above 30 GHz. We find that design engineers at elevated frequencies are hungry to incorporate the beneficial characteristics that our products offer.

MWJ - DLI produces many components for the defense market, especially RADAR systems. I assume that includes mobile RADAR, man pack communications and missile links. What are some of the technical trends that you are seeing in RADAR applications?

DuPell –We have a heritage of supplying high reliability products into RADAR applications, originally with our SLCs, VCs and tuning elements, and now, in addition, with our filter and resonator products. We believe SLCs from DLI continue to populate the vast majority of military and civilian RADARs around the world. Going forward, we see smaller and lighter systems coming to the fore that utilize new technologies. New RADARs have much greater portability. Aerospace systems and anti-RADAR systems have greater overall performance because of the compactness and temperature stability of the sub-systems. Size, temperature stability, and better phase and amplitude tracking are enabling new phased array systems. Smaller, lighter, better.

MWJ - How does DLI’s technology help reduce component size?

DuPell – For our filters, resonators and other custom thin film products, the use of higher K ceramics allows for substantial component size reduction. Most of these products are printed on our proprietary CF, with a dielectric constant of 23, or our CG, with a dielectric constant of 68. When compared to filters printed on alumina or PWB materials, our realization will be many times smaller. When compared to other filter technologies, our ceramic microstrip filters can be orders-of-magnitude smaller. In addition to miniature size, we offer the path to incredibly light-weight solutions, which is a key factor in aerospace and man pack applications.

MWJ - Ceramics are also well known for their temperature stability and vibration insensitivity. Are these characteristics the key differentiators from other technologies in consideration of the harsh environments where many RADAR and communication systems are to be deployed?

DuPell – Temperature stability, together with size reduction, are the two overriding benefits of our new technologies. We do not know of another ceramic material system with the temperature stability that our proprietary CF ceramic has. CF has become the workhorse for our filter, resonator and MLC products. We like to say this material system, which has a temperature coefficient of 0 + 15ppm/°C, is stable from outer space to the desert. For demanding shock and vibration environments, SAW and crystal-based components are difficult to ruggedize. Lumped element, cavity and dielectric resonator-based filters are prone to microphonic effects and ping hits in high shock and vibration environments. Our technologies are not subject to these undesirable effects and, for many applications, this is a major differentiator. Other key benefits of our CF are that it does not age, does not outgas and does not degrade in radiation environments. For some applications, these attributes provide the engineer with performance options not available in other technologies.

MWJ - DLI has been working closely with your sibling companies - Vectron and K&L Microwave. This allows you to respond to customer requests that you and your partnering companies could not solely address. For instance, DLI’s High-Q components allow Vectron to offer frequency sources above 2 GHz expanding Dover’s overall product offerings. Similarly, DLI and K&L are working together to create switched filter banks that offer all the benefits of DLI filters and of K↦L’s integration expertise. How has this collaboration worked out for all parties and do you see expanding it in the future?

DuPell – It is early times for both collaborations, but all three companies are excited about the potential for new products, new applications and new customers.

MWJ - Is there an increasing amount of customer-driven custom engineering taking place at DLI? Is your partnering with companies the result of customer demand?

DuPell – I want to think that everything we do is a result of a customer’s need for new technologies, custom components, special customer or technical service. A substantial majority of our sales are for custom components that result from really getting to know our customers. When we partner with another company, we are trying to solve a common customer problem that we think can be accomplished better by utilizing the partner’s expertise.

MWJ - I read that DLI has partnered with Mimix Broadband and Vectron on a 10 GHz voltage-controlled oscillator (VCO). What is the application and how did all the companies get involved?

DuPell – The potential applications are many, but 10 Gbit SONET and military oscillators were the original targets. When we were ready to take our cavity resonators to market, we talked to Vectron, and in parallel we approached Mimix, with whom we have had a relationship for years.

MWJ - Is this VCO now a standard component in the Dover catalog or is it a Mimix device?

DuPell – Vectron is currently in the Beta stage with several customers and is finalizing the specification requirements. They expect the 10 GHz VCO to be released in 2008.

MWJ - Is this collaborative engineering similar to the design support that your applications group performs for customers of your build-to-print capabilities?

DuPell – For our custom designed components, we control the entire process – design, materials, manufacture and test. Usually our build-to-print customers have a design and material in mind and we simply manufacture the product for them. Because of our design savvy, however, we often make suggestions to potentially improve the performance or the manufacturability of the product. We also can offer to test plates before printing and probe finished parts for the customer, something that our competitors do not do.

MWJ - What can you tell our readers about your internal RF design and test expertise?

DuPell – Over the past ten years or so, we have assembled a talented group of clever senior design engineers. We have more recently added different levels of experience and knowledge of different technologies to our design team. We just recently opened our first remote engineering office to further expand our capability and capacity. We have invested heavily in design tools to support our engineers and are investing heavily in product testing systems to support prototyping and production up to 67 GHz. All this investment is necessary to support and grow the business of our new “Disruptive Technologies for Spectrum Management”.

MWJ - On the topic of technology, can you explain Passive Intermodulation (PIM) from components using non-linear dielectrics? I understand they can support the generation of spurious signals. Is this an issue for DLI’s components and if so what is the impact or preventive measures that need to be considered?

DuPell – All of the material systems we utilize are high Q, so none are non-linear. The use of nickel and other magnetic materials can also cause spurious responses. For capacitor applications in an environment where PIM can be a problem, we offer products with barrier-layers not containing nickel.

MWJ - Do you provide customers with capacitor models? If so, what form are these models, i.e. S-parameters, equivalent circuit models, etc.

DuPell – Yes. We supply S-parameters, Spice models. Our CapCad® system, available on CD or on our web site, helps engineers select the right capacitor for their circuits. We have many of our MLCs available for comparative pad-scalable modeling at Modelithics. We also provide S-parameters and simulations for our filters.

MWJ - DLI has over 100 proprietary and/or patented ceramic formulations. With dielectric constants ranging from 4 to 40,000, what are some of the factors that go into material selection?

DuPell – Some of the factors we consider when selecting a ceramic are frequency of use, temperature range of use, capacitance tolerance, expansion matching issues, size requirements and capacitance requirements. For example, some of our lower dielectric constant (K) ceramics have different temperature coefficients of expansion. Some of our ceramics have positive or negative temperature coefficients of capacitance. Some have wider TCC with very high K and some have very tight temperature coefficients. So, we are very likely to have a ceramic to help us address a customer’s needs – either special or more mainstream ones. On a routine basis, though, most of the products we sell are based on a core group of about twenty ceramic systems.

MWJ - Is DLI’s ceramic technology suitable for direct mounting of active devices and integrated circuits in support of integrated sub-systems such as a front-end module (FEM)?

DuPell – Yes. The miniature 10 GHz VCO mentioned earlier is an example of this.

MWJ - High Q inductors and optimum use of MMIC/RFIC real estate are both very important to RF designers. Designing the inductor off the chip would seem to address both these needs. Are you seeing this approach among your build to print customers?

DuPell – Yes. A high value-added application example is high power amplifiers in the UHF and lower microwave frequencies, where power levels can reach 100s of watts. Here the use of high Q, high dielectric ceramics enable low impedance, small transmission line matching circuits. The result is a miniaturized circuit with fewer discrete parts, excellent repeatability, and higher output power and efficiency than alternative technologies.

MWJ - Any plans to expand the types of passive components offered by DLI beyond capacitors, filters and resonators?

DuPell – Our technology roadmap is always evolving and we share pertinent parts of it with our customers on a regular basis. We do not, though, publicly discuss new products under development until they are fully tested and ready for release.

MWJ - Given the competitive nature of our business, it is certainly understandable that you would need to protect your technology roadmap and choose to keep information about future product developments between you and your customers. I can imagine your customers are fully engaged in these developments - requesting performance capabilities and offering feedback. It sounds like this close relationship has already led to some impressive products. We look forward to seeing what is next. Brian, thanks for your time today. It was a pleasure talking with you.

1. http://www.ttiinc.com/docs/IO/7105/Intro-zogbi.pdf