Paul Blount has been responsible for the design, layout and product introduction of numerous RF, microwave and millimeter-wave products throughout his 15 year career in the industry. Blount received his Bachelor of Engineering from Brunel University in 1991 and a Master of Science in Microwaves and Optoelectronics from University College London in 1994. He has worked at Hittite Microwave Corporation, Alpha Industries (now Skyworks), and Thorn EMI Electronics and has served as a designer, team leader, program manager and business development manager throughout his career. He has taken all of his products through the full development cycle including: customer specification definition, circuit design, foundry liaison, preliminary test, production test, applications support, and returned material handling. He has specified and de-bugged test software for new product evaluation and has designed and maintained models and layout design kits for Libra, ADS, and Cadence Virtuoso. Mr. Blount is a member of the IEEE (Institute of Electrical and Electronics Engineers), has served on the technical program committee for the Gallium Arsenide IC Symposium, and is a member of the UK-based IET (Institute of Engineering and Technology). He is a Chartered Engineer.
MWJ: How did Custom MMIC get started?
PB: After many years of working together in the semiconductor industry, a small group of us decided to venture out on our own in order to focus more heavily on high frequency circuit design using GaAs and related processes. We were incorporated in May 2006, starting out of a basement in Tyngsboro, MA and then three years later moving to an office in Westford, MA, where we are presently located. At first we focused primarily on design services, working as “hired guns” to create MMIC components for a number of well- known component suppliers. Subsequently we have been awarded a number of Phase I SBIR contracts, all of which have been converted to Phase II contracts. For the past five years we have been leveraging these contracts to create our own library of highly competitive standard products.
MWJ: The company name implies that you are focused on developing MMICs tailored to individual customer requirements. Was this a market demand that you perceived before launching the company and have you found that the need for custom engineered MMICs is an industry trend?
PB: Our primary goal is to provide MMIC solutions that not only have exceptional performance at the component level, but also help solve problems encountered by system engineers. Too many off-the-shelf commercial products, while adequate from a performance perspective on their own, are less than ideal when it comes to system integration. In some cases, the input and output ports are poorly situated, or the bias points are in awkward locations, or the designs require too much off chip tuning. In other cases, the design has slightly too much or too little of a particular performance parameter. Therefore, a bulk of our custom work is now based around modifying our standard products for a customer in these regards. This type of work used to be quite common in the industry, but consolidation over the past few years has created a void. The majority of MMIC design companies today are large corporations, and as such, will rarely touch a custom design effort for less than seven figures. Custom MMIC, on the other hand, has streamlined this process to make it very cost effective for our customers. The word “custom” used to always mean “high cost”, but no more.
MWJ: MMICs are the building blocks of all communication systems, performing the functions necessary to control high frequency signals that are responsible for moving information from point a to point b. How would you compare your company’s products and engineering services relative to other chip suppliers in this sector?
PB: From the beginning, we have followed the mantra of avoiding “me too” designs. In all of our MMICs, we strive to add a recognizable benefit over existing products. For example, we have just introduced a new line of distributed amplifiers with a positive gain slope versus frequency. Nearly all currently available distributed amplifiers have a negative gain slope, and as these components are cascaded in a system, they can introduce significant amplitude variation over frequency. Our components will help reverse this problem. Additionally, we have created lines of low noise, wideband, and power amplifiers that only require positive bias, thus eliminating the need for negative voltage generation and bias sequencing circuitry. There is a profound elegance in such simplicity.
MWJ: What is the breath of technology that your designers are able to offer?
PB: We work exclusively with GaAs, GaN, and other III/V derivatives, choosing to avoid silicon and its derivatives such as SiGe. We can handle nearly all RF functions in a typical system block diagram except the oscillators and prescalers. Although we tend to rely on standard packaging, we have developed a few custom packages, mainly for housing multi-chip modules. Finally, we have our own laboratory where we perform all of our characterization and production testing up to 77 GHz.
MWJ: What kind of customers and applications are best suited to your current size and technology offerings?
PB: Our unique three-tier design model, which we call DS1, DS2, and DS3, can accommodate every type of customer. DS1 is a typical custom MMIC development where the customer pays a significant NRE charge but retains full exclusivity to the design. This arrangement is commonly used when we are asked to develop a complex MMIC for a very specific microwave system. DS3, on the other hand, is our most popular option, where the customer pays considerably less NRE than the typical DS1 contract in exchange for exclusivity. In DS3, Custom MMIC retains the intellectual property (IP) and the right to release the component as a standard product and make it available for purchase by others. The benefit of DS3 is the customer gets exactly the component they need in a short amount of time and at a fraction of the development cost. DS2 bridges these two levels by trading some of the NRE development costs for partial exclusivity, whereby we agree to a six month or year delay before releasing the component as a standard product.
MWJ: You’ve had a number of impressive SBIR contract wins, which are you most proud of?
PB: Our proudest moment to date was winning an Army Achievement Award in 2011, which was presented to us in a ceremony at the Pentagon. The award represents a perfect example of what Custom MMIC can accomplish. For this SBIR, which was our first, we considered a 20 GHz commercially available LNA the Army was planning to use in a phased array. Over the course of the Phase I and Phase II grants, we developed an LNA that, when compared to the original component, had increased the gain by 8 dB, lowered the noise figure from 2.2 to 1.2 dB, reduced the DC power consumption by a factor of eight, and lowered the unit price by half. The success of this program led to a further research and development contract which we used to develop forty-five new products across eight products lines including amplifiers, mixers, switches, attenuators, and phase shifters.
MWJ: Do you see Custom MMICs offering design work as a service in a way that isn’t being addressed or is underserved by other suppliers in the market?
PB: As mentioned above, we actively seek smaller development efforts where companies need a modification of one of our standard components but cannot attract the attention of or afford the major NRE levels demanded by larger semiconductor companies.
MWJ: What is a typical turnaround time for delivering a first prototype after you receive the green light to proceed with a new design?
PB: The timing is of course dependent upon the complexity of the design in question. However, a good rule of thumb is six months from contract award to delivery of tested die from the first fabrication run.