Spring 2008 Sabbatical Report

Spring 2008 Sabbatical Report

Kenneth Kihlstrom

Physics Department

Introduction: This Sabbatical was originally applied for 2007 but with the reassignment of Warren Rogers to the Provost office, it was delayed a year to Spring 2008. The professional work was done at Superconductor Technologies (STI) in Santa Barbara in the Spring and Summer of 2008. That time period also involved preparation for Europe Semester 2008 which I co-led with Kim Kihlstrom, Randy and Dana VanderMey. While a lot of time was spent on that, this report will focus on the research work at STI.

STI is a commercial company which (at the time) focused on superconducting filters for the commercial wireless market (fixed frequency filters) and research & development for the government (tunable filters). I was working as part of a group primarily on the government work but some time was devoted to the fixed frequency work especially on a development project for the Chinese market.

Filters are devices designed to allowed desired frequency signals through but block unwanted frequencies. Conventional (nonsuperconducting) filters suffer from an inability to sharply transition from full transmittance at the frequency edge to full blockage. Superconducting filters thus have some real advantages.

Projects during the Sabbatical:

Tunable Filters: DARPA (Defense Advanced Research Projects Agency) contracted with STI to produce tunable filters. These would be filters that would have very exacting performance requirements. Examples would be to be able to tune over a frequency range (roughly 20% around a target frequency) with very sharp filters (quickly going from full transmittance to strong rejection over a very narrow frequency range) be able to change to any frequency in the range in less than a millionth of a second. This was my primary project during the sabbatical.

IED Detection: Improvised explosive devices have been responsible for a large percentage of Coalition casualties in both Iraq and Afghanistan. They are typically detonated by cell phone. Because the cell phone on the device is passive (receiving the call not making it) detection is difficult. We developed a device that put out RF signals that would resonate the circuitry of an RF (radio frequency) device (like a cell phone) even if the device were turned off.

Results of Sabbatical Projects: The tunable filter project came to a successful conclusion in July of 2010 where a unit was delivered to DARPA which met virtually all the specifications (actually goals) of the program. DARPA was very pleased with the work and has used it to demonstrate what is possible with the new technology. But focusing on the work of Spring and Summer 2008, there were several major advances I was involved in. We had made the decision to use varactor technology (these are electrically tunable semiconductor capacitors) as opposed to MEMS (micromechanical) devices. The varactors had significant advantages (very fast, less susceptible to vibration, etc.) but introduced electrical losses that degraded performance. I did a study to the varactors by lot numbers and found significant variations that allowed us to choose good lots and dramatically improved performance allowing us to ultimately meet the program specifications. A graph of the results is below:

Figure 1: Q versus frequency for varactors

Note that for what should be similar results (e.g. all the “Var 4’s”) the Q values (high Q means less loss) vary by almost a factor of 3.

I also developed tuning techniques that allowed us to program the individual varactor voltages for hundreds of frequencies. Below is a table with the specifications for our tunable filters:

Table 1: Target Values for Filter Performanace

These were very ambitious goals and in the end we met virtually all the specifications. Below is a picture of one of the filters that was a key part of the system.

Figure 2: Top view of low band filter assembly microenclosure

This represents a bank of four separate filters that can by electronically addressed (in less than a microsecond) to provide for the lower range. A similar bank provided the higher range of frequencies. Below is a figure with the data for several frequencies for the low band:

Figure 3: Preliminary measured performance (S21 shown) of low band microenclosure

Note that this is a logarithmic scale so each 10 dB represents and order of magnitude reduction of transmission.

The IED detection program led to a successful demonstration of the device that included detection of RF circuits up to 100 feet away. The demonstration was done with a low power transmission signal (to comply with FCC limitations). The return signal (from the hidden RF device) was connected to a circuit which produced an audible beeping that became louder, more frequent and higher pitched as the signal became stronger. The led to a very effective warning and since the transmitter and receiver were very directional it allowed identification of the hidden device with ease. We provided a successful demonstration for DARPA, along with a video. We also outlined a pathway to increasing power output to increase the range should the military decide to pursue this avenue.

Reports from Sabbatical Work: Because this work involves both proprietary techniques and was done for DARPA, there needs to be approvals before it can be published. I am hoping to be able to write this up in one or more papers which will involve a number of co-authors. As an appendix, I am including an outline of what the full paper would contain.