2010

Dr. Robert Romanofsky has over 75 publications and holds five patents in the fields of microwave device technology, high-temperature superconductivity, and the use of thin ferroelectric films in microwave applications. A recipient of NASA’s Exceptional Service Medal, Exceptional Technology Achievement Medal, the Federal Executive Board “Wings of Excellence” award, and the Rotary National Stellar Space Award, he currently serves as senior engineer for the Antenna and Optical Systems Branch at the NASA Glenn Research Center where he works on advanced antenna systems designs.

NASA Tech Briefs: As a senior scientist for NASA’s Antenna and Optical Systems Branch, what types of projects do you typically get involved with?

alt

Dr. Robert Romanofsky: It’s a fairly diverse portfolio. Most of my work for the last several years has been supported by the Space Communications and Navigation (SCAN) project in NASA’s Space Operations Missions Directorate. I’ve been working primarily on the development of an electronically steerable antenna based on some novel thin film ferroelectric devices that were invented here at Glenn. I’m also working on a technology program to accelerate the development of large deployable Ka band antennas. There’s a considerable amount of work that I do for certain DoD interests, as well as industry. We have a number of reimbursable Space Act Agreements with both large and small companies.

NTB: It sounds like a pretty diverse array of jobs.

Dr. Romanofsky: It’s pretty exciting.

NTB: One of your areas of expertise is cryogenic microwave electronics. What is cryogenic microwave electronics, and what are some of its potential applications in terms of both the space program and commercial use?

Dr. Romanofsky: Cryogenic electronics, in general, simply means that the components are deliberately cooled to a certain temperature. The exact threshold of where cryogenic begins is somewhat nebulous, but we do a lot of work at liquid nitrogen temperatures and below.

There are many advantages in cooling electronics. For example, a very important parameter of semiconductors is the mobility, and the mobility has a lot to do with device speed. Of course, there’s a tremendous amount of interest in making logic faster and faster. One way of making that happen is to increase the velocity of the carriers within the semiconductor. These mobilities can increase from perhaps a factor of several thousand centimeters squared per volt second at room temperature to, perhaps, tens-of-thousands of centimeters squared per volt second at 77 Kelvin. So it has an enormous impact on device speeds.

Also, cooling electronics, especially microwave receivers, improves the sensitivity of the receivers, and in a communications system, signal-to-noise ratio is everything. You must be able to extract extremely weak signals from an extremely noisy background. So cooling the electronics, which is exactly what has been done for many decades with the Deep Space Network, allows us to do things like pick up inordinately weak signals from spacecraft like Voyager. That’s a phenomenal accomplishment.

As far as commercial applications, I think they are evolving thanks to the fairly recent development of miniature cryogenic coolers. To the best of my knowledge, only one domestic company is still doing low-temperature superconducting electronics, and that’s HYPRES. They’ve been quite successful in demonstrating an all-digital receiver, and that’s kind of the holy grail of communications engineering, to move digital electronics as close as you possibly can to the antenna. With superconducting technology, you can make extremely fast analog-to-digital converters, and that’s kind of the direction that things are heading in.

NTB: You’ve also done a lot of work in the area of thin film ferroelectric technology. What potential applications do you envision for that?

Dr. Romanofsky: We started working on microwave applications of thin film ferroelectric films roughly ten years ago and it was a novel way of building tunable smart electronics. For example, instead of just having a filter, or an antenna, or something that would just kind of sit there and serve one function, it allowed much greater flexibility.


The main project I’m working on now in this area is the ferroelectric reflectarray. There are many applications for electronically steerable antennas, and people have been trying since World War II to develop a cost-effective, efficient phased array antenna, which has proven to be quite elusive. The low insertion loss provided by our ferroelectric approach allowed us to develop a new concept called the “ferroelectric reflectarray,” which goes a long way to solving the inefficiency issues and the thermal management issues of conventional phased array technology, and also gets the costs down to where it would be of interest to a NASA mission. Many of the phased arrays that are out there now are based on ferrite technology. They require intense magnetic fields, ferrite phase shifters, and that, in turn, means we need a lot of power, and the ferrites are very heavy. All of those things are very inconsistent with space flight. So, the electronically steerable, ferroelectric reflectarray goes a long way to solving all of the shortcomings.

NTB: What makes a cellular reflectarray antenna different than the parabolic reflectarray antenna systems that most people are familiar with?

Dr. Romanofsky: Actually, the cellular reflectarray and the ferroelectric reflectarray are considerably different. The ferroelectric reflectarray is electronically steerable. It can essentially form a collimated beam anywhere in the hemisphere in front of the antenna. The cellular reflectarry is intended for a commercial application. We were approached by a company called Alienworks a few years ago and they were seeking a replacement for a parabolic reflector for the following reasons. Next generation Direct TV is going to be operating at much higher frequencies than we’ve had in the past and the antennas will be much larger, and there is a concern that with the larger antennas and the higher frequencies leading to much finer beam width, wind loading will cause disruptions in the service which, of course, is not tolerable. So they were hoping that a flat antenna could be provided to eliminate most of the wind loading problems. In addition, aesthetics is always an issue with commercial antennas. A lot of people don’t want to have a 30-inch diameter antenna sitting on their roof; they’d rather have something that would blend in, or even be conformal, if that was possible.

A cellular reflectarray does just that. Basically you would tell the provider of the antenna what your zip code is, and I’ve got a program that maps out a reflectarray for that latitude and longitude and all the installer needs to do is make sure the antenna is aligned to magnetic north and that it’s flat, and it will place a beam on the satellite. That’s one of the other big issues with Direct TV installation; every time they’ve got to make a run to do an install, it’s something like a $200 charge. If you’re looking at tens-of-millions of subscribers, that’s an enormous chunk eating away at your profits, so this goes a long way to minimizing the installation costs.

NTB: Another project you’ve worked on is the design of large inflatable antennas that could be deployed aboard relay satellites in deep space. Tell us about that project and some of the unique challenges it presents.

Dr. Romanofsky: We were instructed by NASA Headquarters, by the Space Communications and Navigation project in particular, to initiate a technology development effort of large deployable antennas approximately four or five years ago. Actually, it was initially intended to be part of the JIMO – the Jupiter Icy Moons Orbiter project – and there was a desire to have a large relay station at the Jovian distance to relay enormous data rates back to the Earth and it necessitated, perhaps, a 10-meter diameter antenna. Well, you can’t fit a 10-meter diameter antenna in, say, a 4-meter diameter spacecraft fairing, so you need a way to deploy an antenna. In addition, mass is always an issue, so we started working on gossamer antennas and other types of deployable antennas, and very suddenly the thrust switched from JIMO to exploration.

Of course, there was renewed interest in going back to the Moon and then on to Mars. A similar situation arose, and that is the communications that are going to be coming back from Mars in the future will be at much higher data rates than we’ve had historically. If you do the analysis, you would again need a 5 to 10-meter diameter antenna, possibly in an areostationary relay situation, to get hundreds of megabits per second back to the Deep Space Network. So the same limitation exists – you’ve got to get a very large antenna in a fairly small fairing, and you want to reduce mass as much as possible. So we’ve been developing inflatable membrane antennas, shape-memory polymer antennas, and we’ve helped a few companies with the most popular type of deployable antenna out there now, which is a mesh-type deployable.

NTB: Speaking of technical challenges, what would you say has been the most challenging project you’ve worked on to date at NASA?

Dr. Romanofsky: All of them really have their own challenges and twists. I think the most difficult project I had early on was actually my first space experiment. We were developing a hybrid superconducting, semiconducting receiver. It was a joint effort with NASA Glenn and JPL. It was delivered to the Naval Research Laboratory in the mid-90s and my responsibility was to develop a mixer for that receiver that would work with a starved local oscillator. Starved meant we had a very tight thermal budget because of the limitations on the cooler power, so we needed the mixer to work with as little RF energy from the local oscillator as possible. That, coupled with the fact that it had to be space qualified and we had very hard deadlines, was quite a challenge, and we got it done, and it was quite exciting. It was actually my first nonlinear circuit, and back then I guess I would still consider myself as being a kid, and designing your first nonlinear microwave circuit is sort of a rite of passage I suppose.


Now I’m confronted with the most challenging project to date, and that’s completing the 616 element ferroelectric reflectarray, which is a fairly sophisticated contraption – the first of its kind – and if it works like I’m sure it will, I think it will provide a great solution for low cost, high efficiency, scanning antennas.

NTB: You mentioned earlier the Space Communication and Navigation architecture for future space exploration missions. What can you tell us about this effort and how far along we are on it?

Dr. Romanofsky: I’m probably not the best person to respond to that, but I do know that the Space Communications and Navigation program will eventually be responsible for providing all of the communications and navigation technology for flight missions that will eventually be located throughout the solar system.

NASA Glenn has a very significant role in an important experiment that’s underway called “CoNNeCT” (Communication, Navigation and Networking Reconfigurable Testbed) and they’re basically endeavoring to fly, to demonstrate, a series of software defined radios on the International Space Station, and this will be a prelude to what the SCAN project is trying to accomplish, and that is to establish a common framework of protocols, as well as security methods, and ultimately develop a so-called Internet in the sky.

NTB: You currently hold five patents, one of which covers a MEMS reflectarray antenna for satellite applications. What makes this antenna unique and how does it work?

Dr. Romanofsky: That’s actually the ferroelectric reflectarray antenna that I alluded to earlier. In a conventional phased array antenna, the signal is distributed via some type of manifold to, perhaps, thousands of elements in the array. Each one of the channels is amplified and then it’s applied to a radiating element. The physics of a phased array requires that those elements are spaced nominally half-a-wavelength apart, so it’s an extremely dense electronics box and because of the inefficiency, dissipating heat becomes an issue. In addition, the way they are constructed currently is quite labor intensive and requires very special manufacturing capabilities that lead to high costs.

The ferroelectric reflectarray is a planar structure; it’s fed quasi-optically, meaning that there is no beam-forming manifold, so the manufacturing is simplified. And our novel ferroelectric phase shifter is unique in the following way – the finest feature size on our device is on the order of 10 microns. If you wanted to build phase shifting devices at the frequencies that we’re interested in, using conventional transport devices – that is, transistor devices – the feature sizes would be sub-micron, so the lithography on our antenna is much simpler, and that also leads to lower costs. So it’s a combination of factors that the ferroelectric reflectarray offers that promises a reduction of cost of at least, I believe, a factor of ten, and a significant savings in prime power.

NTB: We’ve been hearing a lot lately about space debris and the damage it can do to satellites and spacecraft. Given their size and shape, how susceptible are satellite antennas to being damaged by space debris, and can anything be done to protect them?

Dr. Romanofsky: That’s a good question. I think we’ve been lucky so far. Obviously, we had an event this past January where there was a Russian communications satellite and, I believe, an Iridium satellite that collided. One of the problems is, we have the ability to track objects that are, perhaps, 10cm in diameter. Things that are smaller than that, I don’t think we really understand the distribution.

I think future events are going to be inevitable. One of the biggest problems is, it doesn’t take a very large particle to do a lot of damage because you’re dealing with relative velocities that are in excess of 10 kilometers per second. A good example of that comes to mind back in the early 80s when there was a paint flake that struck the window of, I believe it was Challenger, and actually caused a “crater”. This was an extremely small particle, but because of the very high relative velocities it did a remarkable amount of damage.

The inflatable antennas are going to be particularly susceptible because if we have a micrometeorite penetration – and again, I’m not even talking about orbital debris or manmade orbital debris; I’m talking about natural phenomenon – but if we have a small micrometeorite piercing the membrane, it would, of course, cause the inflation gas to leak. The differential pressure is actually quite small, so it’s not something that’s going to deflate very rapidly. Nevertheless, rigidization is something that would be required so that if we do have an impact, the antenna would survive.

NTB: With a flexible antenna, how do you make it rigid once you deploy it in space?

Dr. Romanofsky: There are several techniques. We’ve been working with a company that used to be called SRS – it’s now Nexolve, in Huntsville, Alabama. There are certain resins, or polymers, for example, that will rigidize upon exposure to ultraviolet light. That’s one technique.


Again, the inflatable membrane technique is only one of several approaches that we’re looking at. Another very promising technology is the shape memory polymer antenna where you don’t have an inflation gas. You rely on the stored energy of the polymer material. What happens is, if you heat the material up above what’s called the glass transition temperature it becomes elastic, it can be folded and stowed, cooled down, and it becomes plastic. If you heat it up one more time, it deploys to the proper shape. That technology would be a lot less susceptible to these micrometeorite types of incidents.

For more information, contact This email address is being protected from spambots. You need JavaScript enabled to view it..

To download this interview as a podcast, click here

The U.S. Government does not endorse any commercial product, process, or activity identified on this web site.