Test for the digital battlefield at Harris RF Communications
Test Engineer of the Year John Gmitter manages a team charged with developing test stations that support his company’s high-volume manufacturing of tactical communications equipment.
Rick Nelson, Chief Editor -- Test & Measurement World, 3/1/2007
 |
|
In our September 2006 issue, we profiled the accomplishments of six outstanding test engineers from various industries, and we asked our readers to vote for the Test Engineer of the Year. Your choice? John Gmitter of Harris RF Communications. As part of his award, John has designated Monroe Community College to receive a $30,000 engineering grant, courtesy of award sponsors Agilent Technologies, Keithley Instruments, and National Instruments. 2007 AWARDS: Test Engineer of the Year Test Product of the Year Test of Time Award READ OTHER MARCH ARTICLES: |
Those products include the Falcon II series of complete tactical vehicular, base-station, and handheld communications systems. Combined with the Harris RF-6010 Tactical Network Access Hub, the radios form the basis of the Harris Tactical Network, which speeds critical voice, data, and position information to and from the digital battlefield, allowing radio operators to send and receive phone calls and enabling radio outstations to have wireless IP connectivity.
Navy experienceIt was his experience with Harris communications products as an aviation technician in the US Navy that led Gmitter to seek employment with the company. He served in the Navy for four years, he said, where his duties included working on F-14 Tomcats onboard the USS Constellation.
“In my last deployment in the Navy,” Gmitter explained, “I was troubleshooting a station that we used to test the black boxes in the planes, and I had to call Harris customer support”—his first contact with the company. When the Rochester native left the Navy in 1999, it seemed natural to him to apply to Harris.
As for the Navy service itself, that also came naturally: “My father and my father’s father were both in the Navy, and I had heard lots of stories about military life.” On obtaining his degree in electrical engineering, he was drawn toward enlisting. “I figured, if I don’t do it now, I will never do it. And it was a fantastic, outstanding experience. It was hard work, but I really enjoyed it.”
 |
|
Test Engineer of the Year John Gmitter deploys a production test station at Harris RF Communi-cations’ new Carlson Road facility. Courtesy of Harris RF Communications. |
Gmitter was unconvinced. “I said to myself, that really doesn’t sound right—I think I can handle it. So, I went with the team on the spring trip and when I came back, sure enough—he gave me a failing grade—because I’d missed some labs and classes.”
Gmitter had plenty of credits to graduate from MCC at that point, but not enough credits for a degree in engineering. That prompted him to enroll in Utica Tech, where he redoubled his engineering efforts and earned his bachelor’s degree in electrical engineering.
Becoming a test engineer at HarrisWith his college and Navy experience behind him, Gmitter found himself employed as a TE 1 (for test engineer level 1) in product support at Harris, working with Falcon I systems. He was promoted to TE 2 level and became the lead engineer responsible for deployment of a test cell for the new Falcon II.
At that time, Harris had recently acquired a new building—the Carlson Road building—that would allow it to expand its manufacturing capacity to meet the growing demand for Falcon and other Harris products.
 |
|
Engineering manager Kenneth Parfitt recognized the need to revamp the Harris RF Communications test strategy to accommodate higher production volumes. |
It became clear to engineering managers Ken Parfitt and Joseph Zingo that—to effectively use the manufacturing space the company had acquired—it would be necessary to develop a test strategy that would allow for faster and more flexible deployment of test stations. They promoted Gmitter to TE 3 and charged him with heading up the new-product-introduction (NPI) test group’s efforts.
Gmitter and his team faced not only the challenge of expanding capacity. They also had to accommodate new modulation schemes and higher frequencies (up to 2 GHz on the forthcoming Falcon III vs. 512 MHz on the Falcon II multiband version). In addition, they had to find replacements for obsolete instruments and preserve 10 years worth of software-development effort, all while developing small-footprint test stations that wouldn’t take up much production-floor space.
Essentially a player-manager in the process, Gmitter applied his knowledge of RF testing hardware and software architectures and leadership skills to build consensus among the NPI team and deliver upgraded test stations, all within the tight timelines. He found no shortage of talent within the eight- to 10-member NPI test group. “We had a lot of type A personalities in one room who had a lot of great ideas. It could get tense and political,” he said, with respect to whose idea would prevail. “When I came over, it was my job to oversee the overall direction, get these guys to come together, and be responsible for the final decisions.”
Developing a software strategyThe software strategy the team came together on is based on National Instruments’ TestStand 3.1. “We had been using LabView for eight to 10 years and had looked at TestStand, but so far hadn’t had a need for it. But after further review, we decided that, well, we could really start to take advantage of some of its key features, such as resource allocation, instrument locking, and parallel autoscheduling.”
 |
|
Engineering manager Joseph Zingo charged Gmitter with the responsibility of guiding and focusing the talented engineers on the NPI test team. |
One goal of the team was to leverage its old software structure while eliminating some of the drawbacks. “For example, our instrument base structure had a handler and a driver; the top level was a handler, and every piece inside that handler had to be present on the hard drive and be executable, so it was really hard to get a station set up. You could be troubleshooting some digital I/O function that you would never use on the station, but if you didn’t make it work, you couldn’t use anything else. It was really difficult for someone to set up a station just to do a simple thing,” he said.
“So, we decided we were going to change our driver-handler structure to a true object-oriented class-based structure.” To that end, the team enlisted the help of an Endevo OOP wizard, which Gmitter described as a speedy class-based generation tool. The object-oriented programming approach enabled the team to detach functions from the lower-level software so they could execute only the functions they needed. “Everything didn’t need to be present and executable, so if you had a simple function that you wanted to deploy in a half hour, you could do it pretty easily.”
 |
|
The Harris NPI test team based its test-station architecture on the PXI platform, which offers sufficient flexibility to accommodate other types of instruments. Courtesy of Harris RF Communications. |
As for hardware, Gmitter’s team focused on a PXI platform that allowed sufficient flexibility to accommodate legacy GPIB instruments as well as future USB and LXI instruments, should the need arise. A typical test station now might include NI PXI digitizers, Aeroflex PXI signal generators, Pickering PXI switches, Agilent GPIB signal analyzers, Anritsu GPIB RF power meters, Virginia Panel mass-interconnect hardware, and Cytec GPIB switching systems.
Commenting on PXI, Gmitter said, “I like what the PXI chassis does for us. It’s overkill to begin with, but it always provides the ability for production to say, 'we were doing two a day but now we want to do 20 a day.’ If you have a chassis there with spare slots and you designed your station such that it takes advantage of parallel processing, autoscheduling, and resource allocation, you can say, 'Station, I just plugged in two more audio analyzers,’ or 'I put in two more signal generators. Please recognize them and make use of them.’”
 |
|
PXI switch cards from Pickering Interfaces add flexibility to Harris test stations. |
When asked if any one factor contributed most heavily to the project success, Gmitter said, “Teamwork. I just provided the direction and some of the technical expertise.” As for winning the Test Engineer of the Year award, he said, “This is a credit not only to me but to all the engineers and managers at Harris that made the project possible. This was a tremendous team effort.”
