STRATOSPHERE ONLINE

This is the electronic version of the Stratosphere Newsletter. Occasionally final editing is done to the actual layout of the newsletter and spelling checks, and other corrections may not make it into this edition. The content is complete (except for graphics).


OUR FIRST FLIGHT, EOSS(WVN)-1

Can you believe it? In just under three years we have managed to conduct 14 flights. While I hope this is just the beginning for EOSS, I think it is also important to know our roots. This is a summary of how EOSS came about and our first flight.

Dave Clingerman, W6OAL and myself were driving back to Littleton from Boulder where we had helped to kick off the Deep Space Network dish project on Table Mountain. We got talking that it would be a great project but would probably take several years to rebuild the dishes and put them into service. We wanted a project that would not take so long and one that could involve student participation. All of a sudden Bill Brown, WB8ELK, and his balloons came to mind. Both of us thought balloons would be great and why not put a project together?

We approached the local ATV group, Western Vision Network (WVN), to which we belonged and found lots of support. Thus EOSS-1 was born. We scrounged a balloon and helium from Atmospheric Instrumentation Research (AIR). Ken Zawarski, WB9QDC from AIR loaned us a barometer circuit that would interface with the sound subcarrier of the TV. We borrowed a TV camera and ID board from Bill Brown. Bob, WB4ETT loaned us the 147.555 MHz 2-meter beacon. A scant five weeks after Dave's and my ride home, we planned our first balloon flight to be launched from Clement Park.

We had assembled quite a team in so short a time, many of which I am proud to say still participate in EOSS flights. Dave, W6OAL had almost singlehandedly, constructed the payload package. Lessons learned from Dave's first payload still are a part of each EOSS flight today. Merle McCaslin, K0YUK, was yes, Mr. Balloon and Gas. Rick von Glahn, N0KKZ, was our first tracking and recovery lead and enlisted the help of a growing local fox hunting team. I put the ground station together. Eileen Armagost, WD0DGN, was the 40M and 2M net control central and with some help did quite a job trying to manage this first net.

Four days before our scheduled flight, I ended up in the hospital with a kidney stone. I was gonna miss the balloon flight! Somehow, the day before the flight, I recovered enough to come home as long as I took it easy. Easy for them to say!

Early Saturday morning, November 18, 1990, a crew like no other converged near Southwest Plaza in Robert Clement Park. Antennas were put up, radios and power supplies hooked up, TV monitors were placed where hopefully live pictures from the edge of space would soon be viewed. Quite a crowd had started to assemble.

The balloon was removed from its packing, stretched out on a table and soon the hiss of the gas was heard as the balloon started to fill. It was a clear and sunny day.

At 9:30 A.M., we walked the balloon and payload out into the soccer field. I can still remember the excitement as Merle released the balloon. All of our predictions said the balloon would go east but when we released it, west it went. In a few minutes, the balloon did start east and unlike today, the fox hunters went after it from the launch site.

The video was great! After a few minutes, we could see almost the entire city. We used the call of Tim, WB0TUB, and the color ID board video was magnificent. At approximately 93,000 feet the balloon burst, and as the payload tilted over on its side, we saw the curvature of the earth. Recovering the balloon would be our next challenge.

Thanks to Vince Lawrence, N0UA and his airplane, the weak 10 meter beacon was eventually heard and soon the ground fox hunting team recovered the payload. It had traveled 132 miles east of the launch site. Bob Ragain, WB4ETT, Greg Burnett, K0ELM and Ed Boyer, N0MHU, were among the first at the scene and to this day play important roles in each flight.

We learned a lot, what to do and what not to do. The learning continues with each flight. If it didn't, I don't think it would be quite as much fun. The hatbox like payload, with its American flag rudder was taken apart and eventually rebuilt for the next flight but it had a grand flight that day.

In January 1991, EOSS became a separate organization from Western Vision Network. Many WVN (now Colorado radio Amateur Television League) members still participate and I'm glad to say the fox hunters are still with us. We haven't kept all of our members, we lose and gain a few with each new project. From the looks of the crowd at EOSS-14 and the last couple of hamfest tables, it seems EOSS is still exciting. We've come a long way, but I hope, we've got a longer way to go.

Return to Contents


THE CASE FOR GPS

The last several EOSS flights we have been experimenting with the use of LORAN C. Not only did it provide additional insurance for the recovery of our payloads, but it significantly increased our ability to do real science with our projects. Wind speed, direction, position vs time were real additions to our telemetry capability, so much so that we will continue to depend on this information for upcoming projects. However, this capability has not been as reliable as we like.

I believe we can do better. The LORAN C receiver has proven to be quite sensitive but susceptible to electrical noise. During prelaunch checkout, it is necessary to turn off our checkout terminal and most often our ground power supply. Unexplained losses of lock are often experienced, usually in those last hectic moments before launch. We also experience losses of lock during the flight. Depending on how LORAN C locks up, the position data can be over two miles in error. To keep noise out of our preamp we had to separate the preamp from the Shuttle by ten feet. We have broken the preamp coax at least twice during launch and the antenna wire itself has broken twice during flights.

As part of our evolving Shuttle development, I think it is time to seriously consider GPS as a replacement to LORAN C. We have some testing yet to do but it appears GPS is much less susceptible to noise. The antenna is significantly smaller and simpler. The data itself should be much more accurate now that a full GPS constellation is in orbit. Accuracy is specified at 15 to 50 meters depending on the government induced selective availability (error). And, we will get altitude! Our current barometric pressure sensor altimeter has done an excellent job. But due to the very minute pressure above 100,000 feet, our measurement error increases very rapidly. Measurements at or above 120,000 feet are almost impossible. GPS will correct this, the altitude measurement accuracy is the same as the position accuracy stated above. GPS is the right choice.

GPS board receivers are now less than $400.00. I have been surveying the industry and I believe it is time for us to make this move. Will our LORAN C receiver be wasted? I think not. We can repackage it into the original housing and provide it to the Tracking and Recovery team's chase vehicle for use during recovery operations. During the next month or so I will complete my industry survey, and conduct interference testing using GPS receivers available on loan. At the December EOSS meeting, I intend to formalize my proposal for the purchase and integration of a GPS receiver into our system. I invite your opinions and comments.

ed: The membership voted for the addition of GPS capability. We now have a 5 channel magellan gps receiver and are planning integration with Controller Two (second generation computer control system).

Return to Contents


EOSS TO GO METRIC?

Is it time for EOSS to go metric? I believe we should consider this and if there is consensus, convert over with the development of Shuttle II. I have mixed thoughts on this subject myself but it seems that metric is indeed the standard for both science and education and isn't that what we do? What do you think?

Return to Contents


NEW RELEASE-DEVICE TEST ABORTED

During the last few days prior to the EOSS-14 flight, the Balloon and Payload teams worked out details of a secondary experiment involving the balloon release device. Unfortunately, the test was aborted just prior to launch.

The release mechanism was much like that already successfully flown on previous EOSS flights. An enable signal would be sent to the device mounted at the top of the parachute. When activated, voltage from a lithium battery pack would be applied across a two inch length of Nichrome wire. The wire would heat to red-hot and melt through the nylon cord connecting the parachute to the balloon, thus separating the residual balloon material and cord from the *|parachute. This would (eliminate?)preclude the tangle problem that has been experienced on some previous flights.

This time however, instead of sending a second command to activate the device, that activation would occur by the closing of a weight sensitive switch also mounted between the parachute and the balloon. When the balloon burst as a normal event of the flight, the weight on the switch would diminish, closing the switch and activating the melt-through nichrome wire action. To keep launch dynamics and surface winds turbulence from activating the switch, the enable command was to be sent when the balloon was at a safe altitude, i.e., greater than 60,000 feet. The enable signal to the release device required that a small pair of wires connect the shuttle and the device. This is where the problem occurred.

During the launch sequence where the balloon-payload train was raised into the air, stress was put on the wire and it broke. Because this experiment was not critical to the flight and excessive time would be required for repair--putting us out of our FAA window--the decision was made to abort the experiment.

A similar problem also occurred with the LORAN C coax during the same launch but because LORAN was considered more critical to the flight and repairs would be much faster, this problem was corrected. The Balloon and Payload teams have renewed their efforts to develop an adequate corrective action to the launch induced wire stress problem. The experiment will be flown again at the next opportunity.

Return to Contents


NATIONAL BALLOON SYMPOSIUM SUMMARY

The Edge of Space Sciences (EOSS) National Balloon Symposium was held August 20-22, 1993 at the Denver International Airport Holiday Inn. The purpose of the symposium was to bring together amateur balloon groups, students, educational institutions and representatives of the balloon industry to share information and experiences relating to science and education through the use of high altitude balloons.

The activities got underway Friday evening with a welcome and hospitality gathering. Several payloads and other hardware components were on display. Video tapes of various projects were shown as well. Bill Brown, WB8ELK, provided the evening program with a video of both manned and unmanned balloon flights. Bill's humor provided for a very enjoyable program.

On Saturday morning, the formal symposium began. The moderator for the symposium was Marty Griffen, WA0GEH, of EOSS. The morning sessions were:

- Jack Crabtree, AA0P, Welcome, Symposium Purpose, What is EOSS?
- Tim Armagost, WB0TUB, Amateur Radio and Balloons
- Larry Epley, Winzen International, Balloon Physics and Designs
- Rob Kelly, N0SMR, The SSOK Balloon Microcontroller
- Bill Brown, WB8ELK, Balloon Software, Cheap and Dirty Payloads
- Andy Kellett, N0SIS, GPS/LORAN C Performance vs Altitude
- Gil Moore, N7YTK, Utah State U., Balloon Program of USU

Following an excellent lunch in the hotel's atrium, the afternoon sessions got underway. These included:

- Tom Isenberg, N0KSR, EOSS and Education with Balloons
- Ralph Wallio, W0RPK, High Altitude Balloon Education Team (Iowa)
- Rich Volp, N0PQX, Ground Station Hints and Kinks
- Paul Ternlund, WB3JZV, Tracking Balloons with a Mac Powerbook
- Tim Lachenmeier, Raven Industries, Raven Products and Capabilities
- Mike Manes, W5VSI, Payload Construction with Foamcore, VOR Experiments
- Larry Epley, Winzen International, Stratospheric Dynamics

The afternoon sessions concluded with a hour-long open forum where such subjects as liability insurance, project funding, national federation of balloon groups, and follow-up annual symposiums were discussed.

Following a break and social hour, the Symposium Banquet was held. After a delicious dinner, Larry Epley of Winzen International provided the banquet program, "The Pursuit of Science with Balloons." This was an outstanding slide program detailing the history and explorers of scientific ballooning. EOSS President, Jack Crabtree, then presented award plaques for "Outstanding support of the EOSS National Balloon Symposium." The recipients were: Bill Brown for "fathering" ballooning using amateur radio; Winzen International for technical achievements in ballooning and the support of amateur balloon groups; and Ann Trudeau, KA0ZFI, for her efforts in chairing the symposium committee. Attendees were reminded that the symposium was not over as two balloon launches were scheduled for Sunday morning.

Well before dawn the EOSS ground station was housed in the Loveland Repeater Association's emergency trailer. Around it launch teams were busily preparing for the early liftoffs. Because of the closeness of the hotel to the Stapleton Airport, the FAA had approved the launches but only if they occurred before 7:00 A.M. After the normal amount of last minute excitement and confusion, and just before the launch window was to close, the two balloons and their payloads were airborne. The first carried the WB8ELK cross-band repeater. This one would present a challenge because the 2 meter output power was a mere 5 milliwatts. The second balloon carried the EOSS "Shuttle" and tracking beacon. This time however, the separate 2 meter beacon also carried an automatic 35 mm still camera with exposures programmed every 10 minutes.

The balloons drifted northeast and burst at about 102,000 feet. Due to a problem with the LORAN C antenna during descent, the foxhunters converged on the payloads using DFing techniques while maintaining ground communications with the Colorado Repeater Association's many fine machines. With the help of visual sightings and the airborne N0MHU DF platform, both payloads were soon recovered some 60 miles from the launch site. A short recap of the flight back at the hotel concluded the symposium activities.

The symposium had brought 55 people together, from 10 states plus Canada, making the symposium an "international" event. Three universities were represented, two balloon manufacturers and a major aerospace company were in attendance. All agreed that the symposium had been a success and all returned home with newfound knowledge and friends.

Next year, the National Balloon Symposium will be hosted by the group in Des Moines, Iowa. Watch for details later this year or in early 1994.

Return to Contents


NEW MEETING SITE

Feb. 8th, 7:30 p.m.

Directions: South on I-25 to County LIne Road. Turn left and go east two blocks on County Line Road to the first stoplight. Turn left and go north 300 feet to another stop light. Turn right and go east at that light (you are now on Inverness Drive). Follow Inverness Drive a couple of blocks as it curves to the northeast. Stop at the 4-way stop sign, proceed northeast another block to Inverness Place East. The HP Building is at the Northwest corner of this intersection. Park in the HP parking lot, and enter the side door at the southwest corner of the building (no other doors will be available to you). Talk-in 146.64 or 146.94.

Return to Contents


EDITOR'S NOTE

Return to Contents


ELECTIONS

The candidates for office currently are:

Merle McCaslin K0YUK - President
Mike Doherty KB0JYO - Vice-President
Ted Cline N0RQV - Secretary
Greg DeWit N0JMH - Treasurer
Nominations will be accepted from the floor.

Return to Contents


FUN NIGHT OUT, SATURDAY, FEBRUARY 19TH

Return to Contents


OBSERVATIONS ON UHF DIRECTION FINDING

During the hunt for EOSS 14, the WB4ETT and N0QGH--harmonic Colleen Ragain-- team used the 426.25 MHz video transmitter on 70-cm for direction finding. This is a non-scientific comparison with 2-meter DFing systems.

Hardware used on the 70-cm band was an ICOM 7100 all-mode receiver and a 460 MHz commercial band 6 element Yagi antenna. The antenna wasn't retuned for the ham band. A 10 dB per step attenuator was used in this system to reduce received signal level.

Two 2-meter DFing systems were used for comparison:

1) A Clegg receiver with internal attenuation steps of 30, 60 and 90 dB loss (achieved by turning off DC power to various stages in the front end of the receiver).

The antenna was a 4 element Yagi mounted on a mast through the roof of the van. This antenna could not be adjusted for elevation readings during the test.

2) A Kenwood TH215 HT with "DFing converter" to offset the receive frequency from transmit frequency and provide up to 90 dB of variable attenuation. The antenna was another 4 element 2-meter Yagi mounted on a tripod with azimuth and elevation adjustments.

Vertical polarization was used on all antennas.

Signals were acquired at the same time on all three systems. The signals on 2 meters exhibited a characteristic "ground effect" in the first 15 or 20 degrees (estimated) above the horizon. This effect resulted in "mushy" directivity until the balloon was about 20 degrees above the horizon. The performance of the 70-cm DF system was much better within this "ground effect" zone. The bearings became sharper more quickly during the balloon's ascent. The 2-meter signals were useful but not as precise.

The area from where we were DFing was important. This was a rural location with no reflecting surfaces to bounce the 70-cm signal. Some prior DFing events had shown 70 cm to be much more susceptible to reflections from man-made objects.

As the balloon ascended the ease of aiming the 70-cm beam became apparent. Even though this rugged commercial beam is built like a tank it is still much lighter and easier to handle than the 2-meter antenna, even with its two additional elements.

When the balloon was over 30 degrees above the horizon, the fixed elevation 2-meter antenna started giving incorrect readings. This has been demonstrated time after time. Antennas must be elevated toward the balloon to get usable readings. Use of this antenna was discontinued until the balloon was again below 30 degrees from the horizon on descent.

Elevated 2-meter beam vs elevated 70-cm beam:

The 70-cm beam yields more precise (cleaner) bearings than the 2-meter beam. This is probably due to the beamwidth of the 6 element 70-cm beam being much narrower than that of the 4 element 2-meter beam.

The 70-cm beam is easier to handle. Translating beam direction to compass bearings is still the most difficult chore for a hand-held antenna. The tripod mounted 2-meter beam still excelled in ease of translating to compass bearing because you just lock its position, stand back, and take a compass reading.

The team went into the chase mode as the balloon passed overhead. As we drove east we had no way to DF the balloon. It was only a few degrees off of vertical and slightly east of us. We discovered that the 70-cm beam could be held out the window (on the passenger's side!) and pointed UP toward the balloon. Bearings were not plottable due to the very high angle but we knew where the balloon was at all times as we drove. This is a tremendous advantage that we'll use next time.

As the balloon got ahead of us its angle started to drop. The 70-cm beam was used to compare accuracy of the roof mounted 2-meter beam. When the angle to the balloon was again low enough for the 2-meter beam to regain its accuracy, the 70-cm beam was retired for the day. The remainder of the chase was done on 2 meters as my "70-cm directional arm and elbow" recovered.

Next time?

1) Mount the 70-cm beam on the tripod for fixed DFing!
2) Use the 70-cm beam for "moving" DFing.
3) Continue to DF with 70 cm as we approach the balloon beacon.

Looking back on EOSS 14 I wish I had continued to use 70 cm as we approached the balloon. I don't yet know how it will perform on very weak signals such as when we temporarily lost the 2-meter signals. Maybe the 70-cm signal would have done better???? Let's launch another balloon and find out!

Return to Contents


MICROWAVE DF'ING

The EOSS Lost & Found Department has posted a flawless record of tracking and recovering the payloads from all 14 flights. For the most part, these champion bloodhounds have used 2-meter frequencies to track the balloon in flight and to pinpoint its landing. During EOSS 14, some of the team tried DFing the 426 MHz Amateur Television, ATV, signal with favorable results. Why change a good thing? Perhaps the following information will shed a bit of light on this issue.

Many of you have seen Paul Ternlund's computer-generated tracking reports which show how rarely a reported bearing falls directly through the balloon's position. Bearing errors can be caused by a number of things.

Foremost is the fact that 2-meter beam antennas small enough to carry into the field have fairly broad gain patterns; that is changing the antenna compass heading or azimuth angle just a few degrees off of dead on the balloon beacon usually won't yield a detectable change in signal strength. An experienced DFer with a well-calibrated beam probably is reluctant to claim any better than 5 degrees bearing accuracy. Although this seems pretty accurate, a 5 degree error at a range of 30 miles will miss the balloon by 2.5 miles.

Another phenomenon called multipath can cause the signal to peak in a direction significantly different from a true line-of-sight to the beacon. Multipath occurs when the beacon signal reaches the tracking station over more than one path. One of these is usually the direct line-of-sight path, while the other(s) are bounced off large objects such as mountains, power lines and airplanes. Although the indirect signal path is usually weaker than the direct path, its effect is to cause the signal to peak somewhere between the beacon and the reflecting object. This is especially true if the angle between two paths at the tracking site is small enough that both signals are "heard" by the tracking antenna. A tracking station out on the eastern plains of Colorado may report a bearing between the balloon to his northwest and the front range mountains due west.

Another source of error occurs when the signal arrives from a high elevation angle at the tracking site with a beam aimed at the horizon. The directional characteristics of beams become pretty muddy under these circumstances, and its likely that the antenna will hear a reflected signal from a terrestrial reflector more strongly than the direct signal. Some, but not all, tracking stations can adjust both the azimuth and elevation of their antennas. Rigging a 4-element 2- meter quad with a mobile az-el mount is no mean feat, and can create certain difficulties in the presence of low, overhanging branches.

All of the foregoing problems can be reduced, if not eliminated, by using a highly directional antenna, that is one that exhibits a noticeable drop in signal strength with small changes in pointing-angle away from line-of-sight. Obviously this allows the antenna to be aimed along the direct path quite accurately. Not so obvious is the ability to reject even small-angle multipath signals; in fact, a sharp beam can even distinguish between the two or more paths, allowing selection of the true path as that with the strongest signal.

So why not simply build more directive 2-meter tracking antennas? Well, at a given frequency, directivity and gain usually go hand in hand, which is OK. But gain and physical size also increase together. High gain quads and yagis are characterized by their large director count and long boom length. The bottom line is that a 2-meter beam with significantly better directivity than those now in use in the field would be too large to carry into the field, much less mobile. Witness the 17-element monster used at the EOSS ground station!

There's another maxim which works in our favor, however. That is that the physical size of an antenna of given directivity varies directly with wavelength. In other words, you can install three times as many directors on a given length boom at 432 MHz than you can at 144 MHz, and significantly improve directivity. And the elements are only 1/3 the length, to boot.

So the folks who were tracking the 70-cm ATV signal were hoping to enjoy the benefits of a reasonably-sized highly-directional antenna. From what I've heard, they weren't disappointed. Aren't you glad you read this far?

Now, if tripling the frequency is good, won't going higher be even better? Well, the answer's yes, sort of. True, moving higher will give us smaller and more highly directional antennas with the attendant improvements in bearing determination. But as frequency rises, receiver sensitivity drops, and so does transmitter efficiency. It's also possible to have TOO MUCH directivity. If you can't even hear the beacon unless you're pointed directly at it, you may never find a signal to track. And of course, the balloon must be transmitting on the frequency your sooper-de-dooper radio telescope listens to. But the picture isn't really all that grim.

Let's suppose we went up REALLY high in frequency, to 10 GHz; that's 10,000 MHz, folks. This is 70 times 144 MHz, and the wavelength is about 3 cm, or a bit over one inch. And it's one edge of a 500 MHz wide ham band! You could fit ALL of the lower-frequency ham bands side by side in there and still have room left over, but that's not the point. Instead, let's look at the hardware used there.

Although one could build a monster 10 GHz yagi on a six-foot boom, you'd need watchmaker's skills to assemble it, and the boom diameter would be so small, like 1/16th of an inch, that you'd never get it off the bench intact. For such reasons, antennas for 10 GHz look different from those used on UHF and down. Horns and dishes sized for 3 cm, which are impractically huge on 2 meters, easily fit in a car seat. A 17 db gain horn, for example, has a major dimension of less than 4 inches, and fits nicely in the palm of your hand. To get this gain on 2 meters, you'd need a pair of 15-element yagis with 22-foot booms stacked 6 feet apart! Try mobile az-el tracking with that! Then think about using a light-duty photographic tripod or simply pointing with your hand to do the same job at 10 GHz.

Another point to be considered is that you really don't need all that much gain to track a 1-watt 2-meter beacon in flight; you just need directivity. Some of the trackers even crank-in attenuation to get a usable S-meter response. If the beacon power were cut 20 db, to 10 mW, it would still be trackable even with a small 7 db, three-(?)element beam. (Not on the ground, however, so don't fret that this is some perverse weight saving scheme!)

What I'm suggesting is that a 10 mW 10 Ghz beacon should be every bit as trackable in flight using a 17 db horn and a 10 db noise figure (i.e. stone deaf per modern VHF standards) receiver. These performance parameters are right in line with commonly available amateur 10 GHz gear. And you get superior pointing accuracy and field/mobile portability in the bargain.

Some folks may think that microwave technology is a black art wielded only by the Exalted High Propheads of Ham Radio. Despite that some may wish that image preserved, don't believe it for a second. Radio waves are radio waves. The hardware that we use to manipulate them is chosen more for convenience than anything else, as the previous example illustrates.

You might also be surprised to learn that common amateur 10 GHz gear is both simpler and cheaper than we are accustomed to using on HF and VHF. No kidding. True, one can go overboard and build up a 20 watt SSB rig with a 1 db noise figure front end mounted on a 30 foot dish. You'll need such a rig for moonbounce, but it's overkill for the fun stuff we're talking about.

Low cost 10 GHz gear typically performs to standards far below that required on the more crowded bands. Transmitters are simply modulated oscillators which drift with temperature and supply voltage over many MHz. With 500 MHz to roam in, who cares? The transmitters are usually no more than a cavity resonator (hollow box) with a Gunn diode running through it; feed 6 or 8 volts DC to the diode, and you're on the air with 10 to 100 mW. Tweak the DC supply voltage over a 1 volt range with a pot to supply electronic tuning. Add a little audio to the Gunn supply voltage, like putting a carbon mike in series with the DC supply, and voila! FM!

Receivers typically have no front end gain. Received RF is fed directly to a single diode mixer which is pumped with some spare change output from the transmitter. Feed the ungrounded mixer diode terminal to a broadcast FM receiver, and now you can hear. The FM receiver serves as an IF strip, providing all of the receiver's gain and selectivity. Your 10 GHz receive frequency will be that of the transmitter (local oscillator) +/- the frequency your "IF strip" is tuned to.

The configuration described so far is known as a Gunnplexer. And it plays. I've worked over 150 miles using a 10 mW Gunnplexer; of course the 30 db dish antenna and line-of-sight path sure helped! One peculiarity of 10 GHz propagation is that it travels like light. A good sized tree in the path will eat your signal for lunch. So a 10 GHz beacon is clearly useless for locating a payload on the ground; by the time you hear it, you most likely can see it!

Since ordinary coax makes a great attenuator at 10 GHz, transmission lines usually take the form of hollow rectangular tubing, called waveguide. This "mysterious" stuff is really not much more than a fancy form of twin lead. New commercial waveguide prices will bring tears to your eyes, but it's plentiful on the surplus market. You rarely need more than a few feet, since the RF head is small enough to mount right onto the antenna. One-inch copper water pipe from the hardware store makes a perfectly good circular waveguide. A 17 db horn can be built from scraps of PC board stock. A kid's (alu)snow disk makes a fair-to-middling parabolic reflector, and the dish feed can be a quarter soldered on to the end of a piece of waveguide with a couple of slots filed on it.

For those who want to get on the air quickly, first class and with no hassle or homebrewing, ARR sells a full-up 10 GHz gunnplexer-based transceiver for $500; just hook up 12 Vdc, a mike and headphones. If you don't mind a bit of sweat, you can find surplus 10.525 GHz Gunnplexer heads for as little as $15. A little tweak on the cavity tuning screw brings it right into the ham band. These units were mass-produced for the intrusion alarm market before folks got weary of false alarms triggered by cars going down the street and drapes blowing in the wind. Build a simple 8V adjustable dc regulator fed from 12 V battery, add a horn, a wideband FM IF strip and some simple audio circuitry, and that's it. It won't perform like the ARR and it's harder to operate, but you're many bucks ahead.

ARR advertises in the ham rags. Gary Krancher WB1AUA sells 10 mW surplus gunn heads for $15 and horn antenna kits for less than a dollar per db. His address is:

1502 Old North Colony Road
Meriden, CT 06450
(203) 634-3006

I have his free catalog but haven't ordered anything from him yet, so I can't attest to his quality or service. But his are the best prices I've found anywhere.

A notable downside of Gunnplexer gear is frequency stablity. You don't just dial up a frequency and expect to hear the other guy. As the cavity temperature rises, the oscillator frequency drops at a rate of about 300 KHz per degree C. This is the primary reason that wideband receivers are used with these rigs; keeping a signal in a 15 KHz bandpass is a tough chore. The ARR rigs have very effective AFC to keep you locked on once you've found the signal, but you've still got to find it at the outset, and find it again if the signal drops. Before going out to play 10 Ghz with a friend, it's a good plan to "net up" at close range so each of you knows the approximate tuning voltage where you can find the other guy. If you have a highly directional antenna, it must be aimed approximately right so you can hear the signal while you're tuning for it.

Frequency drift may be a real problem in tracking a balloon. The balloon beacon will experience some wide temperature swings, so your tuning voltage will probably have to change quite a bit over the course of the flight. Reading the packet temperature telemetry may help in this regard. Next, you can probably get your antenna aimed close enough to hear the beacon with the aid of a 2-meter or 70-cm beam. Once you've tuned in the 10 GHz signal, you can peak it and read the bearing off to the degree with confidence.

Your choice of antenna can affect the accuracy of your bearing determination. A 17-db-horn has a 3 db (1 S-unit) beamwidth of +/- 11 degrees. You may have to do some swinging to find the center of the peak, but you also won't have much trouble hearing the beacon while tuning for it. An 18" dish yields about 31 db gain and a +/- 3 degree beam at 50% illumination efficiency. The extra gain will more than quadruple your range and improve your bearing accuracy, but finding the beacon will be tougher. If you choose a dish, however, there's another trick you can put in your bag: you get max gain only when the phase center of the feed is right on the dish focal point. Moving the feed in or out will "defocus" the antenna, yielding lower gain and a wider beam. If you're not hurting for signal strength, then you can tune for the beacon while defocussed, then slide the feed to the focal point to take a final bearing.

While 10GHz DFing will never replace "DC-band" hunting, especially for the final recovery phase, properly equipped microwave tracking stations are very likely to produce tightly converging plots during the flight which can be at least as accurate as on-board navigation receivers. Although some of the lower-frequency microwave ham bands may yield greater range with equivalent results, an effective 10 GHz tracking receiver won't deplete your ham gear budget nearly as quickly.

Return to Contents