New - Report of an attempted QSO between KA7OEI and WA7GIE during the June 2005 VHF Contest

Homebrew transverter:
 

On the workbench:  A barely-working 10 GHz-to-70cm transverter.
Click on picture for a larger version.

This is page has been quickly thrown together as a place to stick some pictures and descriptions of this 10 GHz transverter and other 10 GHz-related activity.

This is a semi-homebrew transverter that converts 10 GHz down to a fixed IF in the 70cm band.

Actually, it's IF is not fixed at all:  It could practically be anything from a few MHz to 1 GHz, but with the local oscillator at 9936 MHz, the weak-signal calling frequency of 10368.0 gets converted to 432.0 MHz and the entire 10 GHz amateur band itself goes from 64 to 564 MHz on the IF, but bandpass filtering limits the practical range from about 10050 to 10500.  This also means that a simple WFM/NBFM exciter/receiver could be built that operates at 314/344 MHz (to transmit/receive at the FM frequencies of 10250/10280 respectively) and take advantage of the relatively low IF for frequency stability.

Shown on the picture is the transverter, on the workbench, being tested with a Yaesu FT-817 tuned to 70cm.

Interested in schematics/details of the various blocks?  Go to the Schematic/Explanation page.

How well does all of this work?  At the time that the photograph at the top of the page was taken it barely worked for several reasons:

• There were way too many clip leads and flying wires.  Some of these clip leads/wires carried low-level/high frequency signals, such as the PLL output, reference signals, and power supply lines with circulating currents.
• The loop filter/amplifier was almost nonexistent.  The output voltage swing of the PLL's frequency/phase detector was only +- 1 volt about 2.5 volts:  This wasn't quite enough to provide proper frequency control on the VCXO.  I simply tacked a capacitor and a couple of resistors to make a really cheesy loop filter and was surprised that it worked at all.
• On the VCXO module, nothing had been "tidied up."  The various multiplier stages had combinations of poorly-bypassed power supply leads and un-dressed signal feeds running around, all unresolved remnants of the initial testing/construction.  The result was that the 110.4 MHz output signal wasn't as clean as it could have been.
• There was inadequate output of the 110.4 MHz signal from the VCXO module to the brick.  This signal was at about 0 dbm - but it should have been higher than this for the brick to be happy.  As it was, it would sometimes takes several "power cycles" of the brick before it locked up.
• Nothing was in a box.  Ultimately, these things will be in their own enclosures.  (The main 10 MHz reference will be its own, self-contained unit, however.)
• There was no really convenient way to see if everything has locked up, yet.
• With just the mixer, the maximum output would be very low - just a few 10's of microwatts at the desired frequency.

The "sort of completed" 10 GHz transverter.  The wiring and circuits have been tidied up and the modules have been screwed/secured to a piece of plywood to make it more portable and durable.
Click on picture for a "hugantic" version.

This picture shows the various modules screwed down to a piece of plywood.  This, while not pretty, at least made it possible to transport the unit around so that it could be used for the microwave contest held on the 22nd and 23nd of January, 2005.  All that was required to make this work was to apply 12 volts (from a battery) and allow the ovenized oscillator to stabilize, and then use the 70cm IF radio to transmit/receive.

A few details about the transverter, in its current state:

• At 12 volts, it pulls about 2 amps.  The vast majority of this is from the "Brick" oscillator, which pulls roughly 500 milliamps at 24 volts.  To get 24 volts, a switching upconverter was used to generate 27-28 volts which was then regulated down to 24 (to guarantee a clean, stable supply voltage.)
• There is currently no T/R switching.  In operation, a 10 db pad was placed on the output of the GaAsFET preamplifier  The preamplifier, with about 18 db of gain, still provided a bit of extra signal even with this extra attenuation.  In transmit, the FT-817 was set to the lowest power level (1/2 watt) I simply transmitted into the output of the preamplifier.  The 10 db pad was enough to protect the preamplifier from damage (with only about 50 milliwatts reaching the preamplifier) and the "backwards preamp" provided an additional 20 db or so of loss resulting in approximately 1/2 milliwatt reaching the mixer - which is about the right amount of signal.
• I utilized the "alarm" relay contact on the brick, connecting to it a red LED that lights up if it has lost lock.
• After mixer losses (10 db) and filtering (about 2 db of filter loss and and 3 db more loss due to the lower-side image at 9504 MHz being removed) about -15 dbm (approx. 30 microwatts) emerges from the mixer at 10368 MHz.  A short coaxial jumper (1 db more loss) connected to a 17 dbi horn antenna resulting in a total EIRP of approximately 1.25 milliwatts.
• The noise figure of this system, on receive, is approximately 15 db.  Even so, it is possible to hear the WA7GIE 10 GHz beacon (on Nelson Peak) with the transverter and waveguide flange (without the horn attached) just sitting on the passenger's seat in the car.  (The WA7GIE beacon is horizontally polarized and uses frequency-shift keying - about 10 KHz - with the "Key Up" frequency being above the stated frequency.)
• None of the circuits are well shielded.  This fact was evidenced by the brick unlocking when I transmitted on 2 meters:  Enough 2 meter RF was apparently getting into the 110.4 MHz amplifiers to make the brick oscillator unhappy.  Fortunately, it would re-lock when I stopped transmitting.
• I have added two power switches:  One for the "brick" oscillator, and another for the PLL/multiplier circuit.  This allows the crystal oscillator to remain powered and maintain frequency while minimize (battery) current drain.  These switches also allow some rudimentary troubleshooting to verify lock/operation of the PLL and brick oscillator system.

One method of getting a microscopic amount of RF at 10 GHz...
Click on picture for a larger version.

A really simple 10 GHz signal source:

Did it work in that state?  Yes, it seemed to, once the appropriate incantations and genuflections were done to make everything lock.

Not being able to hear the 10 GHz beacon (I was inside, in the basement, using just an SMA to WR-90 adapter flange for an antenna, not using a GaAsFET preamp - and I didn't know for certain that the beacon's even running at that moment...) I needed a 10 GHz signal source.

Initially, I thought about trying to receive weak harmonics radiating from the 110.4 MHz multiplier - but a quick check with a calculator showed that they happened to end up in the same place on the converted 3cm frequencies as they did at the 70cm IF.

The (successful) attempt is what you see in the picture:  A microwave mixer diode connected, with a set of EZ hooks, to the output of the service monitor and the diode placed near the open end of the waveguide adapter (I observed that polarization of the diode mattered greatly.)  The service monitor (and locked to the GPS reference) was tuned to 942.550 MHz and the output cranked up all the way to +13 dbm.

The WA7GIE 10 GHz beacon Dave, WA7GIE, operates a CW beacon on approximately 10368.232 MHz - plus/minus 3 to 5 KHz, depending on temperature.  The frequency has been adjusted and/or drifted downwards with age over the years that it has been in operation. This beacon is horizontally polarized using a slot antenna and approximately 200 milliwatts of RF and operates from atop Nelson Peak in the Oquirrh mountains.  Keying is actually FSK, with the "Key-Up" frequency being about 10 KHz higher than the "key down" frequency:  This means that the transmitter is keyed continuously - a fact to remember if attempting to receive it using a wider IF (e.g. narrow or wideband FM.)  The keying is somewhat chirpy and the frequency tends to drift a few KHz with temperature, but Dave plans to address both of these issues in the near future.  Click here to listen to a recording of Dave's beacon (.MP3 file, 160k.)  The first portion of this recording contains an entire beacon cycle while the second portion is the "unkeyed" frequency (about 10 KHz up) with the "reversed" keying.  The final portion was recorded in Narrow FM mode tuned midway between the two frequencies:  The "clicking" is the shifting of the beacon frequency during keying.  (This recording was made with the transverter, using a 17dbi horn, sitting in my living room:  The signal had to go through walls, trees, and a neighbor's house.)

At this frequency, the 4-5 inches of wire that connect the EZ hooks at the ends of the coax (and the hooks themselves) are probably radiating most of the 942 MHz energy, but some of it is getting to the diode:  An S-9 signal was registered on the S-meter of the FT-817.  To convince myself that this signal really was on 10 GHz, I waved the diode back and forth and heard a +-100 Hz doppler shift as I did so.

The signal at 10368.05 (the 11th harmonic of 942.55) sounded somewhat ratty with incidental FM (although good enough for intelligibility on SSB) but I suspect that most of this is the service monitor which starts to sound just a tiny bit "warbly" on 70cm (an effect that would be 24 times worse at 10 gig) - and this is not to mention the possible effects of having low-level signal leads flying everywhere.  The FT-817's display showed that 432.0503 (an error of 300 Hz) at zero-beat, but some of that was likely due to the fact that I'd brought the '817 in from the car not too long before.

Note:  As it turns out, the 11th harmonic, direct from the service monitor, could be heard weakly if the EZ-clips were held right up to the waveguide flange:  With the diode they were detectable from across the room...  The harmonics of the service monitor itself (no diode) could be heard from several inches away, as could the higher-order harmonics from an HT on 70cm - and even on 2 meters!  Note, however, that even narrowband modulation on 70cm becomes extremely wide when multiplied to 3cm!

Wideband FM QSOs using Gunn transceivers:
 

Clint, KA7OEI, with the WFM 10 gig setup.
Click on picture for a larger version.

Most activity on the 10 GHz band is undertaken using Gunn transceivers.  These units are self-contained transceivers, capable of both transmit and receive at the same time by virtue of the fact that the transmit oscillator is also the receiver's local oscillator.

It should go without saying that one cannot receive on the same frequency as one is transmitting, so a standard IF (Intermediate Frequency) of 30 MHz has been chosen.  While this frequency could be practically anything in the HF or low VHF range, 30 MHz was chosen because it is still in the HF range, but low enough that the mixer diode's output is still good.  An additional advantage is that there are not usually any strong signals present at or near 30 MHz.  In the past, initial activity was carried out on the bottom end of the FM broadcast band, at about 87-88 MHz but nowadays, every available slot on the FM band is used.

On 10 GHz, two types of Gunn transceivers are commonly used:

• The so-called "Gunnplexers."  This is a brand name, but it generally refers to the fancier gunn transceivers that not only have a mixer diode, but can be electronically tuned using a built-in varactor.  This electronic tuning allows 50-100 MHz of frequency adjustment allowing tuning and selection of either of the two standard wideband FM frequencies.
• Microwave "door openers."  These typically have a mixer diode, but do not have electronic tuning.  Fortunately, varying the Gunn voltage (typically from 7 to 9 volts for 10 GHz Gunn sources) will also vary the frequency by 5-15 MHz, the precise amount depending on a number of factors.  While this is enough to allow modulation of the transmitted signal, it does not allow tuning between the two standard wideband FM frequencies so, unless test equipment is at hand, the mechanical tuning screw cannot be accurately adjusted to another frequency.
• There are also some types of door openers that have appeared that do not have a mixer diode:  Ron, K7RJ was looking at one of these, trying to figure out how they could possibly work, until he realized that the circuit extracted the mix frequency from the Gunn diode itself:  Apparently, the return signal (or a received signal) actually superimposes itself on the Gunn voltage supply line.  Preliminary tests indicate that while this scheme works, it does not result in as good as sensitivity as using a mixer diode, but it is possible that further refinements (such as parallel L/C networks in series with the power supply lead, better receiver impedance matching, amplification, etc.) may be able to improve performance.

My Gunn transceivers are of the "door opener" type, being Solfan units manufactured in the late 1970's and used, until the mid 80's, in the Dee's Hamburger restaurant in Bountiful.  Not having a varactor diode, I'd tuned the one that I take with me in the field to 10250 MHz, relying on the fact that everyone else was either on 10280, or could choose either frequency.

During the contest, the transverter proved to be extremely useful.  Owing to the cold temperatures, my Gunn transceiver drifted up several MHz, putting the intended calling frequency of 10250 MHz (paired with 10280 MHz) outside its tuning range.  Fortunately, I'd brought along a service monitor.

With the transverter's wide range IF capability, I was able to generate a signal at 344 MHz that, when put into the transverter, caused it to emit a signal at 10280 - the receive frequency when using a 30 MHz IF (while transmitting at 10250.)  Using this scheme, I was able to retune the Gunn transceiver at will to verify that I was actually on the intended frequency - something that came in very handy, especially when I worked others whose Gunn oscillators had also drifted off frequency.
 

Top and Middle:  The vantage point of Ron, K7RJ, above the valley fog.
Bottom:  Ron, working N7MLD in DN31 from the traverse range above Draper.
Click on picture for a larger version.

From my location (along Utah highway 111, in line with 6200 south) I was able to work most of the others that were on 10 GHz that day using Wideband FM on the Gunn Transceiver:

• Dale, WJ7L, near-ish Joe's Rock on Wasatch Blvd (a distance of approximately 13.25 miles.)
• Ron, K7RJ, on the traverse range between Utah and Salt Lake counties (a distance of approximately 15.5 miles.)
• Charlie, N7MLD, who was above the Utah State capitol (approx. 13.75 miles) and later moved to the International Center, near the airport (approx. 7 miles.)
• I was also able to hear John, K7JL, (approx. 12 miles) but he was unable to hear me.  It was clear, from the sound of the signal, that there was ground obstruction at my end.  Later, I moved about 100 feet to the east to clear this obstruction, but some unknown problem seemed to stop his Gunn transceiver from working.  (I could work Ron from that location, but I neither Ron or I could hear John at that point.)

10 gig SSB contact:

Just after sunset, we (Ron and I) called it quits and headed down into the valley fog.  Not too long after getting back onto the freeway, Dave, WA7GIE appeared on the 2 meter coordinating frequency, ready to try a 10 GHz SSB contact.  I immediately headed back toward the site that I'd been at earlier that day (along U-111) but the fog had thickened such that visibility was only 100 feet at most - and I could not find my previous spot.  Pulling over, I turned on my GPS receiver and used it to locate that spot - more or less - and set up in the dark about 100 feet off the side of the road.

Coordinating with Dave, we both tuned in the beacon (then, on about 10368.292 MHz,) noted our frequency displays, and then QSY'd down several KHz  with him calling first.  With his 1 watt of RF and small dish pointed in my general direction, I immediately heard his signal very strongly and peaked my horn antenna on his signal.  Despite my 1.25 mW of EIRP, he had absolutely no difficulty in hearing my signal.  This occurred at about 7:15PM on a frequency of approximately 10368.288 MHz..

After several minutes of fiddling with peaking antennas, etc. we decided to try other modes as well, so we tried NBFM - but Dave could not hear me at all.   We quickly realized that we were too close to his beacon frequency and QSY'd down to approximately 10368.275 and had absolutely perfect noise-free communications.  We then switched to AM:  He was able to hear me just fine and I was able to copy him OK, despite a possible misadjustment of his IF rig (also an FT-817:  The factory settings for AM on the '817 radio have been known to be incorrect on some radios) that badly distorted his AM audio.  We then switched back to SSB and shortly thereafter, concluded the QSO at about 7:22 PM.

To our knowledge, this is the first 10 GHz SSB QSO with both ends in Utah, and is most likely the first QSO of any kind (in Utah) on 10 GHz using AM.  (Note:  A 10 GHz distance record was set in the early 90's using SSB with one of the two ends in southwestern Utah near Cedar City.)
 

Other misc. pictures from the 10 GHz weekend.

Report of an attempted QSO between KA7OEI and WA7GIE during the June 2005 VHF Contest

Avantek 3cm preamplifier details - A few details concerning the use of a surplus 3cm military surplus preamplifer.

Go to the KA7OEI main page.
 

This page last updated 20080902