Stress Testing the Alpha 8406 6 meter amp 

After completing the updates to the 8406 to take care of some early issues, I was testing the amplifier on the air, with good results locally. With the ARRL Sweepstakes contest coming up at the end of January, I decided to test it in a real contest. I have only a modest 6m antenna- a four element cubical quad, which has limited azimuth rotation capability. Our QTH is right up against the Rocky Mountains, which subtend an angle of approximately 30 degrees to our west.   

 On the positive side, our takeoff to the east is pretty spectacular. On days when the smog is not too bad, we can see Denver International Airport, 50 miles away to the southeast, and could probably see into Kansas over the Great Prairie. There had been some really good "winter E" openings, and so there was hope for some 6m fun. I had noticed that the rise in background noise when plugging the antenna into my Icom 756 pro 2 was almost imperceptible. This may have had something to do with the 150 foot of LMR400 feedline I use, but whatever the reason, my receive sensitivity was definitely suspect. So I built a preamp to put out at the feed-point of the antenna. 

Now, this presented certain engineering challenges. First of all, remember I am feeding an Alpha into the bottom of the feedline, so just to be on the safe side, the preamp must be able to bypass 2kW safely. I decided to use a pair of the big vacuum relays that we employ in the Alpha 9500. A quick test revealed that they only provide around 40 dB of isolation at 50 MHz. With 2kW to bypass, this could still result in 200 mW of leakage into the preamp- enough to potentially damage it. So I added a second lower-power relay, and this reduced leakage into the sub-milliwatt level. After this, it was necessary to interlock the preamp with the radio and Alpha 8406 so that power never was applied at the wrong time. I added sensing circuits which pass a "phantom" dc signal through the relay contacts so it is clear when they are in a position to receive RF. By feeding this signal back, the Icom's "PTT out" signal was delayed by the exact amount to endure that the preamp relays had closed before the 8406 was keyed. The Icom (and all modern rigs, I am sure) automatically provides a short delay (a few milliseconds, and adjustable in many rigs) between "Key Out" and "First RF". This worked great with my preamp, even allowing QSK operation at "full 8406" power!  With the new 8406 tuning indicator based on amplifier efficiency, obtaining all kinds of power was no problem. Enabling the preamp brought a local beacon from 1 s-unit above the noise to 3 s-units, so all appeared to be in order.

And the first little bit of the contest was fun. However it became apparent that E-skip was going to be scarce and I had other demands on my time, so I was not going to sit there for the whole weekend working rovers and the occasional weak E signal. Fortunately, I had interfaced the Pro 2 to a PC, and could run sound card audio into it. And the moon was in a good situation - moonrise at a convenient time and the EME conditions "good". So I decided to try moonbounce, even though my antenna is considered to be at the very bottom of the performance range for this very demanding mode. The only reason I could consider it at all is due to the existence of JT65A. This is a computer-sound-card weak-signal mode which is part of the WSJT (Weak Signal by K1JT) suite of programs. It allows very weak, very low data rate signals to be decoded. It is possible to decode signals around 25 dB below the average noise you hear in your receiver. There is an excellent web site maintained by its creator, Nobel Prize winner Joe Taylor at http://www.physics.princeton.edu/pulsar/K1JT/.  I have used WSJT in the past to make EME QSOs from Peter 1 Island (3YØX) and Desecheo Island (K5D). But in both those cases, I had a better antenna and a good takeoff over salt water. So I was hopeful, but not optimistic that I would be able to complete an EME contact. I mentioned "good EME conditions" above - for those of you who have not worked EME, you may not think that there would be "conditions". It turns out that there are. They are due to a number of esoteric factors, most of which are predictable. These include range (the moon's orbit is elliptical, and the distance varies noticeably) and sky noise. This is the galactic background noise which varies from point to point in the sky. This background noise gets into your receiver and degrades overall signal to-noise ratio. This weekend, the degradation was going to be small.

Now JT65A, the mode used almost universally for 6m EME, is quite demanding on the transmit chain. The mode has a two-minute cycle. One station transmits during the first minute, and the other listens, and then they swap around. During the transmit minute, the signal is on the air for 45 seconds with 100% duty cycle. So as far as the Alpha 8406 is concerned, it is 1.5kW output for 45 seconds; cool down for 1minute and 15 seconds; and then repeat. This is pretty tough, since it can take quite a long time to complete the sequence of exchanges that constitute a "legal" EME contact.

In the end, I began to try with Ants, ES6RQ. I began to see some decodes from him, but after a half hour or so, I thought it was a waste of time. Next I got a note from him saying "Don't give up" via a chat room. While not giving me any real information, I decided to keep going. After more than one hour we completed a QSO! My first EME QSO from my home QTH!

To say I was happy is putting it mildly - working Estonia on 6m under current ionospheric conditions is pretty cool. A half-million kilometer QSO on a quad!  In retrospect, most of it must have been the amplifier. After all, as I pointed out, my antenna is pretty modest, so it cannot take credit. I was impressed with the Alpha 8406. During the transmit period, I had not much to do but watch its performance. I had a wattmeter in line all the time, and a temperature probe in the exhaust air. During the transmit period, I would see the power slowly sag slightly- 100-150 watts or so- this was due to a combination of (mostly) drive power drift in the rig, and maybe some minor output network detuning due to heating. But it was easily brought back up to full power by adjusting the rig's RF power control. The exhaust air temperature, as measured a few mm above the air holes on the top of the cabinet, would climb to about 270F during the transmit sequence, and then drop to below 150F in the receive period.  So solid performance all round.

Nobody knows exactly how good the next sunspot cycle is going to be, but predictions are that it will be on the low side compared to the last few. This may mean less chance for F-layer propagation on 6, and maybe weaker signals all round. On the bright side, there are many more well designed and built antennas on the market. With a good antenna and some power, the vagaries of the sun are less important. And don't forget the preamp either!

Gordon, WØRUN