One of the pleasures of being a young ham in the late 1950s was building things as cheaply as possible. Some of our schemes did not work out, of course, such as the pole with the homebrew 10-meter Yagi that fell over as we teenagers tried to put it up. Still, when something we tried did work out, the satisfaction was incredible.
A More User-Friendly Station
It's not always as easy to get the same sort of satisfaction these days, given the nature of the available rigs. We no longer need to combine such things as converters, car radios and the venerable Q5er (a military surplus ARC-5 receiver used as a second IF) to work the world. The desire to feel the satisfaction of such accomplishments has never left many of us, however, so we still look for ways to improve our stations through home-built modifications. It is within that context that the W7VP station model was developed. Figure 1 is a block diagram of that concept.
Someone once observed that if you want to design something that's easy to use, turn the job over to a lazy person. Maybe I am that person, because I'm always looking for ways to make my station more convenient to use. While designing my station I had four primary goals in mind.
My first question was, "How few buttons can I push to change a combination of band, amplifier and antenna?" The answer depends on several issues. First, my setup requires the use of two QSK switches. Both are Ameritron QSK-5s. Each in turn is connected to a separate amplifier, and each amplifier is attached to a separate MFJ-986 antenna tuner. Each tuner can be switched between two antennas, for a choice of four antennas.
- Make the station user-friendly for DXing and contesting.
- Keep band switching as simple as possible.
- Incorporate full-break-in CW (QSK) throughout.
- Make it possible to use any permutation or combination of transceiver, amplifier and antenna.
The ICOM IC-756PROIII has two output ports, designated "Antenna Connector 1" and "Antenna Connector 2." This arrangement permits each port to be connected to its own QSK switch. Since each QSK switch is connected through a separate amp/tuner combination, it's conceivable, for example, to have one amp tuned to 20 meters through a tuner to a 20 meter antenna, while the other amp is tuned to 40 meters through the second tuner to a 40 meter antenna. All that's necessary is to push one button on the transceiver to change amp/antenna combinations.
Amplifier T/R Relays
The key to dealing with the amplifiers' T/R relays is dynamic biasing, a feature that cuts off the amplifier tube(s) between words or CW characters, mainly to reduce heat and save power. Since the QSK switches do all the T/R work and are on the whole degrees faster than the typical amplifier T/R switch, the amps can be left in operate mode with their internal T/R relays in a ready-to-transmit state, provided there is no idling current during non-transmitting periods. This is accomplished through dynamic biasing.
While most Ameritron amplifiers have dynamic biasing, my 4-1000A amp was another matter. A few years ago I decided to modify the venerable 4-1000A by making it a grid-driven amp instead of a grounded-grid amp. Figures 3, 4 and 5 show the result. The amp is driven through a 4:1 balun terminated by a 200 ohm non-inductive resistor. For input tuning, I removed the input switching and replaced it with an LDF AT-100 Pro automatic antenna tuner. This permits tuning for each band's input without switching in different inductances to balance out the inherent capacitive reactance of the tube.
The input bias was designed to switch between approximately -90 V dc and -62 V dc to accomplish cutoff during non-transmitting periods. In order to make the bias switching as instantaneous as possible I inserted the circuit board shown in the Figure 6 schematic. Switching was so fast that for QSK purposes it was virtually instantaneous - at least for the CW speeds I'm used to. Figure 7 depicts the RF envelope during a series of dits at about 27 WPM. In fact, the switching was so fast that most SSB contacts could not tell that the bias was being switched during breaks in modulation.
For the complete version of this article as published in the NCJ, view the pdf version.