I'm interested in building a VFO based single tube CW transmitter.
These devices are known to have issues with frequency drift and chirp; frequency drift can apparently be mitigated by ensuring constant temperature for certain components (coils and capacitors), but chirp is an effect of changes in current within the active element(s) -- in this case, the vacuum tube.
My plan is to use a pentode, since I have a couple miniatures on hand and they offer more gain and more options than triodes. In order to get more than milliwatts to the antenna, however, I need to run about 90V on the plate (better than 400+ on some larger tubes, which might be a second iteration for this project). The question, then, is whether there's a way to key the output of a single pentode while keeping it running continuously as an oscillator, while avoiding running high voltage to the key.
Obviously it would be easy to key a low voltage to one of the grids to start and stop the oscillator, or to do the same for a second tube acting as an RF amplifier -- but is there a way to key a low voltage and keep the single tube oscillating stably?
One way that comes to mind is to key a relay in the output side -- the relay coil runs on 6V or 12V, but depending on the relay part, can switch several hundred volts. With a double throw relay, this would also let me automatically switch the antenna from TX to RX (normally necessary by one means or another anyway to protect the RX from TX power) and be able to hear between code pulses. The output from the TX, in this case, could be routed to a dummy load when the key is up.
Yet this doesn't seem to have ever caught on in the general amateur radio world -- despite setups I've seen that automatically switch antenna on key down for commercial transmitters. Burned contacts in the relay? Other reasons?