On 11/9/22 3:54 PM, Manfred Mornhinweg wrote:
Don, I have seen that RF rectification technique used to bias switching diodes. But personally I don't like it much, at least not generally. Maybe in a specific application. Bias development by that method will always lag a little behind the RF signal. This means late switching, and distortion (harmonics, IMD).
It's easy enough to make a small self-oscillating flyback converter to generate a high bias voltage. It might take just seven parts or so. So that's what I would rather do. Or use a 555 to drive a small pulse transformer, with a voltage doubler at its secondary. 10 parts. But after adding all the chokes and capacitors and other parts of the switching circuit (with FETs or diodes), the total parts count gets into the range where a relay looks really tempting...
I'd buy an isolated DC/DC integrated converter something like Cui PQME1-S12-S15-M -- 12V in 15V out. $5 each
You can also get the 5V in. So you can drive them with all manner of switching circuits to turn antennas on and off.
They stack and come in different voltages.
I've used a bunch of their other parts in various configurations, often to take 12V battery (which is really 9-18V) and make regulated voltages. Like the PQP3-D12-S5-M - they're a bit pricier, $9.
They're fairly quiet RF wise if you put the right capacitors around them. Because all the current loops are inside a tiny package, they don't radiate much H field noise. It's hard to do that with a discrete design, just because of the physical size of the components.
If you need to knock down the noise, then a LT3042 post regulator (>70dB PSRR at 10 MHz) works.
Packaging and layout are the key.
