[hpsdr] Switches for Mixers

[Chris Bartram]

I did a lot of work on switches in conjunction with a professional HF project 
I was involved-in a couple of years ago, and the best devices I found were 
microwave power HEMTs. Devices in the 2W region seemed about optimum. 
Unfortunately, for obvious reasons, I can't say much more...

Also, although I'll also have to be a little circumspect for similar reasons, 
the use of a (well designed!) microwave PLL divided down to HF in a simple 
divider IS capable of impressive phase noise performance: the limitation 
being the phase noise of the dividers. Their contribution is similar to the 
phase noise contribution of the logic of a DDS... The discrete spurious 
levels generated by a divided microwave PLL are a lot smaller though...

Incidentally, be cautious about the use of commodity ECL in synthesisers and 
receivers. It can be very noisy. One of the reasons that the arrival of CMOS 
was greeted with a sigh by those of us designing synthesisers at that time 
was the very significant reduction in synthesiser phase noise of the then new 
MOS synths wrt the early bipolar devices. Current mode logic frequency 
dividers - which are still used in current synth. chips - are specifically 
designed for low phase-noise performance.

Chris
GW4DGU