[Moon-net] T/R relays

[Chris Bartram]

John

>  the relays have inductance so you have to tune this inductance
> away. But on the output circuit of your PA it's just a part of the circuit
> and it won't bother that much.. (my guess)

Matching isn't a problem - in the case of the OMRON relays, I used in the old 
muTek amplifier, the loaded Q was of the order of 0.3 and the return-loss was 
significantly better than 20dB. 

> RX wise it can pickup noise just as good as the relay can radiate RF power
> on TX..

Of course it can, but with proper design, it's not a big problem! If you 
consider the power transmission case, the losses in those relays can easily 
be demonstrated to come from dissipative effects (skin effects, dielectric 
losses in the capacitors required to perform the impedance match, PCB losses 
etc.). With a total insertion loss of ~0.05dB/relay, radiation was a very 
small percentage of the total. 

In the receive case, the effective antenna gain was probably around -30dBi. 
Even with a polycarbonate case (chosen for durability) external noise sources 
were simply not a problem.

> But do you realy want such a "radiator" in your system ?? it needs to be
> screened certainly when it is INSIDE your Power amplifier with who know
> what other electronic regulation parts.

In my amplifier, the input circuit of a very high performance GaAsFET 
amplifier was located within a few centimetres of the relays. Were the relays 
significant radiators we'd have seen a _lot_ of failures or device 
degradation! We didn't!

I'm not going to make any cheap comments about other people's products - 
although I'm tempted! At muTek we tried to make very high quality products 
with honest specifications. It now seems that many were ahead of their time. 
To the best of my knowledge nobody else has since tried to use noiseless 
negative feedback around GaAsFET technology in commercially produced 2m 
preamps. We were doing that in 1982! I'm told by people who have made the 
measurements that our now 25 year old 144MHz transverter design still 
measures very well indeed against current products.

There's an old saying: 'any fool can design a product, it takes a good 
engineer to design an affordable product which works well!' I tried - and try 
- to do the latter.

Vy 73

Chris
GW4DGU