Re: Attenuators - theory and the design of.
Purely-resistive attenuators, matched to the ssource and sink impedances [assuming these are frequency-independent] are really rather simple.
If the source/sink impedances are inconsistent, these can be 'tamed' by making the attenuation something like 10dB or so - downstream of that you'll really only be seeing the characteristic-impedance of the attenuator, not the source.
Slug the incoming side with a load-resistor that can easily dissipate whatever the source can throw at it.
In my experience the biggest problem with attenuators has been dissipating tens/hundreds of Watts of peak RF energy with 50,000:1 duty-cycles.
"It's only averaging a Watt; what's complicated?!"
|