Discussion in 'Homebrew and Kit Projects' started by VK4FFAB, Apr 18, 2017.
That is a very good read. Thanks Wolf
FETS make more sense in that application.
Small signal switching almost never needs a PIN diode.
What is important is junction capacitance and reverse recovery time for some applications, the two are intimately related to each other.
In the K-2 I still have to question using the 1N400X series diodes given the high junction capacitance, but I guess it is good enough, they certainly serving as ordinary switches, not as PIN diodes, if it wasn't for the higher breakdown voltage needed a 1N60 would have worked just as well.
As far as the original discussion goes the OP can do what he needs to do with the relay he already has and an 1N5711 to act as a shunt switch across the DBM's input transformer with the appropriate coupling cap..
But inexpensive small-mass reed relays can last 500 million operations and have no cutoff frequency....they'll work right down to DC, which to me is a huge advantage.
That's what Ten Tec used in most of their excellent QSK rigs which have keying envelopes that are nearly perfect at 50 wpm. $2 reed relay vs. $1 FET, I'd usually pick the reed relay -- especially since they can easily handle a couple of amps, good enough for 100W-200W transceivers.
Good PINs that work down to 160m and can handle 200W are much more expensive than either! Microwave PINs, on the other hand, can be really cheap...they're in most cellphones.
As a small point of clarification, I should note PINs don't have "reverse recovery time," they have "minority carrier lifetime," which is tested differently and more important.
A "perfect" PIN for HF use would have a damned near infinite minority carrier lifetime.
After all, a PIN is intended to be a current-controlled resistor; nothing more, nothing less. We used them not only as variable attenuators but also modulators in signal generators even 40 years ago (like the 8640B project). In that design, the variable output level control and the AM modulator were the same set of PINs.
And as a current controlled resistor they make great attenuators, and amplitude modulators at UHF and above.
They can, but not only at UHF...even at HF and MF.
The 8640B has <1% THD (AM) through the entire spectrum, using PIN modulators. And at the same time, using DC bias as the output level control, was adjustable over a 20 dB output range while still providing <1% THD for AM.
That was the goal, and we pulled it off.
I need to pull that manual and look at it. I always liked that signal generator....
It's going to have to wait...
All my spare time is consumed rebuilding front panel assemblies for HP-8569B spectrum analyzers.
Ran into a similar problem. 40dB of isolation matched that of the small reed relays I used for filter switching. Perhaps I don't fully understand your problem but solved mine by putting two relays in series.
You have to select the right reed relays.
Well, also, there's more to it than that. The reed relay coil itself can induce current into the reeds, so that must be very well isolated and decoupled for RF.
And those folks have actually home-designed and -built amateur radio rigs - and used them on the air. They know what they're doing, and talking about.
I've worked W6JL. Beautiful 40 meter CW signal. His design is right here on his Zed page. It uses 1N4007 and 1N5408 diodes.
73 de Jim, N2EY