> Building a stable amplifier that has high gain becomes increasingly
> difficult the higher ones goes in frequency. Things such as layout
> issues, bypassing and genuine parasitics become critical. When I say
> "genuine parasitics" I don't mean the kind Rich talks about. I am talking
> about the inherent, real world, stray inductances and capacitances in
> components.
The real thing that separates amps are the tubes. Tubes used at
VHF and UHF typically have almost zero length grid leads because
they are brought to a ring or flange. If you look at tubes that are
problematic for VHF stability, they are tubes with long skinny grid
(and sometimes anode) leads and long grid structures inside the
tube.
Tubes that are problems include 572B's, 811A's, 4-1000A's, etc.
Tubes that are almost unconditionally stable when installed
correctly include 8877's, 4CX250's, 3CX3000's, 3CX800's,
3CX1200Z7's, etc.
Some tubes are in the middle, like 3-500Z's and 3CX1200A and D7.
If you want an easy to stabilize HF amp, use a compact tube with
a grid flange. If you want a tube that is a problem, use a 4-1000A or
572B.
The problem is virtually always, for the size of tubes we deal with,
the lead from the anode...out of the envelope...and to the plate
tuning capacitor and the length of the grid lead from the grid to the
grounding point and then through the sheet metal back to the
tuning capacitor ground. And no, you can not evaluate that path
with a grid dip meter. That is total nonsense, because it is an
impedance problem not a resonance or Q problem.
VHF amplifiers live happily and are totally stable with anode Q's in
the hundreds, because the grid is simply resonant far above the
frequency where anode impedance increases. Q isn't even the
major issue.
The shorter we make both those leads, the easier the amp is to
stabilize. The smaller the impedance of the suppressor can be, and
in many cases a suppressor isn't even needed. That's why no one
can offer a formula to calculate the proper size of a suppressor,
despite people who "claim" they can.
It's a shame this entire subject, and the topic of why tubes fail, has
been taken to the low level it is at, with failures caused by tube
manufacturing or operational problems blamed on parasitics. But
then I guess the reason so many people keep swallowing the bait
is it's easier to tell people it's almost always one single problem
and to give one magic cure for everything than to give an honest but
more complex answer.
73, Tom W8JI
W8JI@contesting.com
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|