unconditionally stable because the grid has a positive input
resistance. If the grid is tuned LF of the plate, it has a
negative input resistance. On this basis then, a tube with a
good short grid cone (8877?) with a collet type connection
has a good chance of being stable without any suppression. A
2C39 should be pretty good, too..>>
The most stable tubes and easiest to tame if they are
unstable are normally the tubes with the highest operating
frequency range. This is why tubes like 572's and 811A's
are so problematic, while compact ceramic tubes good at VHF
or UHF are the easiest to stabilize.
>.As an aside, why were there amplifiers in the 1950s with
>series inductors in the grid for parasitic suppression?
>That appears, on the face of it, to be ridiculous....
Adding grid inductance isn't generally a good idea, but
sometimes moving the grid resonance much lower than the
anode can still stabilize a tube just by separating the
resonances. In parasitic mode, nearly every case is where
the amp acts like a TPTG oscillator. You are simply moving
the grid away from the anode resonance, although moving it
higher would indeed be better.
.
> The 'conventional' parasitic approach is, if I understand
> the various arguments, to lower the circuit gain at VHF by
> reducing the plate load impedance. I don't see how a
> resistance wire suppressor does that, since a resistance
> and inductance in series have an increasing impedance, at
> least until the parallel resonance caused by the self
> capacity of the inductor is reached.
Actually it increases impedance up to as much as 1.414 times
the resonant frequency. Consider an inductor in parallel
with a capacitor. If Xc=XL at F the value of reactance at
anything less than 1.4 F increases. This is why some shunt
capacitance in a balun or choke can actually increase
effective inductance over a very wide frequency range.
So the circuit gain would increase, while with the shunt
inductance and resistance, it tends towards whatever the
impedance of the resistor with its strays is. However, the
plate parasitic frequency would decrease, thus meeting the
criteria above for parasitic suppression. It would act as a
sort of glitch resistor though....
> Or do 'conventional' parasitic suppressors actually do
> both? The inductance lowers the parasitic frequency, but
> provides another VHF parasitic possibility, damped by the
> shunt resistors?
Actually multiple things are required Peter.
The first is the suppressor must have an impedance that is
significant compared to the impedance of the whole anode
system from tube anode to the input reactance presented by
the tank at the parasitic frequency. Without that, anything
you do has little effect on the system.
The second thing is it must look resistive enough to have
enough bandwidth to affect any possible resonance, the Q
can't be too high.
The third thing is you don't want the anode impedance to be
too high at the frequency of instability.
The fourth thing is you generally want the anode to be
parallel resonant below the grid parallel resonance.
Anyone who claims one suppressor system fits all is clearly
just using junk science.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|