Amps
[Top] [All Lists]

Re: [Amps] Parasitics & Filament Sag

To: "Steve Thompson" <g8gsq@eltac.co.uk>, <amps@contesting.com>
Subject: Re: [Amps] Parasitics & Filament Sag
From: "Tom W8JI" <w8ji@w8ji.com>
Date: Wed, 30 Aug 2006 19:45:04 -0400
List-post: <mailto:amps@contesting.com>
>> During the Grate Parasitics Debate in 1996, a member of 
>> the audience
>> hauled his SB-220 to work and loosely coupled the anodes 
>> to a HP
>> spectrum analyzer.  When the 220 was keyed, he observed 
>> damped-wave
>> ringing at 112MHz at on And at off.
> I wish I could reproduce that. I've tried repeatedly with 
> spectrum
> analyser and 'scopes and only ever find operating 
> frequency ringing,
> even with suppressors removed. I've tried with a couple of 
> amplifiers,
> and also with a tank circuit I put together where I can 
> apply pretty
> quick steps of up to 10A.
>
> Can anyone help point me in the right direction?

I can't, except  my comments below:

> hauled his SB-220 to work and loosely coupled the anodes 
> to a HP
> spectrum analyzer.  When the 220 was keyed, he observed 
> damped-wave
> ringing at 112MHz at on And at off

It sounds illogical or false to me.

One thing we do know for sure is the bias circuit can only 
charge and cut the tube off at the rate the filament choke 
common mode impedance in conjunction with the bypass 
capacitors allow. It's easy to see that rate with a scope, 
or to understand it with a simple low pass filter analysis.

The relay contact generally has a capacitor across it or to 
ground. There is also a couple capacitors in parallel that 
bypass the cold end of the filament choke to ground. Then we 
have the series common mode impedance of the choke, and at 
least one more coupling capacitor that may have to charge.

One thing for certain, that filament choke in conjunction 
with the bypass capacitors has to form a very effective 
lowpass filter for the lowest band the amp is used on or 
considerable RF will appear at the filament transformer. 
This means the rate of bias change HAS to be less than the 
lowest frequency the amp is used on.

The highest rate of anode current or voltage change can't 
possibly occur when the relay switches. This is because, by 
definition, the filament choke and bypass caps have to 
isolate energy at the lowest operating frequency. In other 
words if I had a amp that effectively isolated the tube 
cathode or filament from the relay at 3.5MHz, it would be 
physically and electrically impossible to perturb the bias 
at a rate anywhere approaching 3.5MHz. It simply cannot 
happen.

The "imaginary" switching transient has to be at much less 
than 3.5MHz in this case, there is no way around it. That 
means the tank is subjected to a greater transient level 
change and at a much higher rate during normal amplification 
than it ever could be while going into and out of standby.

Any claim a relay closure or opening could cause an anode 
transient even approaching the speed and level of normal 
operational waveform rates is obviously false. It just can't 
happen. It's as silly as the oscillation bending the 
filament theory, and the oscillation melting a hole in the 
glass theory. Anyone should be able to see that.

73 Tom 


_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>