Amps
[Top] [All Lists]

[AMPS] neutralization

To: <amps@contesting.com>
Subject: [AMPS] neutralization
From: G3SEK@ifwtech.demon.co.uk (Ian White, G3SEK)
Date: Mon, 19 May 1997 16:21:28 +0100
Rich Measures wrote:

>>If it isn't in "C&F" or some other 30-year-old publication, does that
>>mean that it can't exist or can't be correct? A few new things have
>>happened since then.

>IMO, the laws of physics do not change substantively.  For example,
>'Cold Fusion' doesn't.

Hey, isn't "Lawsa Physics" supposed to be my line?   :-)

The "Laws of Physics" as they apply to electronic engineering haven't
changed, and are most unlikely to. But there can still be vast numbers
of new engineering developments within those laws. "Care & Feeding" was
a milestone in the engineering of vauum tube amplifiers, but it's still
30 years back along the road.


>>> I liked Ian White's comment that, due to phase shift, every 
>>>negative-feedback [neutralization] circuit (at the fundamental frequency) 
>>>becomes a positive-feedback [regeneration] circuit at some higher 
>>>frequency.   
>>
>>No, not "some higher frequency": I said "some DIFFERENT frequency".
>>My own main concern is with VHF/UHF amps that need to be stabilized on
>>LOWER frequencies. It just shows that you have to consider each case
>>very carefully on its own merits.

>As long as capacitive reactance decreases with frequency and capacitive 
>susceptance increases with frequency, the chance of parasitic 
>oscillations on lower frequencies seems a bit unlikely to me.  How many 
>amplifiers do we see with VLF suppressors?  

As I said, just because you haven't heard of it, that doesn't mean it
ain't so.

Come and look at my UHF amp. As well as the intended - and loaded -
input and output resonances at 432MHz it has unloaded input and output
resonances down at VHF. The capacitance of the tube and the short input
stripline resonates with the inductance of the RF choke, and there's
another VHF resonance at the output side. The potential instability is
handled by keeping the frequencies of these two unloaded resonances well
apart. 

That's not all. The screen bypass capacitor will parallel-resonate with
the inductance of the smoothing capacitor in the voltage regulator,
leaving the screen completely floating at some frequency way down in the
HF region. The cure for that is a small RFC wound on a 100-ohm resistor,
connected between the screen bypass and the outside world.

If you don't take these precautions to avoid LOWER-frequency resonances,
a VHF/UHF amp will go bang just like any other.


>>I still have to smile about the alleged damage caused by VHF
>>oscillations to tubes such as the 3CX800 and 8877, which will happily
>>pump out 1kW or more in amplifiers for 50, 144, 220 and even 432MHz.
>In a VHF amplifier, there is one resonant circuit connected to the 
>anode--i.e., the VHF tank itself.  

That's a fair point: the VHF resonances in a VHF amp are loaded - which
they aren't in an HF amp. What I'm trying to emphasize is that the
damage is caused by the current being too big, and not (or not
primarily) by being VHF. 

I know a reaearch institute that could very easily resolve some of these
questions by modeling transient DC and RF heating in a gold-plated
tungsten wire. Anybody care to sponsor this debate for $50k?

>However, in an HF amplifier, two 
>resonant circuits are connected to the anode.  One circuit is 
>HF-resonant.  The other circuit is VHF-resonant.  This can be checked out 
>yourself with a dipmeter, Ian.  If you find that I am in error, please 
>publish your findings here.   

Don't need to - I'm sure you're correct that those resonances exist, and
that they need to be suppressed. What I'm still not convinced about is
the need to use nichrome suppressors.

>Too bad you don't live a bit closer.  I 
>have a decapitated 8877 on the shelf that--when viewed under a 
>microscope--provides a pretty good example of early-stage gold 
>sputtering.  There are five areas of gold pooling and a number of gold 
>blisters on the grid that have burst their tops and strewn gold 
>melt-balls about the cathode--as Eimacs' Mr. Foote described.  
>And.........if you get smiled-out on that one, I have another gold 
>sputtered 8877 in the garage that WE can decapitate and examine for 
>another round of laughs.

I'm absoltely certain that the sputtering exists, having seen your
photographs, both in QST and on your web pages. What I'm still not sure
about is WHY that sputtering took place, and  especially what frequency
- VHF, HF or even DC - was responsible. All we truly know is that the
grid has been very hot, and probably that there was a loud bang.

Similarly with the bandswitches: all we truly know is that they arced.
They don't tell us WHY they arced, or what frequency was responsible.

I fully accept the raw evidence, but am not yet convinced about the
specific causes.



73 from Ian G3SEK          Editor, 'The VHF/UHF DX Book'
                          'In Practice' columnist for RadCom (RSGB)

--
FAQ on WWW:               http://www.contesting.com/ampfaq.html
Submissions:              amps@contesting.com
Administrative requests:  amps-REQUEST@contesting.com
Problems:                 owner-amps@contesting.com

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