[Top] [All Lists]

Glitch protection (was Re: [AMPS] amazing doubletalk...)

To: <>
Subject: Glitch protection (was Re: [AMPS] amazing doubletalk...)
From: (Michael J. Tubby B.Sc. (Hons) G8TIC)
Date: Fri, 14 Sep 2001 16:54:19 +0100

----- Original Message -----
From: "John T. M. Lyles" <>
To: <>
Sent: Thursday, September 13, 2001 4:54 PM
Subject: [AMPS] amazing doubletalk...

> I am amazed at the lack of  thought (I mean science) behind
> statements which frequently are repeated here. In the past years,
> this has been discussed and validated on this forum, and in
> professional RF power practice its not even a question to
> regurgitate, and yet, here it is again.
> As a professional involved in RF amplifier designs, I completely
> support and agree with Ian White's reply and references listed at the
> bottom of this note, and have also presented physical evidence in the
> way of photographs of the insides of tubes which have arced (2 weeks
> ago here). They were not due to parasitics. Energy is energy. Whether
> we have 3kV with 80 uF or 30 kV with 80 uF of capacitance. I pointed
> out that in the latter case, one MUST use an active shunt crowbar
> device in addition to series resistance. In ham sized tubes, series
> resistance alone may work. But the cause of an unexplained discharge
> within the envelope of a vacuum tube is NOT always because of an
> undesirable oscillation. Gas pressures inside a tube may change from
> effects from high voltage conditioning, changing RF level, DC
> current, load VSWR. We should ask other EE's and RF practicioners who
> design amplifiers to speak here; unfortunately, most  will shun the
> unprofessional approach of the 'scientific' method as practiced on
> many internet discussions and won't bother.
> I have attempted to give my own experiences to this forum. I
> certainly haven't built as many amateur amplifiers as many here. Or
> troubleshooted them. It is extremely frustrating to read people
> preaching stuff which they believe in like religious zealots. We have
> enough of them in the news in the USA right now.
> On 8/30, Richard Measures <2> responded this way:
> JL  first stated:
> "I recommend that hams consider other methodologies which can cause
> arcs besides an 'all inclusive' parasitic theory....
>   <2> replied to this:
> "// The all inclusive parasitic theory originated with ....."
> JL stated:
> "in industry we don't subscribe to that theory."
> <2> replied:
> "file://nor do I"
> OK,  we finally hear that there are other possibilities in mind?
> 73
> John

Surely there are a range of causes for arcs in tube-based power amplifiers,
for example when I was younger (much younger, like 20 years ago) my first
W1SL 2 x 4CX250B amplifer for 144MHz had a poorly designed power
supply, the screen volts were switched from +330VDC to ground between
transmit and receive - the common contact of each relay was connected to
the tube's screen grid.

In this circuit during T/R change-over the screen grid would float for
(the physical change-over time for the relay) - this would allow the screen
rise in potential (towards the anode) and cause a flash-over on occasions -
maybe only once in 6 months but it used to happen. A simple fix was a 100K
resistor from the screen grid to ground.

Very poor, or no loading - for example the antenna falling off or becoming
disconnected could result is an instantenuous 30:1 VSWR with resultant high
RF voltage on the anode(s) and again... flash bang... this has happened to
during a contest and probably many others.

There are yet other causes for flash-overs, like out-gassing and some would
suggest cosmic rays and/or radio-active decay of heavy elements used in, or
accidently introduced into the tube construction. It is alledged that these
release enegetic nuceli which can strike other gas atoms and cause a break-
down which in turn can result in a flash-over.

All of the above are good reasons for a "glitch resistor" - which reminds me
that I must fit one in my 8877 on 6m.

I used 5 x 220ohm 14W Vitreous Enamel (glass) wire-wound types in parallel
in my last amp (3500V at about 1.2A) - this gives about 44ohms at 70W. Now
the specs say they are rated for 750V AC but in testing they didn't appear
mind a direct short to ground (ultimate glitch) but then again I also have a
16A magnetic circuit breaker (MCB) in the transformer primary which drops
out and saves the resistors from continued stress...


FAQ on WWW:     
Administrative requests:

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