wb4mnf wrote:
> Not needed with RL filter. Only RC filter. Right?
>
Do you mean:
"Not needed with LC filter, only C filter. Right?"
If so, the answer's No - wrong. They both have a C on the output, and
there is more than enough energy stored in a typical C to damage a tube,
if all of that stored energy gets dumped into the tube (eg due to an
internal arc).
For example, 30uF charged up to 3kV stores 135 joules of energy (E = 0.5
* C* V^2). Eimac's limit for energy dumped into tubes up to 1500W anode
dissipation is only 4 joules [1] - so you really do need to take some
precautions.
The purpose of the glitch resistor is to absorb the energy of a current
surge, so that the tube doesn't have to. The voltage drop across an arc
is probably only about 50V [1] so that in an arc from 3kV, 2950V appears
across the glitch resistor and only 50V across the tube. This means that
only 1.7% of the total energy goes into the tube - all the rest goes
into the resistor.
For 135J stored in the capacitor, this would mean that only 2.25J gets
dumped into the tube, which is comfortably less than the 4J limit.
But the mains power is still switched on, and still pouring energy in!
This means that as well as the glitch resistor, you need a *fast* way of
interrupting the mains supply.
A fast, low-rated mains fuse or breaker is generally good enough (but it
requires a step-start in order to be able to switch the thing on without
blowing the fuse or tripping the breaker).
Better is a fast mains relay, powered by a circuit that senses excessive
anode current.
Best of all is a circuit that senses excessive anode current, switches
off the mains, and at the same time fires a thyristor "crowbar" that
shorts the B+ supply to the tube. This means that almost all the stored
energy in the capacitor gets dumped into the glitch resistor. A good
example is at:
http://www.qsl.net/oe5jfl/flashover.htm
It's the usual deal - you can get adequate protection very simply
(glitch resistor and fuse) *if* you select the right components. Better
protection is possible, but also more complicated. As ever, it's your
choice.
[1] Eimac Application Bulletin #17, 'Fault Protection'
I'm keeping this whole thread (thanks also for the e-mails about sources
of glitch resistors) and by the end of this should have enough material
to throw together that promised "flashover FAQ".
--
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.com/g3sek
|