Richard Wrote:
>>> >I am working on an amp with a pair of grid driven 4-400's. What size
>>> >glitch resistor would be recommended for the plate circuit ?
>>>
>>> 1, 10 -15 0hms, 10w. vitreous enamel for up to 3kV. 2 in series for more
>>> than 3kV or >30uF.
>>
>>This is a wire wound resistor ?
>>
>Yes, with nichrome wire, glass-coated.
This question comes up regularly (and I remember foolishly volunteering
to prepare a FAQ web page about it) so can we take this opportunity to
pull together everything we've learned from one another?
Some of the following I remember reading from Rich, and others...
The glitch resistor also needs to be physically long, with connections
at opposite ends. This is important because the full B+ appears across
the resistor in the event of an arc, so a shorter resistor is more
liable to breakdown directly between turns.
Can we identify a specific range of US-made resistors, for the record?
The British equivalent would be the Welwyn 14W "W24" series which are
just over 2 inches long. The specification for sustained maximum voltage
is 750V but they seem to work fine for arc protection at 1-1.5kV per
resistor.
Assuming your tube(s) draw a respectable fraction of an amp, the glitch
resistors will run quite hot. Like Rich, I'd use two (or more) in series
to multiply the power dissipation as well as the voltage capability -
maybe two in series for 2-3kV, and three or more for 3kV upwards.
The metal-clad power resistors are not good as glitch resistors. They
don't have the voltage capability, and the metal case is undesirable
because it shorts out most of the insulation path from end to end. Also
the square ceramic-cased type are not good - not even the ones with
connections at opposite ends - because their surge capability is poor.
Both these and the metal-clad resistors have a resistive element would
on a fibreglass-epoxy core, which can burn, swell and break the resistor
when overheated even briefly. The vitreous (glass) enamel types are
recommended because they contain nothing burnable.
Best of all are the silicon carbide "Glo-Bar" type, because these are
physically long and have the most uniform voltage distribution from end
to end. However, they are expensive and probbaly not necessary for any
but the largest amps. Please can someone remember/find the current
manufacturer and URLs?
--
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.com/g3sek
|