W8JI wrote:
>
>The best way is to use the blower to turn on bias and filament, the
>bias to turn on HV, and the HV to turn on screen. Then if you loose
>one of the mess nothing explodes or cooks inside the PA.
That sounds like a logical interlock sequence.
>
>> It seems to be the best fix for this would be a high impedance resistor
>> across the fuse to drain whatever leakage current is pulling up the
>> B-. I was wondering if anyone else had any good ideas on a cure for
>> this?
>
>Yes, let the negative rail stay grounded, or near ground and FIRMLY
>clamp it there. Put the fuse in the HV, where it belongs. Be sure to
>use a HV type fuse!
The worst-case event for the HV supply is an internal or external arc
which almost short-circuits the anode to chassis. Follow Eimac's advice
and use a resistor in the HV+ supply to limit that short-circuit current
until the fuse blows. About 25 ohms per kilovolt is a good figure,
limiting the current to 40A (in practice it will be less than that,
because there are other in-circuit resistances in the filter cap and in
the arc, but who's complaining?).
The limiting resistor has to be capable of withstanding the voltage and
also the surge of power until the fuse blows. DO NOT use the resistor as
a fuse - it will pull a continuous arc and will not interrupt the
current! To avoid internal arcing between turns of wire, the resistor
needs to be long-bodied. As recommended several times in this group, a
Globar type is ideal, but you can also wind your own resistor using
resistance wire on a 1-inch former with turns spaced one wire diameter
(the inductance does somme good too). Don't forget the power dissipation
of this resistor in normal operation, which can be in the 50-100W
region.
When you have made sure that the maximum surge current cannot possibly
exceed 40A, you can also design for the rest of the components in the
HV-minus rail. The diode (or two in series) and the resistor that clamp
the HV-minus rail close to ground MUST NOT FAIL until the fuse or other
breaker has interrupted the supply to or from the HV transformer. If
they do fail, the HV-minus rail will zoom away negative with obvious
implications for equipment damage and human safety. Since these diode(s)
and resistor are safety-related components, I personally would not rely
on transient surge ratings, because we cannot guarantee the duration of
the transient. It costs very little to use diodes that are over-rated to
handle the full 40A continuously - they come in fours, as packaged full-
wave bridge rectifiers. Since safety is involved, I see absolutely no
reason to economize!
Finally, consider using a fast interrupter to turn off the main supply
to the transformer, as described by K5PC in the last 24-36 hours. Solid-
state mains relays seem to be developing a good reliability record in PA
usage, but I've also had good experience with conventional contactors.
If you have limited the maximum HV surge current, and have used
components that can handle that current comfortably, at least up to
about a hundred milliseconds, then you probably don't need a fuse in the
HV supply itself.
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
www.ifwtech.demon.co.uk/g3sek
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|