[Amps] HV fuse

[Ian White, G3SEK]

Joe Subich, K4IK wrote:
>
>
>The US broadcast television transmitters using "inductive
>output tubes" (IOT) - essentially a single cavity output
>with a gridded electron gun - used triggered spark gaps to
>discharge the beam (anode) supply.  It's been a while but
>I seem to remember 40 KV anode supplies with 1.0 - 1.5 Amp
>average anode current ... probably around 4 Amps peak.
>If I recall correctly, the power supplies had three phase
>bridge rectifiers and capacitor input filters with about
>10 uF of capacitance.
>
>The triggered gaps were made by EEG among others.  I saw
>the aluminum foil of #20 wire test a couple of times ---
>wrap a piece of bare #20 around the end of the "dead man
>stick" leaving a 5 cm piece sticking off the end and
>reach into a live HV cabinet.  Worked like a charm (just
>be very certain you touched AFTER the sense resistor -
>a current transformer in one case).
>
>> Bottom line: if you're going to copy OE5JFL's circuit,
>> copy *all* of it.
>>
>> http://www.qsl.net/oe5jfl/flashover.htm
>>
>
>I would be concerned about timing issues in a design that
>uses a cascade of devices to withstand the higher voltages.
>If they don't all trigger at exactly the same time there
>is a real chance of a cascade failure.

Thyristors are not like regular transistors. The voltage rating of a 
thyristor is the maximum that it can hold off before it triggers - not 
fails, but *triggers*. Also, thyristors can be triggered by a sudden 
step-up in voltage (dV/dt triggering). Both of these facts are relevant 
to how the circuit operates.

Due to production spreads, one device in the chain will always trigger 
first, and this particular circuit is designed to make the rest of the 
devices trigger correctly in cascade.

Initially, when all the devices are turned off, the chain of resistors 
and zeners ensures that the total voltage is equally divided between all 
the devices. When the first device triggers, it suddenly steps up the 
voltage on all the others. I believe it's mostly the dV/dt effect that 
then makes the next device trigger... and so on until the whole chain is 
conducting.

Bottom line: OE5JFL's circuit works. It has been widely copied in the 
EME community, to protect large UHF transmitting tubes that are becoming 
very hard to replace at any price.


-- 
73 from Ian G3SEK         'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek