Peter Chadwick wrote:
>It is often said in amateur circles that the choke is there to provide a safety
>ground if the feed capacitor from the plate goes short circuit. In that case,
>the choke has to be pretty 'hefty' in that it will need to blow the HV fuse or
>whatever without burning up itself, and the usual 2.5mH choke will literally be
>toast long before that happens with a big amplifier. As Rich says, to just
>bleed the dc of the tuning and loading capacitors, resistors are adequate.
>
>I've never had a case of the feed capacitor going short circuit, although that
>doesn't mean it hasn't happened.
Probably when TV doorknob caps were pushed beyond their limits.
But it is a real risk - although the probability is low, the
consequences of B+ voltage getting onto the antenna line could be very
nasty indeed.
The advantage of a lower-value choke is that it can be wound with
thicker wire, so it would blow the B+ fuse more reliably.
At first glance, a reactance of several time 50 ohms at 1.8MHz should be
OK (about 60uH)... but it would modify the C2 setting in a pi-tank, and
there would be significant circulating currents in the choke.
It's probably better to go for several hundred uH. For example, a 250uH
choke would carry about 100mA of circulating current at 1.8MHz/1500W,
and requires about 30pF increase in C2 to compensate for its reactance.
Resonances don't matter here, because the first resonance is always a
parallel one. The only potential problem is that above this resonance,
the choke looks increasingly capacitive, so above the resonance you'd
have to *reduce* C2 to compensate.
A series resonance would be disastrous, but we can confidently say that
traditional pie-wound 2.5mH chokes don't have any on the HF bands - if
they did, all the ones in amps would be toast; and that ain't so.
--
73 from Ian G3SEK 'In Practice' columnist for RadCom (RSGB)
Editor, 'The VHF/UHF DX Book'
http://www.ifwtech.co.uk/g3sek
|