[Amps] 4CX1600B

Ian White, G3SEK G3SEK at ifwtech.co.uk
Fri Nov 14 18:05:23 EST 2003


R. Measures wrote:
>
>
>>R. Measures wrote:
>>>>The bang would have come from the external circuit, where that same
>>>>large current was also flowing (for example from a fuse blowing).
>>>>
>>>**  I don't buy it because with 0V on the grid, the HV drops to a low
>>>potential because the 1600B is drawing c. 3A, until the mains breaker
>>>opens uneventfully.
>>>
>>Why only 3A? With no grid bias, the tube will draw all the current that
>>its cathode and the HV supply can possibly deliver... and more if it
>>could.
>
>€  All cathodes are emission-limited.  According to Svetlana's
>characteristic curves, a 4cx1600 will draw c. 3A with 0V on the grid and
>350V on the screen.

So "all that its cathode can possibly deliver" in this case is 3A... 
which proved sufficient to make the power supply go bang.

>>
>>Been, there, seen it and heard it. This was only on a little VHF amp
>>using a 5894 but the white-hot anodes lit the whole room, until
>>something in the HV supply failed with a very loud bang! The cause was
>>very simple: the grid bias zener had failed short. The solution was
>>equally simple: replace it with a larger one that didn't run so hot.
>
>**  Zener ratings should be divided by three, with the exception of
>Antarctica in the Winter.

So I learned from that experience!

>>The surprising good news was that the tube seemed completely unharmed.
>>
>**  He tried the tube again?
>
No, still talking about my amp.

>>RF-wise, the amplifier worked fine before the zener failed, and fine
>>again afterwards. KL7RA's amp did the same after his component problem
>>was fixed. He told us that his problem was due to the bias rectifier
>>going short, so the original cause was probably something perfectly
>>simple like a mains spike.
>
>**  Did he report a shorted rectifier?

Yes, he did.

>.
>--   1Ampere pi-section 3-500Z grid to ground chokes are known to implode
>at the same time that a filament-grid short occurs.  Some have speculated
>that the tube shorted, which blew the choke, but when a clip-lead is used
>to short a filament pin to a grid pin on a healthy 3-500Z's tube socket,
>nothing extraordinary happens, and the normal 0V-bias anode-I
>uneventfully flows.   One clue to the cause of the implosion and short
>lies in the VHF parasite suppresor resistor because it typically
>undergoes a large change in R during a tube-shorting event, but without
>exhibiting any signs of external damage.

That's an interesting statement to interject at this point - but in a 
real inquest, you'd be challenged to prove both its truth *and* its 
relevance to this particular dead amplifier. Otherwise it would be 
excluded as irrelevant.

And so:
>>
>>That's why I say there's not a scrap of hard evidence or logical reason
>>to suspect *any* RF-related cause in this particular case.
>>
>**   The evidence may well be found in the autopsy.

You and I have very different ideas about the valid uses of evidence, 
and indeed about what "evidence" even is...

and we'd better end it there, before the moderators become nervous.


-- 
73 from Ian G3SEK         'In Practice' columnist for RadCom (RSGB)
                            Editor, 'The VHF/UHF DX Book'
http://www.ifwtech.co.uk/g3sek


More information about the Amps mailing list