A comrade at another national laboratory uses 4CX350A by the dozens.
They have 4 driving 4 in a class A high level feedback preamp circuit
at 2-4 MHz. And many installations of this setup. This month we were
chatting about the short (< 1 year) tube life they are having, and I
decided to check with CPI/Eimac. I learned that this tube had been
specified for a certain filament voltage, but it was for certain
military SSB gear (i.e., Rockwell Collins) and that if run at the
datasheet nominal voltage, short emission life would result. However,
if backed off 10% on the fil V., then normal long life would be
found. With adequate cathode current for peaks.
They sell as different version, the 4CX350S (I think that was it?),
which runs the filament at the proper temperature at the rated
datasheet value, for long life.
73
John
K5PRO
>
>Having managed Telefunken 20 kW HF transmitters at work using
>parallel 4CX250B's as drivers, the average filament hours we got
>usually were around 2 years (16000 h) before stage gain decreased and IMD
>increased to a point where the tubes had to be replaced (the PA
>servo tuning required a quite high stage gain to
>work properly). These transmitters were used for voice SSB services,
>with a VOX relay reducing
>plate current to about 5 % of the normal idling value. (for a photo,
>take a look at the end of this page:
>http://home.t-online.de/home/Hbusch/sdj.htm)
>
>On the other hand, the filament hours obtained on 4CX250B's in the
>Standard Radio ST1610 ship's
>transmitters usually were less than 8000 when the transmitters were
>used for automatic radiotelex service.
>A lot of effort was spent to find the causes, and poor filament
>voltage regulation together with long intervals
>of stand-by without plate current was deemed to be the culprit.
>
>However, it seems that some oxide cathode transmitting tubes can
>have exceptional life spans
>if properly handled. The otherwise infamous Eimac P290A comes to my mind
>where a few tubes surviving "infant mortality" sometimes would
>perform according to
>specifications for more than 70000 h. It appears that filament
>regulation, proper cooling
>and flash-over current limiting can be of influence for prolonging
>the oxide cathode life.
>
>I would find it interesting to know the date code on the "retired" 4CX250B
>
>73/
>
>Karl-Arne Markstrom
>SM0AOM
|