DC on pilot lamp filaments = short life

<< < (3/3)

WA1GFZ:
the problem with bulbs is the cold resistance. The typical cold resistance is 1/10 that of hot.
This causes a large current transient at turn on. Ths shock is pretty hard on the fine wire inside. I suspect the reason for the coiled up wire. Also running bulbs at a slightly lower voltage will help it to have a long life.
RMS AC voltage .707 peak  is the same heating as DC..

w3jn:
I've run DC on the fhyelliments of a quad of Russian 300Bs in my HB audio amp for a thousand or so hours with no failures.

k4kyv:
I seem to recall reading somewhere that it is preferable to run directly-heated thoriated tungsten filaments on a.c., but if you have to run them on DC, to periodically reverse the polarity of the DC voltage applied to the filaments.

With both light bulbs and tube filaments, it wouldn't be a bad idea to soft start them with a series resistor. 

I use the lowest voltage primary tap available on my filament transformers, usually something like 105 volts, and place a filament rheostat in series, to bring the measured filament voltage down to its normal value with the rheostat set at midrange, at normal a.c. line voltage.  This gives me conrol of the  filament voltage when the line voltage is unusually high or low, and the series resistance limits the surge current at initial startup.

The same thing would work with pilot lamp filaments.  Put in enough series resistance to dim them down noticeably.  This will cause the bulb to give off a softer glow, extend the life of the bulbs, and limit the surge current.

Another approach that wastes less energy and produces less heat would be to use a variac, and install a shaft extension to the rear of the variac (most of them are designed to accomode this), and install a rotary on/off switch to the rear of the variac so that it works just like the classic on/off-volume control that was so common on broadcast radios from the mid-30's until vacuum tubes were phased out.

If you need to switch something over about 5 amps, it might be hard to find an appropriate rotary on/off switch.  Use a heavy duty contactor relay, and let the on/off switch activate the contactor.  Before installing the relay, test it to make sure it runs quietly without a loud buzzing noise.

Just turn the  knob all the way to the left until the power clicks off.  Turn on by clicking the knob to the right, and continue rotating until the filament or line voltage meter reads the correct value.

W2XR:
Don,

Very enjoyable QSO 0n 75M last night-thanks!!!

Directly heated triodes should always be run with AC on the filament. This is desirable so as to maintain essentially a constant bias potental across the entire filament, relative to the grid. With DC on the filament, the bias potential will vary significantly from the positive to the negative leg of the filament.

The only reason DC would be applied to the filament of a directly heated triode would be to minimize the hum modulation caused by AC; but with either a reasonably well balanced center-tapped filament transformer or a hum-nulling pot installed across the filament circuit, the hum is usually down far enough so as not to cause audible noise in most applications.

I run the 833As in my modulator on AC, with the center-tap of each of the two filament xfmrs returning to ground, and there is zero noticable hum in this arrangement. The class A P-P 845 audio driver currently under construction to drive the 833As will use the same filament supply scheme. It was always done this way in BC xmtrs.

73,

Bruce

Navigation

[0] Message Index

[*] Previous page