I became interested in filament voltage control after learning how
excessive voltage shortens tube life.
Then with the power company incrementally increasing line voltage in most
areas the concern grew.
Some tube manufacturers recommend using the lowest voltage consistent with
good cathode emission. One note suggested reducing filament voltage until
the emission drops slightly and then increase the voltage about .2 volts.
No mention of observing the voltage or current was made.
What makes sense is to follow that procedure and make note of the resulting
filament voltage. Then monitor the voltage until something changes, at
which time another calibrate cycle can be done.
The question in my mind is how to measure cathode emission. Can this be
done observing plate current? Drive on? Drive off? Pulse the bias circuit to
simulate peak current under full drive? Peak current is normally about 3X
average cathode current. Does emission effect peak the same as average
cathode current?
Perhaps power output is a better indicator of cathode emission?
So many questions; hopefully there are some answers!
73,
Gerald K5GW
In a message dated 7/14/2011 10:18:48 A.M. Central Daylight Time,
wlfuqu00@uky.edu writes:
If you really want to control the filament, make some optics, lens,
filters and detectors and observe the
light from the filament and use that as feedback to the controller. Use
the
color temperature or the black body radiation curve to do it.
Regulating the "average" AC Voltage rather than the RMS Voltage is just
as good as long as
the waveform remains the same. Gee, how did we get by all these years
without filament voltage regulators?
Even all those broadcast transmitters with adjustable (variac) filament
voltages used average or
peak measuring meters. And the variacs and human intervention had a much
slower response.
73
Bill wa4lav
At 09:38 AM 7/14/2011 -0400, Ron Youvan wrote:
>Hsu wrote:
>
> > You are correct. But a problem ,AD's true RMS chip is not a
> inexpensive device.
> > I have an idea, using a inexpensive CdS photoresistor-LED or
> micro-bulb
> > photocoupler( like N110 in ICOM PS-35 power supply) because
bulb
> and CdS
> > photoresistor with very bad frequency response
> > so the sampling voltage will associate with true RMS volt (I
> guess) ...
> > If the noise is not a problem and the tube's cathode is oxide-coated ,
> > the switching regulator is the best. but monolithic IC maximal out
> put current
> > only 10A (ST's L4970) others need external power MOS FET.
>
>
> Hsu said his idea is to use the CdS photo-resistor and a micro-bulb
to
> create a home made
>photo-coupler to get a "slow response" version of a
>"opto-isolator" to measure a tube's heater Voltage, but that will result
>more in an
>averaged value.
>
> The best idea is to use a true RMS Volt meter such as almost every
> Fluke DVM ever
>made to SET the RMS Voltage "at the tube," then use THE CURRENT Amperage
>reading to maintain
>that Voltage. The advantage is if a connection becomes lose and develops
>an IR drop, the current
>will drop and you will turn up the Voltage control for the current that
>gets you
>the correct current which is the correct "at the tube Voltage" until you
>can no more make that
>current, then you will look for a bad connection, find and fix it. (as
>you tighten connections
>the current will rise) A lot of commercial equipment works this way.
>--
> Ron KA4INM - Mistakes are often the stepping stones to utter
failure.
>_______________________________________________
>Amps mailing list
>Amps@contesting.com
>http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|