[Boatanchors] NEED TUBE INFO

Tue Jul 24 02:51:00 EDT 2007

I think Glen's point that vacuum tube rectifier filaments are, in 
general, powered from independent transformer windings gets to the root 
of the earlier confusion about the filament voltage of the 280 
rectifier.  The rectifier filament is on a different transformer winding 
than the rest of the tube filaments because the rectifier winding has to 
float at high DC potential.  It follows that the rectifier tube filament 
voltage is not necessarily the same as that of the rest of the tubes in 
the rig.

 The rectifier has a directly heated cathode because it's more efficient 
than an indirectly heated cathode, and any hum contributed to the 
rectified DC by the filament AC voltage is filtered out in the power 
supply ripple filter.  The original reason for indirectly heated 
cathodes in the signal tubes was to eliminate AC hum induced on the 
signal by the fluctuating voltage and current of the AC filament supply 
(recall the earliest radio tube filaments were powered by batteries).  
For a directly heated cathode, the AC voltage is constantly switching 
back and forth across the cathode.   The cathode current will be most 
heavily concentrated at the part of the cathode that is most negative in 
voltage, so this means the physical origin of the electron beam on a 
directly heated cathode is constantly moving around as the filament 
voltage fluctuates.  In addition, the magnetic field generated by the AC 
heater current is constantly changing direction, which caused the DC 
electron beam in the tube to sway back and forth a bit.  All this 
physical motion results in a significant line frequency hum being 
impressed on the electron beam.

The indirectly heated cathode was designed to help this.  Since it is 
electrically and thermally insulated from the heater, the entire surface 
of the indirectly heated cathode is at one potential and temperature, so 
the electron beam emission is uniformly distributed on the cathode 
surface.  The body of the indirectly heated cathode is usually made from 
nickel or nickel alloy, which has a magnetic shielding effect to reduce 
the magnetic field of the filament current.

However, as Brian said, there is sometimes a virtue in having the heater 
in indirectly heated cathode tubes biased at a DC potential that is a 
bit more positive than the cathode itself.  The heater is at a higher 
temperature than the cathode, because it is separated from the cathode 
by an alumina sleeve, which is both an electrical and thermal 
insulator.  The heater is usually made of pure tungsten, which is a good 
enough emitter of electrons to make the heater itself behave like a 
cathode, and some of the electrons it emits can make their way to the 
cathode through or around the alumina insulator, resulting in noise 
current adding to the electron beam current.  In some of the old RCA 
vacuum tube design books, they recommend maintaining the AC heater power 
off of ground and floating it at +15 to +40 VDC  to reverse bias the 
diode formed by the heater filament and the indirectly heated cathode.  
This is done in a lot of audiophile amplifiers.  It is not practical in 
a radio having so many tubes, where each tube filament would have a 
different optimum DC bias, and any improving effect would be overwhelmed 
by the usual HF atmospheric noise.

Note that even if DC bias voltages are used on indirectly heated 
cathodes, the voltages are usually modest.  If the voltage between the 
heater and cathode exceeds more than about 100V, there is a significant 
risk of heater/cathode insulation breakdown, depending on the tube type 
- some special tubes are designed with higher voltage cathode 
insulation, especially those used in series voltage regulator circuits.  
Cathode insulation doesn't break down often, but if it does the results 
can be spectacular - and catastrophic.

Bob K4ERR

Glen Zook wrote:
> In most sets one side of the filament/heater is
> connected to ground.  If the filament of the rectifier
> is connected in parallel with the rest of the tubes
> then the rectified high voltage is going to be applied
> to ground through the filaments before the rectifier
> tube shorts out due to the ground connection from the
> filament circuit.
>
> Even if one side of the filament circuit is not
> connected to ground the tubes still won't work because
> in directly heated tubes (where the filament is the
> cathode) you will be applying the positive voltage to
> the cathode instead of a negative voltage.  In
> indirectly heated tubes (which have a separate
> cathode) the filaments can act as the plate (anode)
> and will can burn out because of the electron flow
> from the cathode to the heater.  It depends on the
> actual tube but this is possible in certain cases.
>
> That is why when directly heated rectifiers (which
> most tube type rectifiers are directly heated) are
> used they always have a separate winding on the power
> transformer since the B+ comes off of the filament
> (cathode).
>
> Glen, K9STH
>
>
> --- Brian A Clarke <brianclarke01 at optusnet.com.au>
> wrote:
>
> Would you please explain how having the filaments of
> other than the rectifier at B+ shortens the life of
> those other valves - especially if they are heaters?
> Just as an aside, I regularly design and build valve
> amplifiers with the heaters at quite high dc
> potentials compared with their respective cathodes -
> gets my hum-type noise down around -110 dB wrt the
> output.
>
>
>
>
>
>
>
>
>
> Glen, K9STH
>
> Website:  http://k9sth.com
>
>
>        
> ____________________________________________________________________________________
> Yahoo! oneSearch: Finally, mobile search 
> that gives answers, not web links. 
> http://mobile.yahoo.com/mobileweb/onesearch?refer=1ONXIC
> _______________________________________________
> Boatanchors mailing list
> Boatanchors at mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/boatanchors
> ** List Administrator - Duane Fischer, W8DBF/W9WZE **   
> ** For Assistance: dfischer at usol.com **         
> $$ For vintage radio info, see the HCI web site $$      
> http://www.w9wze.org    
>
>
>   