Directly heated cathodes

[boiss]

A quick question here, If one end of a directly heated cathode is taken to ground instead of using a grounded centre tap are there any negative effects for the valve and/or circuit operation? This is with a DC heater supply.
Mike.

[Duke_Nukem]

Depends on power levels.

Since the cathode and heater are now one and the same the anode current goes into the heater. Since one end of the heater is at a different volts to the other, more of the anode current will go to one end than the other.

With an AC heater the effect averages out, but DC can lead to one endof the heater being hotter than the other and hence earlier failure.

Of course, depends on anode current ; for example a 10mA anode current will have naff-all effect on a 2V 1A heater.

TTFN,
Jon

[boiss]

Jon, Thanks, I guess that an Ia of 100mA could be considered naff all with a Heater current of 5A, 2%.
Mike.

[boiss]

Jon, I read an article today where valves with directly heated cathodes were used and the design used relays to switch the polarity of the heaters each time the equipment was switched on. Maybe a bit OTT!
Mike.

[cathoderesistor]

Hello Mike,
I'm not sure I follow your arrangement.
So you've got DC heating, and one side of the cathode grounded; where are the cathode resistor and bypass cap? In the other leg?
What valve is it?

If using DC heaters, it's easier to wire each side of the amp the opposite, then just swap the bottles every now and again.
Happy new year.

[boiss]

The heaters will have one end grounded. Valves are 813s and PSU is a modified PC unit. I like the idea of swapping the valves every now and then. Simple is what works best!
Happy New Year,
Mike.

[Duke_Nukem]

Thread to do with 813 amps moved here :-

https://www.vintage-radio.net/forum/...6603#post96603

TTFN,
Jon

[Chris55000]

Hi!

Its really only a problem with 1.4V battery valves, which usually had the bases annotated (F+) and (F-), since these operated mostly at or near zero bias, with the exception of output types. Incorrect polarity of the LT would sometimes markedly alter the anode current, possibly introducing distortion as well!

The heavier mains types really *were* designed for operation on centre-tapped a.c., and if you tried to run these off d.c., you might well find the stated grid bias is unsuitable, as the effective GB the valve "saw" would depend on which physical end of the filament went to +ve or -ve LT, the actual rated filament voltage for the valve and which end of the LT you earthed!

My suggestion would be to connect two equal resistors of about 100R 2W each, OR a suitably rated pot. of about 200R across the LT you propose to try, and connect the return of the grid leak, transformer secondary, etc., to the GB- source, with the positive GB returned to the pot. slider *and chassis earth*.

This would effectively need a floating LT for the valve, balanced roughly halfway by the pot, with roughly half the LT volts between each end of the filament and chassis earth, adjusting the pot and neg GB to give an anode current as near to the Valve Maker's rated figure as possible.

I'll send a diagram if this isn't clear what I've said!

Chris Williams

[Ray Cooper]

This would effectively need a floating LT for the valve, balanced roughly halfway by the pot, with roughly half the LT volts between each end of the filament and chassis earth, adjusting the pot and neg GB to give an anode current as near to the Valve Maker's rated figure as possible.

You certainly would be able to set up the anode current in this way - but unfortunately there'd be a whole range of settings of the 'heater' pot for which adjustment of the 'bias' pot would give the desired value - if I make my meaning clear...

The arrangement also doesn't remove the basic problem that one end of the cathode will be running at a higher voltage than the other, and that the higher end will therefore be delivering more cathode current. It's really only a problem with large transmitting type valves, which can run at amperes of anode current. What was usually done with those was that the polarity of the heater supply was periodically reversed - usually on a weekly basis. Since most of these high-current valves were (historically) fed from a motor-generator set, this was merely a matter of reversing the field current in the generator. Home constructors are highly unlikely to be such sticklers for historical accuracy...

The problems with DC heater supplies of this nature were one of the main reasons for going over to AC heating - life becomes so much simpler. An arrangement more or less identical to the one you describe was then used - and the (variable) resistor across the heater was coyly known as a 'hum bucker'. By critically adjusting this, almost all of the injected hum problems inherent in AC heating could be tweaked out, and it was a poor TX indeed that had any audible hum caused by its heater supply.

[PJL]

Just to summarise mainly for my own peace of mind.

DC heater supplies on directly heated valves will result in varying potential differences along the heater wire. This will result in varied anode current flow to the heater wire and premature failure. One option is to switch the heater connections but at best this is a compromise.

This is not a problem for battery set valves as they are designed for the purpose and have the polarity of the heater marked up.

Heating from the anode current path has been attributed as a likely cause but I would assume that uneven ageing of the coating is also very probable.

When using an AC supply it must be equally split at the heater connections wrt grid or hum will be induced. If cathode bypass is required then the only possible arrangement is a centre tapped LT.

[oldeurope]

Originally Posted by PJL

Just to summarise mainly for my own peace of mind.

DC heater supplies on directly heated valves will result in varying potential differences along the heater wire. This will result in varied anode current flow to the heater wire and premature failure. One option is to switch the heater connections but at best this is a compromise.

This is not a problem for battery set valves as they are designed for the purpose and have the polarity of the heater marked up.

Heating from the anode current path has been attributed as a likely cause but I would assume that uneven ageing of the coating is also very probable.

When using an AC supply it must be equally split at the heater connections wrt grid or hum will be induced. If cathode bypass is required then the only possible arrangement is a centre tapped LT.

Good morning,
the whole thing is a bit more complicated than expected.

AC heating:
Splitting the heater connections equally does not give perfect results.
One reason is that the cathode emission is not the same at every point of
the cathode.
The 'hum bucker' Ray discribed is a must here.
Note that this methode is not able to compensate all harmonics on the mains sine wave.
An other not seen aspect is the hum caused by the low thermal inertia of the cathode.
This blog shows two examples how to eliminate it.
http://rl12t15.blogspot.com/2007/11/...zung-300b.html.

DC heating:
The influence of the anode current in the cathode temperature
can be minimized by using a voltage stabilized heater supply.
Do not use a hum bucker or two equal resistors here,
the hum should be a warning that something is wrong in the stabilizing arrangement.
For example large riples are breaking through the stabilizer.
Warning: Do not use CRC arrangements.
They always make undeheating caused by aging of the cap after the rectifier.
In most valves the heater current differs this gives wrong heater voltages.

Kind regards,
Darius

[oldeurope]

Hi, I searched for other solutions than mine

http://rl12t15.blogspot.com/2007/11/...zung-300b.html

in the www to eliminate the double mains frequency hum caused
by the low thermal inertia of the cathode of an direct heated tube.

I found a discussion in a japanese(?) forum:

http://www.tubebbs.com/redirect.php?...oto=nextnewset

And this one from Steve Bench

http://members.aol.com/sbench/humbal.html

It definately seems that there is a lot of pioneer work to do...
If you find more related links, please let me know.

Kind regards,
Darius

[boiss]

Hi again All,
And I thought it was simple!!!!
Cheers,
Mike