The AM Forum

This got me thinking about it, but still do not want to believe DC is acceptable for a 3CX3000 because part of the filament would have a different bias than the other part.

A guy is selling a 7.5V 140A DC "Switching mode filament power supply for 3CX3000A7/3CX6000A7/4CX5000A/4CX10000D"
271423653948 on the place.

I could see it vaguely in some heavily biased class C application where 7.5V is a tiny fraction of the p-p drive volts, but not in linear amp service.
Also, the costly supply vs. a simple transformer makes little sense. So, unless you had a trailer load of them, why?

Putting aside the important issue of the cost of the filament power supply...

It seems to me that whatever problems would arise due to the voltage drop across the filament, in the case of a DC filament supply, would arise along with the additional problem of hum in the case of a filament supply that reverses polarity 120 times per second.

Stu

I'm sure a call to Eimac will clear it up, how to do it or not.

Most all of the big transmitters up to the mid-40's ran DC on the final fils to lessen the hum problem.  Always from a MG set, some with commutator ripple filtering.

I have one of those MG's from the late 20's, makes 7.5V @ 140 A from 230V 3ph.  Still runs fine, might use it someday in a retro rig for fun.

73DG

Typically tubes of this type and size run AC filaments. For hum reduction, a center tap return is used for cathode current. In cavity circuits where one side of the filament may be earthed in the socket, then DC is recommended as a center tap would short out part of the filament when trying to cancel hum. With tubes in broadcast transmitters like 4CX3500A, 4CX10,000D, 4CX15,000 series, and 4CX20,000 series, AC is used. Ditto for the triode versions from Eimac.

For very large tubes in cavities, such as the Photonis (RCA) 7835 super power triode, or the Thales (Thomson) TH781 tetrode, DC is required not only for the aforementioned cavity connection issue, but also to prevent vibration on the massive filament that could cause a mechanical short. For the Thales tetrodes, a particular polarity is specified, positive ground (cathode). There is a requirement to slowly ramp up the filament voltage on these tubes, over minutes. This is fairly easy to do with DC, using a voltage-controllable switcher but not so easy with AC except to use either saturable reactors, AC/DC/AC converters or motorized variacs.

I remember reading a note from GE that DC on filaments (including lamps/lightbulbs) will reduce the life because localized hot spots are more likely.  They suggested reversing polarity fairly often to reduce this effect.  I wouldn't want to do anything that would reduce the life although typically these big tubes die from lack of emission or other problems long before the heater or filamentary cathode opens.

    Just Picked up a good 3CX3000/A7 off epay for $25..

   "Let'er Rip tater chip"

/Dan

     Yo'

          A good load test for that would be 3   3Cx3000's with
filaments tied in Parallel..  I have the MG set for the ARC-58
that made it a TRC-75.  120VDC in at 110Amps.. melted the
"Drop" to my friggin' house...    "Keep em' guess'in"....


/Dan

I have ran DC on 3cx3000's, 3cx10000's and 3cx20000's with no problems. I have also seen it done on 4cx5000's and 4cx10000.

John N8QPC