• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Single heater power supply for 2 tubes

I have an adjustable regulated 10A DC power supply, that I'm using right now for heaters of 300b tubes. It works, but both 300b end up with common cathode as they are connected in parallel to the PS. What would be the best way to "isolate" them? Would a Schottky diode and a capacitor do the trick? It will have a small voltage drop, which I can adjust for in the output of the PS.
Joined 2004
Paid Member
If you connect two 300B filaments parallel and heating its with common 5V, the crosstalk between two channels will be about -40dB.
So one channel generates -from its cathode current variation- -40dB disturbance signal on the other channel output transformer.
This corresponds to 1% distortion.

Not a good idea.
  • Like
Reactions: 1 user
I would like to clear up the confusion about this subject.

The results of using a single 5VDC power supply to power two single ended 300B tube filaments, depends entirely on the following:

1. Use individual fixed bias or individual adjustable fixed bias on the 300B grids. Ground one end of the 5VDC filament power supply (which also grounds one end of each of the filaments too).
If the tubes are Extremely well matched, the negative grid bias can be the same (Individual grid bias not needed). But the match may change with tube aging.
Result: One end of the filaments are grounded, there will be No Loss of Stereo Separation.

2. Use a single (common) self bias resistor, and put a large capacitance across the common self bias resistor.
You must use Extremely well matched 300B tubes, since they are sharing the exact same self bias voltage.
Because of the bypass cap, there will be No Loss of Stereo Separation.

3. Note, Stereo Separation:
I built a SE Stereo 45 amplifier, Only one filament winding, so I paralleled the 45 filaments.
There was a pair of 25 Ohm AC balance resistors, each filament end connected through 25 Ohms to the top of the single (common) self bias resistor and self bias bypass cap.
That resulted in 12.5 Ohms of un-bypassed resistance in the filament circuit.
The 45 Stereo Amp Result: The channel separation was -40dB all the way from 20Hz to 20kHz.
Is that a problem?

Let's be realistic, consider a Real Music Signal source:
The best ever separation of an LP Phono Cartridge is -36dB Only at a few mid-range frequencies.
At Bass frequencies, the separation may be only -10dB;
And at high frequencies from as early as 5kHz or 7kHz, and all the way to 20kHz, the separation may be only -10dB.

Do Not tell the LP fans that their LPs are no good.
Have you ever heard an Excellent LP playback system. I Have!

Build your amplifier. Listen. Enjoy!
You will be surprised and pleased.

There were many people who heard my SE Stereo 45 amplifier.
They all liked the sound of it.
Nobody ever complained about the -40dB 20Hz to 20kHz channel separation.
Last edited:
  • Like
Reactions: 2 users

Did you ever listen to good LP playback systems?

Until Ampex and others put 1/4 inch tape players into peoples living rooms . . .
The only living room stereo music sources were:
1. AM FM Stereo Simulcasts (un-matched L and R channels: frequency response and distortion, AM versus Mono FM)
2. FM Multiplex. Without local strong RF signals, the S/N was terrible (30dB quieting versus 50-60dB quieting for Mono FM), and channel separation was not very good either).
3. What you seem to think has terrible separation . . . Stereo LP records.
The causes:
Phono cartridge graphs of frequency versus channel separation (both early cartridges, and today's best phono cartridges).
And, the channel separation of the Record Cutters.

If anyone ever enjoyed listening to # 1 or # 2 above, just remember that those stereo broadcasts used Record Cutters, Phono Cartridges, and LP Records! . . .
Channel Separation?

Just My Opinions (and I bet others might agree too).

4. There was one more Stereo Music source back then . . . Singers, Band, or even a Bosendorfer in the living room.
Sorry, Bosendorfer is 'misspelled", I can not find the umlaut over the first o. Software is not my favorite, locating special characters included.
Last edited:
Joined 2004
Paid Member
"and I bet others might agree too"
Sorry, no.

Explaining a fundamental design flaw by saying that the source is c.ap anyway, IMO the ignoring of the problem.
During planning, we strive to create the best possible product, taking into account the possibilities.

BTW the "other" 300B filament supply cost is a fraction of tube or OPT costs.

p.s. I have few thousand LP, and I listen its ... but not with mediocre listening chain.
Motto: All as weak as its weakest element.
  • Like
Reactions: 1 users
What would be the best way to "isolate" them?
The ends of each DHT filament should also be isolated from each other, as well as from other filaments.

Why the isolation? The music signal flow along at least a part of the filament's wire substrate. This has a resistance, so there is a small signal-voltage across the ends of the filament. Attaching power supplies with electrolytic output capacitors across the the filament crushes the signal; if the power supply is a voltage regulator, it regards the music signal as an "error" which must be eliminated, and in doing so it adds its own corruption to the signal.
Connecting the filament for another channel corrupts the signal by the same mechanism. My filament regulators are used widely here to prevent these problems, and in a stereo chassis, (with fixed bias, so that the negative ends can be grounded), you can feed two of them from one DC feed (preferably unregulated).

Although the 300B - even the cheaper ones - can sound superb, it's better to settle for indirectly heated KT-88s than run two stereo DHT filaments in parallel.
  • Like
Reactions: 1 user
Euro21 and others,

Euro21, you said: "Motto: All as weak as its weakest element."

Here is a good phono cartridge:

I admit, there are better phono cartridges out there, So . . . would you please attach a link to a detailed review of your favorite cartridge.
We want to hear about your phono cartridge's channel separation, channel balance, frequency response, ringing frequency, and harmonic distortion, etc.
"Motto: All as weak as its weakest element."

Please do not misunderstand me, I love the sound of good LP playback systems.
But just like Dirty Harry says: "A man has got to know his limits".

The SE Stereo45 amplifier I designed and built had Channel Separation of -40dB. The most important factor of that is:
The channel separation is Exactly Equal from 20Hz to 20kHz.

Please find us a phono cartridge that has Equal channel separation across the complete audio band.
Let me know when you find one (A Pearl of Great Price).
Last edited:
XY pictures of the channel separation of Stereo Music CDs, to show how little some stereo CDs are,
and how good other CDs are.

TEK0026.JPG Very Little channel separation Kurt Elling "Night Moves", cut # 1 song = album title

TEK0027.JPG Lots of channel separation Walton: Belshazzer's Feast, cut # 5 "Thus in Babylon" Naxos

TEK0025.JPG Medium channel separation

You can see the difference in channel separation.
And, the sound difference is easily heard. TEK0027 was used in group listening sessions. Someone commented that the recording had real good separation.
Get the Naxos CD listed above, and check out the sound of the channel separation (you will be pleasantly surprised at the spaciousness).

As always, Have Fun!


  • TEK0026.JPG
    48.9 KB · Views: 5
  • TEK0027.JPG
    63.6 KB · Views: 5
  • TEK0025.JPG
    54.6 KB · Views: 10