How to bias DH dual triodes?

[leadbelly]

Andy Jevans' thread on DHT preamps got my juices flowing, so I started looking at some designs. There is something about the implementation of DH dual triodes like the 3A5 that I don't quite understand: how do you cathode bias them individually, or can you? In other words, since the cathode is continuous for both halves with just a center tap, do you just bias the cathode and does that effectively bias both triodes? Or do these tubes require grid bias? How do you implement a cathode follower? If there is a good design guide for these I would appreciate a link.

[revintage]

Look at the tube as a dualplate monotriode. This way you double Gm. Go for ca 90V/8mA(both halves). Have a few sims that maybe could interest you. CCS on top and parafeed through a 4:1.

[revintage]

I am building the upper one at the moment to try with a 1,5V R6 cell for the heaters.
If this works I will also try the lower where the heaters are fed from 12V DC via CCS. Transformer is LL1676 4:1.

[pedroskova]

Originally posted by revintage
I am building the upper one at the moment to try with a 1,5V R6 cell for the heaters.
If this works I will also try the lower where the heaters are fed from 12V DC via CCS. Transformer is LL1676 4:1.

Hi Lars,

Please excuse my ignorance regarding the bottom circuit that you attached, but wouldn't adding a CCS to the cathode/heater end up fighting with the plate-loading CCS? This is where my ignorance enters, as I blindly hold to someone else's advice that you can have one, but not the other. Can this be done with a DHT? I semi-understand why some use a CCS to set heater voltage, but I'm getting confused by the addition of the cathode into the
equation.

[leadbelly]

Originally posted by revintage
Look at the tube as a dualplate monotriode. This way you double Gm. Go for ca 90V/8mA(both halves). Have a few sims that maybe could interest you. CCS on top and parafeed through a 4:1.

OK, but how could it be done? Look here:

Originally posted by pedroskova
Please excuse my ignorance regarding the bottom circuit that you attached, but wouldn't adding a CCS to the cathode/heater end up fighting with the plate-loading CCS? This is where my ignorance enters, as I blindly hold to someone else's advice that you can have one, but not the other.

This has been discussed before in other threads. Basically, the higher impedance CCS dominates.

[45]

You can cathode bias them. The cathode is obviously in common between the halves.
However, as stated in the data sheet, they have enough separation to be used independently.

To employ the cathode bias you can tie pin 1 and 7 for 1.4V filament supply. Then, instead of just 1.4V, you will supply and additional voltage that will be your bias at the cathode resistor connected between pin 4 (center filament) and ground.
For example, if you have a more typical 6V from a 7806 regulator, you can drop it to 3.9V (using an additional RC filter) so that you will have 3.9 - 1.4 = 2.5V which is the bias at 90V plate voltage and 3.7 mA anode current. In this case the cathode resistor should be approx 11 ohms.
Assuming a 39K load (approx. 5x the rp of the tube) you will need 90+144 = 243V anode supply voltage.
Obviously you can (should, IMO) bypass such resistor but you will need a huge cap...

45

[Eli Duttman]

There is something about the implementation of DH dual triodes like the 3A5 that I don't quite understand: how do you cathode bias them individually, or can you?

LB,

Twin triodes that have common cathode connections, like the 3A5, 6J6, and 6SC7 can be self biased only by the grid leak technique. GE's active EQ phono preamp, which also relies on cartridge inductance for a time constant, illustrates the method.

If using the same bias voltage on both grids is acceptable, LEDs and rechargeable cells under the cathode work.

[45]

Originally posted by Eli Duttman


If using the same bias voltage on both grids is acceptable, LEDs and rechargeable cells under the cathode work.

It doesn't look so practical for 3A5.

45

[Eli Duttman]

It doesn't look so practical for 3A5.

Series wire 2X alkaline "D" cells and connect to the ends of the filament. Connect the filament's CT to the biasing system.

[45]

Originally posted by Eli Duttman



Series wire 2X alkaline "D" cells and connect to the ends of the filament. Connect the filament's CT to the biasing system.

Yes Eli but they won't last so long.
You stay always there worried about charge of the battery.....

45