6DJ8 cathode heater voltage question.

[Morse]

Hi everyone and thanks for all your help in advance!

I'm in the process of building a variant of the Morgan Jones headphone amp, and a question has come up about 6DJ8 heaters and the heater/cathode voltage limit.

Here's the situation:

My 1975 edition of the RCA Receiving Tube Manual gives the heater cathode voltages for the 6DJ8 as:

Peak Value..............unit 1 (--) unit 2 (-150V)
Average Value.........unit 1 (50V) unit 2 (-130V)

Now, the circuit I'm building looks a lot to me like a single gain stage followed up by a White Cathode Follower. Okay, if I use the centre tap on my 6.3V trafo and float it to about 47V, it's dead centre between the voltages on the 2 halves of each of the White Cathode Follower which was my first thought. However, it doesn't fit well with the data in the RCA manual.

So, what do all of you think on this? I've run into problems with cathode/heater potential differences in the past, so I would really like to minimize the potential differences between the cathodes and heaters of all 3 valves.

Thanks again and all the best!
Morse

[dhaen]

Hi Morse,

The figures you quote from the RCA tube manual suggests that there is a difference between the 2 halves.
Likewise this sheet:
http://www.triodeel.com/6dj8_p2.gif mentions that it is for use in cascode amplifiers.
I think therefore, that the H/K insulation on one triode is better than the other.
You will need to observe the specification, and use the "input" section as the bottom valve, and the "output" section as the top.

Since I don't have access to Morgan Jone's design schematic, I can't comment on the actual best heater bias point.

Cheers,

[EC8010]

Heater/cathode insulation voltage limits are very soft, so they vary from valve manufacturer to valve manufacturer despite the fact that they all use essentially the same insulation (aluminium oxide).

A White cathode follower has an upper valve with its heater at an elevated voltage, and the lower valve is a conventional common cathode amplifier. Presumably, you are making a stereo amplifier, so why not split each envelope between channels and use a separate elevated heater supply for the upper White cathode follower valve?

[fdegrove]

Hi,

In the case you stick to the anode voltages as presented in the articles there is no need to bias the heaters.

The insulation between the f/c is given as 150 V,most of the tubes have a similar value BTW.

Cheers,

[Morse]

Hi everyone and thanks for the thoughts;

I've considered dedicating each valve for the different functions and floating each heater as needed, but wouldn't I need to match the sections on the different valves pretty closely? In other words, if I had a 'lower half' that was significantly different electrically from the 'upper half' of the White Cathode Follower wouldn't be a problem?

If I do just build the circuit without floating anything, it will make for an 'interesting' layout. I'll have to work on that.

As far as trusting to published limits, like EC8010 pointed out, each individual valve has it's own limits - and they've caused problems in some of my kit in the past which is why I've been exploring these options.

Other thoughts on this?
Thanks again,
Morse

[dhaen]

Hi Morse,

There is no need to match the top and bottom parts of the stage. They're doing quite different things.

I would suggest you follow EC8010's advice as the best option in this case.

Bit if you must have 1 valve per channel, observe the manufacturers data on input/output sections, biasing the heaters up a little.

Remember, this valve was designed for an RF cascode amp, so these postentials were expected. There should be no H/K insulation problem in this case.

Cheers,

[Morse]

Thanks John;

Okay, I'll do my wiring with each valve split right/left and dedicated to it's function then add a separate 6.3VAC filament trafo to float the heaters on the 'high' part of the circuit. There's plenty of room in this one - I'm building a 8-3/4"x8-3/4"x3-1/2" cherry wood enclosure so I'll just bolt the filament trafo directly underneath the PT.

Anyway, I need to order some new 7 pin ceramic sockets in a week or two since I like the idea of using 6X4's to make a bridge rather than solid state rectification. Come to think of it, I'm down to my last three 6X4's and since I'll need two for a bridge, I'll probably buy half a dozen or so of those while I'm at it!

All the best,
Morse

[Sch3mat1c]

Um, how do you use less than three seperate 6X4s in a FWB?

Tim

[fdegrove]

Hi,

Q:

Um, how do you use less than three seperate 6X4s in a FWB?

A:You'll need four per bridge.

Cheers,

[Morse]

Hi fdegrove and Tim;

I'm planning on either using a single 6X4 or a pair of them.

If a single, I would build a hybrid bridge, with the anodes of a pair of 1N4007's connected to ground and the cathodes of the 6X4 connected to the B+ out. If a twin 6X4 bridge, then I'd just go ahead and use the 4 diodes of the two 6X4's to replace the four 1N4007's of the original schematic.

Since the current draw is very small (less than 30mA) I don't think I'll need to parallel the diode sections inside the 6X4.

All the best,
Morse