6J6-based SSMH?

[videoman5]

Anyone have any suggestions for a DIY headphone amp? I was looking into the SSMH, but I'd rather use something with a 12VDC power supply instead of a 48VDC one. I toyed around, and designed this:



I substituted 6J6's for the 19J6's called for in the original SSMH design, and replaced the 48VDC PSU with a 12VDC PSU. From there, I adjusted the resistors accordingly to keep the gate-source voltage at 5v. I am fairly new to DIY headphone amps, so I wanted a "pro" to take a look at it first. So, be nice?

The picture has a few typos; you need a ~1-2A @ 12VDC PSU, not a .5A one, and the "48V" should obliviously read "12V".

[kevinkr]

Not sure how well the 6J6 is going to operate on 12V.. The main reason for the 48V supply was to have sufficient B+ for proper operation of the tubes.

[Frank Berry]

It's not going to work with a 12 volt supply.
If you must use 12 volts, why not use a transistor rather than a tube as your first stage amplifier?

[DF96]

That design is a bit marginal even at 48V supply. The 6J6 is not brilliantly linear, although a headphone amp has quite small signals. The big problem at 12V supply will be distortion created in the source impedance (volume control, plus grid stopper) by grid current.

[aardvarkash10]

12v B+ on a 6J6? Are you going to get enough (or any...) anode voltage movement?

Edit - Ooops beaten to it while checking the 6J6 spec...

[DF96]

12AU7 can work at low voltages, but it won't be very linear.

[videoman5]

Originally Posted by Frank Berry

It's not going to work with a 12 volt supply.
If you must use 12 volts, why not use a transistor rather than a tube as your first stage amplifier?

Gwah, I thought the "6" in "6J6" meant 6v. Anywho, not to be a bother, but would you have a suggestion on what to change out? Pardon my noobiness; I am willing, and wanting, to learn.

[aardvarkash10]

the first 6 refers the the heater voltage...

The 6J6 and 19J6 are essentially the same tube with different heater circuit characteristics.

Change nothing out in the signal circuit - you will need AT LEAST 48V B+, as discussed more would actually be better, although Pete may disagree.

[kevinkr]

Note that you will need to add resistance in series with the filaments as they form part of the output circuit, and you want to drop ~ half of the supply voltage across the FET and the other half across the resistor-filament combination for best linearity, and at a certain point excess voltage across the FET means excessive dissipation.. Go back the original circuit and calculate the series resistance based the filament current and as designed source voltage.. I think 51 ohms might be close enough **but check it** 5W resistor min...

Also Pete is a regular here, you might just send him a PM and see what changes he suggests to use the 6J6...

[videoman5]

Originally Posted by kevinkr

Note that you will need to add resistance in series with the filaments as they form part of the output circuit, and you want to drop ~ half of the supply voltage across the FET and the other half across the resistor-filament combination for best linearity, and at a certain point excess voltage across the FET means excessive dissipation.. Go back the original circuit and calculate the series resistance based the filament current and as designed source voltage.. I think 51 ohms might be close enough **but check it** 5W resistor min...

Also Pete is a regular here, you might just send him a PM and see what changes he suggests to use the 6J6...

So like this? I was wondering exactly what goes where; I can't visualize it without a diagram.