How to power transistors in a tube amp

[FREYES_07]

Hello all!

Want to build a tube amp with a ss fx loop,

How should I power these transistors?

Should I rectify the heaters 6.3V ac?

Thanks for the insight!

[pdf64]

Why not use a different loop with high voltage transistors? There's a lot available using eg LND150s.

Do you realise that a 250k send level pot totally negates the low output impedance of the preceding buffer?

[lonote]

Do you realise that a 250k send level pot totally negates the low output impedance of the preceding buffer?

That is an interesting point that I hadn't thought of.

This looks like a drop-in MOSFET conversion of the Dumbleator schematic (that has a 250K pot in that postition)

[martin manning]

Here's a popular loop that runs from the HV: https://store.metropoulos.net/products/ ... x-loop-kit

[pdf64]

Do you realise that a 250k send level pot totally negates the low output impedance of the preceding buffer?

That is an interesting point that I hadn't thought of.

This looks like a drop-in MOSFET conversion of the Dumbleator schematic (that has a 250K pot in that postition)

Yes, whether by accident or intention who knows, the Dumble loop (and anything based on it, with that high value pot in the send output) isn't transparent.
Unless setting the send level control to max CW, the cathode follwer seems kinda pointless.

[nuke]

Dumble just dropped the high voltage plate supply through a resistor to power the FET input board.

You could also add a half-wave (one diode) rectifier to derive a positive voltage from the bias supply.

I don't think I would rectify the 6.3VAC heater, since it may not always be isolated from the tubes it powers, and we usually have a CT to ground, or a virtual (resistors) center tap, or sometimes, the filament supply is biased +20 to +40 volts above ground to control hum and relieve H-K voltage margins for cathode followers or some phase splitters.

[FREYES_07]

Do you realise that a 250k send level pot totally negates the low output impedance of the preceding buffer?

Nope, how so?

[FREYES_07]

Dumble just dropped the high voltage plate supply through a resistor to power the FET input board.

Cool! I'll try this hehe thanks!

[pdf64]

Marshall use a 47k 10W dropper from the screen grid node, then 30V of zeners, to supply their TL072 fx loop
https://www.drtube.com/schematics/marsh ... 2_iss4.pdf
https://www.drtube.com/schematics/marsh ... -60-02.pdf

[R.G.]

Do you realise that a 250k send level pot totally negates the low output impedance of the preceding buffer?

Nope, how so?

Ohm's Law.

When the pot is at maximum, the buffer impedance drives the output directly, and the 250Kpot resistance is just a resistor to ground (neglecting any wiper resistance). As you turn the pot down, the top end of the pot resistance appears in series with the buffer impedance (and we can imagine that the buffer impedance is zero to a first approximation), so more and more of the pot resistance appears in series between the buffer and output on the wiper.

To a second order approximation, the output sees the parallel combination of the pot resistance above and below the wiper. At the top and bottom of travel, this is zero. As the pot is turned to the middle of the pot resistance, the impedance to the load increases until it is the pot resistance divided by 4 at the resistive middle. So in this setup, the impedance from buffer to output increases from (nearly) zero to 250K/4 = 62.5K.

This both negates the advantages of the low buffer impedance and also causes any tonal effects that a high source impedance might cause in whatever happens to be plugged into the load.

You might like reading "The Secret Life of Pots" at geofex.com.