1. Home
  2. Electronics Design
  3. Tube preamp

Tube preamp

I'm playing around with my amp's tube preamp(12AX7). I have set to pots for the bias for both degeneration and cathode, i.e.,

R1 |

+-C-Gnd | R2 | Gnd

R1 is the degeneration resistance and R2 + C is the cathode bias.

What I don't understand is that when R1 and R2 are almost 0(< 100 ohms) I still get a rather decent signal even with the grid has a signal of at least

1V p2p. I would expect a significant amount of clipping on the positive half cycle. I have checked the voltage on the grid and it is 0 and the cathode at around 200mV or lower.

I get the input signal and output signal on and they are rather very close. Changing the bias from around 0 ohms to 2.5k ohms essentially just starts clipping the peaks. This is fine except I would expect at around 0 that the positive part of the signal would be completely clipped.

What am I missing? (I've checked everything and all the measurments are correct). All I can think of is that the grid is being driven positive but the tube's plate resistance isn't dropping a significant amount to create clipping(the tube could be bad).

Reply to
Jeff Johnson
Loading thread data ...

or

ast

alf

at

e.

he

t

You expect a 12AX7 to ever have a low plate resistance? :-).

At zero bias and 150V or so on the plate through a you're probably still seeing plate currents like two or three mA.

"Clipping" depends very much on the resistor you have between the plate and B+ at this point. Running with no bias you'll get less gain than with the right amount of bias but not a lot. And zero bias is just not that far away from -1 or -2V bias which is where the tube would normally be run.

You're not seeing optimal gain, dynamic range or lowest distortion with no bias but the tube works as a voltage amplifier just fine. This is true for most all the intentionally-very-wimpy triodes.

Tim.

Reply to
Tim Shoppa

Except at very low plate currents (relative to normal operating levels), triodes have quite a linear transfer with a positive grid. With zero bias, a high source resistance could result in a distorted waveform at the grid input due to grid current on the positive half cycle. Otherwise, the input-output characteristics should not show serious clipping just because the grid goes positive.

Reply to
Pimpom

also if you do drive a big enough signal into the grid, and you are using a series C as a DC block, the grid will conduct momentarily and charge the cap and in effect create a negative grid bias. The AVERAGE grid voltage will go down so that the peak positive will just start conducting. That kind of "compression" is what gives overdriven tube guitar amps the desired distortion.

Mark

Reply to
Mark

Are you sure that the R1 potentiometer is actually working properly when in the extreme position and is not erratically open ?

Indirectly heated small signal tubes have such strong electron emission from the cathode that it will drive a floating grid slightly negative.

Look at various small signal RF/IF/audio (especially pentode) stages from previous decades, they often have the cathode directly grounded and a few megohms resistor from grid to ground, with the signal AC coupled to the grid.

The electron charge cloud will bias the grid to several volts negative respect to ground if completely floating. The resistor just leaks excessive charge in order to make the bias more predictable., resulting typically to -1 .. -2 V, which is enough for small signal operation.

This grid to ground potential can be measured with a voltage meter with sufficiently high input impedance ( > 10Mohm).

Reply to
upsidedown

Grid-Leak Bias.

Hope This Helps! Rich

Reply to
Rich Grise

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.