• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Experiment with 12AU7 pre stage + CSS servo

Hi all,
Here is schematic of my 12au7 pre, it use dual polar hv and an opamp to sense the output node and output current through 2 current mirror (q1,q2 and q7, q8) to set dc offset at output voltage.
You may ask why i make thing so complicated, instead of cap at the input, it can be moved to output so i can get rid of negative hv and opamp servo. But my next stage can have input impedance as low as 10k (common pot value) so i can't use low value capacitor. I want to use capacitor as low as 0.01uf so i can afford hi-end Duelund cap at that position.
With this schematic, it simulate perfectly 0V at output but when i try to change R1 and R19 different value instead of 300R, it no longer work correct, output jump between supplies.
Does anyone have an idea of what's wrong here? Thanks!
12au7 pre + css servo.jpg
 

Attachments

  • pre + css servo.asc
    3.7 KB · Views: 37
O.K.

So your 12AU7 is set to about 15mA of plate current at 200V by cathode resistor R4.
Which means that the NE5532 has to provide the same current through R18, Q7, R20.
The mirror Q1 / Q8 carries 1/3 of that, which is 5mA.
Which is again upped to 15mA through R2 / Q2.

That means that the current mirrors have to stay close to their ratios of 1/3 and 3/1 to allow the opamp to stay within its +/- 15 V swing capabilities.
The maximum tolerance for R1 and R19 is therefore about +/- one E12 step or between 270 and 330 ohms.
Otherwise the opamp output goes rail to rail.

Another issue: the NE5532's output short circuit current is 10mA min.
Thus it may not be able to do +/-15V at 15mA.
If you triple resistor values in current mirrors except R2, the requirement for the opamp output drops to 5mA.
e.g.: R18=3K, R20=300, R19=1K, R5=68K, R1=1K, (R2 stays 100).

Nevertheless R1 and R19 still have to stay within limits (910 ... 1200 ohms) for this to work as intended.
 
  • Like
Reactions: 1 user
O.K.

So your 12AU7 is set to about 15mA of plate current at 200V by cathode resistor R4.
Which means that the NE5532 has to provide the same current through R18, Q7, R20.
The mirror Q1 / Q8 carries 1/3 of that, which is 5mA.
Which is again upped to 15mA through R2 / Q2.

That means that the current mirrors have to stay close to their ratios of 1/3 and 3/1 to allow the opamp to stay within its +/- 15 V swing capabilities.
The maximum tolerance for R1 and R19 is therefore about +/- one E12 step or between 270 and 330 ohms.
Otherwise the opamp output goes rail to rail.

Another issue: the NE5532's output short circuit current is 10mA min.
Thus it may not be able to do +/-15V at 15mA.
If you triple resistor values in current mirrors except R2, the requirement for the opamp output drops to 5mA.
e.g.: R18=3K, R20=300, R19=1K, R5=68K, R1=1K, (R2 stays 100).

Nevertheless R1 and R19 still have to stay within limits (910 ... 1200 ohms) for this to work as intended.
Hi,
12au7 need about 15ma to get 0V at output with 270r cathode resistor, so the opamp need to be able to supply > 15ma of current, i just pick ne5532 for convenient, if really do it, i will choose an opamp with current capability of at least 40ma. Opamp can swing between two rails - 0.7V minus diode drop of Q7 that mean it can output from 0 to (18+18 - 0.7* 3) / 1k (R18) = 34ma so it's fine.
I want r19 and r11 to bigger value so that branch carry less current so Q1 and Q8 have to dissipate less heat. It turn out that the ne5532 model is faulty, it's a custom model that i download from long time ago, just use it for convenient. I change it to LT1028 model that come from with LTSpice and now it working fine, i can change R19 and R11 to bigger value and the output still zero.
I Just update R3 at Q7 collector to initial set the css at 13ma when startup then opamp just need to add another 2ma to zero output offset.
12au7 pre + css servo update.jpg
 
This is just for science or something practical ?
That is not a good operating point for 12AU7 ... current too high , cathode resistor too low , power dissipation over specs . If you want some sort of output power stage I don't think 12AU7 is a wise choice
 
  • Like
Reactions: 1 user