I did quick layout for BJT-FET linestage. I should have the PCB in ~2 weeks to test it in my DAC. Not yet decided if the GND will be a plane or just a wire...as well as missing few caps for V+/- decoupling.
Thank you Joachim for great thread!!
Attachments
I am too lousy at bread-boarding -> the circuits usually work but when I look at them I got scared. JG makes really neat prototypes.
Joachim - I might have missed it - what is the approx. gain of the line stage? I wanted to use k246/j103 so I guess I will not get over sane values for linestage?
Joachim - I might have missed it - what is the approx. gain of the line stage? I wanted to use k246/j103 so I guess I will not get over sane values for linestage?
Thats the trick ! Use the lower Gm parts. Amplification in the first stage is approximately the source resistor times 2 the drain resistor. Because 2SK246, 2SJ103 have lower conductance and higher Vgs you end up with higher value source resistors hence the gain is less. Gain of the output stage is 2 x Gm x RL. The term x2 comes from the parallel symmetric nature of the circuit. That is one advantage plus lower K2 distortion and 6dB less noise then a LTP ( long tail pair ).
Ultimately you can adjust the gain with the loading resistor at the output. The only limitation is the current that the Hawksford cascode can supply into that resistor. That gets harder when you lower gain because RL has to be lower.The lower the gain, the lower RL and so you get less dynamic range. In an NFB circuit this is very different. Lower the gain and the output voltage that can be thrown around stays the same provided that the feedback network is high impedance enough to not steal much current from the output stage. For example the cascode can supply 10mA in Class A. That´s what we want because we do not have global feedback to lower distortion in a Class AB or Class B circuit. If we make RL 220 Ohm we get plus-minus 2.2V. You can drive the circuit on less voltage, say plus-minus 10V and raise the current in the cascode to 20mA. Then you get plus-minus 4.4V.
Last edited:
Cannot find your name on the list of artists, Joachim.
May be I should insist for listing you too.
Viele Grüße
Sam
Equilization by tone controls #2658
Mr Burchard I am new here, and read this thread with much interest.
Reply #2658
That is what I have in mind, to use the 3 Tone controls of my REVOX amp to equalize my "Panansonic EPC450-451 strain gauge cartridge".
Cartridge resistance is 1000 ohm, and need 4 mA, Vo= 3mV.
Manual says the Tone controls, low, mid, high are exactely working where equalization for the cartridge is required for correction., surprised.
Controls have steps of 2 dB up/down, 8 dB max. I have to use the AUX input therefore.
Aux inp sensitivity is 200mV at 100 kohm. Cartridge Vout = 3mV.
Looking on this thread for a suitable amp with a gain of 70, cannot make a choice.
Regards
Mr Burchard I am new here, and read this thread with much interest.
Reply #2658
That is what I have in mind, to use the 3 Tone controls of my REVOX amp to equalize my "Panansonic EPC450-451 strain gauge cartridge".
Cartridge resistance is 1000 ohm, and need 4 mA, Vo= 3mV.
Manual says the Tone controls, low, mid, high are exactely working where equalization for the cartridge is required for correction., surprised.
Controls have steps of 2 dB up/down, 8 dB max. I have to use the AUX input therefore.
Aux inp sensitivity is 200mV at 100 kohm. Cartridge Vout = 3mV.
Looking on this thread for a suitable amp with a gain of 70, cannot make a choice.
Regards
First you need a constant current source. That could be a J-Fet with source resistor or a current mirror for example. I assume that the cartridge has then to be AC coupled into the phono amplifier. The most easy way would be to use a good Opamp. If that works we could construct a discrete solution. Concerning the tone controls a strain gauge cartridge does not need much equalization because it has a 6dB/oct falling response by itself. That does not work 100% perfect so you can use your tone controls for fine tuning.