Hope it don´t end up to be a oscillator, you have to slow it down, do not forget.
I have decided to follow Calvin suggestion of using a current source to cancel the dac offset current.
The current offset and dc out is now control by a R29 (that will be a 1k resistor in serie with 1k potentiometer ).
I have added C8/R25 C10/R26 C4/R24 for stability.
And now every thing seems to be very stable, the most sensible nodes are the nodes named UP and DOWN.
If you wish, the input zobel can be omited, but i keep them, because i feel it add stability.
The main current sources ware also changed (v5 and V4 will be two green led in serie).
The current offset and dc out is now control by a R29 (that will be a 1k resistor in serie with 1k potentiometer ).
I have added C8/R25 C10/R26 C4/R24 for stability.
And now every thing seems to be very stable, the most sensible nodes are the nodes named UP and DOWN.
If you wish, the input zobel can be omited, but i keep them, because i feel it add stability.
The main current sources ware also changed (v5 and V4 will be two green led in serie).
Attachments
What type of troubles ?
I will be glad to help if i can.
I start already to make the pcb for the circuit above.
Below are the frequency response of the circuit before the modifications and after.
As you see the old version start to misbehave above 20Mhz With two enormous peaks at 33Mhz and 88Mhz .
The new version has a much stable frequency response
Note that this frequency response are measure without the output capacitor C1. The last figure is the response with the output capacitor.
As you see the old version start to misbehave above 20Mhz With two enormous peaks at 33Mhz and 88Mhz .
The new version has a much stable frequency response
Note that this frequency response are measure without the output capacitor C1. The last figure is the response with the output capacitor.
Attachments
if you post a schema maybe i can help.
Edit: I have to take a better look to the sen/cen circuit as it seems are a lot of people interested.
If i understand you . you change the jfets by the bjt´s.
Last edited:
You have to make the voltage stable in the up and down nodes if you want to achieve very low distortion, i think i know how to do that, but today i have no time. Next week.
Joachim,
Re post #208, you are basically back to the original Leach design :
Moving Coil Cartridge Head Amps
With all its pro's and con's.
Cheer,
Patrick
Re post #208, you are basically back to the original Leach design :
Moving Coil Cartridge Head Amps
With all its pro's and con's.
Cheer,
Patrick
The bias transistor is like that of Hawksford, right ?
http://www.diyaudio.com/forums/digital-source/168093-hawksford-discrete-iv-twist.html
Patrick
http://www.diyaudio.com/forums/digital-source/168093-hawksford-discrete-iv-twist.html
Patrick
As i see circuit pots 208 the bias transistors sense the base current of the input transistors and add them to the other polarity of the output.
In the Leach the bases of the input transistors are simply held on a constant potential.
I use an optimized Leach input stage in my Nobrainer Phono. That does not have a floating supply though and filters and clamps the base current of the input transistors to ground for a 3dB noise advantage. Here the big electrolytic does that job.
In the Leach the bases of the input transistors are simply held on a constant potential.
I use an optimized Leach input stage in my Nobrainer Phono. That does not have a floating supply though and filters and clamps the base current of the input transistors to ground for a 3dB noise advantage. Here the big electrolytic does that job.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.