at the moment i am making a layout for the Bal.Linestage and i am thinking about to use one or two CCS instead of the resistors to bias the FETs. Does somebody has any (listening) experience with that? When using a CCS is it possible to reduce the neg. railvoltage to, lets say -15V or even to zero? Would you recommend a simple FET+Resistor-type for the CCS or a more sophisicated design? With respect to noise it might be better (i think) to use only one CCS to bias both FETs, Nelson Pass uses one CCS for each FET in his commercial designs. Is there a difference in sound?
I haven't found a lot of difference in actual sound
between the variations of current sources. A current
source is preferred, but has not been used in some of
the Zen projects because of the additional complexity.
The biggest difference is that the input common mode
rejection with balanced inputs is very much better
with current sources, and with an unbalanced signal, the
current source gives lower distortion.
Dual current sources, one for each of the input devices
allows for an arbitrary amount of resistance to be
inserted between the Source pins of the differential
circuit without concerns about altering the DC point.
On the Aleph P circuit, the resistance was made adjustable,
which allowed gain trim and also higher input overload
when set to higher values.
The choice of JFET current sources vs MOSFET vs Bipolar
comes down to what works best at various current and
voltage levels. When the voltage and current is high,
I usually use MOSFETS. If the voltage and current is
low, JFETs, which are simpler and quieter.
|All times are GMT. The time now is 11:33 AM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright ©1999-2015 diyAudio