Hi,please gan you give me some suggestion to obtain a little gain improvement,on the imput stage, for my John Linsley Hood class A amplifier,coupled to a cd player.I used the following Transistor
2N2907 -2N1711- 2N3055.
many thanks
2N2907 -2N1711- 2N3055.
many thanks
Attachments
Last edited:
Or change the 2k7, more increases gain, less decreases it, and the lower cutoff frequency
doesn't change.
If you change the 220 in the feedback you should really alter the 220uF to match (although you might just increase it anyway to reduce its risk of affecting THD).
doesn't change.
If you change the 220 in the feedback you should really alter the 220uF to match (although you might just increase it anyway to reduce its risk of affecting THD).
If you reduce the 220 ohm resistor, you will have lower distortion compared with increasing the 2.7k because the input stage is largely dominated by the 220 ohms. Going to 100 ohms will increase the open loop gain by almost double (but not quite), hence the closed loop distortion will be similar to (but not exactly) the same while giving you your higher gain.
I would also recommend increasing the capacitor value as Mark suggests, perhaps to 1mF for both reasons he mentions.
I would also recommend increasing the capacitor value as Mark suggests, perhaps to 1mF for both reasons he mentions.
Last edited:
Perhaps you can suggest a value for the extra gain you want? ie What does "a little gain improvement" mean? 3dB? 10dB? 15dB? No way to tell.
Do you have a lot of volume from the CD or not enough? Do you have a pot at the input?
The power of the amplifier is limited to 10W RMS and cannot be increased in a simple way. Use speakers with high sensitivity.
The power of the amplifier is limited to 10W RMS and cannot be increased in a simple way. Use speakers with high sensitivity.
It's hard to believe you need more gain with a CD player. This amp was designed for sources with lower output.
Firstly not all CD players are high level outputs, secondly not all tracks are compressed to death and actually have wide dynamic range.
The JLH Simple ClassA does not have much open loop gain a figure of 600 being mentioned and that with the original high gain output transistors now unobtainable.
This level of 600 has to be split between closed loop gain and nfb so increasing the former will reduce the latter and with less to spare there the distortion will increase.
There is a saying the more you need money for something the less likely the bank will loan money to you.
JLH was quite happy to use IC op.amps with low closed loop gain numbers as low as two. With reasonably efficient speakers that should work with a volume pot at the Class A input.
With a closed loop gain of two there is a huge level of feedback available from IC op.amps. You will find these circuit blocks in the analogue output stages of CD players - they are more reliable than discrete component layouts and make it easy to build something on a hobbyist project board
This level of 600 has to be split between closed loop gain and nfb so increasing the former will reduce the latter and with less to spare there the distortion will increase.
There is a saying the more you need money for something the less likely the bank will loan money to you.
JLH was quite happy to use IC op.amps with low closed loop gain numbers as low as two. With reasonably efficient speakers that should work with a volume pot at the Class A input.
With a closed loop gain of two there is a huge level of feedback available from IC op.amps. You will find these circuit blocks in the analogue output stages of CD players - they are more reliable than discrete component layouts and make it easy to build something on a hobbyist project board
Last edited:
The open loop gain of the JLH amplifier is, like all amplifiers, dependent on the gains of the individual transistor stages.
The input transistor of the JLH operates at about 350uA. Therefore, the emitter impedance will be, from 25/.35, about 70 ohms. Therefore, the gm of the input stage is about 1/(220+70) or 3.4mA/V, taking into account the feedback grounding resistor. If you reduce that to 100 ohms, the gm will become approximately 1/170 or 5.9mA/V. This is almost doubling the open loop gain, and therefore the closed loop gain can be increased while maintaining the same distortion figures (almost), as I mentioned in my earlier post.
Conversely, the closed loop gain can be maintained if the feedback resistor is also reduced, with a reduction in distortion.
However, the caveat is that this may change the frequency response too and alter the stability, but this can be compensated for by small capacitors (in the region of 33pF) connected between the collector of the driver and the input emitter. Or used as a Miller if that does not work.
The distortion differences with high or low gain transistors (assuming that the minimum is at least 100) is not quite so important when the gain is dominated by re and the feedback network. A high gain transistor will help, of course, by reducing the base current which, if non-linear, due to a non-linear collector current being needed, has an effect but that depends on the output impedance too of any preceding preamp.
The input transistor of the JLH operates at about 350uA. Therefore, the emitter impedance will be, from 25/.35, about 70 ohms. Therefore, the gm of the input stage is about 1/(220+70) or 3.4mA/V, taking into account the feedback grounding resistor. If you reduce that to 100 ohms, the gm will become approximately 1/170 or 5.9mA/V. This is almost doubling the open loop gain, and therefore the closed loop gain can be increased while maintaining the same distortion figures (almost), as I mentioned in my earlier post.
Conversely, the closed loop gain can be maintained if the feedback resistor is also reduced, with a reduction in distortion.
However, the caveat is that this may change the frequency response too and alter the stability, but this can be compensated for by small capacitors (in the region of 33pF) connected between the collector of the driver and the input emitter. Or used as a Miller if that does not work.
The distortion differences with high or low gain transistors (assuming that the minimum is at least 100) is not quite so important when the gain is dominated by re and the feedback network. A high gain transistor will help, of course, by reducing the base current which, if non-linear, due to a non-linear collector current being needed, has an effect but that depends on the output impedance too of any preceding preamp.
Last edited: