I was just browsing and came across this pre-amp project:
http://mitglied.lycos.de/MaikHerzog/audio_projects/preamplifier/preamp_text.html
I was wondering if anyone knew why this guy decided to:
A) Have the first stage as a buffer, then the second stage with gain. I thought if you have the first stage as gain, then the second stage as a buffer you will achieve a lower output noise. So why has he done it in this way?
B) In the first stage (buffer) he uses two separate current sinks and says that they must be matched very closely. Why not use a current mirror?
C) In the second (gain) stage, he ditches the idea of using two current sinks and just has a single one, with twice the current. What was wrong with two current sinks, and if a single one is better, why not do that to the first stage.
If anyone could explain. I would be most grateful!
Thanks
TED
http://mitglied.lycos.de/MaikHerzog/audio_projects/preamplifier/preamp_text.html
I was wondering if anyone knew why this guy decided to:
A) Have the first stage as a buffer, then the second stage with gain. I thought if you have the first stage as gain, then the second stage as a buffer you will achieve a lower output noise. So why has he done it in this way?
B) In the first stage (buffer) he uses two separate current sinks and says that they must be matched very closely. Why not use a current mirror?
C) In the second (gain) stage, he ditches the idea of using two current sinks and just has a single one, with twice the current. What was wrong with two current sinks, and if a single one is better, why not do that to the first stage.
If anyone could explain. I would be most grateful!
Thanks
TED
maik - premamplifier
maik is a member here at forum
maik profile
Best would be if he could comment himself.
His latest post was 15 november.
halo - waits for maik
maik is a member here at forum
maik profile
Best would be if he could comment himself.
His latest post was 15 november.
halo - waits for maik
(thoughts) on this 
It isn't always easy to figure out why people have been inconsistent. It may be they have a theory and it may be they don't know why but are making a choice based on someone elses design.
A) Have the first stage as a buffer
No idea.
B)/C) Why not use a current mirror?
I would have used a common CCS, as is used in the 2nd stage. It may be that he finds it easier to trim the currents when there are separate sources. But this would be equally true for the 2nd stage. I don't know.
The psu design, using FET followers, does not add to my confidence that this designer is using a well informed methodology. Perhaps he can explain it and put me right.
A) Have the first stage as a buffer
No idea.
B)/C) Why not use a current mirror?
I would have used a common CCS, as is used in the 2nd stage. It may be that he finds it easier to trim the currents when there are separate sources. But this would be equally true for the 2nd stage. I don't know.
The psu design, using FET followers, does not add to my confidence that this designer is using a well informed methodology. Perhaps he can explain it and put me right.
Hi Helix,
let me try to anser your questions.
1) I used the first stage as buffer and the second for
the gain because in some experiments I found out,
that I like it more to get the gain from the power
MOS transistors and not from the dual-j-fet I use
in the first stage. Means, it just sounds better to
my ears to have it his way.
2) Using a current mirror in the first stage is possible
but make the stage more complicated. With this preamp
design I followed somehow the road of Nelson Pass to
make the gain stages as simple as possible. Have you
checked if the differential input stage will still provide a
matched balanced output from a single ended input if
you use a current mirror? That was also one goal of
the circuit design, to get a very well matched balanced
output from single ended inputs.
3) Again the decision for a single current source in the
second gain stage was taken by listening experiment.
I found that the configuration with a single current source
results in the same quality of sound compared to two
current sources but it is the more simple circuit (no matching...).
I really think, that this preamp sounds very good and the
volume control works great. Nevertheless I have some more
ideas to further improve the volume control but so less
time to do all the experiments and work
Please see the description on my homepage as a starting
point for your own ideas and let us know what you
work on.
Regards,
Maik
let me try to anser your questions.
1) I used the first stage as buffer and the second for
the gain because in some experiments I found out,
that I like it more to get the gain from the power
MOS transistors and not from the dual-j-fet I use
in the first stage. Means, it just sounds better to
my ears to have it his way.
2) Using a current mirror in the first stage is possible
but make the stage more complicated. With this preamp
design I followed somehow the road of Nelson Pass to
make the gain stages as simple as possible. Have you
checked if the differential input stage will still provide a
matched balanced output from a single ended input if
you use a current mirror? That was also one goal of
the circuit design, to get a very well matched balanced
output from single ended inputs.
3) Again the decision for a single current source in the
second gain stage was taken by listening experiment.
I found that the configuration with a single current source
results in the same quality of sound compared to two
current sources but it is the more simple circuit (no matching...).
I really think, that this preamp sounds very good and the
volume control works great. Nevertheless I have some more
ideas to further improve the volume control but so less
time to do all the experiments and work
Please see the description on my homepage as a starting
point for your own ideas and let us know what you
work on.
Regards,
Maik
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.