[MRC01]

This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

[Eeyore]

MRC01 wrote:

This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

Typically it adjusts the signal level at the input to the next stage.
Implementations which put the control in the feedback loop do not typically have
the ability to reduce the gain/level to zero.

Graham

[[email protected]]

On Sep 11, 10:56 am, MRC01 wrote:
This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

Most often, it is done via the latter method: it's between two gain
stages as an attenuator. While doing it in the feedback loop
may seem a neat way to do it, it's got its own set of problems.
FOr example, the stability and compensation of the gain stage
is gain-dependent. As you decrease the feedback resistance
thus reduce the gain, your compensation may need to change
so that you maintain unconditional stability. That's not so easy
to accomplish in practice. Further, things like the input offset
voltage and input bias currents may end up causing DC to
flow through the pot resulting in noise as you're changing
gain.

[MRC01]

On Sep 11, 8:06 am, Eeyore
wrote:
MRC01 wrote:

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

Typically it adjusts the signal level at the input to the next stage.
Implementations which put the control in the feedback loop do not typically have
the ability to reduce the gain/level to zero.

So a typical preamp, for example, would have fixed gain and the level
of the input to the opamp would also be fixed, and the volume control
would attenuate the opamp's output signal? Is there an advantage to
attenuating the output rather than the input?

[MRC01]

On Sep 11, 8:09 am, wrote:

Most often, it is done via the latter method: it's between two gain
stages as an attenuator. While doing it in the feedback loop
may seem a neat way to do it, it's got its own set of problems.
FOr example, the stability and compensation of the gain stage
is gain-dependent. As you decrease the feedback resistance
thus reduce the gain, your compensation may need to change
so that you maintain unconditional stability. That's not so easy
to accomplish in practice. Further, things like the input offset
voltage and input bias currents may end up causing DC to
flow through the pot resulting in noise as you're changing
gain.

If the volume knob attenuates the input, how does one design it to
maintain a constant input impedance? For example I can imagine the
input, as seen by the source/input device, as R1 in series and R2 to
ground (in parallel with the opamp input). Assuming the op amp's input
impedance is high, the magnitude and ratio of R1 and R2 determine the
input impedance and the attenuation. Which - R1 or R2 - would be the
volume knob? Whichever one it might be, either way, rotating the knob
would change the input impedance at the same time it changed the
volume.

When I built my ladder stepped attenuator, I individually selected R1
and R2 for each position of the knob to get 2 dB per click with every
position having the same load impedance to the source device. With a
pot I don't see how this is possible since it's only setting one
value, not both.

[Arny Krueger]

"MRC01" wrote in message
ups.com...

This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

If you're talking mic preamp, then the gain control usually does control the
gain of an active amplification stage.

If you are talking hifi preamp, the gain control usually attenuates the
signal right in front of the line level gain block. The signal is either
coming from the outside world, or the output of the phono preamp.

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain.

Frequently done in mic preamps. This approach generally can't reduce gain to
minus infinity dB.

Or I could imagine the volume control attenuating the signal before it
reaches the opamp.

Frequently done in hi fi gear.

How is it typically done?

Unnh, it depends! ;-)

[Michael Black]

MRC01 ) writes:
This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

You forgot the third a variable gain block controlled by a varying
DC voltage.

I don't have a clue how common those are, but the devices are out
there to do it.

Michael

[[email protected]]

On Sep 11, 11:28 am, MRC01 wrote:
On Sep 11, 8:09 am, wrote:



Most often, it is done via the latter method: it's between two gain
stages as an attenuator. While doing it in the feedback loop
may seem a neat way to do it, it's got its own set of problems.
FOr example, the stability and compensation of the gain stage
is gain-dependent. As you decrease the feedback resistance
thus reduce the gain, your compensation may need to change
so that you maintain unconditional stability. That's not so easy
to accomplish in practice. Further, things like the input offset
voltage and input bias currents may end up causing DC to
flow through the pot resulting in noise as you're changing
gain.

If the volume knob attenuates the input, how does one design it to
maintain a constant input impedance? For example I can imagine the
input, as seen by the source/input device, as R1 in series and R2 to
ground (in parallel with the opamp input). Assuming the op amp's input
impedance is high, the magnitude and ratio of R1 and R2 determine the
input impedance and the attenuation. Which - R1 or R2 - would be the
volume knob? Whichever one it might be, either way, rotating the knob
would change the input impedance at the same time it changed the
volume.

That's one reason why I said it's between two gain stages.
You typically have an input gain stage: it provides a gain
fixed to suit the particular input signal range and it provides
a specific and fixed input mpedance. Then you have your
variable attenuator, and then you have your output buffer
stage. Thus both your source and your sink are completely
isolated from the vagueries of the attenuator. The two
buffer stages together with the attenuator can be designed
for optimum performance

When I built my ladder stepped attenuator, I individually
selected R1 and R2 for each position of the knob to get
2 dB per click with every position having the same load
impedance to the source device. With a pot I don't see
how this is possible since it's only setting one
value, not both.

When it's buffered properly, you simply don't care. The
preamp presents a constant and proper input impedance
regardless of volume control setting, and the it presents a
suitable and constant output impedance regardless of
volume control setting.

That's why it's done the way it is.

[MRC01]

On Sep 11, 8:40 am, wrote:

That's one reason why I said it's between two gain stages.
You typically have an input gain stage: it provides a gain
fixed to suit the particular input signal range and it provides
a specific and fixed input mpedance. Then you have your
variable attenuator, and then you have your output buffer
stage. Thus both your source and your sink are completely
isolated from the vagueries of the attenuator. The two
buffer stages together with the attenuator can be designed
for optimum performance

When I built my ladder stepped attenuator, I individually
selected R1 and R2 for each position of the knob to get
2 dB per click with every position having the same load
impedance to the source device. With a pot I don't see
how this is possible since it's only setting one
value, not both.

When it's buffered properly, you simply don't care. The
preamp presents a constant and proper input impedance
regardless of volume control setting, and the it presents a
suitable and constant output impedance regardless of
volume control setting.

Aha! So a preamp has two opamps in series with the volume knob in
between them? Input to opamp1 is fixed, comes from input jack. Output
from opamp2 is fixed, goes to output jack. That means constant input
and output impedances. Volume knob with variable impedance and
attenuation is in between them.

[Arny Krueger]

"MRC01" wrote in message
ps.com...

Aha! So a preamp has two opamps in series with the volume knob in
between them? Input to opamp1 is fixed, comes from input jack. Output
from opamp2 is fixed, goes to output jack. That means constant input
and output impedances. Volume knob with variable impedance and
attenuation is in between them.

It is quite feasible to build a preamp with just a gain control and a gain
block (e.g. op amp).

The gain control has a low enough impedance that the input impedance of the
gain block has negligable effect on it.

You still have constant impedances in and out, but you also have a preamp
that is virtually impossible to overload with a large input signal, given
proper setting of the gain potentiometer.

[Eeyore]

MRC01 wrote:

Eeyore wrote:
MRC01 wrote:

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

Typically it adjusts the signal level at the input to the next stage.
Implementations which put the control in the feedback loop do not typically have
the ability to reduce the gain/level to zero.

So a typical preamp, for example, would have fixed gain and the level
of the input to the opamp would also be fixed, and the volume control
would attenuate the opamp's output signal?

A typical preamp would have at least 2 stages. An input stage that may have some gain
(although little is needed in these days of CD levels) followed by a potentiometer
which then feeds an output stage. The pot is thereby 'isolated' from both input and
output.


Is there an advantage to attenuating the output rather than the input?

You don't want the output to come from the pot wiper itself as the impedance will be
variable and uncomfortably high at most settings. That's why a pot *ought* to be
followed by a 'buffer' stage.

Graham

[Eeyore]

Arny Krueger wrote:

"MRC01" wrote:

This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

If you're talking mic preamp,

I had presumed he wasn't.


then the gain control usually does control the gain of an active amplification
stage.

In this case absolutely yes.

Graham

[Eeyore]

Michael Black wrote:

MRC01 ) writes:
This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

You forgot the third a variable gain block controlled by a varying
DC voltage.

I don't have a clue how common those are, but the devices are out
there to do it.

They are moderately common in pro-audio equipment for applications other than
volume controls usually. I can't say I've ever met one in a hi-fi application
though.

Graham

[Eeyore]

MRC01 wrote:

wrote:

Most often, it is done via the latter method: it's between two gain
stages as an attenuator. While doing it in the feedback loop
may seem a neat way to do it, it's got its own set of problems.
FOr example, the stability and compensation of the gain stage
is gain-dependent. As you decrease the feedback resistance
thus reduce the gain, your compensation may need to change
so that you maintain unconditional stability. That's not so easy
to accomplish in practice. Further, things like the input offset
voltage and input bias currents may end up causing DC to
flow through the pot resulting in noise as you're changing
gain.

If the volume knob attenuates the input, how does one design it to
maintain a constant input impedance?

You don't. The input passes through a buffer stage before hitting the volume
control. The buffer stage provides the required constant input impedance.

Graham

[Eeyore]

MRC01 wrote:

When I built my ladder stepped attenuator, I individually selected R1
and R2 for each position of the knob to get 2 dB per click with every
position having the same load impedance to the source device. With a
pot I don't see how this is possible since it's only setting one
value, not both.

No, a pot adjusts 2 values simultaneously. Only in 'rheostat mode' is a pot used
to provide only one variable resistance.

Graham

[Eeyore]

MRC01 wrote:

On Sep 11, 8:40 am, wrote:

That's one reason why I said it's between two gain stages.
You typically have an input gain stage: it provides a gain
fixed to suit the particular input signal range and it provides
a specific and fixed input mpedance. Then you have your
variable attenuator, and then you have your output buffer
stage. Thus both your source and your sink are completely
isolated from the vagueries of the attenuator. The two
buffer stages together with the attenuator can be designed
for optimum performance

When I built my ladder stepped attenuator, I individually
selected R1 and R2 for each position of the knob to get
2 dB per click with every position having the same load
impedance to the source device. With a pot I don't see
how this is possible since it's only setting one
value, not both.

When it's buffered properly, you simply don't care. The
preamp presents a constant and proper input impedance
regardless of volume control setting, and the it presents a
suitable and constant output impedance regardless of
volume control setting.

Aha! So a preamp has two opamps in series with the volume knob in
between them? Input to opamp1 is fixed, comes from input jack. Output
from opamp2 is fixed, goes to output jack. That means constant input
and output impedances. Volume knob with variable impedance and
attenuation is in between them.

Exactly.

Graham

[Richard Crowley]

"MRC01" wrote ...
This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

By "preamp" do you mean a piece of stereo gear, or do you mean
a mic preamp as used for recording?

If you mean the stereo hi-fi system preamp, the main volume
control is somewhere in the middle of the audio path and adjusts
the amplitude (level) of the signal. It may have op-amp stages
before and/or after the volume control.

If you mean a mic preamp as used in a mixer for recording or
sound reinforcement (PA) systems, there is most often TWO
controls, one of each. That is, there is typically a "Trim" control
which indeed adjusts the feedback gain through the first op-amp
stage. Then there is the main "Fader" which adjusts the mix level
of that channel. The Fader is typically a voltage-divider potentiometer
which adjusts the "output level" of the preamp stage. Although
there is frequently more than that: (bus switching, submix levels,
pan pots, etc.)

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

Both methods are used, depending on what kind of equipment you
are looking at.

[Arny Krueger]

"Eeyore" wrote in message
...


Michael Black wrote:

MRC01 ) writes:
This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

You forgot the third a variable gain block controlled by a varying
DC voltage.

I don't have a clue how common those are, but the devices are out
there to do it.

They are moderately common in pro-audio equipment for applications other
than
volume controls usually.

The use of VCAs in place of quad pots for the state variable filters in some
Rane crossovers comes to mind.

I can't say I've ever met one in a hi-fi application though.

There was a high end preamp that used to the old Crystal 3310 chip for its
gain control. The CS 3310 was an interesting chip - almost CD quality
dynamic range, not that expensive, and almost impossible to overload.

Then 3310s became unobtainable...

[Eeyore]

Arny Krueger wrote:

"Eeyore" wrote
Michael Black wrote:
MRC01 ) writes:
This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

You forgot the third a variable gain block controlled by a varying
DC voltage.

I don't have a clue how common those are, but the devices are out
there to do it.

They are moderately common in pro-audio equipment for applications other
than volume controls usually.

The use of VCAs in place of quad pots for the state variable filters in some
Rane crossovers comes to mind.

Several of the better quality DJ mixers use VCAs for the crossfader function
too. Including Rane too as it happens !


I can't say I've ever met one in a hi-fi application though.

There was a high end preamp that used to the old Crystal 3310 chip for its
gain control. The CS 3310 was an interesting chip - almost CD quality
dynamic range, not that expensive, and almost impossible to overload.

Then 3310s became unobtainable...

That rings a bell. I think I once noticed it in a short-form catalogue. It was
an audio DSP IIRC. I was evaluating a DSP reverb chip from Crystal at the time.
The support was hopeless (the guys working on that product had been pulled off
itt o work on something else IIRC) and then that product vanished too.
Thankfully we hadn't spent that much time on it.

Graham

[Arny Krueger]

"Eeyore" wrote in message
...


Arny Krueger wrote:

"Eeyore" wrote
Michael Black wrote:
MRC01 ) writes:
This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its
output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

You forgot the third a variable gain block controlled by a varying
DC voltage.

I don't have a clue how common those are, but the devices are out
there to do it.

They are moderately common in pro-audio equipment for applications
other
than volume controls usually.

The use of VCAs in place of quad pots for the state variable filters in
some
Rane crossovers comes to mind.

Several of the better quality DJ mixers use VCAs for the crossfader
function
too. Including Rane too as it happens !

That rings a bell. Well, they were pretty practiced with it and were already
getting quantity pricing! ;-)

I can't say I've ever met one in a hi-fi application though.

There was a high end preamp that used to the old Crystal 3310 chip for
its
gain control. The CS 3310 was an interesting chip - almost CD quality
dynamic range, not that expensive, and almost impossible to overload.

Then 3310s became unobtainable...

That rings a bell. I think I once noticed it in a short-form catalogue. It
was
an audio DSP IIRC.

No, the 3310 was all-analog, designed to sit between an analog input and an
ADC.

[Eeyore]

Arny Krueger wrote:

"Eeyore" wrote
Arny Krueger wrote:
"Eeyore" wrote

I can't say I've ever met one in a hi-fi application though.

There was a high end preamp that used to the old Crystal 3310 chip for
its gain control. The CS 3310 was an interesting chip - almost CD quality
dynamic range, not that expensive, and almost impossible to overload.

Then 3310s became unobtainable...

That rings a bell. I think I once noticed it in a short-form catalogue. It
was an audio DSP IIRC.

No, the 3310 was all-analog, designed to sit between an analog input and an
ADC.

I should have read your post more carefully. I wonder what that chip was that
I'm thinking of ?

Graham

[MRC01]

On Sep 11, 9:03 am, Eeyore
wrote:
MRC01 wrote:
When I built my ladder stepped attenuator, I individually selected R1
and R2 for each position of the knob to get 2 dB per click with every
position having the same load impedance to the source device. With a
pot I don't see how this is possible since it's only setting one
value, not both.

No, a pot adjusts 2 values simultaneously. Only in 'rheostat mode' is a pot used
to provide only one variable resistance.

Right. But the pots I've seen adjust both values the same. Attenuating
the input while keeping impedance the same requires varying R1 and R2
at different values and different rates.

For example at -6 dB R1 equals R2 (more or less, if you're not
compensating for the non-infinite impedance of the opamp) but at -12
dB R1 is 3x R2. Are you saying there are pots that provide both R1 and
R2 at independent rates such that these ratios are maintained for
constant input impedance and linear attenuation in dB?

[[email protected]]

On Sep 11, 1:44 pm, MRC01 wrote:
On Sep 11, 9:03 am, Eeyore
wrote:

MRC01 wrote:
When I built my ladder stepped attenuator, I individually selected R1
and R2 for each position of the knob to get 2 dB per click with every
position having the same load impedance to the source device. With a
pot I don't see how this is possible since it's only setting one
value, not both.

No, a pot adjusts 2 values simultaneously. Only in 'rheostat mode' is a pot used
to provide only one variable resistance.

Right. But the pots I've seen adjust both values the same. Attenuating
the input while keeping impedance the same requires varying R1 and R2
at different values and different rates.

For example at -6 dB R1 equals R2 (more or less, if you're not
compensating for the non-infinite impedance of the opamp) but at -12
dB R1 is 3x R2. Are you saying there are pots that provide both R1 and
R2 at independent rates such that these ratios are maintained for
constant input impedance and linear attenuation in dB?

No, it's MUCH simpler than that.

A volume control gets hooked up in what's called a
"potentiometer" configuration: one end of the resistance
is the input, the other end is grounded. The wiper moves
between these two points and provides the output.

Now, assuming the the load impedance of the output is
substantially larger than the total resistance of the
potentiometer, and you consider the value of R1 to
be the reststance between the input and the wiper, and
R2 to b the resistance between the wiper and ground,
the following will hold:

Gain = R2/(R1 + R2)

and

Rtotal = R1 + R2

But since Rtotal IS the total value of potentiometer,
that value defines the input impedance, and, again, as
the output load is greater than the total resistance of
the potentiometer, the input impedance is constant
regardless of the volume setting.

[Eeyore]

MRC01 wrote:

On Sep 11, 9:03 am, Eeyore
wrote:
MRC01 wrote:
When I built my ladder stepped attenuator, I individually selected R1
and R2 for each position of the knob to get 2 dB per click with every
position having the same load impedance to the source device. With a
pot I don't see how this is possible since it's only setting one
value, not both.

No, a pot adjusts 2 values simultaneously. Only in 'rheostat mode' is a pot used
to provide only one variable resistance.

Right. But the pots I've seen adjust both values the same. Attenuating
the input while keeping impedance the same requires varying R1 and R2
at different values and different rates.

That variation in 'rates' is the result of the way the pot's track is manufactured. An
audio taper pot does not have constant resistance change vs rotation.


For example at -6 dB R1 equals R2 (more or less, if you're not
compensating for the non-infinite impedance of the opamp) but at -12
dB R1 is 3x R2. Are you saying there are pots that provide both R1 and
R2 at independent rates such that these ratios are maintained for
constant input impedance and linear attenuation in dB?

Provided the load impedance on the pot (the following op-amp stage) is fairly high,
the load that a pot presents is by its very nature nearly constant.

The (typically) carbon track of the pot is screen printed using several inks of
different resistivity to provide the required audio taper. The wiper of the pot simply
taps off the signal in a potential divider manner.

Graham

[MRC01]

On Sep 11, 11:35 am, wrote:
Now, assuming the the load impedance of the output is
substantially larger than the total resistance of the
potentiometer, and you consider the value of R1 to
be the reststance between the input and the wiper, and
R2 to b the resistance between the wiper and ground,
the following will hold:

Gain = R2/(R1 + R2)

and

Rtotal = R1 + R2

But since Rtotal IS the total value of potentiometer,
that value defines the input impedance, and, again, as
the output load is greater than the total resistance of
the potentiometer, the input impedance is constant
regardless of the volume setting.- Hide quoted text -

If I understand, you're saying R2 (the pot) is in parallel with the
load. Thus Rtotal = R1 + R2 only if the load impedance is infinity, or
at least significantly higher than R2 (the pot) - then R2 in parallel
with the load is approximately R2.

But this would mean the total impedance R1 + R2 would vary as you turn
the knob, since R2 is the knob and R1 is a resistor of fixed value.
This (R1 + R2) is the impedance seen by the upstream device.

Now let's suppose this preamp has its volume control sandwiched
between separate input and output opamps. Then the device upstream
from the pot is an opamp connected to the input jacks of the preamp.
The original source (CD Player) sees that upstream opamp's input
impedance, not R1 + R2. This upstream opamp sees a load of R1 + R2,
which can be as low as R1 (when the volume is turned all the way down
R2 = 0), or as high as R1 + R2 (when the volume is turned all the way
up, R2 = max). But that's OK because this opamp is designed to be
immune to its load impedance. That is, the frequency response,
distortion and noise should hold constant (more or less) as the load
varies from R1 to R1 + R2.

Is that right?

[[email protected]]

On Sep 11, 3:37 pm, MRC01 wrote:
If I understand, you're saying R2 (the pot) is in parallel with the
load. Thus Rtotal = R1 + R2 only if the load impedance is infinity, or
at least significantly higher than R2 (the pot) - then R2 in parallel
with the load is approximately R2.

But this would mean the total impedance R1 + R2 would vary
as you turn the knob, since R2 is the knob and R1 is a resistor
of fixed value. This (R1 + R2) is the impedance seen by the
upstream device.

No, you've got it wrong.

The resistance R1+R2 is the TOTAL resistance of the
pot, and does NOT change at all. You start with a pot,
say, whose resistance is 10kOhms. The wiper moves
between one extreme and the other. That wiper is
simply a tap somewhere in that 10 kOhm resistor,
and the sum of resistance between the wiper and
one side (call it R1), is ALWAYS equal to 10 kOhms
minus the resistance between the wiper and the other
end (call that R2), so that:

10 kOhms = R1 + R2

and thus:

R1 = 10 kOhms - R2

and

R2 = 10 kOhms - R1

Always, always, always.

Now, let's assume, simply for ease of visualisation,
that it is a linear pot. When the pot is at its full ON
position, R1 = 0 and R2 = 10K. IN it's full OFF position,
R1 = 10K and R2 = 0. Halfway up, R1 = 5k and R2 = 5k.
In more details:

0% R1 = 0 R2 = 10K
10% R1 = 1k, R2 = 9K
20% R1 = 2k, R2 = 8k
30% R1 = 3k, R2 = 7k
...
90% R1 = 9k, R2 = 1k
100% R1 = 10K R2 = 0

The sum, or total of R1+R2 is ALWAYS 10K.

To use it as a potentiometer require that you
connect the signal input to one end of the
pot. The wiper is the signal output going into
a high-impedance load, and the other end
of the pot is grounded.

In such a configuration, the input impedance
WILL ALWAYS BE 10 kOhms, no matter what the
position is. All the way "up" the resistance between
the wiper and input is 0, and between the wiper and
ground is 10 K, and thus the total is 10K. All the way down,
the resistance between the input and the wiper is 10K,
and between the wiper and ground is 0, and thus the total
is 10K.

[MRC01]

On Sep 11, 1:47 pm, wrote:
... That wiper is
simply a tap somewhere in that 10 kOhm resistor,
and the sum of resistance between the wiper and
one side (call it R1), is ALWAYS equal to 10 kOhms
minus the resistance between the wiper and the other
end (call that R2), so that:

OK, that makes sense. So the pot has three connection points and
replaces BOTH resistors R1 and R2. R1 is between points 1 and 2. R2 is
between points 2 and 3. Wire it up so R1 is in series with the load
and R2 is in parallel with the load. Each position of the knob gives
complementary values for R1 and R2 such that their sum is a constant
and of course the gain is based on their ratio.

What confused me was, this is what I suggested a few messages ago to
Graham but you said it was much simpler than that.

[MRC01]

P.S. If that's how it works, why buffer the input stage? If the input
jack goes directly to a 10k pot attenuating the signal into an opamp,
the source device would always see the same 10k (or 20k, or whatever)
regardless of the volume knob position.

[Geoff]

Eeyore wrote:
MRC01 wrote:

This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its
output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

Typically it adjusts the signal level at the input to the next stage.
Implementations which put the control in the feedback loop do not
typically have the ability to reduce the gain/level to zero.

And are highly susceptible to nose when adjusting.

geoff

[Eeyore]

MRC01 wrote:

P.S. If that's how it works, why buffer the input stage? If the input
jack goes directly to a 10k pot attenuating the signal into an opamp,
the source device would always see the same 10k (or 20k, or whatever)
regardless of the volume knob position.

That would be a perfectly valid option if no additional input gain was required
and a 10 kohm input impedance was acceptable.

Graham

[Eeyore]

Geoff wrote:

Eeyore wrote:
MRC01 wrote:

This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its
output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

Typically it adjusts the signal level at the input to the next stage.
Implementations which put the control in the feedback loop do not
typically have the ability to reduce the gain/level to zero.

And are highly susceptible to nose when adjusting.

When badly designed (someone forgot a cap).

Graham

[[email protected]]

On Sep 11, 5:39 pm, MRC01 wrote:
P.S. If that's how it works, why buffer the input stage? If the input
jack goes directly to a 10k pot attenuating the signal into an opamp,
the source device would always see the same 10k (or 20k, or whatever)
regardless of the volume knob position.

A number of reasons:

1. The load impedance on the wiper connection of the
pot may NOT be sufficiently large,

2. The signal level may be not be of a suitable level. One
way of ensuring good noise performance is to get the
signal to a fairly high level first, and then attenuating,

3. Whatever the load presented by the pot may be, even if
constant, it might NOT be the right load for the source,

4. The source may need equalization (like a phone input).

[MRC01]

On Sep 11, 5:25 pm, wrote:
On Sep 11, 5:39 pm, MRC01 wrote:

P.S. If that's how it works, why buffer the input stage? If the input
jack goes directly to a 10k pot attenuating the signal into an opamp,
the source device would always see the same 10k (or 20k, or whatever)
regardless of the volume knob position.

A number of reasons:

1. The load impedance on the wiper connection of the
pot may NOT be sufficiently large,

2. The signal level may be not be of a suitable level. One
way of ensuring good noise performance is to get the
signal to a fairly high level first, and then attenuating,

3. Whatever the load presented by the pot may be, even if
constant, it might NOT be the right load for the source,

4. The source may need equalization (like a phone input).

Let's suppose we're talking about a solid state line level source
device like a CD Player that would be perfectly happy driving a 10k
load. The signal is big enough, doesn't need amplification. No
equalization needed, and the 10k pot is sufficiently large. Why not
skip the input opamp and thus eliminate a stage, simplify the design
and perhaps gain some sonic advantages?

Are there other drawbacks to using too high a value for the pot? Why
not just always use a 100k pot? Anything happy with a 10k load should
be happy with a 100k load, right?

For example in my passive attenuator I didn't want to go above 10k
because the worst case output impedance at -6 dB is already high
enough at 2.5 kOhm. Going to a 20k attenuator would double that which
would accentuate any cable issues. But that is a passive design. With
an active preamp this would not be an issue, but there may be other
issues associated with high impedance pots. What are they?

[JANA]

In the normal configuration for most amplifier devices the volume control is
an attenuator. The amplifier stages after the volume control are at their
full gain. The volume control in this case would be attenuating the signal
to control it.

What is ideal, is to design an amplifier where a combination of both
attenuation and gain are used. In this case, the gain of the amplifier
device is controlled, and at the same time, the attenuation of the signal is
controlled. In this case, the SN should be much higher. With this type of
design, to have proper linearity and stable response would be fairly
critical.

--

JANA
_____


"MRC01" wrote in message
ups.com...
This is a simple question yet I'm having a hard time finding an
answer.

In a typical preamp with opamps, does the volume control change the
gain, or does it adjust the level of the opamp's input, or its output?

For example, the I could imagine the volume control as a variable
resistor in the opamp's gain feedback loop, adjusting the gain. Or I
could imagine the volume control attenuating the signal before it
reaches the opamp. How is it typically done?

[isw]

In article om,
MRC01 wrote:

If the volume knob attenuates the input, how does one design it to
maintain a constant input impedance?

Except for phono preamp inputs, and *some* professional gear, constant
input impedance is just not important. Most consumer gear has very high
input impedance and very low output impedance; that's all it takes.

That said, you can hook the ends of the pot to the input and ground;
then take the output off the wiper. If the following amplifying stage
has a high input impedance, everything works fine *and* you have a
constant input impedance.

Isaac

[Geoff]

Eeyore wrote:
Geoff wrote:


And are highly susceptible to nose when adjusting.

When badly designed (someone forgot a cap).


Ooops noise not nose !

Which is why mixers have faders separate from input trim controls, cap or
not.

geoff