Conceptual design of a constant volume control opamp circuit

[MRW]

Good day:

I don't know how to word these properly. I'm really confused right
now. I'll try. If I were to design a single supply constant volume
control opamp circuit, would it be right to assume the following:

- In coming up with a conceptual block diagram, I take a sample of the
input signal and compare it to some threshold value. If the signal is
not equal to the threshold value then change the gain so that it
becomes equal to the threshold value. However, I'm not sure if this is
a valid assumption, since I would need to have a very fast opamp for
the gain to change instantaneously before the next instantaneous
input. I guess for audio inputs most opamp would be fast enough to
change the gain.

- The VCA that I have has its gained controlled by two voltages via
the following relationships: Vcn+ - Vcn-. I was thinking of setting
Vcn- at a fixed voltage, and use Vcn+ for the controlling voltage.

- For threshold circuit, I was thinking of using a differentiator. The
output of this differentiator would be sent to another opamp circuit
with an output voltage reference to a value equal to Vcn-. The output
of this opamp would be connected to Vcn+.

 

[MRW]

Hmm... I don't think my differentiator idea will work in single supply
operations. I was running thru my conceptual block again, and I just
realized that the differentiator output will stay at 0V if my input
voltage is higher than the threshold voltage.

 

[Jon Slaughter]

Good day:

I don't know how to word these properly. I'm really confused right
now. I'll try. If I were to design a single supply constant volume
control opamp circuit, would it be right to assume the following:

- In coming up with a conceptual block diagram, I take a sample of the
input signal and compare it to some threshold value. If the signal is
not equal to the threshold value then change the gain so that it
becomes equal to the threshold value. However, I'm not sure if this is
a valid assumption, since I would need to have a very fast opamp for
the gain to change instantaneously before the next instantaneous
input. I guess for audio inputs most opamp would be fast enough to
change the gain.

All you really need is for the gain two chain in the opposite direction that
the signal is going. What this does is create a negative feeback look. If
the signal increase you decrease it. If you decrease it to much then it the
it will drop below the threshold and hten your circuit will amplify it
causing it to increase.

The issues you have to worry mainly about is oscillation. If you change gain
is to large you can overshoot to much and then it will try and compensate by
overrcompensating. Then you get excess major issues. Ofcourse any
oscillation is bad and you want to find the critical point that creates an
overdamped system(or the closest you can come to it).

- The VCA that I have has its gained controlled by two voltages via
the following relationships: Vcn+ - Vcn-. I was thinking of setting
Vcn- at a fixed voltage, and use Vcn+ for the controlling voltage.

- For threshold circuit, I was thinking of using a differentiator. The
output of this differentiator would be sent to another opamp circuit
with an output voltage reference to a value equal to Vcn-. The output
of this opamp would be connected to Vcn+.

Maybe. But why not just use the difference between the threshold and the
signal as an estimation of the gain needed for feedback?

suppose your signal is s(t) and your threshold is T.

then s(t) - T tells you the amount of the signal above/below T. if s(t) > T
then s(t) - T > 0. Therefore you need to have a negative gain to reduce s(t)
down.


if s(t) < T then s(t) - T < 0 so you need a positive gain.

i.e., T - s(t) is your gain needed.

So here all you need is a opamp to add the two. Ofcourse ultimately you want
to build in some stability into the design so you'll want some function f
that has odd symmetry and som hysteresis to combat the circuit being
underdamped.

(e.g., you could use the (T - s(t))^(1/3) or something similar. This way you
get your gain which works fine for s(t) close to T but doesn't over
compensate if s(t) is far from T. Actually this might be to much. I'm sure
there are functions specifically for this type of stuff if you look harder)

 

[bw]

Good day:

I don't know how to word these properly. I'm really confused right
now. I'll try. If I were to design a single supply constant volume
control opamp circuit, would it be right to assume the following:

- In coming up with a conceptual block diagram, I take a sample of the
input signal and compare it to some threshold value. If the signal is
not equal to the threshold value then change the gain so that it
becomes equal to the threshold value. However, I'm not sure if this is
a valid assumption, since I would need to have a very fast opamp for
the gain to change instantaneously before the next instantaneous
input. I guess for audio inputs most opamp would be fast enough to
change the gain.

- The VCA that I have has its gained controlled by two voltages via
the following relationships: Vcn+ - Vcn-. I was thinking of setting
Vcn- at a fixed voltage, and use Vcn+ for the controlling voltage.

- For threshold circuit, I was thinking of using a differentiator. The
output of this differentiator would be sent to another opamp circuit
with an output voltage reference to a value equal to Vcn-. The output
of this opamp would be connected to Vcn+.

I can't help with your specific application.
In the old radio days it was called Automatic Gain Control (AGC)
Also consider searching the phrase "foldback limiting" used in DC
power supplies.

 

[John Fields]

Good day:

I don't know how to word these properly. I'm really confused right
now. I'll try. If I were to design a single supply constant volume
control opamp circuit, would it be right to assume the following:

- In coming up with a conceptual block diagram, I take a sample of the
input signal and compare it to some threshold value. If the signal is
not equal to the threshold value then change the gain so that it
becomes equal to the threshold value. However, I'm not sure if this is
a valid assumption, since I would need to have a very fast opamp for
the gain to change instantaneously before the next instantaneous
input. I guess for audio inputs most opamp would be fast enough to
change the gain.

- The VCA that I have has its gained controlled by two voltages via
the following relationships: Vcn+ - Vcn-. I was thinking of setting
Vcn- at a fixed voltage, and use Vcn+ for the controlling voltage.

- For threshold circuit, I was thinking of using a differentiator. The
output of this differentiator would be sent to another opamp circuit
with an output voltage reference to a value equal to Vcn-. The output
of this opamp would be connected to Vcn+.

---
You have the whole thing backwards.

Consider: If you have a DC threshold level that you want to
reference something to and you can change the gain of the circuit
instantaneously, then its output will be DC.

Most AGC circuits work by rectifying and integrating the AC output
and using that voltage to change the gain of the stage(s) preceding
the detector. That way, if you have a really loud passage of, say,
music, the integrator will charge up and lower the gain of the amp
for a while, and then when the amplitude gets lower it'll up the
gain of the amp. The trick is getting the timing right.

There's also compression...

Google is your friend.

 

[bw]

---
You have the whole thing backwards.

Consider: If you have a DC threshold level that you want to
reference something to and you can change the gain of the circuit
instantaneously, then its output will be DC.

Most AGC circuits work by rectifying and integrating the AC output
and using that voltage to change the gain of the stage(s) preceding
the detector. That way, if you have a really loud passage of, say,
music, the integrator will charge up and lower the gain of the amp
for a while, and then when the amplitude gets lower it'll up the
gain of the amp. The trick is getting the timing right.

There's also compression...

Google is your friend.

For some reason I guessed the OP wanted something like Automatic
Volume Control (AVC) but I typed AGC due to fatigue. Hope he gets what
he wants. "Feedback" is a big part of electronics.