Controlling volume with a CV

[drolo]

Following up on the idea shown here: http://www.diystompboxes.com/smfforum/index.php?topic=103364.0
I was looking for ideas for a volume controll operated by a CV (0-5V preferably). I would like to replace the 2 out of phase LDR's  I was using as a pot, with something more linear, simple, cheap, predictable ...
I did some googling and reading around and what appears to be the most convenient, is an OTA like the LM13600/700. Not having used one yet, I know from rumors only that they are quite noisy and prone to distort easily if the input is too hot. Is that so?

Would there be any way to use regular opamps where the volume could be controlled with a 0-5v CV voltage from total silence to unity gain?

[GibsonGM]

You could do this using an optocoupler and 'silencer circuit' (aka Noise gate).  But I'd like to see how someone with more knowledge than I would set up a variable resistance using a CV, WITHOUT using an opto  ;o) 

[samhay]

Total silence is hard. Would -20, -40 etc dB be enough?
Is it Ok if you have more than unity gain and then a knock-down voltage divider or volume pot?
I assume you would like a nice linear or logarithmic volume change with CV voltage?

You want a voltage-controlled amplifier, which is what the LM13600/LM13700 are, but you don't want to use a VCA. Fair enough - this is not a criticism, as I have puzzled this on a few occasions myself.
Can you use the CV to drive a voltage controlled 'resistor' (LED/LDR, FET...) in e.g. an op-amp feedback loop or to a (-) input? Tremolo designs are probably a good place to start, but then you just designed a tremolo, so you probably knew that already.

On the back of an extant thread of Rob's about the Escobedo Q&D VCF (http://folkurban.com/Site/QDVCF-714.html), I tried to re-jig this topology to give a flat response. This will give you gain as the CV voltage increases - not what you want, but perhaps something different to try.

[drolo]

Hi Sam,

Total silence is hard. Would -20, -40 etc dB be enough?

I would need to check if there is not too much bleed through, especially if placed after a distortion/overdrive

Is it Ok if you have more than unity gain and then a knock-down voltage divider or volume pot?

That's absolutely OK

I assume you would like a nice linear or logarithmic volume change with CV voltage?

I think either one would be OK

Can you use the CV to drive a voltage controlled 'resistor' (LED/LDR, FET...) in e.g. an op-amp feedback loop or to a (-) input?

If I find no other way, I will keep my LED/LDR arrangement. It works but I was trying to get around without using optocouplers. For the price but also especially for the current needed to drive them.

On the back of an extant thread of Rob's about the Escobedo Q&D VCF (http://folkurban.com/Site/QDVCF-714.html), I tried to re-jig this topology to give a flat response. This will give you gain as the CV voltage increases - not what you want, but perhaps something different to try.

I will have a look at you pic when i'm at home. dropbox is blocked here at work ... ;-)

[samhay]

All sounds fair enough.
I have tweaked and updated the schematic slightly. This hasn't been breadboarded, but the generally topology is sound, if not a little unusual - Tim Escobedo has certainly had a few very interesting ideas. You can adjust the gain in a number of ways, but you will have to take some care so as not to hit the power rails.
This is probably not the best approach for a tremolo, but hopefully it might stimulate some discussion.

[R.G.]

I was looking for ideas for a volume controll operated by a CV (0-5V preferably). I would like to replace the 2 out of phase LDR's  I was using as a pot, with something more linear, simple, cheap, predictable ...

The effect of some of Mother Nature's Laws appears to be summarized in the standard quip: good, fast, cheap - you may pick any two.

I did some googling and reading around and what appears to be the most convenient, is an OTA like the LM13600/700. Not having used one yet, I know from rumors only that they are quite noisy and prone to distort easily if the input is too hot. Is that so?

OTAs are differential amplifiers on their input. All differential amplifiers have a linear range of about +/-25mV before they are overdriven. Opamps get around this by using the gain of the rest of the amp to enforce the inputs only seeing a few millivolts. OTAs suffer from this in that inputs have to be attenuated down to the linear level, then amplified back up, adding noise at each step. It takes care and thoughtful design to keep the noise down. Thoughtful design is relatively rare.

Would there be any way to use regular opamps where the volume could be controlled with a 0-5v CV voltage from total silence to unity gain?

Go read about the VCA in the SE571.

[merlinb]

I did some googling and reading around and what appears to be the most convenient, is an OTA like the LM13600/700. Not having used one yet, I know from rumors only that they are quite noisy and prone to distort easily if the input is too hot. Is that so?

You can improve the usefulness of an OTA by using it as a variable resistor in the feedback loop of an opamp. It is then doesn't see large enough signal to distort (much), and its noise contribution is reversed (contributes most noise when the circuit gain is least).
One possibility: The gain control law is not super linear (though it can be made so). Also the gain is highest when the control voltage is lowest, and it can't drop below unity. I suppose you could reconfigure it as an inverting amplifier instead, but it would be difficult to get the gain truly down to silence.

[drolo]

All sounds fair enough.
I have tweaked and updated the schematic slightly. This hasn't been breadboarded, but the generally topology is sound, if not a little unusual - Tim Escobedo has certainly had a few very interesting ideas. You can adjust the gain in a number of ways, but you will have to take some care so as not to hit the power rails.
This is probably not the best approach for a tremolo, but hopefully it might stimulate some discussion.

I had a look at your schematic and it does look interresting, unusual indeed ;-) but at least it would be quite a simple solution.
I am actually not giong to use this in a tremolo but a volume pedal, not that it matters actually

You can improve the usefulness of an OTA by using it as a variable resistor in the feedback loop of an opamp. It is then doesn't see large enough signal to distort (much), and its noise contribution is reversed (contributes most noise when the circuit gain is least).
One possibility: The gain control law is not super linear (though it can be made so). Also the gain is highest when the control voltage is lowest, and it can't drop below unity. I suppose you could reconfigure it as an inverting amplifier instead, but it would be difficult to get the gain truly down to silence.

Funnily I was going to look up your engineer's thumb schematic again before you posted this :-)
You just made my life even easier.

I will give these a try as soon as I get some time to breadboard

Thanks already for your useful suggestions

[electrosonic]

Looking at Merlin's schematic - what is the relationship between CV and attenuation? Linear or logarithmic or something else?

Andrew.

[PRR]

Linear; +5V is low gain.

[drolo]

Now that I think of it, I doubt my CV is linear to begin with, as it is produced by a magnet and hall effect sensor.
I need to test some of these solutions and see what works best for a volume pedal sweep.

[amptramp]

Now that I think of it, I doubt my CV is linear to begin with, as it is produced by a magnet and hall effect sensor.
I need to test some of these solutions and see what works best for a volume pedal sweep.

A Hall-effect sensor and a voice coil may be used to make a multiplier - either audio into the coil and voltage to control it or voltage into the sensor and voltage out with the coil to control it.  You may already have what you need.