Hi, I have a few questions on a simple matter . Most of the amps I've built never used pots for volume control, I used the signal source's volume control ( digital ) for that .
Now using the classic volume control with a pot like this :
would this affect the quality of the signal ? or , change the input impedance based on the pot's location ?. Using a buffer before the pot would help ?.
What about this , would this be a better approach ? besides the output being inverted and not being able to have " zero volume ". would this affect the input impedance as well when pot is turned ?.
What is the best approach for this ? Best way to use potentiometer for volume control ?. ( besides using digital volume control ic's )
- Bruno.
Now using the classic volume control with a pot like this :
What about this , would this be a better approach ? besides the output being inverted and not being able to have " zero volume ". would this affect the input impedance as well when pot is turned ?.
What is the best approach for this ? Best way to use potentiometer for volume control ?. ( besides using digital volume control ic's )
- Bruno.
Not a good approach.
For a volume control inside the power amp, use a 10k pot (unless used with a tube preamp/source).
Generally no buffer following the pot will be needed for a 10k inside the amp.
For a volume control inside the power amp, use a 10k pot (unless used with a tube preamp/source).
Generally no buffer following the pot will be needed for a 10k inside the amp.
Do you want the pot inside the amplifier to be your ONLY volume control?
If not, why do you want or need it at all?
If it is the only volume control, use a 10k (higher if tube source) switched resistor control, with only high quality resistors and switches.
Tubecad has many interesting options. https://glass-ware.stores.turbify.net/atandsise.html
If not, why do you want or need it at all?
If it is the only volume control, use a 10k (higher if tube source) switched resistor control, with only high quality resistors and switches.
Tubecad has many interesting options. https://glass-ware.stores.turbify.net/atandsise.html
There will always be compromises. Any volume control will add noise, for example. A good designer will ensure that the noise added is well below audible.
Your second schematic will result in a quite noisy volume control, especially at the low volume settings where the noise will be the most offensive.
If you don't like the traditional volume control, have a look at the Baxandall volume control: https://www.ti.com/lit/ug/tidu034/tidu034.pdf
Tom
Now using the classic volume control with a pot like this :would this affect the quality of the signal ? or , change the input impedance based on the pot's location ?. Using a buffer before the pot would help ?.
A buffer is not needed. If the potentiometer value is kept low the noise will also be low, and 10k Ohms or 20k is common in this position. The input impedance is constant at any attenuation position, although the source resistance seen by the amp’s input changes. That has no consequence here.
The 10k load is a good compromised between low noise and still having a high enough input impedance for most all preamps that may end up driving the amp. Most SS preamps could drive 1k easily. Only preamps having unbuffered outputs and an output impedance of several kiloOhms will have rolled off bass in this scenario.
Your first approach is best, and just needs an RC filter to attenuate out of band radio frequencies from getting into the amplifier. The second approach appears to employ the pot like a variable resistor. As the volume knob is turned to attenuate the signal and the resistance increases, the noise will also increase and be amplified. Instead, always connect the low side of the pot to the signal ground plane like in your first image.
Sadly not such a thing.
All real world solutions to problems involve some kind of compromise or another
A Baxandall volume control uses a linear pot and has a better log response than a so called log pot.
Something to consider about the Baxandall active volume control is, it gives only 20dB to 30dB of user-friendly attenuation. It was designed for use on mixing consoles and precise recording level adjustment panels.
Better than 270 degrees of the top righthand control range provides 20-30dB of level adjustment, but the bottom 30 degrees of the lefthand volume knob range is sensitive. Even a slight adjustment at the bottom of the control knob range causes large changes in output levels, like 60dB level change is squeezed into 30 degrees of rotation. It’s effectively the opposite of a log volume potentiometer in the lower control range.
This is fine if the plan is to do the actual low listening volume adjustment from a preamp having a conventional logarithmic control. However, if he ever plans to connect a source directly to the amp inputs, the Baxandall may prove frustrating. If the wiper momentarily loses contact with the track the result is a full scale burst.
My preamp had such a volume control and I disabled that curse and converted it to a conventional volume control.
Better than 270 degrees of the top righthand control range provides 20-30dB of level adjustment, but the bottom 30 degrees of the lefthand volume knob range is sensitive. Even a slight adjustment at the bottom of the control knob range causes large changes in output levels, like 60dB level change is squeezed into 30 degrees of rotation. It’s effectively the opposite of a log volume potentiometer in the lower control range.
This is fine if the plan is to do the actual low listening volume adjustment from a preamp having a conventional logarithmic control. However, if he ever plans to connect a source directly to the amp inputs, the Baxandall may prove frustrating. If the wiper momentarily loses contact with the track the result is a full scale burst.
My preamp had such a volume control and I disabled that curse and converted it to a conventional volume control.
And better stereo tracking, since linear stereo pots track better, especially at more attenuation.
I've never had an issue with a plain ALPS RK271-series "Blue Velvet" volume pot. I'd get a 5-10 kΩ logarithmic one.
Tom
Tom
In your second diagram, short-circuit R2 and connect the NFB to the wiper of the pot. Then you have an active gain control. Use a linear pot. Many advantages. Baxandall is even better but you need to add a prior buffer opamp.
Good for you. That doesn't make it a more robust design, though.
At some point in grad school I mentioned to my advisor that a circuit I designed "shouldn't fail" in a certain way. His response was, "design it so it can't fail that way".
Tom
At some point in grad school I mentioned to my advisor that a circuit I designed "shouldn't fail" in a certain way. His response was, "design it so it can't fail that way".
Tom
I'll throw in rod elliott's article for "better volume control":
https://www.sound-au.com/project01.htm
https://www.sound-au.com/project01.htm
Not with a properly designed Baxandall circuit, you simply lose the signal for the period of the drop-out (just like a passive volume control). Its imperative not to have opamp bias current flowing through the wiper with a BJT input opamp - neglecting that important rule will lead to massive noise bursts on a scratchy pot as the opamp slams into the rails, even with a simple buffered passive volume control.
It is wise to use a FET opamp when pots are involved as you can reduce the AC loading on pot wipers substantially, not just the DC current.
For an example see: https://sound-au.com/project01.htm, or the circuit used by Doug Self: https://www.diyaudio.com/community/threads/need-volume-control-design-recommend.295823/#post-4816231
Hi Mark,
Thanks for pointing that one out by Doug. I think that design drops to about -60dB if the wiper contact is interrupted, much like the pre in Linear Audio Issue 5.
The input bias needs a fixed, uninterrupted current path, but there are still designs surfacing without a guaranteed path. I’ve worked on two devices recently wherein the volume/balance pots were being used as input bias paths, the solution being to add a resistor to ground close to the op amp input.