7.1 surround volume control

[RussellW]

Hi,
RK16816MGA04 ALPS Potentiometers

Minimum order: 1
$9.91 + $7.00 Freight + $0.92 Tax = $18.33

Part number RK16816MGA04
Number of resistor elements 6 ganged-unit
Shaft types Flat
Length of the shaft 25mm
Total resistance 100k ohms
Resistance taper 3B (audio)
Motor print terminal With
Motor voltage 4 to 6V DC (standard: 4.5V)
Potentiometer rotational speed 12±3 sec/300° (4.5V DC)
Maximum operating current At rotation: 100mA/At end slip: 150mA
Can be turned with shaft or motor.
ARP RK16816MGA04

This unit has 6 gangs which could be used for center, F-right, F-left, R-right, R-left, and subwoffer.
As I understand it for 7.1 surround I would need 8 gangs.
I ordered two and am going to make one track the other using 2 of the 12 gangs for references. I would need to send current through the 2 references and use a comparator to rotate the second unit. My concern is that it would be so sensitive that the second unit would always be running back and forth.
I would like suggestions about how to reduce the sensitivity so that the second unit is usally at rest.

Thank-you,
Russell Williams

 

[macboy]

You can't reduce the sensitivity of a true comparator. It is either hi or low. On or off. No in between. If you want three states (turn right, stop, turn left), you need at least 2 comparators or something else.

Have two comparators, one for the positive leg of the motor, and one for the negative. Add a little hysteresis to one or both. Now there will be a dead zone where both comparators will be either + or - output, and the motor will not turn. You can use a single supply with this arrangement.

Or try a rail-to-rail opamp set up for high gain, but not open-loop. When the two pots are very different, the opamp will swing to one rail or the other, driving the motor full strength to correct this. When they are close, the output tends toward zero to give less, then no motor activity. The higher the gain, the closer the match, until you just get oscillations. You will need a dual supply for this arrangement since one leg of the motor will always be grounded. Use transistors to buffer the output of the opamp to give sufficient drive for the motor. If you do not bias the transistors you can take advantage of the dead zone they will create.

Neither approach can give you 100 % match between the two pots.

You might look into adding some extra resistors to the sense gangs of those pots to try to linearize the output a little. You can probably get an S-curve, but definitely not a linear output. Still, that's better than a log curve.

 

[SoNic_real_one]

You need a PI(D) regulator - that is a facy name for two OpAmp's - one that will do the Proportional part and the other for Integrator part. You can add the Derivative part if you feel like
PID controller - Wikipedia, the free encyclopedia
The Op Amp PID Controller

 

[RussellW]

Is hysteresis negative feedback; placing a resistor from the output to the negative

Is hysteresis negative feedback; placing a resistor from the output to the negative input?

You can't reduce the sensitivity of a true comparator. It is either hi or low. On or off. No in between. If you want three states (turn right, stop, turn left), you need at least 2 comparators or something else.

Have two comparators, one for the positive leg of the motor, and one for the negative. Add a little hysteresis to one or both. Now there will be a dead zone where both comparators will be either + or - output, and the motor will not turn. You can use a single supply with this arrangement.

Or try a rail-to-rail opamp set up for high gain, but not open-loop. When the two pots are very different, the opamp will swing to one rail or the other, driving the motor full strength to correct this. When they are close, the output tends toward zero to give less, then no motor activity. The higher the gain, the closer the match, until you just get oscillations. You will need a dual supply for this arrangement since one leg of the motor will always be grounded. Use transistors to buffer the output of the opamp to give sufficient drive for the motor. If you do not bias the transistors you can take advantage of the dead zone they will create.

Neither approach can give you 100 % match between the two pots.

You might look into adding some extra resistors to the sense gangs of those pots to try to linearize the output a little. You can probably get an S-curve, but definitely not a linear output. Still, that's better than a log curve.