A simple way to select between a local control signal and a remote control signal is sketched below.
When the DPST switch sections are open, local control is obtained using potentiometer R3 which feeds an op-amp voltage follower via R4 to that follower's non-inverting input. While this is going on, R2 keeps a feedback path closed for its associated op-amp to keep that op-amp from going to either of its rail limits.
When the DPST switch sections are closed, the R2-associated op-amp becomes a voltage follower which overrides the effect of R4 on the other voltage follower which allows the external control voltage to set the output.
The accuracy of all this depends on the op-amp offsets and gains, the exact values of the resistances and the exact values of the on/off switching resistances of the switch sections. In most cases, these will not be critical numbers, but if you are concerned with such things, you can do the worst case analyses to establish the error limits.
This cat has been put to bed in many different ways - mostly simpler than ths.
I.e. - the function of R2 would appear to be very subtle.
Please explain (and the two-pole switch)?
Posted by: george storm | March 01, 2012 at 09:39 AM
What is the purpose of the lower pole of the switch?
Posted by: Howard Edelman | March 01, 2012 at 11:15 AM
When the switches are open, potentiometer R3 simply feeds the output voltage follower. In this case, R2 keeps a finite feedback loop in place around the first op-amp to keep that section from saturating to either rail. Such saturation could upset the second op-amp if they are part of the same chip.
When the switches are closed, the first op-amp is brought to very nearly unity gain since R2 gets shunted by the switch sections. Also, the upper switch's on-resistance get's "buried" in feedback which ensures that the equivalent circuit source voltage and source impedance that is presented to the second op-amp is not materially affected by either R4 or R3.
Posted by: John Dunn | March 01, 2012 at 12:00 PM
The first op amp has unity gain even without the second switch pole closed and the offset voltage due to input current can be better balanced (R1 = R2 assuming that the External Control voltage input impedance is much lower). You may wish to put a resistor in the feedback loop of the second op amp for the same reason.
Posted by: Howard Edelman | March 02, 2012 at 02:39 AM