Outputting the larger of two voltage sources.. how?

[eyAyXGhF]

Hey guys,
I haven't stumbled across a circuit to do this in any of my books, maybe because I don't know the name of it. I have two voltage sources, and I'd like to output the greater of the two.

Either will be going from approx. -1 -> +10V

For example: If V1 = 3V and V2 = 5V, the output should be 5V.

I suppose this would be considered an analogue OR circuit?

 

[crutschow]

If you don't mind a small voltage drop you could use a couple of Schottky diodes.

Otherwise you could use a differential comparator controlling a couple of P-MOSFETs in series with each power source. The comparator turns on the MOSFET with the highest voltage. (Note: For this scheme you need to use the P-MOSFET source as the output, not the drain as you normally would, to prevent reverse conduction through the substrate parasitic diode. That works because MOSFETS are bidirectional).

 

[Diver300]

Two diodes would do it, but the voltage would be reduced by the diode voltage.

You could use a precision rectifier circuit if you want the maximum voltage accurately.

 

[alec_t]

What current would the voltage sources be expected to drive?

 

[3v0]

Either will be going from approx. -1 -> +10V

The two diode thing will not work if you are using negative voltages.

 

[eyAyXGhF]

The two diode thing will not work if you are using negative voltages.

Would the precision rectifier method work with negative voltages? Thinking about it for simplicity, I -can- have the output as 0 - +10V and forget about the negative portion.

To answer some above questions, the diode drop wouldn't be an issue... I'm concerned with the final output voltage, so I can calibrate the two voltage sources to overcome the diode drop no worries. As for current draw, this will be for an output so I will be buffering it anyways. Hope that helps!

Mike

 

[Grossel]

look at this circuit - ned slightly modifications

Hi. This simple opamp circuit will output the most negative of two (or add as many as you would like) inputs.

To get your circuit to work, you have to flip the diodes and connect the resistor to negative voltage supply instead.

[edit]
It will work for voltage ranges limited by:
* Maximum and minimum voltage from opamp, wich in most cases is close to the supply voltages to the opamp
* The voltage at the end of the resistor. Should be connected to negative voltage supply to the opamp to ensure maximum voltage range. If you forget that and just connect it to ground, the circuit will not work for negative voltages.

 

[ChrisP58]

Would the precision rectifier method work with negative voltages? Thinking about it for simplicity, I -can- have the output as 0 - +10V and forget about the negative portion.

To answer some above questions, the diode drop wouldn't be an issue... I'm concerned with the final output voltage, so I can calibrate the two voltage sources to overcome the diode drop no worries. As for current draw, this will be for an output so I will be buffering it anyways. Hope that helps!

Mike

If the two voltages are just control/sensor signals, I would use a voltage comparator to detect which is greater, and use it's output to drive an SPDT relay or analog switch.

 

[3v0]

If both inputs never go negative two Schottky diodes will work as good as any solution.