Do mux-es need overhead voltage to work correctly

[Eraldo]

My question is if muxes can be supplied by the same voltage they are switching and if they can be used in an extreme case such as inside the feedback loop of a dc dc converter.
 
I am trying to build a switchable voltage using a mux to control the feedback (switch resistors) of a dc-dc converter while being powered by said converter output (the mux supply with be connected with the output of the dc dc converter) and i wanted to know if anyone else has ever tried it. I could easily use discrete p channel mosfets but i liked the simplicity of a mux.

Logically, the mux itself is made by discrete mosfets itself back to back, plus the control logic and assuming they are driving p channel mosfets inside the chip, the circuit wouldn't need any overhead voltage (imo).

The bigger problem i see with this design approach that I'm trying to use is the behaviour of the mux during transients and if it would be able to keep up with the dc dc converter.

Tldr: can you switch feedback resistors ina dc dc converter using a mux which itself is powered by said voltage it is controlling.

[TimFox]

Read the data sheet for the individual mux circuit to see its useful range.
The CD4051 CMOS mux can switch voltages between the power pins Vdd (positive) and Vee (negative), while the logic control signals are between Vss (ground) and Vdd.
There are limits on the three supply voltages:  see 
https://www.ti.com/lit/ds/symlink/cd4051b.pdf?ts=1732830697816&ref_url=https%253A%252F%252Fwww.mouser.at%252F
A DG508 also operates with an analog voltage between the supply rails, typically +15 V and -15V, which is a higher operating supply voltage than the CD4051.  There is also a "GND" pin for the digital control.
https://www.analog.com/media/en/technical-documentation/data-sheets/dg508a-dg509a.pdf

The "GND" or Vss pins on these devices are not actually power pins, but reference inputs for the digital control.
Be careful about power supply sequencing:  if the positive supply turns on while the negative supply is a high impedance (such as a 7905 negative regulator with low input voltage), the device may "latch up" and self-destruct.
To avoid this, caused by the GND pin being more negative than the negative power input due to current flowing from the positive supply into the high impedance temporarily present on the negative power pin, one should add a Schottky diode from the positive and negative pins on the IC to ground (normally reverse biased).  A conventional PN diode can sustain enough voltage to latch up the circuit.
(Ask me if I learned this the easy way or the hard way.)

[mawyatt]

The "On Resistance" of the various Analog Switch types are usually not constant across the input voltage range, some vary significantly which can introduce non-linearity in the signal chain. Check the data sheet in detail before utilizing.

Best

[TimFox]

When that variation is important, a differential connection of two switches will show matching between ON resistances better than the variation of resistance, over the common-mode voltage range.

[Eraldo]

Thanks for the replies. I especially didn't know that fact about mux-es latching up and burning like that.

In my application i will be controlling the voltage of a dual output boost converter (+-15V) so I was considering a 30+V mux for the job. But maybe might be better for 2 independent 16+V single supply muxes for this application. The annoying part would be to control the negative voltage mux digital pins since it would need a discrete inverter.

I guess the only way to know if it will work is to simulate it and build a prototype 

Also matwyatt thanks for the tip. Thankfully the voltage stability criteria is not that high and the feedback resistor values are quite high.

[TimFox]

When my similar circuit was latching up, a factory engineer told me that the important thing was to identify the substrate (most negative pin) and the most positive pin (Vdd or equivalent) and ensure that no other pin (including the "GND" pin for a mux) could go outside of the range of voltages by as much as a PN diode drop (call it 0.6 V).  Hence the need for Schottky diodes.

[mawyatt]

I am trying to build a switchable voltage using a mux to control the feedback (switch resistors) of a dc-dc converter while being powered by said converter output (the mux supply with be connected with the output of the dc dc converter) and i wanted to know if anyone else has ever tried it. I could easily use discrete p channel mosfets but i liked the simplicity of a mux.

If the feedback is a series resistor and shunt resistor, then using a Mux to ground with different shunt resistors to set the output might reduce the voltage stress across the Mux, rather than a Mux that switches the series resistor.

Anyway, just a thought.

Best