Can i switch between multiple resistors using MUX/switch? As i am trying to do this using MUX HEF4051, at the output i am getting extra 337 ohm resistance in actual resistance at input. I have connected 1 terminal of resistors at the input of 1 MUX while other terminal at the input of other MUX and i am checking resistance across output of both MUX but i am getting 337 Ohm of added resistance.

 

Search datasheet for NX3L4051 and MAX4734.

Allen

 

All muxes have resistance but some have much less than others.

What are the resistors doing in the circuit?

 

Can i switch between multiple resistors using MUX/switch?

Yes.

As i am trying to do this using MUX HEF4051, at the output i am getting extra 337 ohm resistance in actual resistance at input.

The 4051 data sheet indicates that there is a series resistance through the switch:

I have connected 1 terminal of resistors at the input of 1 MUX while other terminal at the input of other MUX and i am checking resistance across output of both MUX but i am getting 337 Ohm of added resistance.

Please post a schematic showing your application and the connections between the resistors and the muxers.

 

Search datasheet for NX3L4051 and MAX4734.

Allen

Tank u allen, will check it for sure n i will gt back to you

 

All muxes have resistance but some have much less than others.

What are the resistors doing in the circuit?

Actually those are base resistors of fixed value for another chip which i need to switch between according to my other input signal.

 

Yes.

The 4051 data sheet indicates that there is a series resistance through the switch:
View attachment 127120

Please post a schematic showing your application and the connections between the resistors and the muxers.

please check the attached image showing the connection though its not much detailed but i think it will show you very much about what i am trying to do.

 

What's a 1/p circuit?

 

The contact resistance of single mux is 165 Ohm. So toatal 330 Ohm of resistance for both muxes getting added in my input resistance. So is there any way that i can nullify this error.

 

What's a 1/p circuit?

That is input circuit and its nothing but RTD to digital converter MAX31865.

 

please check the attached image showing the connection though its not much detailed but i think it will show you very much about what i am trying to do.View attachment 127121

Indeed. What's the highest mux ON resistance you can tolerate, what's an i/p circuit, and what does its input look like?

 

Indeed. What's the highest mux ON resistance you can tolerate, what's an i/p circuit, and what does its input look like?

I cant tolerate any error in the input resistance to the input circuit as small error will also cause the huge deflection in output of input circuit.
I want input resistance as it is to the input of input circuit.

 

I cant tolerate any error in the input resistance to the input circuit as small error will also cause the huge deflection in output of input circuit.
I want input resistance as it is to the input of input circuit.

Since any switch will introduce a series resistance error term, you MUST be able to tolerate some level of inaccuracy. What is that level?
BTW, can you not do something like this?:

 

You need to use a four-wire measurement to eliminate wire and mux resistance errors, as shown in this first example circuit in the MAX31865 data sheet:

Thus the force (measurement) current to the RTD goes through two mux channels and the measurement inputs (which take only nA of leakage current) go through a separate pair of mux channels.
Any effect of the resistance in the force (current) channels is thus not seen by the measurement channels which carry no current and look directly at the voltage across the RTD.

It requires double the mux channels and wires, but eliminates the problem of wire and mux resistance on the measurement accuracy.
A pair of CD4097 dual 8-channel muxes will work well for that.
That same 4-wire mux scheme was used on a NASA weather satellite I worked on to accurately measure temperatures at various points in the satellite.

 

The contact resistance of single mux is 165 Ohm. So toatal 330 Ohm of resistance for both muxes getting added in my input resistance. So is there any way that i can nullify this error.

There may be, but in order to determine whether there is or not, you need to be able to accurately determine the dynamic series resistances of each of the switches when the machine is operating and compensate for them.

 

You need to use a four-wire measurement to eliminate wire and mux resistance errors, as shown in this first example circuit in the MAX31865 data sheet:
View attachment 127122
Thus the force (measurement) current to the RTD goes through two mux channels and the measurement inputs (which take only nA of leakage current) go through a separate pair of mux channels.
Any effect of the resistance in the force (current) channels is thus not seen by the measurement channels which carry no current and look directly at the voltage across the RTD.

It requires double the mux channels and wires, but eliminates the problem of wire and mux resistance on the measurement accuracy.
A pair of CD4097 dual 8-channel muxes will work well for that.
That same 4-wire mux scheme was used on a NASA weather satellite I worked on to accurately measure temperatures at various points in the satellite.

The MAX31865 can only tolerate up to 50 ohms Rcable:

so the 4097 won't work. TI, AD, and NXP, among others, have muxers with Ron way less than 50 ohms.

 

You might be able to get around the resistance of the switches by using an opamp in this circuit which is relatively insensitive to the resistance of the switches. Looking into port A you will see the selected resistance with respect to port B. The drawback is that unless you double the number of parts if you look into port B you will always see a high resistance. Is this compatible with your application?

A 74HC4066 is used as an example. A '405x mux should work well also.

 

You need to use a four-wire measurement to eliminate wire and mux resistance errors, as shown in this first example circuit in the MAX31865 data sheet:
View attachment 127122
Thus the force (measurement) current to the RTD goes through two mux channels and the measurement inputs (which take only nA of leakage current) go through a separate pair of mux channels.
Any effect of the resistance in the force (current) channels is thus not seen by the measurement channels which carry no current and look directly at the voltage across the RTD.

It requires double the mux channels and wires, but eliminates the problem of wire and mux resistance on the measurement accuracy.
A pair of CD4097 dual 8-channel muxes will work well for that.
That same 4-wire mux scheme was used on a NASA weather satellite I worked on to accurately measure temperatures at various points in the satellite.

Thanks for the brief but for me restriction is to use 3-wire RTD. Can you please show me how to do it with doubling the MUX?

 

There may be, but in order to determine whether there is or not, you need to be able to accurately determine the dynamic series resistances of each of the switches when the machine is operating and compensate for them.

Is there any way to compensate for that resistance through hardware?

 

Since any switch will introduce a series resistance error term, you MUST be able to tolerate some level of inaccuracy. What is that level?
BTW, can you not do something like this?:
View attachment 127124

Actually i want to switch between multiple 3-wire RTDs.