0
\$\begingroup\$

enter image description hereI would like to multiplex battery voltages used for impedance measurements. The battery cells are not connected. My idea is to have one multiplexer chip for the positive poles, and one chip for the negative poles. Then the outputs are compared to get the battery voltages.

Is it OK to multiplex floating voltages like this, and then compare them afterwards? What factors might affect the measurements? Are there better alternatives?

I am planning on using the mux MUX506IDWR. The outputs of the muxes will go to a measuring device (Gamry Reference 3000) which I assume has high impedance and will not draw current in the voltage measurement. I don't really know what goes on inside of the Gamry device. All I want is to deliver the battery voltages without distorsion, as if a battery is directly connected to the Gamry.

4-point measurement is used, with a relay multiplexer used for the current cables. Relays are used for the current wires, since 3 A will be used in the measurement. Using mux ICs for the voltage measurements are used in order to save money and space.

\$\endgroup\$
  • 3
    \$\begingroup\$ You should draw a circuit to show what you intend to do. The mux you selected has a typical series resistance of 125 ohms that will need to be accounted for if you want your measurements to be accurate. You might think you can just subtract 125 ohms but it's not that simple, the 125 ohms is a typical value, it can be much less or a bit more. In a properly designed circuit you would not let any current flow through the mux used to measure the voltages. Then the current is low and the series resistance can be neglected. Use a 2nd mux to switch the currents. Read up on 4-point measurements. \$\endgroup\$ – Bimpelrekkie Jun 26 '18 at 13:56
  • \$\begingroup\$ It is very important that you tell us how you intend to "compare" the outputs of the multiplexers. Please provide a schematic and links to the components you plan to use. \$\endgroup\$ – Elliot Alderson Jun 26 '18 at 14:01
  • \$\begingroup\$ No links for the multiplexer or Gamry? How is the relay multiplexer connected and why don't you just use another one of those? You are really making it hard to help you. \$\endgroup\$ – Elliot Alderson Jun 26 '18 at 14:29
2
\$\begingroup\$

It is not OK to think each battery is floating, when your concept of sharing some 0V between all of them due to stray capacitance.

They must be have some common mode 0V even when you have a MUX.

Therefore if there the extras system impedance between these cells is always high and <100pF you can use a Two Pole MUX with an INA using a twisted balanced wire pair that has more shield capacitance to 0V analog than the capacitance between cells. You still need to consider ESD and current limit protection on all inputs to protect from cap discharge or crosstalk capacitance of your MUX . and settling time.

On long cables with this criteria where common mode input impedance is low and differential is high , you can still get noise on SAR type ADC’s so a cap snubber on inputs with ADC trigger delays can suppress noise to <1mV.

Whereas integration type ADC’s automatically remove noise during slower conversion.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.