2
\$\begingroup\$

How can I make simple model and simulate the crosstalk between two adjacent channels of a multiplexed DAQ inputs? I observe that in its most exaggerated form, a nearly exact duplicate of one channel appears on an adjacent channel to which nothing is connected. I need to demonstrate in LTspice or any simulator that both sampling rate and the load impedance has effect on the crosstalk.

What I want to simulate that how the sampling rate and the load impedance effects the crosstalk on the other channel. Im a bit confused how I can model and see the coupling effect between channels.

\$\endgroup\$
  • 1
    \$\begingroup\$ I would start by just putting some impedance between the channels; a small capacitor, maybe a mutual inductance. \$\endgroup\$ – Hearth Oct 6 '19 at 19:00
  • \$\begingroup\$ Crosstalk depends on coupling impedance/source impedance ratio for high Zload , which depends on cable lengths, gap and relative orientation for pF and uH. e.g. Ribbon wire 20nF/m 0.5uH/m \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Oct 6 '19 at 19:58
0
\$\begingroup\$

consider this.

By the way, as the analog-mux switches from one source to another source, the new source MUST provide charge to erase the voltage of the previous source.

schematic

simulate this circuit – Schematic created using CircuitLab

by the way, I ignored ON_resistance of the FETs. Also ignored inductance of the IC package, of bondwires, of PCBtraces, of VDD bypass caps for the Multiplexor power (+5, +15, -15)

Sampling rate will not affect the crosstalk, unless your rate is so fast that the LOGIC gates that drive the analog-switches are not able to fully control the analog-switches state to fully on or to fully off. This likely will occur at some multiplexing rate well above 1MHz. Also depends on how clean YOU maintain the GND and the VDD of the IC; will you simulate that? Place 10nanoHenry in each VDD connection.

Search around; I already provided a fully-detailed schematic for this type of question, months ago.

\$\endgroup\$
  • \$\begingroup\$ Here it says high sampling rate will boost crosstalk: "Even under this nearly ideal impedance situation, a high sample rate can boost crosstalk by minimizing the capacitive discharge time on the multiplexer's channels. In effect, the capacitance has less time to bleed off its charge before the analog-to-digital conversion takes place, resulting in crosstalk where none existed before." Do you understand what they mean? (reference: dataforth.com/catalog/pdf/an116.pdf) \$\endgroup\$ – cm64 Oct 6 '19 at 23:15
  • \$\begingroup\$ Are V1 and V2 sources coupled to channels? But where is the multiplexer here switching through the channels? NOT1 and NOT2 are for multiplexer switching? Can you yourself simulate this? I need something I can simulate and plot the crosstalk effect. \$\endgroup\$ – cm64 Oct 6 '19 at 23:18
  • \$\begingroup\$ If one input is floating, there is no charge source available to change the charge on the output 10pF capacitor Cload. \$\endgroup\$ – analogsystemsrf Oct 7 '19 at 12:48

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.