6
\$\begingroup\$

As far as I know, analogue multiplexing is used to take voltage samples from many channels in a short time. It passes the voltages from an input to the ADC without amplifying them. I guess buffer them?

But is that called analogue because of the inner circuitry or because of its operation? And what is a digital multiplexer used for?

\$\endgroup\$
  • \$\begingroup\$ A typical analog multiplexer is a FET switch, and it doesn't buffer the signal. \$\endgroup\$ – Nick Alexeev Oct 23 '17 at 1:51
  • \$\begingroup\$ It matters more how it is made than what it is labeled. Some "bus switches" sold as digital devices actually make exceptionally good analog parts, used for example as mixers in high dynamic range radio receivers. \$\endgroup\$ – Chris Stratton Oct 23 '17 at 4:15
8
\$\begingroup\$

The operation of any multiplexer (MUX for short) is conceptually the same, be it analog or digital.

In essence you have a set of numbered data inputs (usually a power of 2, say \$N=2^n\$ inputs named, say, \$X_0, X_1,\$ ... etc.), a set of digital selection inputs (in number of \$n\$) and an output. A mux works by sending one (and only one) data input signal to the output. The data input to be "routed" to the output is the one selected using the selection inputs, i.e. the one whose number, expressed in binary, is put on the selection inputs.

In other words, a mux acts like a digitally-selectable single-pole/N-throw switch.

The difference between analog and digital muxes, seen from the outer world, is that the data inputs and the output are digital (two-level) for digital muxes, whereas in analog muxes the data signals can be analog.

The common implementation, in the two cases, is quite different. Analog muxes are made using FET switches (most commonly MOSFET transmission gates) and they act more like actual switch arrays, without amplification of the switched signals.

Take as an example the common 74HC4051 analog mux:

enter image description here

Here S0, S1 and S2 are the selection inputs, /E is the enable input (all four pins are digital), whereas Y0..Y7 are the analog data inputs and Z is the analog output. Note that in this case you can also reverse the role of the analog inputs and output and get a demultiplexer (a.k.a. DEMUX).

In particular, see how a single switch is implemented:

enter image description here

Digital muxes, on the other hand, use a bunch of digital logic gates to steer signals from (digital) inputs to the (digital) output. Digital muxes amplify the signal, since the internal gates are, after all, buffer amplifier.

You can refer to the Wikipedia article about the mulitplexer to understand how a digital multiplexer works internally. Actual implementations don't differ too much.

Take as an example the common 74HC151 digital mux and its internal logic diagram:

enter image description here

In this case they use I0,I1,...,I7 for the data inputs and Y for the output (they also provide a negated output for free).

\$\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.