0
\$\begingroup\$

I need a very simple circuit to connect a signal to one of two ways like this:

enter image description here

My com signal is actually a microphone input, so very low voltage, <200mV, how can I make a solid state switch that not interfere with the low voltage signal? So com can go to L1 or L2 with no distortion. Im planning to actuate the switch with TTL voltage.

\$\endgroup\$

3 Answers 3

3
\$\begingroup\$

Two suggestions:

  • You can use e.g. an analog switch/MUX like the 4053.

  • If the signal you want to switch is a pure AC signal you might also be able to use a diode that is switched between forward and reversed biased. See for example here in section "Analog Switch". This method is often used with signals in the RF range.

\$\endgroup\$
2
\$\begingroup\$

The electronic solutions proposed are fine and will do what you want. However, if you want a more generic and flexible solution I'd consider using a decent quality reed relay. You are not limited with supply rails in this method and the input can be totally floating and isolated from your control voltage.

Here is a decent offering from Pickering relays. It can be driven from TTL and is suitable for switching thermocouple inputs (this means it will be fine for low level audio signals too).

enter image description here

\$\endgroup\$
2
  • \$\begingroup\$ I also thought about (reed) relays but the OP explicitly asked for a solid state switch. \$\endgroup\$
    – Curd
    Oct 12, 2013 at 7:58
  • \$\begingroup\$ @Curd that's why I began my answer suggesting the electronic proposals are fine. I'm offering this answer up in case the OP hasn't considered that sealed reed relays are a viable alternative to solid-state switching - given that he wants to switch a microphone signal (which might also be a balanced signal and contain 48V phantom power) I thought it appropriate but yes, it aint solid-state for sure. \$\endgroup\$
    – Andy aka
    Oct 12, 2013 at 8:20
2
\$\begingroup\$

Using a dedicated analog MUX, as stated above, would probably be the best solution. You can, however, build it from discrete components. Following circuit should do the job. Each switch (otherwise known as a transmission gate) consists of two MOSFETs, P- and N-Channel respectively. It is activated by applying VCC to N-FET's gate and by pulling P's gate to GND. Applying voltages in the opposite order will disable the switch. That's what the inverter's doing (it can be realized by two MOSFET's as well).

Note that this is only guaranteed to work when used in the form of an off-the-shelf IC. It may or may not work when bodged together using discrete CMOS

analog MUX schematic

It will function properly - more or less linearly - if the INPUT voltages remain below VCC - Vth, where Vth is the switch-on threshold voltage of the MOSFET's.

\$\endgroup\$
1
  • \$\begingroup\$ You should mention that these are called transmission gates; building a transmission gate on a wafer makes sense because you have localized impurities/etc, but out of discrete components there might be an issue with matching - not sure, would have to check. \$\endgroup\$
    – trayres
    Oct 12, 2013 at 6:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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