Momentary Switch to Latching - Audio

Thread Starter


Joined Sep 16, 2014
Hi, this isn't an easy question to phrase, but essentially I'm looking at using momentary switches to control which headphone jack (3 total) the input audio will be diverted to. Ideally, I would like the selected output to be the only output - ie, if output 1 is active, then the button for output 2 is selected, output 1 becomes inactive and output 2 stays on.

I understand the basics of turning a momentary switch into a latching one (have "achieved" this before using circuits found online), but that method was used as a power switch, where the calculations were simpler. On top of this, the plan I have also requires the original latch do be deactivated. I understand that this might be seen as a bit of a jump in terms of complexity, but I was wondering if anyone would be able to point me in the right direction on this project I have planned.

Thanks in advance


Joined Aug 21, 2008
At one time, you could find an assembly of buttons that would allow only one to be latched on at a time, mechanically unlatching every other button in an array when one button was pressed. These days, a button array like that would be difficult to find. The most practical arrangement would be (assuming stereo) a 3P3T or 3P4T rotary switch.

I you really need to use buttons, then you can use an array of relays to do latching; a microcontroller or discrete logic circuit or even discreet components (eg resistors and transistors) to do the logic, the output of which would drive a relay of one sort or another.

What is your preference?


Joined Apr 11, 2010
Hi Benby,

You may use 3 magnetic latch relays. Each momentary switch is connected to a 'set' coil which is in parallel with the other two 'reset' coils. The contact of each relay connects the source to the corresponding output.


Without seeing a schematic, I think you'll need steering diodes on the reset inputs, to prevent a sneak path to the selected relay. This would result in trying to both set and reset the selected relay, resulting in an unknown condition.

Consider three positions for an example. Relay 1 would go to Relay 2&3 reset inputs. Relay 2 would go to Relay 1&3 reset inputs. But when you select Relay 1, the reset signal would go to Relay 2 and then through the parallel connection, back to Relay 1's reset.


Joined Apr 11, 2010
I've sketched up what I think you are proposing. (BTW, I've used these latching relays in model railroading).

Note I've omitted the Set inputs for this example. A similar problem to what this illustration describes occurs with the set inputs. In the picture below, I've highlighted the current path when the first momentary switch is pressed. Note that relays 2&3 are reset (in yellow highlighter). However, due to the parallel connections, there is a path to reset on relay 1 (shown by the orange highlight). Set and Reset are going to fight each other on this relay, resulting in an unknown condition and perhaps relay damage.

Diodes on each path are necessary to prevent such sneak paths. Here is what I mean:

You can see how they block unintended current paths in the picture below.