It seems like a switched jack would be a straightforward way to make pins do "double duty" in a uC-based application. That is, the "normally closed" connections of the jack could connect the programming-related pins to the application circuit, but when the "programming cable" was connected, you wouldn't have to worry about interference from or damage to the application.

Is this done? If not, is it b/c switched jacks are just way too expensive for this use? Is the wiring just too long and noise-sensitive? Other reasons?

  • 1
    \$\begingroup\$ Solid-state switches are more reliable than moving parts, and designing them to avoid interference with the application circuit is not any less necessary with mechanical switches. Open mechanical connections still acts as antennas, cause reflections, etc. \$\endgroup\$
    – Ben Voigt
    Sep 16, 2017 at 21:28
  • \$\begingroup\$ Jacks are actually used as programming ports sometimes. Three, but even two dropping full-duplex, wires including GND are quite enough for bidirectional communications. Icom radio transceivers have been using this for decades. \$\endgroup\$
    – carloc
    Sep 17, 2017 at 7:45

1 Answer 1


There are several issues with switched connectors in an ISP app:

Too few pins!

Switched connectors are only available in configurations with a few (2, 3, maybe 4) contacts -- many in-circuit programming interfaces need more connections than that.

Too big!

Switched connectors are usually of the audio-jack style, and many of the more exotic configurations require a chunkier connector body, typically only seen in 6.35mm, which makes the connector quite large and costly, compared to more typical ISP connectors.

Too bouncy!

Like any other switch, the switches in a switched connector contact bounce -- this could throw off the circuit if a "live" connection or disconnection was made. While the connector contacts experience contact bounce as well -- the programmer can work around this by tristating its side of the connection, allowing hotplugging of the programming connector safely, something that'd go away in your proposal.

  • \$\begingroup\$ I imagine if there was sufficient cause, connectors could be made that get around at least the first two of these problems, though switch bounce is always going to be a concern. At least unless you use mercury-wetted switches, which have their own much more serious problems. \$\endgroup\$
    – Hearth
    Feb 27, 2021 at 16:37

Your Answer

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

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