2
\$\begingroup\$

I'm building a eurorack synth module, and I'm about to do the final tests on the design to make sure that the module cannot be destroyed (electronically, using what's available in the rack, so protection against sledgehammers etc is not relevant here) by the user.

Since the system power rails are \$\pm12\mathrm{V}\$, I've taken as my guideline that all front panel inputs and outputs should be able to withstand an indefinite short to either power rail, or anything in between. I believe, but have not yet tested, that my current design should withstand any of the signal voltages shorting to the power rails. All connections are on TS or TRS minijacks.

To test this, I've built a "module" which has outputs with power rails connected directly to the signal pin, i.e. I can get either the \$12\mathrm{V}\$ or \$-12\mathrm{V}\$ rail connected to a miniplug lead, which I'd then stick in to the various inputs to make sure nothing gets destroyed.

However, I realized there's a snag: since the TS (and TRS) -jacks momentarily short the signal pin to ground when inserting, this means that when sticking my test lead in the the jack, I will be shorting the system power rail to ground. I'd imagine this can't be good for my module, which has sensitive parts such as a microcontroller. Also, I'm not sure if the other modules or the PSU can withstand the short.

Question: I see a few ways forward:

  1. I simply plug the test lead to each jack when powered off, and then see that the system survives powering on. This of course includes the implicit assumption that the user also does not have actual access to the power rails (which no sane module should provide, but people do DIY etc etc).
  2. I conclude that (unless I make all inputs differential) I cannot protect against the power rails themselves, since the jacks always provide direct connection to the system ground. Therefore I relax my requirements and decide on some finite minimum impedance which I'll use to test the input protection. If so, how do I choose a reasonable impedance?
  3. I stick to my guns, and connect also the ground pins of the jacks via some finite size resistors that will survive the full \$\pm12\mathrm{V}\$. However, this will degrade my signal quality and/or require full differential input stages. I haven't seen any other eurorack module go this far.

Which of these approaches, if any, sound reasonable? Maybe both 1. and 2.?

\$\endgroup\$
  • \$\begingroup\$ Instead of power cycling the system, why not put a switch (on+12/off/on-12) on the test lead that runs to the +/- rails? So: switch OFF position- insert test wire into jack. switch to +12V rail feed, check for smoke. switch to -12V rail feed, check for smoke. switch OFF move test lead to next jack, repeat. I would think it would be faster than cycling the DUT (device under test). Also you can add a simple resistor/LED to the test lead output to make sure it's off when removing/inserting wire into jack. \$\endgroup\$ – isdi Oct 18 '18 at 12:32
  • \$\begingroup\$ @isdi not a bad idea at all, I'll certainly do that for the next revision of the testvoltage -module! Anyway, speed is not that important right now, as I'm just testing the prototype to make sure the design is right. Of course, it might not be bad to test this in production too for each individual module... \$\endgroup\$ – Timo Oct 18 '18 at 12:44
  • \$\begingroup\$ "I will be shorting the system power rail to ground. I'd imagine this can't be good for my module". Isn't this part of your reliability test? Shouldn't it survive a temporary short or abrupt disconnection of power? \$\endgroup\$ – pipe Oct 18 '18 at 14:26
  • \$\begingroup\$ @pipe that's exactly what I did in the end, and the test passed :) The reason I was hesitant was that the PSU is not under my control (i.e. the user buys a separate PSU), but I ended up testing with a low-power PSU. If you make an answer out of your comment, I'll accept it. \$\endgroup\$ – Timo Nov 1 '18 at 9:35

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.