Author Topic: Common mode supply noise in ungrounded modules, does it exist?  (Read 148 times)

0 Members and 1 Guest are viewing this topic.

Offline huub8

  • Contributor
  • Posts: 11
  • Country: nl
Hi all,

I'm currently designing a circuit/module that will be powered from long leads (several meters). In the future, there may potentially be some daisy-chaining going between the modules too. The system will be powered from between 5 and 9 volts DC. So effectively, the power leads could act as an antenna.

Apart from differential noise, I am now wondering if I should also be worrying about any common-mode noise due to the long supply wires. The modules themselves are only connected to the + and gnd of the (unshielded) supply wires. The modules are encased in plastic, so I do not see how there would be any reference to earth. Am I correct in thinking that even though the supply wires are reasonably long, the lack of any shielded encasing means that common-mode noise would have no return path to the source and thus would not be superimposed on the dc of the supply wires?

If not, what kind of frequencies would one expect to mainly have to filter out?

Offline TheUnnamedNewbie

  • Frequent Contributor
  • **
  • Posts: 937
  • Country: 00
Re: Common mode supply noise in ungrounded modules, does it exist?
« Reply #1 on: May 13, 2020, 10:02:52 am »
You always have a return path to ground, in the form of capacitance.
The best part about magic is when it stops being magic and becomes science instead

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 14991
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Common mode supply noise in ungrounded modules, does it exist?
« Reply #2 on: May 13, 2020, 04:15:04 pm »
Offensive frequencies are on the order of the electrical length of the cabling.  Typically 30MHz+, but long cables can be problematic much lower.  For mains, telecom and so on, they use conducted emissions more as a convenience I assume (otherwise, you'd need to do a more sensitive, open-air test, with very long cables and antennas).

Yes, any induced RF between grounds going in and out of a module, is a problem.  Or more generally, the common mode, taking the average over all signals in the cable, weighted appropriately of course (since signals inside a bundle, or inside a shield, will count less).

Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
The following users thanked this post: huub8

Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo