0
\$\begingroup\$

It's my understanding that a load which has a SMPS can put noise back onto the power cord, which could reach another component that's plugged into the same outlet.

Is it true that a 1:1 transformer can prevent SMPS noise of one load from polluting the power that goes to other loads?

If so, must that transformer have a floating-neutral secondary or can the secondary be grounded (to block this "backwash" noise) and must it be positioned between the mains and the offending load that has a SMPS or should it be positioned between the mains and the other load that you are trying to protect?

Thanks!

\$\endgroup\$
  • \$\begingroup\$ Maybe if you wire it as a common mode choke. In which case, it will not provide galvanic isolation. This can be considered a crazy idea for now, unless someone chimes in to say that they have done it. \$\endgroup\$ – mkeith Aug 18 '17 at 1:51
6
\$\begingroup\$

It's my understanding that a load which has a SMPS can put noise back onto the power cord, which could reach another component that's plugged into the same outlet.

Yes, that is true, but if the device conforms to current EMi legislation, this noise will be quite low. If it's a chinese $1 USB charger, maybe not. Never use a chinese $1 USB charger while in your bathtub, by the way. I've found this helpful to stay alive.

Is it true that a 1:1 transformer can prevent SMPS noise of one load from polluting the power that goes to other loads?

A transformer has a lowpass effect due to its inductance, so it will filter out some of the HF noise, if it is differential. As for common mode noise, that would depend on transformer interwinding capacitance.

So you'd want a rather bad transformer, with lots of leakage inductance, and low interwinding capacitance (ie, not a toroid).

As Tony Stewart. EE since '75 hints, a transformer is an expensive, clumsy, and ineffective way to do this. A much better way would be to:

  • Increase common mode impedance, with a common mode choke
  • Filter out differential noise with a LC filter

This is basically what a IEC mains filter does.

enter image description here

But the best is to increase noise rejection in the sensitive equipment.

\$\endgroup\$
1
\$\begingroup\$

No , it does not reduce conducted noise. (significantly, except for RF Band)

For this, you need a CM choke with X1 & Y2 caps ...aka "line filter such as below.

enter image description here

\$\endgroup\$
0
\$\begingroup\$

Yes, a transformer is both lossy and limited (by skin-depth of the iron parts) in high frequency response, so can eliminate a conducted high frequency signal. Some isolation transformers have VERY low coupling capacitance for the windings, and any capacitive coupling of RF current is significantly reduced.

The usual commercially supplied switching power supplies, though, don't require such an expensive solution. One typically sees ultraisolation transformers only in a belt-and-suspenders approach to minimizing interference during experimentation, not in a complete designed product. A modern SMPS, well shielded and carefully designed and tested, needs no accessory hunk of iron.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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