-1
\$\begingroup\$

When I want to protect an electronic board from a magnetic field generated by a transformer, I put a circle of vias (as a jail grid) connected to GND planes to stop the magnetic field.

How does a current transformer work? Normally, a magnetic field should be also blocked with copper wires so it should not work. What is this magic?

enter image description here

\$\endgroup\$
8
  • 1
    \$\begingroup\$ I think you need to explain how you see your circle of vias doing what you say they do. That strikes me as the problem here i.e. you might be thinking that a CT and the vias are related but, it's unlikely that they are. You should also link to an internet reference source for "jail grid". \$\endgroup\$
    – Andy aka
    Dec 20 '21 at 10:49
  • 1
    \$\begingroup\$ It sounds like you should be posting a picture of your PCB. \$\endgroup\$
    – Lundin
    Dec 20 '21 at 10:57
  • \$\begingroup\$ A magnetic field could not be blocked with copper, you need a special ferromagnetic material with high permeability like mu-metall. Read en.wikipedia.org/wiki/Mu-metal \$\endgroup\$
    – Uwe
    Dec 20 '21 at 11:04
  • \$\begingroup\$ MU-METAL ok for the earth's magnetic field. If magnetic field too high, ... would be saturated, so inefficient ... \$\endgroup\$
    – Antonio51
    Dec 20 '21 at 11:26
  • \$\begingroup\$ I will do a draw of my "via grid". Here is a picture of a magnetic field which seems to be blocked with copper (here it's PCB planes) : media.monolithicpower.com/wysiwyg/… So my question can only be : is magnetic field block with copper as shown on this picture ? (I believed it was) \$\endgroup\$
    – zian
    Dec 20 '21 at 11:57
5
\$\begingroup\$

Conductors, like copper, can attenuate AC magnetic fields, which is the field you will get from a transformer. It depends on the configuration of the conductor, and the frequency of the field.

Those CT windings, when short-circuited by the ammeter, certainly do block the magnetic field in the CT core. Compare the CT core flux with the short-circuit load and without.

A circle of vias may or may not attenuate an AC magnetic field, depending on how they are configured.

\$\endgroup\$
4
  • \$\begingroup\$ Sorry, but I don't understand what you mean with "those CT windings, when short-circuited by the ammeter, certainly do block the magnetic field in the CT core. Compare the CT core flux with the short-circuit load and without". If we forget the ammeter and just consider a short-circuit load, why this short-circuit would block the magnetic field in the CT core created with the red wire ? \$\endgroup\$
    – zian
    Jan 8 at 18:30
  • \$\begingroup\$ @zian There are several ways to explain it, if one doesn't work we'll try another. The core flux is due to the sum of all currents linking it. If the ammeter is missing, we only have the current in the big red wire. If the secondary is short-circuited, now a secondary current flows, which mostly (ideally completely) cancels out the magnetic effect of the primary current, to reduce the core flux to near zero. Another is to say flux is proportional to volts per turn on the core. when you s/c the secondary, volts drop to near zero, so the flux drops as well. \$\endgroup\$
    – Neil_UK
    Jan 8 at 19:13
  • \$\begingroup\$ We are here only working in AC currents : so to cancel out the magnetic effet, I understand that it's only due to a de-phasis of the magnetic fields : to cancel 100% of these magnetic fields, dephasis should be precisely 180° ? It is the case here ? \$\endgroup\$
    – zian
    Jan 9 at 10:52
  • \$\begingroup\$ @zian Yes, 180 degrees phase and equal magnitude will give you 100% cancellation. Unequal magnitude, or phase not 180, will result in a residual. In the case of a non-ideal transformer, the sum of primary and secondary currents has to be finite, as this total current is causing the flux in the core. Transformers tend to have high permeability, especially ones designed to be used as current transformers, so the residual current is low. Then when it's used as a current transformer with a short on the output, the voltage and hence flux is low, so the currents are very nearly balanced. \$\endgroup\$
    – Neil_UK
    Jan 9 at 11:11
2
\$\begingroup\$

No magic at all. Current transformers do work well just like other versions of transformers.

Magnetic fields could not be blocked with copper, that is your error. Electric fields may be blocked with copper.

To reduce magnetic fields you need a special ferromagnetic material with high permeability like mu-metall. As Andy aka commented, the field is diverted into the ferromagnetic material.

An electromagnetic field of higher frequency may be shielded by a copper enclosure. The electric component of the field is blocked and the magnetic field component generates eddy currents cancelling the magnetic field locally.

If you want to shield a current transformer, you have to avoid closed loops around the torus core. The shield should not contain short circuited windings on torus core.

\$\endgroup\$
1
  • 4
    \$\begingroup\$ Strictly speaking a ferromagnetic material doesn't block; it diverts i.e. it supplies a preferable path for the field to travel. \$\endgroup\$
    – Andy aka
    Dec 20 '21 at 12:11

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.