Sometimes in app notes i see something like @assume that magnetizing current value is 10% of the primary current". Why? How to choose the magnetizing current value the right way?


Magnetising current will be practically irrelevant in a good mains transformer design. As it's in quadrature to the load current, a mag current of 10% of the full load will increase the full load current of the transformer by 1% (Pythagorus). 10% is a very conservative estimate, it will typically be smaller.

Magnetising current is sensitive to the transformer inductance. Transformer iron varies in relative permeability from a few thousand to many thousands, depending on price, and strain history. If a transformer is built from Es and Is, or two C cores, variations in size of the inevitable airgap could cause a 10:1 change in inductance, together with a similar change in magnetising current. So magnetising current is measured on a transformer as an assembly quality test.

There are mains frequency transformers which have a different typical ratio of magnetising current to full load. Current ratio transformers will use a special high permeability core to reduce mag current even further, as it represents an error term in the current ratio. Microwave oven transformers will have a mag current in the same order as the full load current, perhaps only 1/3 or even 1/2, as they are built down to a price, run hard, and fan cooled.

Ferrite core transformers will tend to have higher relative magnetising current than iron transformers, due to the lower permeability.

| improve this answer | |
  • \$\begingroup\$ is it ok if i choose 10%? It is 2-switch forward smps transformer working on 84khz \$\endgroup\$ – silkyre6xtenz Mar 16 '18 at 19:48
  • \$\begingroup\$ Nicely written. I also learned about using magnetizing (British spell things so wrong!) current as a quality test and that makes so much sense to me. Thanks. \$\endgroup\$ – jonk Mar 16 '18 at 19:48
  • \$\begingroup\$ @silkyre6xtenz If I was designing a ferrite cored transformer system, I would not choose 10%, I'd do the sums. If you run at relatively high field, so accepting large core losses for perhaps limited duty cycle or low copper losses, your mag current could easily be much higher than 10%, and worth designing for properly. But it's up to you. 10% makes a good 'back of the envelope' figure for gross feasibility. \$\endgroup\$ – Neil_UK Mar 16 '18 at 20:33

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.