# Does operating an iron core near its saturation field reduce performance?

I am sizing an iron core for an electromagnet, and I am uncertain if there are disadvantages to operating close to the saturation flux density. If the saturation field is X, would it be better (more efficient / less heat / other parameters that increase performance) to operate significantly below that, say at 0.5X, or does it not really matter? Can I comfortably size the system to operate at 0.9X?

• Is it a one-off or mass production? Sep 15 at 19:39
• One-off, though I would be happy to hear the differences for both cases
Sep 15 at 19:42
• @Arad Understood. Less efficient. If you look at the B-H curve you can see it shallow out as it approaches saturation which means you are getting incrementally less out for what you put in, magnetism wise. Sep 15 at 19:48
• Given that an EM has a gap, have you calculated what the lowest value of reluctance you will ever see? I say this because it's really hard to saturate a core with a significant air-gap. Sep 15 at 19:50
• @DKNguyen Thanks, it makes sense.
Sep 15 at 19:50

Less efficient. If you look at the B-H curve you can see it shallow out as it approaches saturation which means you are getting incrementally less out for what you put in, magnetism wise.

Note that saturation flux density is simply defined at some magnetization for the given material, and you may get more or less simply by operating at a different level. For example if it's rated at 10kA/m, that's pretty intense magnetization, and you might not do that in sustained operation (air cooling) anyway. Conversely, you can get more (indeed, arbitrarily more) flux density at higher magnetization, but it's probably not worth it (unless the test value was rather low; which might be the case for very high μ materials I suppose?).

Note also that flux is proportional to magnetization and permeability: $$\B = \mu H\$$. If effective permeability $$\\mu_\textrm{eff} = \frac{\mu_r l_e}{l_e + l_g \mu_r}\$$ is dominated by air gap (because the pole pieces don't quite mate flat, or have to hold gap against a retaining force, etc.), then material properties don't matter until such point where saturation reduces $$\\mu_r\$$ enough that $$\\mu_\textrm{eff}\$$ suffers. Which doesn't mean that magnetization should be raised to such a point: just that the limit is independent of core properties until such a point. So the technical answer would be "no", because efficiency is low independent of core performance, above some minimum value. The limiting factor, therefore, would depend only on resistive losses, and whatever $$\\mu_\textrm{eff}\$$ is.

If efficiency (idle power) is a huge priority, you might consider a mechanical solution instead. Magnetism by itself doesn't consume power; the only reason we dissipate power is out of convenience, because electromagnets can't be made from superconductors.* Use a permanent magnet for the holding force, and switch it either by sliding pole pieces out of the way (magnetic chucks for machine tools operate this way), or by momentarily demagnetizing it with coils so the pole pieces can be moved away before snapping back together.

*For just casual everyday use. Even when something's big or important enough to be worth using superconductors, the AC losses during magnetization may increase cooling costs more than is worth; careful design is required. Offhand, I don't know of any superconducting magnetic cranes for example, but I suppose one could be made.

• Thank you! So if I understood you correctly, in a system with an airgap it doesn't matter, because it would be really hard to saturate the iron core in such case anyway. Conversely, if there is no airgap (which is the case in my system), material properties matter and I should be cautious about approaching the saturation point. Thank you for the advice about the permanent magnet, I'll consider that.