Electrical and Calorimetric Power Loss Measurements of Practically Ideal Soft Magnetic Cores

In this paper, we demonstrate two methods for measuring power losses of soft magnetic cores made of nanocrystalline
$({\rm Fe}_{100-a-b}{\rm Co}_{a}{\rm Ni}_{b})_{96-y-z}{\rm Cu}_{1}{\rm Nb}_{3}{\rm Si}_{y}{\rm B}_{z}$
alloys at the magnetic induction level
${\rm B}_{\rm max}=300~{\rm mT}$
and at the measurement frequency
$f=100~{\rm kHz}$
. We investigated the possible sources and consequences of measurement errors for the both electronic and simplified calorimetric methods. While the electrical core loss measurement showed more than a factor of 2 higher value for
$\mu_{r}\sim 300$
sample than for
$\mu_{r}\sim 2600$
, the calorimetric method proved that it is not the case; it is only about 30% higher. This result is interpreted with significantly higher measurement uncertainty of the electrical measurement when the phase angle
$\phi$
is very close to 90
$^{\circ}$
, i.e., when the magnetic core behaves like an almost ideal inductor. The calorimetric measurement error is also gradually increasing as
$\mu_{r}$
is lowered, due to the increasing fraction of the copper wire's Joule heating. The measured data are consistent with the eddy current loss component. It has been demonstrated that the tape thickness plays the most important role for the power loss of these soft magnetic cores.