B-field Exposure From Induction Cooking Appliances 2
1 Executive Summary
Induction cooking uses the fact that alternating magnetic fields generate heat in the ferromag-
netic cooking vessel due to magnetic hysteresis and induced eddy currents. Safety concerns
about exposure to magnetic stray-fields have arisen. Studies by Suzuki [1] and Yamazaki [3]
have assessed these magnetic fields with contradicting findings.
The objective of this study was to assess the maximum exposure that arises during use of
induction cooking devices. Three devices currently available on the Swiss market were selected:
the built-in appliances 1 and 2 (Electrolux GK58TCi and Gaggenau CI 261 110) and the portable
appliance 3 (Inducs SH/BA 5000). The appliances were mounted on wooden supports allowing
measurement in close proximity to the hobs without disturbing the B-field.
Fifteen pots and pans of different sizes and shapes, as well as of various materials were
evaluated in single and multi-hob use in order to select a worst-case set of pots corresponding
to the worst-case B-field exposure. A standard set of pots was also defined according to the
European Norm EN50366. The Narda probe ELT-400 specifications were validated, and the
probe was used for the characterization of the induction cookers in the time- and frequency-
domains (cooking signal frequency, dependency on the heat setting, etc.).
In the first step, the spatial B-field exposure was evaluated according to EN50366 [4] (i.e.,
at a measuring distance of 30 cm using the standard set of pots). All three appliances met the
compliance criteria of ICNIRP [2] for incident B-fields by a margin larger than 14 dB (see Table
11).
In the second step, the worst-case exposure was evaluated as a function of pot and heating
configurations. It was demonstrated that different pot and and heating configurations can
result in exposures that well exceed +10 dB of the standard EN50366 configurations at the same
distance (see Figure 55). In addition, the field distribution has a strong negative gradient in the
direction of larger distances. Therefore, DASY4 was enhanced to enable 3D field scanning using
the NARDA probe. At the very short distance of 1 cm, the fields can be more than 30 dB larger
than at 30 cm (see Table 14). Combining the results from worst-case configurations and short
distance measurements, the standard EN50366 values can be exceeded by 37 dB (see Figure 55).
The uncertainty of the evaluation was determined to be 1.5dB (k=2).
The third task was to evaluate the findings with respect to compliance testing. Assuming
the distribution of Appliance 3, exposure close to the appliance could be as much as 37 dB or
a factor of 70 above the ICNIRP safety limits. A very simple approximation suggest that the
induced currents for such a worst-case compliant appliance would exceed the basic restrictions
by nearly a factor of 10. In other words, the current standard EN50366 for compliance testing
does not prevent exposures far above the basic restrictions and therefore needs revisions.
To obtain the scientific basis for a sound and reliable compliance test procedure, systematic
evaluations of induced currents as a function of human anatomy and field distributions are
necessary and recommended.
