Curtis 1212S Manual,
Rev. B
37
Radiated paths are created when the controller circuitry is immersed in
an external field. This coupling can be reduced by placing the controller as far
as possible from the noise source or by enclosing the controller in a metal box.
Some Curtis controllers are enclosed by a heatsink that also provides shielding
around the controller circuitry, while others are partially shielded or unshielded.
In some applications, the vehicle designer will need to mount the controller
within a shielded box on the end product. The box can be constructed of just
about any metal, although steel and aluminum are most commonly used.
Most coated plastics do not provide good shielding because the coatings
are not true metals, but rather a mixture of small metal particles in a non-con-
ductive binder. These relatively isolated particles may appear to be good based
on a dc resistance measurement but do not provide adequate electron mobility
to yield good shielding effectiveness. Electroless plating of plastic will yield a
true metal and can thus be effective as an RF shield, but it is usually more
expensive than the coatings.
A contiguous metal enclosure without any holes or seams, known as a
Faraday cage, provides the best shielding for the given material and frequency.
When a hole or holes are added, RF currents flowing on the outside surface of
the shield must take a longer path to get around the hole than if the surface
was contiguous. As more “bending” is required of these currents, more energy
is coupled to the inside surface, and thus the shielding effectiveness is reduced.
The reduction in shielding is a function of the longest linear dimension of a
hole rather than the area. This concept is often applied where ventilation is
necessary, in which case many small holes are preferable to a few larger ones.
Applying this same concept to seams or joints between adjacent pieces or
segments of a shielded enclosure, it is important to minimize the open length
of these seams. Seam length is the distance between points where good ohmic
contact is made. This contact can be provided by solder, welds, or pressure
contact. If pressure contact is used, attention must be paid to the corrosion
characteristics of the shield material and any corrosion-resistant processes applied
to the base material. If the ohmic contact itself is not continuous, the shielding
effectiveness can be maximized by making the joints between adjacent pieces
overlapping rather than abutted.
The shielding effectiveness of an enclosure is further reduced when a wire
passes through a hole in the enclosure; RF energy on the wire from an external
field is re-radiated into the interior of the enclosure. This coupling mechanism
can be reduced by filtering the wire where it passes through the shield boundary.
Given the safety considerations involved in connecting electrical components
to the chassis or frame in battery powered vehicles, such filtering will usually
consist of a series inductor (or ferrite bead) rather than a shunt capacitor. If a
capacitor is used, it must have a voltage rating and leakage characteristics that
will allow the end product to meet applicable safety regulations.
The B+ (and B-, if applicable) wires that supply power to a control panel
should be bundled with the other control wires to the panel so that all these
wires are routed together. If the wires to the control panel are routed separately,
a larger loop area is formed. Larger loop areas produce more efficient antennas
which will result in decreased immunity performance.
Keep all low power I/O separate from the motor and battery leads. When
this is not possible, cross them at right angles.
APPENDIX A: EMC DESIGN CONSIDERATIONS
