REPOSITORY > RESULTS

Abstract

With the increasing use of electricity powered devices and processes in our daily lives,
the need for the precise measurement of electrical currents across more than 15 decades
from below nA to tens or hundreds of kA has become inevitable both for billing and more
optimal energy distribution purposes as well as for the operation and automatization of
electrical devices.
A particularly noticeable increase in measurement demand can be observed in direct current
(DC) systems due to the massive deployment of renewable energy sources, such as
wind and solar power plants and grid storage systems, and the transition of the automotive
industry to electric mobility. Some of these systems may require high-voltage isolation and
peak currents in excess of kA. The existing standard compact and lower cost current sensing
solutions hardly ever achieve an overall measurement uncertainty below 1%, mainly due
to offsets and hysteresis; their typical bandwidth is in the range of 250 kHz, and they may
also be noisy.
The main goal of this doctoral dissertation was thus to design and characterize a new
method and device for the high-precision wide-bandwidth isolated measurement of direct
and alternating electric currents. A particular focus was on achieving key performance
parameters, such as high accuracy, precision, sensitivity, wide bandwidth, low noise, and
low temperature drift dependency by a simpler construction when compared to the current
best state-of-the-art implementations, thus targeting the use of a single gap-less core,
low power consumption, and small form factor also suitable for typical networked devices
constituting the Internet of Things.
As part of this doctoral dissertation, we first conducted an overview of the widely used
methods and principles of current measurement with their corresponding benefits and
drawbacks, and developed the necessary models of magnetic circuits for the open-source
ngspice simulation environment upgraded into a full co-simulation framework with Verilog
and embedded firmware. This knowledge base and research environment were subsequently
used to develop, simulate, and verify a new basic element for magnetic circuits with a
current controlled variable reluctance (CCVR) used for changing the amplitude or the
direction of the magnetic flux and thus making it measurable. This element consisting
of a gap-less core and a specific winding combined with the control electronic circuitry
constitutes the Platiše Flux Sensor. The new wiring and compact gap-less implementation
deliver low offset and hysteresis, a bandwidth in the MHz range, low power consumption,
and low noise operation.
Finally, we integrated the novel Platiše Flux Sensor in a functional prototype of a
closed-loop zero-flux DC current transducer (CT). This prototype has been thoroughly
tested, validated and characterized in a reference setup. We have shown that a 40 A range
low current transducer based on the novel method achieved an overall superior performance
compared to representative comparable devices based on alternative technologies. More
recent research and developments using very high permeability and mechanically stronger
materials for core extended such performance up to 2000 A.