Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

A Differential Measurement Probe with High Common Mode Rejection

Franzén, Björn LU (2018) In CODEN:LUTEDX EIEM01 20182
Industrial Electrical Engineering and Automation
Abstract
Measuring signals with high frequency common mode components are typically prone to measurement errors due to low common mode rejection at high frequencies. The purpose of this master thesis is to design a measurement probe that presents sufficient common mode rejection to measure current on high voltage motor drives directly at the switching stage, before filtering.

Outlined in this work is the design process, on a block diagram and simulation level, culminating in the construction of a prototype. The measurement probe consists of a sensor head and a receiver that are fiber optically isolated from each other. It has a signal chain that splits the signal into a high and a low frequency path. The high frequency path is frequency... (More)
Measuring signals with high frequency common mode components are typically prone to measurement errors due to low common mode rejection at high frequencies. The purpose of this master thesis is to design a measurement probe that presents sufficient common mode rejection to measure current on high voltage motor drives directly at the switching stage, before filtering.

Outlined in this work is the design process, on a block diagram and simulation level, culminating in the construction of a prototype. The measurement probe consists of a sensor head and a receiver that are fiber optically isolated from each other. It has a signal chain that splits the signal into a high and a low frequency path. The high frequency path is frequency modulated and the low frequency path is digital. In the receiver, these two paths are joined before being output as an analog signal.

A signal chain is simulated that shows a DC-100 MHz response with an amplitude accuracy of 2.9 %, dynamic range of 53 dB, dropping to 38 dB at the end of the passband and, a spurious free dynamic range greater than 40 dB. Furthermore, a prototype is constructed but has not been debugged to a degree that permits meaningful measurement results. (Less)
Popular Abstract
When electrical signals are measured there are always errors. One of these errors is related to the difference in voltage between the measurement system and the signal being measured. If the voltage difference is large, care must be taken to limit the amount current that is allowed to flow between the two systems. Typically, this is done by use of an isolation transformer. However, if the voltage difference is also of high frequency, the amount of isolation the transformer can provide is greatly diminished. This is mainly because of how close the windings in the transformer are to each other. Increasing the distance between the windings means that leakage currents are lowered, and thus isolation improved, but also implies a less efficient... (More)
When electrical signals are measured there are always errors. One of these errors is related to the difference in voltage between the measurement system and the signal being measured. If the voltage difference is large, care must be taken to limit the amount current that is allowed to flow between the two systems. Typically, this is done by use of an isolation transformer. However, if the voltage difference is also of high frequency, the amount of isolation the transformer can provide is greatly diminished. This is mainly because of how close the windings in the transformer are to each other. Increasing the distance between the windings means that leakage currents are lowered, and thus isolation improved, but also implies a less efficient transformer. Since these are contradictory requirements, an isolation transformer is limited to moderate frequencies. A way around the problem is to replace the isolation transformer with a fiber optic cable. Then, the separation distance can be made as long as the optical fiber itself.

In this thesis, a prototype is constructed, and the theory used in its construction is derived. The prototype consists of two devices, a sensor head and a receiver. The sensor head is connected to the signal that is to be measured, and the receiver is connected to an external measurement system, for example, an oscilloscope. The two devices are interconnected by optical fibers. By doing so, it is expected that small, high frequency signals can be measured even when the voltage difference between the signal, and the system measuring it, is both large, and of high frequency. (Less)
Please use this url to cite or link to this publication:
author
Franzén, Björn LU
supervisor
organization
alternative title
A Fiber-Optically Isolated Measurement Probe
course
EIEM01 20182
year
type
H3 - Professional qualifications (4 Years - )
subject
keywords
Common mode rejection, Fiber-optic measurement probe, Current Sensing, High voltage measurements, Optical isolation
publication/series
CODEN:LUTEDX
report number
5415
other publication id
LUTEDX/TEIE
language
English
id
8962420
date added to LUP
2019-06-27 12:40:30
date last changed
2019-06-27 12:40:30
@misc{8962420,
  abstract     = {{Measuring signals with high frequency common mode components are typically prone to measurement errors due to low common mode rejection at high frequencies. The purpose of this master thesis is to design a measurement probe that presents sufficient common mode rejection to measure current on high voltage motor drives directly at the switching stage, before filtering. 

Outlined in this work is the design process, on a block diagram and simulation level, culminating in the construction of a prototype. The measurement probe consists of a sensor head and a receiver that are fiber optically isolated from each other. It has a signal chain that splits the signal into a high and a low frequency path. The high frequency path is frequency modulated and the low frequency path is digital. In the receiver, these two paths are joined before being output as an analog signal.

A signal chain is simulated that shows a DC-100 MHz response with an amplitude accuracy of 2.9 %, dynamic range of 53 dB, dropping to 38 dB at the end of the passband and, a spurious free dynamic range greater than 40 dB. Furthermore, a prototype is constructed but has not been debugged to a degree that permits meaningful measurement results.}},
  author       = {{Franzén, Björn}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{CODEN:LUTEDX}},
  title        = {{A Differential Measurement Probe with High Common Mode Rejection}},
  year         = {{2018}},
}