The invention relates to the field of integrated current sensors having a potential separation.
Potential separated current sensors are widespread in power electronics, medical engineering, etc. Considering multiple-purpose current sensors, it is desirable, that current can be measured in a broad frequency range including direct current (DC). Hall sensors, which are arranged in a gap of a toroidal magnetic core, are often used as current sensors in a frequency range from 0 to approximately 50 kHz. Such current sensors often require magnetic field concentrators that are arranged in a voluminous package together with the current sensor. Such sensor arrangements are expensive and cumbersome to manufacture, so that there is a need for potential separated current sensors which on the one hand are small in their geometric dimensions and inexpensive to manufacture, and on the other hand allow for a current measurement in a very broad frequency range.
One aspect of the invention relates to a measurement circuit for a potential separated current measurement that includes: a primary winding for carrying a primary current generating a magnetic field, a secondary winding that is magnetically coupled with the primary winding and provides a first signal dependent on the primary current, a Hall sensor that is arranged such that it is pervaded by the magnetic field of the primary current for providing a second signal depending on the primary current, and an evaluation circuit for determining a measurement signal representing the primary current from a first signal and a second signal.
Another aspect of the invention relates to a sensor arrangement that includes a substrate, at least one Hall element integrated in or arranged on the substrate, a first coil which is spaced apart from a surface of the substrate in a vertical direction, a second coil that is spaced apart in a vertical direction from the first coil, and an isolation layer that is arranged between the first coil and the second coil.
According to one example of the invention, the aforementioned evaluation circuit comprises a low-pass receiving the first signal and providing a third signal, the third signal being a low-pass filtered version of the first signal, and an adder for a weighted addition of the second and the third signal, the adder providing a measurement signal dependent on the weighted sum.
The invention can be better understood with reference to the following drawings and description. The components in the figures are not necessarily to scale, instead emphasis being placed upon illustrating the principles of the invention. Moreover, in the figures, like reference numerals designate corresponding parts. In the drawings:
Hall sensors can achieve very good properties in terms of offset and temperature stability, if they are connected to circuits that are suitable for compensating offset and temperature drifts. However, the bandwidth of Hall elements is too low for many applications. The behavior of Hall sensors can be described approximately by a transfer function of a first-order low-pass. The combination of Hall elements with a planar transformer having planar primary and secondary windings allows for the construction of a novel current sensor, which on the one hand has the desirable DC-properties of a Hall sensor, and on the other hand has a cut of frequency much higher than common current sensors using Hall elements. Furthermore planar coreless transformer windings can be integrated on a semiconductor chip.
UT=M·di
In the Laplace-domain (i.e., in the frequency domain) the above equation (1) yields:
UT(s)=s·M·iP(s) (2)
The reference symbol M denotes the mutual inductance of the primary winding 10 and the secondary winding 20, the expression diP/dt the first derivation of the measurement current IP with respect to time t, UT(s) the Laplace transformed first signal UT, IP(s) the Laplace transformed measurement current IP. The variable s denotes the generally complex Laplace-variable.
The secondary winding 20 provides—according to Faraday's Law—as a first signal UT a voltage which is proportional to the first derivation of the measurement current through the primary winding, wherein the constant of proportionality corresponds to the mutual inductance M. This differentiating behavior is expressed in the model of
The secondary winding 20 is connected to a low-pass 25, which receives the first signal UT. The output signal of the low-pass 25 is denoted as “third signal” UT,′ which is a low-pass-filtered version of the first signal UT. The transfer function of the low-pass 25 can be, for example, written as follows:
The symbol UT′(s) denotes the Laplace transformed third signal UT. The equation (3) describes a first-order low-pass having a time constant TTP.
The block diagram further comprises a Hall sensor 30, providing a second signal UH being dependent on the measurement current IP. As already mentioned, Hall sensors can have a rather low bandwidth. However, the behavior of the Hall sensor 30 may be described by a sensor element 30a which supplies a signal proportional to the measurement current IP (primary current) to a low-pass 30b. The output signal of the low-pass 30b then corresponds to the actual output voltage of the Hall sensor 30. In the following this output voltage is denoted as second signal UH. The relation between the second signal UH and the measurement current IP can be written as follows:
The expression UH(s) denotes the Laplace transformed second signal UH (i.e., the output voltage of the Hall sensor 30), the expression KH denotes the Hall constant, i.e., the constant of proportionality between the measurement current IP and the second signal UH in the case of direct current. The transfer function of equation (4) also has a low-pass characteristic, wherein the low-pass 30b of the Hall sensor 30 has a time constant TH. As it will be become apparent below, the time constant TTP of the low-pass 25 should be chosen such that it is approximately equal to the time constant TH of the low-pass 30b of the Hall sensor 30 (TTP=TH).
By means of the weight-factors 26 and 34 and the adder 40 a weighted addition of the second signal UH and the third signal UT′ is performed. The weighted sum is the output signal Uout of the measurement circuit. The weight-factor 26 is in the present example THKADJ/M and the weight-factor 34 is KADJ/KH. The weighted sum UOUT is in the present example defined as follows:
By substituting equations (3) and (4) into equation (5) the following expression can be obtained for the Laplace transformed of the output voltage UOUT(s) for identical time constants TTP=TH:
It can be seen from equation (6), that a frequency independent relation between the output signal UOUT and the measurement current IP can be obtained by combining the second signal UH of the Hall sensor 30 and the third signal UT′ derived from the output signal of the secondary winding 20. The measurement circuit of
The low-pass 25, the weight-factors 26 and 34, as well as the adder 40 are inter alia components of an evaluation circuit 200. The evaluation circuit of
The above described arrangement of
Like the measurement circuit of
A calibration of the measurement circuit may, for example, be performed once after switching on the power supply or can be performed repeatedly, for example, after a defined time span has elapsed since the latest calibration. A predefined current is injected into the further winding 21 by the control circuit 50. The further winding 21 is also arranged on the substrate 100 described above such that the magnetic field of the further winding 21 penetrates the Hall sensors in their sensitive direction. The further winding 21 is also realized as a planar coil and isolated from the other planar coils.
Calibration is performed under the assumption that the measurement current IP through the primary winding 10 equals zero. In this case the output signal UH of the Hall sensor 300 is only dependent on the current flowing trough the further winding 21. This output signal is also provided to the control circuit 50. Since the control circuit 50 “knows” the current through the further winding 21 and the resulting output signal UH of the Hall sensor, the control circuit 50 is able to calculate an offset error and a gain error of the Hall sensor 300. The gain of the Hall sensor is defined as the change of the output voltage (second signal UH) over the measurement current IP. This expression is also referred to as sensitivity.
A calibration can also be done without the further winding 21. In this case the secondary winding 20 is connected to the control circuit 50 during the calibration (cf. dotted line in
The gain and the offset of the Hall sensors can be programmable, i.e., adjustable. The Hall sensor 300 therefore comprises inputs for receiving respective control signals CTG and CTO. The gain and the offset of the Hall sensors can be adjusted according to the control signals CTG and CTO. During a “normal” measurement operation the further winding 21 carries no current and does not influence the measurement operation.
The Hall sensor of
The Hall sensor 300 has a control unit 35 which may comprise an analog-to-digital-converter, a multiplexer which is adapted for combining (multiplexing) the output signals UH1, UH2, UH3, UH4 of the single sensor elements to one signal. The control unit 35 of the Hall sensor can, partially or as a whole, be a component of the control circuit 50. For example, the control circuit 50 may be a microcontroller with an integrated analog-to-digital-converter. In this case, the control unit 35 does not need an extra analog-to-digital-converter.
The control unit 35 of the Hall sensor 300 is, for example, adapted for combining the output signals of the single sensor elements 30 to 33 to a single output signal of the Hall sensor 300 (i.e. the second signal UH). This combining may be a weighted addition to the output signals UH1 to UH4 of the single sensor elements 30 to 33. If four sensor elements 30 to 33 are arranged as depicted in
An additional temperature sensor (not shown) may be connected to the control circuit 50 for measuring a possible change in temperature and for compensating for the resulting drift of the offset and/or of the gain of the single sensor elements 30 to 33 via adjusting the offset and the gain by means of the control signals CTO and CTG. The weighted addition of the adder 40 and the weight factors 25 and 34 can also be performed by the control circuit 50. This may especially be desirable if a microcontroller is employed as control circuit 50. In this case the functional blocks 26, 34 and 40 can be implemented fully digitally in a microcontroller. An external control signal UTRL received by the control circuit 50 may be used to trigger a calibration operation and/or a measurement operation.
Number | Name | Date | Kind |
---|---|---|---|
4200814 | Tanaka et al. | Apr 1980 | A |
4596950 | Lienhard et al. | Jun 1986 | A |
4881031 | Pfisterer et al. | Nov 1989 | A |
5049809 | Wakatsuki et al. | Sep 1991 | A |
5200701 | Siebold et al. | Apr 1993 | A |
5493211 | Baker | Feb 1996 | A |
5831431 | Gottfried-Gottfried et al. | Nov 1998 | A |
6366076 | Karrer et al. | Apr 2002 | B1 |
6429651 | Choi et al. | Aug 2002 | B1 |
6593841 | Mizoguchi et al. | Jul 2003 | B1 |
6690164 | Fedeli et al. | Feb 2004 | B1 |
6861838 | Kawase | Mar 2005 | B2 |
6879145 | Harris | Apr 2005 | B1 |
6927662 | Kahlmann et al. | Aug 2005 | B2 |
7015691 | Kang et al. | Mar 2006 | B2 |
20040056749 | Kahlmann et al. | Mar 2004 | A1 |
20060028313 | Strzalkowski et al. | Feb 2006 | A1 |
Number | Date | Country |
---|---|---|
2621302 | Dec 1976 | DE |
10100282 | Jul 2002 | DE |
Number | Date | Country | |
---|---|---|---|
20090001964 A1 | Jan 2009 | US |