The invention relates to an assembly and a method for determining the angular position of a rotating object, for example.
The angular position for example of a rotating machine is often recorded by using a sensor, which by means of inductive coupling forms, from a temporally varying periodic input value, two output values dependent on the angular position of the for example rotating electrical machine, the amplitudes of the time-periodic output values depending in different ways on the angular position to be recorded. Preferred technical embodiments of such sensors are for example resolvers or variable-reluctance (VR) sensors, in which an excitation usually occurs by means of a temporally sinusoidal electrical signal, and the amplitudes of the two likewise temporally sinusoidal output signals are modulated, dependent on the angular position to be recorded, on the one hand sinusoidally for example, and on the other hand cosinoidally (as for example for resolvers with secondary windings arranged turned through 90 degrees to each other).
The sinusoidal voltage on the first secondary winding is calculated here as
Us(t)=a·U0−sin(we·t+β)·sin(α), and
the cosinusoidal voltage on the second secondary winding is calculated as
Uc(t)=a·U0·sin(we·t+β)·cos(α), where
we is the exciter circuit frequency, t the time, U0 the excitation amplitude, α the rotation angle of the machine, β the phase angle between excitation and the taps on the secondary coils, and a the transfer factor of the amplitude between the primary winding and the secondary windings.
By evaluation of the sensor voltages, the required angle α is determined from the values US(t) and UC(t) measured on the secondary windings. Here:
US(t)/UC(t)=sin(α)/cos(α)=tan(α),
which yields the current rotation angle α. A disadvantage here is that this approach only leads to suitable results if sin (we·t+β) is not close to the zero crossing. If sin (we·t+β) is close to the zero crossing, then Us(t) and Uc(t) take on very small values or tend to zero, and a suitable evaluation is no longer possible for this angular position.
One solution possibility according to the present state of the art is to choose a sampling time for the two output signals of the secondary windings such that sin (we·t+β) takes on a sufficiently large value. A consequent disadvantage is that the time for determining the angular position can therefore no longer be freely selected, but must be executed in a particular phase relationship (and hence in a particular time relation) to the excitation voltage, in order to obtain a sufficient signal amplitude for the evaluation.
A further solution possibility according to the present state of the art is to perform a calculation of the current angle, from values determined beforehand for US(t) and UC(t), in an assembly preferably implemented as an integrated circuit, by means of phase-locked loops (PLL). A consequent disadvantage here is that so-called lag errors arise in PLL circuits with changing angular velocities. If angular accelerations are too great, the PLL no longer runs synchronously, and a suitable signal for determining an angle can no longer be found.
It is the object of the present invention to specify an assembly and a method for determining the angular position, in which assembly and method the aforementioned disadvantages are avoided.
The object is achieved with an assembly for determining a current angular position of an object rotating around a stationary point. The assembly includes an inductive sensor with at least one primary winding being fed an excitation signal and at least two secondary windings inductively coupled thereto. The inductive coupling causes signals to be generated with a periodic and phase-shifted waveform in the two secondary windings depending on a rotation angle. The at least two secondary windings include a first secondary winding and a second secondary winding. The assembly further has three phase shifters each with one input and one output, and the phase shifters have the same phase shifts. The phase shifters each are electrically coupled on the input with the primary winding, the first secondary winding or the second secondary winding respectively. An evaluation unit is provided for determining the current angular position from the signals at the inputs and at the outputs of the phase shifters.
The object is achieved in particular by an assembly and a method for determining the angular position by means of an inductive sensor, in which assembly and method three further signals are formed from the excitation signal of the primary winding of the sensor and the signals tapped in its secondary windings by phase shifter units, and are supplied after a sampling to an evaluation circuit which has for example polarity sign determination units. For example, in the evaluation circuit, the current angular position of the rotating machine can be determined at any time, for example using an appropriate calculation rule, from the now six available values, without the aforementioned disadvantages appearing.
The invention will now be described in detail with reference to the embodiments represented in the figures of the drawings, in which
The assembly further possesses a generator unit 5 for excitation of the primary winding 2 with an excitation voltage 17 which is generally temporally sinusoidal. The assembly further possesses three subtractor units 6, 7 and 8, three phase shifters 9, 10 and 11, two polarity sign determination units 12 and 13, a sample and hold unit 14 provided for signal sampling, and a processor 15.
As shown in
A sinusoidal voltage 17 fed from the generator unit 5 into the primary winding 2 generates via electromagnetic coupling a floating sinusoidal voltage 18 in the secondary winding 3 and a floating sinusoidal voltage 19 in the secondary winding 4, the amplitude of 18 being modulated as a function of the rotation angle according to a sine function and the amplitude of 19 being modulated as a function of the rotation angle according to a cosine function, i.e. according to the terminal relationships on a resolver or a variable reluctance sensor respectively. By means of the subtractor unit 6, the sinusoidal excitation voltage 17 is converted into a sinusoidal excitation voltage 20 referred to ground. By means of the subtractor unit 7, the temporally sinusoidal voltage 19 of the secondary winding 3 is converted into a temporally sinusoidal voltage 21 referred to ground of the assembly.
By means of the subtractor unit 8, the temporally sinusoidal voltage 20 of the secondary winding 4 is converted into a temporally sinusoidal voltage 22 referred to ground.
Furthermore, the temporally sinusoidal excitation voltage 20 referred to ground is shifted in phase by the phase shifter 9 through 90 degrees, the temporally sinusoidal signal 21 is shifted in phase by the phase shifter 10 through 90 degrees, and the temporally sinusoidal signal 22 is shifted in phase by the phase shifter 11 through 90 degrees. In the polarity sign determination unit 12, a signal is generated with value 1 when the excitation signal 20 is greater than zero, otherwise a signal with value 0 is generated. In the polarity sign determination unit 13, a signal is generated with value 1 when the excitation signal of the phase shifter 9, referred to ground and shifted in phase through 90 degrees, is greater than zero, otherwise a signal with value 0 is generated.
In the sample and hold sampling unit 14, samples of the signals supplied from the subtractor units 7 and 8, the phase shifters 10 and 11 and the polarity sign determination units 12 and 13 are formed at the time of receiving a trigger signal 16 from the processor 15. These six signals are supplied for further processing to the processor 15 as a computing device, which calculates therefrom the current angular position of the rotating machine at the time of receipt of the trigger signal 16 by the sample and hold unit 14. The current angular position is for example calculated as
α=arctan(ab1/ab2)+π·sign(ab2)·VZ_EXITS,
a and 2b show as an example a flow diagram of the method for determining the current angular position of a rotating machine. In a step 1, a primary winding of a resolver is energized with a temporally sinusoidal excitation signal. In a step 2, this excitation signal is converted by means of a first subtractor unit into an excitation signal referred to ground. In a step 3, a temporally sinusoidal floating signal induced in a first secondary coil of the resolver is converted by means of a second subtractor unit into a temporally sinusoidal signal referred to ground. In a step 4, a temporally sinusoidal floating signal induced in a second secondary coil of the resolver is converted by means of a third subtractor unit into a temporally sinusoidal signal referred to ground.
In a step 5 of the method, the excitation signal referred to ground is shifted in phase through 90 degrees by means of a first phase shifter. In a step 6, the temporally sinusoidal signal referred to ground is shifted in phase through 90 degrees by means of a second phase shifter. The temporally sinusoidal signal referred to ground is shifted in phase through 90 degrees by means of a third phase shifter in a step 7. By means of a first polarity sign determination unit, in step 8 a signal with the value 1 is formed if the excitation signal referred to ground is greater than zero, otherwise a signal with value 0 is generated. In a step 9, by means of a second polarity sign determination unit, a signal is generated with value 1 when the excitation signal, referred to ground and shifted in phase through 90 degrees, is greater than zero, otherwise a signal with value 0 is generated.
In a step 10 of the method, sample values of the signals formed from the second and third subtractor units, the second and third phase shifters and the first and second polarity sign determination units are formed at the time of the trigger signal by a sample and hold unit on receipt of a trigger signal, the point in time of which is freely selectable. In a step 11, the sample values formed by the sample and hold sampling unit are received by a processor, and the current angular position of the rotating machine is calculated from these sample values according to the mathematical procedure described above.
In the present method described as an example, the order shown for the steps is not mandatory, and can be varied as desired. In the previously explained example of an assembly according to the invention, the subtractor units 6, 7, and 8 can also be omitted, for example, if a floating measured value acquisition is not required.
Number | Date | Country | Kind |
---|---|---|---|
10 2009 032 095 | Jul 2009 | DE | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/EP2010/059689 | 7/7/2010 | WO | 00 | 9/7/2011 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO2011/003928 | 1/13/2011 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
2496920 | Seeley | Feb 1950 | A |
5646496 | Woodland et al. | Jul 1997 | A |
5796357 | Kushihara | Aug 1998 | A |
6191550 | Yoshihara | Feb 2001 | B1 |
6278388 | Kushihara | Aug 2001 | B1 |
6426712 | Hare et al. | Jul 2002 | B1 |
20020175645 | Fujimoto et al. | Nov 2002 | A1 |
Number | Date | Country |
---|---|---|
29519716 | Jan 1996 | DE |
10162196 | Dec 2002 | DE |
0779607 | Jun 1997 | EP |
2448350 | Oct 2008 | GB |
Entry |
---|
Gasking, “Resolver-to-digital Conversion—A Simple and Cost Effective Alternative to Optical Shaft Encoders”, 2002-2006, Analog Devices, Application Note AN-263 pp. 1-4; http://www.analog.com/static/imported-files/application—notes/394309286AN263.pdf. |
Number | Date | Country | |
---|---|---|---|
20120209562 A1 | Aug 2012 | US |