SU1205237A1 - Variable reluctance pickup of turn angle - Google Patents

Description

The invention relates to automation and can be used in systems for automatically regulating and converting information, in particular in converters of the angle of rotation of the shaft into a digital code.

The aim of the invention is to improve the accuracy of converting the angle of rotation into an electrical signal.

Fig, 1 shows a sensor, a longitudinal section; in fig. 2 shows section A-A in FIG. one; in fig. 3 is an electrical diagram of the induction sensor in FIG. 4 is a block diagram of a conversion unit I in FIG. 5 - output characteristics of the induction sensor.

The induction sensor contains a common rotor 1 and two stator 2 and 3, installed in the housing 4 with a relative turn relative to each other (see Fig. 1). The field windings and the output windings are lumped and arranged mutually perpendicularly in the four slots of each stator. Field windings 5 and 6 (Fig. 3) are connected in parallel and connected to an alternating voltage source 7 with an output voltage.

and,

ipch

and sin U t.

On the output windings of the sensor 8 and 9, signals are generated

and

4oi

0 (

and sincotf (-by)

 .and.

sin s tf (otAix

2

TO.

where K, is the coefficient taking into account the temperature errors (of the sensor itself and the source of alternating voltage) 5

f and f are functional dependences of the sensor output voltages on the angle of rotation of the rotor 1.

Since these dependences are determined only by the geometry of the magnetic conductors, f, fg f can be accepted with a high degree of accuracy. The signals UQ and Uo2 from the output windings 8 and 9 of the sensor are fed to a conversion unit 10, which, first, determines the separation of the sum of voltages

UQ, and Ugj to their difference in digital form

{

i ". , /

ot + -5-j + (,

iiA T)

(2)

L,, 1 uotN, / s.

 Pt | -N -t-)

and, secondly, multiplies the obtained result N. |, () by the correction factor Kp (N):

N (cu) Kp (N) Np ("t),

where N (ot) is the output n-bit

conversion block binary code 10.

From expression (2) it can be seen that the value of Np (",) does not depend on the frequency with t, nor on the amplitude U of the output voltage of the alternating voltage source 7, nor on the coefficient of temperature error. For example, with a linear dependence of the voltage of the output windings on the angle of rotation of the rotor 1, i.e. with Eq. (2) follows: N (about) with sinusoidal N. (about) ctg-tgei, with

cosine N. (ei) ctg- ctgcu.

The considered partial cases confirm the conclusion that the code value NP (o () obtained as a result of the transformation by formula (2) is determined only by the initial turn angle A od of stators 2 and 3 and the turn angle l, and Over a wide range, vary the steepness of the output characteristic Np (oc). Therefore, the dependence N. (oi) is a stabilized (reduced) characteristic of an induction sensor,

Since, with a simple geometry of the magnetic cores and concentrated sensor windings, the high linearity of the output voltage from the rotation angle cannot be obtained, to reduce errors, the resulting value N (, (06) is multiplied by the correction (calibration) coefficient Kp (M), which is experimentally determined, and recorded into a persistent storage device.

Let us consider in more detail the operation of the conversion unit 10 (Fig. 4).

The conversion unit contains two .transformers 11 and 12, the primary windings of which are connected to the output windings of the sensor, each having two secondary windings connected in series to receive the sum signal and UQI + Ug2 and counter to receive the difference signal U

and

och

- Ug-, phase sensitive

rectifiers 13 and 14 for demodulating the voltages U and Uj, respectively:

and Sign (ii) fU)) -f (- -) 1

2. J.

and, K, u, fU.) - fU-f :).

The rectifier rectified by the 15 output voltage U of the phase-sensitive rectifier 13 is fed to one input of the comparator 16:

and, | of1 f (i +) + fU), (3)

ysb g

The other input of which receives the voltage from the output of the phase-sensing rectifier 14 via a controllable divider 17 and a controllable multiplier 18:

Ub

N (c6)

 () g

Kp (N)

(four)

The controlled divider 17 (its transmission coefficient) is controlled by the output code of the persistent storage device 19, which is connected to the output buses of the reversible counter 20 and to the sign bit formed by the null authority 21, i.e. divider 17 serves to introduce a correction (calibrated) coefficient K. (N). The controlled multiplier 18, together with the reversible counter 20 (the counting input of which is connected to the master oscillator 22, and the input of the counting direction 41 to the output of the comparator 16) and the comparator 16 is an analog-to-digital balance circuit (analog-digital converter) at

steady-state voltage on both inputs of the comparator 16 are equal to within the price of a junior

Yes,

i.e.

V Ug. Accounting (3) and

(4) and expressing the output value N (oc,),

we will receive:

AndAo 1 /

..: ... ., ... (five)

1 / yL, / io,

iU .-) - i (oc--)

,

Thus, the output N () conversion block, taking into account the sign bit (at the output of the H011 block), represents the ratio of the sum of the voltages Up, and Uo2 of the output windings of the sensor to their difference and, therefore, does not depend on the excitation voltage of the sensor and other destabilizing factors, and the introduction of the calibration coefficient Kp (N) to the stabilized (illustrated) characteristic of the calibration factor makes it possible to significantly reduce the conversion error from nonlinearity (see Fig. 5).

FIG. Figure 5 shows the output characteristics of the sensor, illustrating the process of introducing the correction factor (characteristics for simplicity are shown only in one square):

N (oi,) - dependence of the output

code from the angle of rotation without the introduction of a correction factor j N ,, a () is ideal (linear)

ABOUT

addiction

Kj,. (N) is the dependence of the correction coefficient, obtained experimentally by dividing the dependence (ot) by the dependence N (ei) | Kp (N) is real (synthesized in the ROM) dependence of the correction factor on the output code value, uN (fti) - error of the conversion Np (oC) - Nj, () without taking into account the correction factor j

iN (oi) - pb error of conversion of an induction sensor,

that is, taking into account the correction factor Kp (N). The induction sensor, with its constructive simplicity and simplicity of the electronic components included in it, has a high linearity and insensitivity to destabilizing effects.

(ful.l

VA / J

Figm

FIG. five

VNISHSH Order 8537/56 Circulation 645 Subscription Branch ShSh Patent, Uzhgorod, Projecto st., A

Claims (1)

TURN ANGLE INDUCTION SENSOR containing a stator with lumped windings, one of which is an excitation winding, and the other is an output and non-winding explicitly pole rotor, characterized in that, in order to increase the conversion accuracy, the stator is made in the form of two identical axially displaced packets , each of the field windings and the output winding are made of two sections, each section has one section of the field winding and the output winding, the packages of stators are deployed around the axis by an angle of about 1-2 °, sections of the field winding are connected in parallel, the sensor is equipped with a conversion unit, to the inputs of which sections of the output winding are connected, realizing the dependence where r is the angle of rotation of the rotor; L ("b) is the dependence of the output code on the angle of rotation ^ the functional dependence of the output voltages of the sections of the windings on the angle of rotation. where " _Eid (about) - ideal (linear) you. running dependence of the conversion unit on the angle of rotation of the rotor; N p (<<) the output dependence of the conversion unit without the correction factor. about ΖΠ s u m> ί