SU978174A1 - Displacement to code conversion device - Google Patents

Description

A device for converting movement into a code is related to converters of non-electric values into a discrete electric signal and can be used in multichannel data acquisition systems, in digital devices for indicating the position of elements of a controlled object. In closed numerical control systems (CNC) machine tools for position control movable elements of the machine.

A calculating device is known, which contains two digital-analogue positive devices whose inputs receive analog output signals of a sine-cosine sensor (inductosyn, selsyn, rotating transformer) and compensation signals presented in binary code and selected from two permanent memories of their devices.

The differential device generates an error signal, which is analyzed by the control circuit of the counter, the JlBVM levels of the signal amplitude of the error signal correspond to two fixed values of the resolution that cause the state of the reversible counter. The output of the counter is the address of a sample of compensation signals recorded in the persistent storage devices. The operation of the device is carried out until the error signal is equal to zero, and the electrical equivalent of the mechanical movement of the moving parts of the sensor relative to each other, expressed in the dual code Ts, is established at the output of the counter.

The disadvantage of the device is its relatively low speed, due to the fact that this device is analog-to-digital

15 is of the follower type and implies low speeds and accelerations of the measured values of displacements 2.

A device for analog-to-digital converter of trigonometric functions is known. It contains two digital-analogue multiplying devices, the outputs of which are connected to the inputs of the analog-digital integrating device, the output of which is connected to the inputs of digital-analogue multiplying devices GZ through the converter of digital signals.

In addition to the displacement output, this device has a speed output, however, the device performance is limited, since it can be attributed to analog-digital conversion of bodies with double integration of G2. Besides. Convert by definition. the division of the measured value is performed not by reversible counters, as in the device 13, but by successive shift registers. The closest technical solution to the present invention is a device for converting movements into a code containing a sensor, the outputs of which are connected through the amplifier unit to the first inputs of digital-to-analog converters, the outputs of which are connected respectively to the first and second inputs of the sign selector, signal distributor, first output which through the filter is connected to the sensor input, the second, output to the first input of the error analyzer (comparator), the output of which is connected to the second input of the block Discreteness, the third output of the signal distributor is connected to the first input of the discrete selection block, the output of which is connected to the input of the reversible counter and the first input of the sine-cosine digital converter, the fourth output of the pulse distributor is connected to the second input of the sine-cosine converter, the first and second the inputs of which are connected to the second inputs of the corresponding digital-to-analog converters, and a third of the output is connected; with the third input is a block for selecting a sign (quad switch), the output of which is connected to the first input of the error analyzer 4. Compared to devices of the next type and devices that operate on the principle of double integration, the speed of the known device is higher due to the dynamic program implemented in the device. However, this increase in speed is achieved by decreasing the measurement accuracy due to the increase in discreteness. Therefore, the need for measurement with high accuracy imposes significant limitations on the permissible speeds and accelerations of the eNBdx displacements. In addition, for all the above devices that have a reversible counter or shift register in the error channel, a relatively low noise immunity and, therefore, low reliability, especially for digital display systems in shop conditions, is characteristic. The purpose of the invention is to increase the speed, accuracy and reliability of the device for converting movements into code. The goal is achieved by the fact that in a device for converting movement into a code containing a sensor, the first and second outputs of which are connected to the first rt second inputs of the amplifier unit, respectively, a distributor, the first output of which is connected through the filter to the sensor input, the first and second digital-to-analogue converters, the outputs of which are connected respectively to the first and second inputs of the comparator, a register of successive approximations, an output code generator, a quadrant identifier, a decoder, the first are entered and a second permanent memory blocks, and the sampling switch storing unit and first and second outputs which are respectively connected to first and second inputs of the switch, which outputs soedine.ny to the first inputs of digital-analog converters; the outputs of which are connected to the first and second inputs of the comparator, the output of which is connected to the first input of the register of successive approximations, the first output of which is connected to the first input of the output code generator and to the first inputs of permanent storage units whose outputs are connected to the second inputs of the corresponding digital-analogue converters , the second output of the register of successive approximations is connected to the second input of the output code generator and the input of the distributor, the second output of which is connected to the first m input of the quadrant determinant and the first input of the sampling and storage unit, the first and second output of the amplifier unit are connected to the second and third inputs of the sampling and storage unit and the quadrant determinant, respectively, the output of the quadrant determinant is connected to the third input of the quadrant switch and to the input of the detifrator, the first output of which connected to the third input of the output code generator, and the second output is connected to the second inputs of the permanent memories, the third and fourth outputs of the distributor are connected to volts ring and the third input register of successive approximations respectively. FIG. 1 shows a block diagram of the device; in fig. 2 is a diagram of the operation of the distributor; in fig. 3 - quadrant determinant; in fig. 4 - quad switch; in fig. 5 - diagrams explaining the principle of addressing permanent storage devices. The proposed device contains a distributor 1, filter 2, sensor 3, block 4 amplifiers, block 5 sample

and storage, a switch 6 quadrants, a determinant of 7 quadrants, a decoder 8, digital-to-analog converters (UAnj 9 and 10, permanent storage units (SDB) 11 and 12, a comparator 13, a register of 14 successive approximations (DFS), a driver. 15 of the output code.

The outputs of the sensor 3 are connected to the inputs of the amplifier unit 4, the first output of which is connected to the first inputs of the sampling and storage unit 5 and the determinant of 7 quadrants, the second output of the amplifier unit 4 is connected to. the second inputs of sampling and storage unit 5 and quadrant determinant 7, the output of which is connected to the third input of quad switch 6 and to the input of the decoder 8. The first output of the decoder B is connected to the third input of the output code converter 15, and the second output is connected to the second inputs of the ROM 11 and ROM 12. The first and second inputs of quad switch b are connected to the corresponding outputs of the storage sample block 5, and the first and second outputs are connected to. responsibly with the first inputs of the DAC 9 and 10, the outputs of which are connected to the first and second inputs of the comparator 13. The output of the comparator is connected to the first input of the BCP 14, the first output of which is connected to the first input of the output driver 15, and also to the first inputs of the OTL 11 and 12 The outputs of which are connected respectively to the second inputs of the DAC 9 and 10. The second output of the switchboard 14 is connected to the second input of the output code generator 15 and the input of the distributor 1, the first output of the distributor through the filter 2 is connected to the input of the sensor 3, the second output of the distributor 1 is connected to the first input of the determinant 7 quadrants and the first input of the sampling and storage unit 5, the third and fourth outputs of the distributor 1 are connected respectively to the second and third inputs of the switchboard 14.

The determinant of 7 quadrants (fig. 3) consists of two comparators 16 and 17. The switch 6 of quadrants (fig. 4) contains operational amplifiers 18, 19 and switches 20, 21.

The device works as follows.

Consider the operation of the device on the example of working with a sensor of the type circle. Inductosin (Fig. 2).

The pulse signal of the frequency of the distributor 1 (Fig. 2a) is converted by the filter 2 into a sinusoidal signal. (Fig. 2b). This signal energizes the rotor winding of sensor 3. The information signals U ebK t and (Fig. 2c and 2d), modulated in amplitude according to the sine and cosine laws as functions of the angle of rotation, are removed from the stator windings of the sensor.

Ujb ,,. Urn sinpe-srn (2Jft-90 ° (1) UebW2 and „, cospe.sin (2 ft-90), (2)

where p is the number of pairs of winding poles

rotor or electrical reduction ratio; 9 - mechanical angle of rotation

rotor relative to the stator. In particular, for circular inductosine, p is 180, and the sensor pitch is 2. For a sensor inductosin-type, the stator winding signals are out of phase by 90 ° relative to the rotor winding signal. These signals through the block 4 amplifiers enter the block 5 of sampling and storage and the determinant of 7 quadrants. In block 5 of sampling and storage, consisting of two typical schemes of sampling and storage, the instantaneous values of the amplitudes of the signals (1) and (2) are recorded at time points

t -t- + 2k 5 (to 0.1) on the signal T1 from the distributor 1 (fig.2d).

The purpose of converting the sensor output signals is to determine the oL-electric equivalent of the angle p and represent it in binary code. The two most significant bits of the claim code indicate the Hcwep quarter of a sensor step (quadrant number) in which the measured angle p® is located.

The determinant of 7 quadrants (fig.Z) together with the decoder 3 serves

45 for determining the quadrant number. The distributor of 7 quadrants consists of two comparators 16 and 17, gated with a signal of distributor 1, and in fact is a phase analyzer

50 output signals of the Ue sensor (, ni7 The first quadrant corresponds to the output code of the quadrant 7 00, the second is code 01. The third is code 11 and the fourth is code

55 10 (in the higher order the phase of the signal Cl is analyzed, in the lowest the phase of signal 2 according to the table) 1

01

01

the code of determinants of quadrants in the binary code of the number of the quadrant.

This code is fed to the first input of the shaper 15.

Blocks 6, 8-14 determine the lower

bits of the desired binary code by analyzing the instantaneous values of the amplitudes of the signals (1) and (2). As you can see, and further examine the operation of the device, the analyzed amplitudes should be converted to positive polarity signals regardless of the quadrant number. Obviously, the information about the sensor signals is not lost in this case, since their phase has already been analyzed by the determinant of quadrants 7.

FIG. 4 shows a circuit of quad switch 6, consisting of two operational amplifiers 18 and 19 switched on in inverter mode, and two switches 20 and 21, which are controlled by the output code of 7 quadrants (see table).

Taking into account the response time of the switch b quadrants, the distributor 1 generates a signal 12 Allow conversion to RPF 14 (Fig. 2a). From this moment, the work of RPP 14 starts from the clock pulses of the TK of the distributor 1 with the frequency F (Fig. 2g). RPP 14 operates according to the dynamic program 2 and issues an n-bit test binary code where the frequency of the DAC is 9 and 10. This code, together with the quadrant code, makes it possible to uniquely represent any angle (inside the sensor step 1fig. 5a). PZB 11 and 12 are stored, respectively, with sot and sinoL within 1 and P quadrants, represented in binary code Msob and quantization step 2 (11g. 5b). The values of NcosoL and Msind are the same for the I and II and II and IV quadrants respectively. Therefore, the decoder 8 ensures the equivalence of addressing the cells of the PZB in these quadrants. In this case, the I and m I quadrants correspond to the code O, and II and IV correspond to the code 1 (see table), and the address of the sample of the PZB cells is (n + 1) bit word.

The outputs of the DAC 9 and 10 are formed voltage

(3)

- Uyf, sin p dcos oL

and

YOU «1

Table continuation

Comparator 13 analyzes the difference of signals (3) and (4)

iS pecosot-i, „co8pb5 by (., (pe-ot) (5-)

Depending on the sign (plus, minus) and the corresponding value of the comparator D (9. Or 1), the RPF 14 captures O or 1 in the bit that is analyzed in a given tact of the dynamic program.

Upon completion of the dynamic program of the register of successive approximations, a binary equivalent of the absolute value of the angle inside the quadrant is established at its output with an accuracy of one of the least significant bit of the input DAC code and the condition

(.. (b)

By signal End Conversion,

5 arriving at the third input of the imaging unit 15, in the latter the quadrant code, coming from the output of the decoder 8, and the angle value inside the quadrant are fixed. As a result, a binary equivalent of the absolute value of the measured angle within the sensor pitch is stored in the output register. Signal End of Conversion to PFR 14 also returns to

5 the beginning of the cycle, and the distributor 1 stops the issuance of TK clock pulses to the input of the PDP 14 (Fig. 2g).

From the analysis of the equation I) it follows that to fulfill this condition in

Q of all four quadrants requires that both input voltages on the comparator have a positive polarity. To do this, it is sufficient to manage the switch keys of the six shifts using the method described above.

As RPP 14, a typical microcircuit can be taken. Let the rotor of the sensor 3 move relative to the stator by such a value that the phase correlation of the signals on the stator winding corresponds to the third quadrant. The instantaneous values of U sinpO and U icospe of negative polarity are stored on the signal of the distributor 1 T1 in block 5 of the sample and storage, the computer 5 of the controllers 16 and 17 (Fig. 3) sets the quadrant control switch 11 code 11, which connects the inputs of the DAC 9 and 10 to the outputs of the operational amplifiers 18 and 19. The decoder 8 converts code 11 to O (quadrant II) and sets it at the first input of output register 15, and also converts code 10 to O to form the two most significant bits of the sample address of the OTL 11 sample and 12.This code remains are constant throughout the entire measurement cycle. On a T2 signal, allow the conversion of the RPP 14, in the first cycle of operation, the n-bit test code 011 ... 1 is generated, with the sample address of the PZB 11 and 12 0011 ... 1. At this address, the PZB 11 and 12 are selected with an accuracy of one of the youngest bits of a cos 45 ° and sin 45 °. In the next cycle, the RPP 14 generates the code D01 ... 1, the sampling address of ELVs 11 and 12 - OD01 ... 1. On the (n + 1st) cycle of operation of the PAD 14, the condition (5) is fulfilled also by the signal KoHet; the conversion the desired code is fixed in the driver 15.

Analysis of the structural scheme and operation of the functional parts of the device confirms that the introduction of the sampling and storage unit, the register of successive approximations of the output register, two permanent memory devices, the decoder, the quadrant determinant, and the new connections allow one to achieve the goal.

Since the frequency F of the clock pulses of the FAR can significantly exceed the frequency f of the supply voltage of the sensor, the processing of any error within one sensor step is completed in one period of frequency f, which significantly increases the speed of the device compared to the prototype.

At the end of each measurement cycle, equal to -J-, in the source register 15, the absolute value of the angle within the sensor step, is stored in binary code. moreover, the quadrant code is formed by the determinant of 7 quadrants and the decoder 8.

An important characteristic of measuring systems is the settling time with a jump-like change in the measured quantity.

Let us take as an example a sensor of the circular inductosin type. The type frequency is f 10 kHz, the sensor step is 2..

Let the minimal 3jia4eHHe discreteness is 3.6. Consider the settling time at a step difference of 1/4 sensor {& - 0,5 1000l). By plotting for this mismatch a graph similar to that shown in the prototype, it can be concluded that the dynamic program implemented in the C4 device provides the establishment time for 86 periods of the frequency f, i.e. for 8.6 ms.

A dynamic program, implemented in the proposed device, provides the establishment time for 100 ISS.

When moving at a constant speed, the measurement accuracy in the proposed device is significantly higher than in the prototype, since the measurement resolution remains minimal while in the prototype it varies depending on the magnitude of the speed.

When driving with constant acceleration, the accuracy in the proposed device is also significantly better than in the prototype. With the accelerated movement, the device 4 is forced to determine in each measurement cycle an increasing amount of movement. In this case, the discreteness selection unit at the time of the end of the next measurement cycle increases the compensation value by and, This value is not equal to the displacement value in the previous measurement cycle. Therefore, at the time of measurement, an additional dynamic

the error due to the lag of the compensation value from the measured.

In the proposed device, this error is minimal.

The increased noise immunity of the proposed device as compared with the prototype is due to the fact that in each measurement cycle it is not the increment of displacement that is determined, but the absolute value of the position within the step,

which eliminates the loss of information under the influence of destabilizing factors (the change in the gap in sensors of the type of inductosin during vibrations, the pulsation of the supply voltage and

etc.). The increased noise immunity of the proposed device ensures its higher reliability compared to the prototype.

The proposed device provides the possibility of outputting and 11 formations in a sequential code, without additional hardware, costs, since RPT 14 allows, upon completion of

To transmit each cycle of a dynamic program through a communication channel with external devices a serial code, starting with the highest bit. At the same time, the transmission time over the communication channel

remains the same as when transmitting a parallel code, but the number of communication lines is significantly reduced. The output of information in a sequential code is especially useful in multichannel systems for collecting and processing information, in which information from each sensor is processed by its device, and then the outputs of these devices are combined into a common bus. liMeHHo in this case, the greatest economic effect is obtained.

One of the most promising applications of the proposed device is the creation on its basis of a multi-channel system for the collection and processing of induction, in which one proposed device sequentially processes signals from several sensors.

When switching to the next channel, the error value can reach the value of the sensor step. At the same time, the speed of such a system built on the basis of a device selected as analogues and a prototype does not satisfy a large class of tasks encountered, for example, when controlling machine tools. This leads to the necessity of complicating the multichannel system through analog or digital memory (storing information from the i-sensor in the (1-1) -th measurement cycle) or the need to design only one, as a last resort, two-coordinate measurement systems.

In addition, analog memory reduces measurement accuracy, and digital memory lowers performance.

The ability to measure any discrepancy within one sensor step over a single frequency period f makes it possible to create, on the basis of the device proposed, a high-speed multi-channel information collection and processing system without digital or analog memory. Processing signals from several sensors with one device will give a significant economic effect.

Let us estimate the economic efficiency of the proposed device for a particular application - in closed CNC systems by the position of the movable elements of machines with sensors of the “1 and inductance type.”

For most drives used in CNC machines, the frequency of change of control signals is no more than 1 kHz.

Thus, at the speed of the proposed device for converting movements into a code equal to 100 ISS, it is possible to implement a multichannel system that sequentially processes information from displacement transducers. The average number of coordinates in machines with CNC devices is four. At a cost of the proposed device of 1.5 thousand rubles, the economic effect from the use of

one device for converting earthquakes into code instead of four in the CNC device will be about 3600 rubles.

. The economic effect of the increase

reliability is ensured by increasing the noise immunity of the proposed device, for example, in digital indication systems under general conditions, which reduces the likelihood of defective products.

Claims (4)

Formula of the invention 5 A device for converting movement into a code containing a sensor, the first and second outputs of which are connected to the first and second inputs of the amplifier unit, respectively; a distributor, the first output of which is connected through a filter to the sensor input; the first and second digital-analog converters, whose outputs are connected respectively to the first and 5 second inputs of the comparator, characterized in that, in order to improve the speed, accuracy and reliability of the device, a register of consecutive approximations, the driver is inserted into it output code, quadrant determinant, decoder, the first and second permanent storage blocks, the switch and the sample and storage unit, the first and second c outputs of which are connected respectively to the first and second inputs of the switch, the outputs of which are connected to the first inputs of the first-to-analog converters whose outputs are connected to the first and second inputs of the comparator, the output of which is connected to the first input of the register of successive approximations, the first output of which is connected to the first input of the output code generator and to the first moves the permanent storage units whose outputs are connected to corresponding second inputs. The second output of the register of approximations is connected to the second input of the output code generator and the input of the distributor, the second output of which is connected to the first input of the quadrant determinant and the first 5 input of the sampling and storage unit, the first and second outputs of the amplifier unit are connected to the second and third the inputs of the block of sampling and storage and the determinant of quadrants, respectively, 0 the output of the determinant of quadrants is connected to the third input of the switch, and the quadrants and to the input of the decoder The first output of which is connected to the third input of the output code and the second output is connected to second inputs of the permanent memory unit, the third and fourth distributor are connected to the outputs of the second and third inputs of the register of successive approximations respectively. Sources of information taken into account in the examination 1. US patent 3641565, cl. H 03 K 13/02, published 1972 9 4 2. Bakhtiarov G, D., and others. Analog-digital converters. M., Soviet Radio, 1980, p. 158, 167, 143/174. 3. US Patent 3896299, 1975 ° ° 13/02, published. 4. Author's certificate of the USSR. 742999, cl. G 08 C 9/04, 1980 (pro; otip). i s Ufrj-with rv Tifuh a gd 7 /  T-j (rig 2: one t