UDC 621.17

E.V. Abramchenko, V.S. Gribakin, G.P. Kolesnik.

Interference Immunity Increasing In Digital Measuring Devices

 

The possibility of the measuring process automation is discussed, considering simultaneous increasing of the interference immunity and the accuracy of a digital measuring device in high currents switching and interference conditions due to the strong electromagnetic fields.

Keywords: measuring device, interference, electromagnetic compatibility, electrical machine, electronics, frequency.

 

Electromagnetic compatibility problems are very important for high current electromagnetic power converters (e.g. electrical machines), their research and testing. According to the standard ГОСТ 7217-2003 «Rotating electrical machines, synchronous motors. Test methods», during the testing of an induction machine the slip value must be estimated using the formula:

                                              (1)

where are the stator and rotor electromagnetic field rotation frequencies.

Usually it is recommended to estimate the slip by one of the following means:

1)  Stroboscopic analysis. The motor shaft is marked according to the number of ports. When the shaft is lighted with a strobe light plugged in the same power grid as the tested machine, the time, in which specified number of marks passes the fixed indicator, is calculated.

2)  Using an inductive coil. The coil is placed at the motor’s end, where the needle of galvanometer oscillates the most. After that the number of oscillations during the specified period of time is calculated.

3)  Using the rotor current frequency. The shunt connected with millivolt meter (with a permanent-magnet metering system, i.e. the zero mark is at the center of the gauge) is placed in the wound rotor circuit. The slip S is calculated using the formula:

where N is the number of marks passed by the needle or the number of full needle oscillations (N = 10 - 20); T is the test duration in seconds;  is the supply voltage frequency, Hz.

The standard also allows the slip estimation using the measuring devices for a shaft rotation frequency and the slip measurement, which have error less than 5%. The additional load of these devices must be less than 1% of the output power. If needed, the measured slip values could be reduced to the stator coil working temperature with the formula: [1]

where   is the slip at the working temperature ,  is the slip at the temperature ; α is the temperature resistance coefficient of the coil.

Analyzing the slip measurement methods given above it is seen that these realizations make the measuring of the slip value of an induction motor possible with 5% error, though it is open to question. An additional point is that the automation of the measurement process remains to be implemented as it is only can be done by using devices based on (1).

The solution is really simple at a first sight – to design a microcontroller-based device (with a frequencies to time periods transition). However, in a test lab different experiments are conducted at the same time, with a currents switching and electromagnetic field interference; thus it could be difficult to measure the slip with precision.

The problem can be solved if a designed device will have high interference immunity (i.e. multiple integrating) and required measurement tolerances. For that it is necessary to estimate the error of slip measurement of an automated device based on (1).

Using the typical procedure of inferential measurement, next formula is obtained:

                                                         (2)

where  is the maximum error of power supply frequency measurement;  is the maximum error of rotor rotation frequency measurement.

It is important to note that usually synchronous generators and induction generators supply power for induction motors.

Formula (2) can be reconstructed using slip equation:

                                                 (3)

It is known that nominal slip values for motors built in Russia are estimated with accuracy from 1 to 10 percent, so (3) will be:

 

In this manner, if it is needed to measure the slip S with  error, the stator electromagnetic field frequency and the shaft rotary frequency must be measured with (0,05 – 0,5)% combined error. If , then this error must be much less.

Let us remark here that a measuring device has to be able to measure the slip in synchronous machines with different rotary frequencies - 3000, 1500, 1000, 750, 600 and 500 rpm (according to the number of ports р = 1, 2, 3, 4, 5, 6).

Hence this work sets the problem of designing an automated slip measurement device for induction motors with high interference immunity and fundamental error of measurement around 0,05 %.

This could be solved by realization of device (Fig. 1) which meets the requirements mentioned above. Nodes of the schematic diagram are powered by the AC power supply connected to a phase winding of an induction motor through the transformer T1. Next symbols are used:

          1) D1 and D2 are voltage dividers;

          2) CD1 and CD2 are comparison devices (analog comparators);

          3) FM1 and FM2 are frequency multipliers with phase-locked-loop frequency control (PLL);

          4) DLPF1 and DLPF2 are determined length pulse formers;

          5) BLF1 and BLF2 are balanced low-pass filters;

          6) DA is difference amplifier;

          7) AD is AD converter with double (or triple or quadruple) integration;

          8) S is rotation frequency sensor of induction motor;

          9) PS is testing machine power supply.

 

Fig.1.

Let’s take a detailed look at these nodes. Microchips 1401CA1 with an open collector (or equal LM139) can be used as the comparison devices. The mandatory requirement is that the differential gap circuit must cover the 100-200 mV zone; resistors which form that circuit must be precise, too (like C2-29).

The frequency multipliers could be designed on the CD4046 microchip with a phase detector which works on a wave surface. There are the binary counters MC14520A in the feedback circuit of the each frequency multiplier (in the rotor circuit the CD4022 is followed by the MC14520A). Supply frequency multiplication factor in the stator circuit is 1024 (Area 1), thus nominal output frequency of multiplier is 51200 Hz (considering nominal supply frequency equals to 50 Hz). It is supposed that supply frequency fluctuation is inside the ±2 Hz range. Correction circuit (Fig. 2) between the phase detector and the voltage-controlled generator (VCG) has the following elements: R3 = 4,3 MOhm; R4 = 330 MOhm; C2 = 2,2 µF. The rest of the elements are: С1 = 2,3 nF ; R1 = 10 MOhm; R2 = 200 MOhm and Cф = 4,7 µF. A multiplier with discretely controlled multiplication factor (Fig. 1, switch SB) is used in the rotor circuit. The second electric board (not showed) enables changing of the division factor of the frequency multiplier FM2 with the CD4022 microchip.

Fig. 2

After experimental tests and calculations had been conducted in MATLAB, it was clear that the changing of the multiplication factor does not require the changing of elements’ values in the correction circuit (Tab. 1); it can be done using different timing capacitors C1 – C6 (Fig. 1) considering the PLL astatic system stability. However, it should be noted that by utilizing these types of microchips the theoretical test results may differ from an actual data by about 20%.

Table 1

Number of pole pairs

1

2

3

4

5

6

Stator field nominal rotation frequency, rpm

3000

1500

1000

750

600

500

Rotor PLL frequency control multiplication factor

1024

2048

3072

4096

5120

6144

Capacities , pF

С1=1500

С2=1160

С3=820

С4=570

С5=360

С6=330

 

The so-called Case method is implemented in both channels after the PLL frequency control circuit. It is based on forming and averaging a sequence of potential pulses of multiplied frequencies  and  with constant volt-second characteristics. This method provides high accuracy (less than 0,01% error). Additionally, an operating speed is only limited with the low-pass filter characteristic time τ:

,

where  is the permissible relative error, which is defined by the pulsation of   (BLF output voltage);  is the pulse height at DLPF output terminals;  f is the multiplied rotor or stator field frequency.

In our situation, fast operating speed isn’t essential because the frequencies  and  change their values relatively slow during tests of an electrical machine.

For example, a component element of one ARC filter with two operating amplifiers can be represented by a neutral point in the passive RC circuit (Fig. 3). Other modifications of the corresponding circuits are also possible. It is reasonable to utilize the most common and known asymmetrical ARC component elements [2,3].

Fig. 3

Signal sources connected to the terminals of the balanced filter (dashed and solid lines represent different connection types) represent next active signals:

1)              is the paraphase (friendly) signal

2)              and  are the in-phase signals (filter input interfering signals)

3)                and  are the inductions in the symmetrical nodes i and j of the balanced filter.

Under the operating conditions of the symmetrical ARC circuit component it is considered that all signals mentioned above effect simultaneously. That is why the configuration of the passive RC circuit will be determined by the type of signal source in the same transfer function calculation. In this way, properties of this transfer functions will depend on sources that represent friendly and interfering signals.

    A symmetrical ARC circuit component (Fig. 3) voltage transfer, an interference voltage suppression in the operating amplifier supply circuit or a reduction of supply voltage ripples in the input signal are all connected with the transfer function:

                                                       (4)

where  is (Laplace transform from now on) either the voltage transfer function of the friendly signal and the supply circuit interference suppression or the transfer admittance under the conditions of the supply voltage ripples reduction.

 are the friendly signal transfer functions of two ARC circuit components with asymmetrical input combined into one symmetrical ARC circuit component;

    , if  and  are the supply circuit interference voltage transfer functions;

    , if  and  are the friendly signal voltage transfer functions;

            – Laplace transform of the operating amplifier supply circuit interference voltage.

    A stability of the considered circuit should be estimated using sensitivity functions and disturbing factors. The sensitivity of the transfer function (4) concerning  (a parameter of asymmetrical ARC circuit components combined into one symmetrical ARC circuit component) can be defined as follows:

                                           (5)

Additives (5) represent sensitivities of each asymmetrical ARC circuit component concerning  parameter. Considering relative errors correlation of alike elements in asymmetrical ARC circuit components, transfer function (4) tolerances of symmetrical ARC circuit component in the steady-state operating conditions can be described as:

          (6)

where  is the influence coefficient (temperature dependable, for example);   is the tracking factor of influence coefficients of like  and   elements.

Let us remark here that determined length pulse formers in both channels consist of two circuit components: a circuit component of a pulse former (CD4098B microchip or similar can be used) and a circuit component of an integral key (HI401 or similar). Such configuration helps to avoid instability in electrical levels of logical zeroes and ones in the output voltages of CD4098B like microchips.

Now, the subtractor (based of OP-07 microchip) represents a common formula for an AD converter with the push-pull integration:

where  and  are the output BLP1 and BLP2 voltages, the difference amplifier gain  so all gage dial GD can be used (Fig. 1). Because of that, maximum slip value is 19.99%, which allows the device to measure all possible slip values, even critical, for all types of induction motors.

For even stronger suppression of AC line and multiple frequencies, the voltage divider D2 and the frequency multiplier FM1 can be used to form a clocking rate of the AD converter. The clocking rate can change according to  variations, so the slip measurement doesn’t receive any interference from such variations.

Proposed device might use only one adjustable resistor in BLF1 (or BLF2) feedback circuit. In that case, balanced low-pass filters can be based on OP-37 microchips.

Experiments showed that test device had 0.06% instrument error when the AD converter was based on the IL 7107 N microchip and the DKS 40 was used as a rotary encoder. As an alternative to the DKS 40, an opto-isolator could be used for cheapening and enhancement of the device. In that case motor shaft needs one mark regardless of its power or size.

A natural application of a proposed device is a RPM percent indicator; it would lack the nonlinear conversion response error while working in the heavy industrial interference environment.

 

References

1. ГОСТ 7217-2003 «Машины электрические вращающиеся. Двигатели асинхронные. Методы испытаний». ИПК, издательство стандартов, 2003. 39 с.

2. Грибакин В.С., Грибакин А.С., Колесник Г.П., Колесник П.Г. Синтез помехоустойчивых активных RC-звеньев с изменяемой частотой настройки для медицинской аппаратуры. В кн.: Физика и радиоэлектроника в медицине и экологии. Доклады V МНТК ФРЭМЭ 2004, Владимир. - 2004. - с. 222-225.

3. Грибакин В.С., Грибакин А.С., Колесник Г.П. Точность метода вольтметра-амперметра при наличии в каналах напряжения и тока селективных устройств. Проектирование и технология электронных средств. №4, 2004 г. - с. 26-29.

4. Грибакин В.С., Грибакин А.С., Колесник Г.П. Реализация активного полосового фильтра с изменяемой частотой настройки. В кн.: Физика и радиоэлектроника в медицине и экологии. Доклады V МНТК ФРЭМЭ 2002, Владимир. 2002. - с. 140-143.

5. Колесник Г.П. и др. Канал обработки информации при измерении электрической емкости конденсаторов. Проектирование и технология электронных средств. №1, 2007 г. - с. 71-74.

 

Аннотация

Рассматривается возможность автоматизации измерения при одновременном повышении помехоустойчивости и точности цифрового измерительного устройства, работающего в условиях коммутаций больших токов и помех, обусловленных сильными электромагнитными полями.

Keywords: измерительное устройство, помехи, электромагнитная совместимость, электродвигатель, электроника, частота.