DonNTU   Masters' portal

Abstract

Introduction


Currently thermometer anemometers are widely used to measure various physical and chemical parameters of the gas streams. In particular, the TA is successfully used to control the speed and expenditure of ventilation flows in constructions of various architectural buildings and in the course of their operation, to conduct research works to study the dynamic parameters of turbulent flows, and for measuring the concentration of explosive gases in the automotive industry for the optimization of internal combustion gasoline engines , in the study and control gas stream flow-around of compound bodies, and for gas flow measurement in large pipes, etc. [4].

1. Relevance of the topic

Today is a very urgent task to develop technical means for the experimental determination of the dynamic characteristics of speed measuring devices and temperature gas streams. At the Donetsk National Technical University department of electronic engineering jointly with the Donetsk National University Technological Bureau "Turbulence" begun works of creation corresponding test equipment in the form of a specialized number of aerodynamic of stands. In this regard, there is a need to develop fine structure of turbulence classic measuring device to perform works on the metrological certification of developed stands.
    Analysis showed, that one of the few devices satisfy the requirements set, imposed to the means of measuring the characteristics of of turbulent currents, is the hot-wire anemometer (hereinafter – the TA) of. The TA  Sensor substantially does not disturb the flow, has a high spatial resolution for measuring the microscale of turbulence and low inertia while measuring pulsation [9].


2. The purpose and objecties of the research

The aim is to develop and study hot-wire TA for measurement the fine structure of turbulent gas flows in specialized aerodynamic stands. To achieve this goal following tasks are formulated in this work:

1. Overview the methods and means of measurements of the local velocity and temperature of the gas stream;

2. Development of constant temperature hot-wire anemometer to study the fine structure of turbulent gas flows;

3. Static calibration of hot-wire anemometer and refinement of the heat balance equation;

4. Evaluation of dynamic characteristics of the anemometer;

5. Justification of the structural design of the electrical anemometer signal recording system;

6. Use the method of three superheatings to separate signals of velocity and temperature.

    

3. Overview of Research and Development

In 1996 in Russia was established Research and Production Association "Turbulence-DON". It specializes in design and manufacture of commercial facilities for gas, liquid, thermal power, telemetry systems, and also carries out metrological service of flow meters, a wide range of services for the design and construction of accounting systems energy resources. By the volume of the products the company is a leader in the country.

   LLC SPA "Turbulence-DON" – develops and manufactures such metering devices: stationary and portable gas hot-wire flow meters, municipal gas meters, gas jet flow; stationary and portable liquid flow meters (including sewage) systems for Heat; telemetry systems, pressure sensors, testing equipment, and also provides a wide range of services for calibration, maintenance, repair, testing of all types of gas flow meters [6].

In Ukraine, it is necessary to determine the dynamic thermal performance of air flow to solve a wide range of practical problems in industry, environment, health and safety. This parameter control of HVAC systems, the microclimate in residential and industrial buildings, cabins and saloons vehicles, fume collection channels and pipes.

The instruments used for these measurements should have a wide range, both the velocity (from tenths to tens of meters per second) and the temperature (from tenths to several hundred degrees), a high sensitivity at low pressure differentials (unit Pascal). They must be reliable and easy to use, provide an indication of the measured parameter in units of physical quantities. These requirements are fully satisfy modern hot wires.

In the Special Design and Technological Bureau "Turbulence", in Donetsk National University, developed and produced several versions of such hot-wire devices:

- Thermoanemometer AIST-5. Autonomous measuring instrument of air flow velocity and temperature.

- Thermoanemometer IRIT-4. Flow indicator and traction.

- Thermoanemometer ISRV-2.Thermoanemometric gas meter [2].

At the local level, in Donetsk National Technical University similar issues related to the development of hot-wire anemometers for determination of the fine structure of turbulent gas flows were studied by such masters as Chujko V. A., Morozov A.A., Czibulka V.S., Tymoshenko I.N. 

4. Сurrent outcome

The concept of turbulent and laminar flow was introduced in 1883 by the English physicist O. Reynolds, studying the motion of the fluid in the pipe. At low speeds the movement is regular, but the ratio of inertial forces to viscous forces (Reynolds number Re = ud/v, where u – characteristic velocity, d – characteristic dimension of flow, in this case diameter of the tube, v – kinematic viscosity) exceeds a critical value (ReK = 103), the movement becomes stable and more or less random. Thus in the flow appear irregular whirls of different sizes, and flow rate at each point varies randomly with time. These vortices can be crushed or sometimes merge together. The more so-called supercritical, i.e. the more Re excides ReK, the more intensive are the processes [7].

     To observe the turbulence directly , we should make recognizable movement of the flow of water or air. In the air, it is easy to implement using smoke. On the figure 1 it is shown  the turbulent motion of the gas [8].


Animation of the turbulent motion of the gas (the number of frames - 7, volume - 142 KB, the number of cycles of repetition - 6 times, the delay between shots - 100 ms, the delay between repetitions - 200 ms)

Figure 1 –  Animation of the turbulent motion of the gas (the number of frames - 7, volume - 142 KB, the number of cycles of repetition - 6 times, the delay between shots - 100 ms, the delay between repetitions - 200 ms)

    Circuit technique of the thermometer anemometer. Functional diagram of the designed  constant temperature TA is shown on Figure 2. The main element of the TA is a Wheatstone bridge with sensor Rt, included into one of its arms, and hard bar with feedback and with high amplification factor. For measurement of turbulent pulsations hard bar must provide even frequency characteristic in the band from 0 to 20 kHz. Schematic electric curcuit of the developed prototype of the TA is presented on Figure3.
2e
Figure 2 – Functional diagram of the developed hot-wire TA of the constant temperatur
3e

Figure 3 – Functional diagram of the developed hot-wire TA of the constant temperatur

 Measuring bridge TA is built according to the classical scheme. Wire sensor Rt of thermometer anemometer is joined into one of the arms. Sensing element is made of tungsten wire of a diameter in 8 microns. The thread is welded to holders located at a distance of 4.5 mm by spot welding to provide a reliable electrical contact and mechanical connection strength. The sensor is fixed on a tool box that provides rigidity.
    To balance the zero bridge and to overheat the sensor on the stage of experimental studies used a resistance multiplier Rvar. After choosing the optimum value of the resistance it should be replaced by a fixed resistor. The ratio of Rt and Rvar arms is 1:10. In-feed of  the measuring bridge is made by a constant current.
    The hard bar of TA is made according to the scheme of​​ the measuring amplifier on three operational amplifiers (hereinafter – the OC). This circuit solution allowed to gain a high amplification factor, high input impedance and good common-mode rejection [3]. In case of equality R7/R6 = R5/R4 the output voltage of the amplifier is given:

 f1,                                                                                               (1)

 

 The use of discrete CO made possible to select passive components for optimal parameters of the scheme. In particular, implemented the possibility of device frequency compensation.
    A distinctive feature of the TA is using a field transistor FET IRF840 with insulated gate in the feedback circuit that has a low drain-source resistance in the ON state (0.850 ohm) and fast switching time (21-35 ns), which improved the frequency response of the TA as a whole.

Methods of TA initial setup. During TA initial setup the following operations must be done:

            1)    To enter the setup mode S1 is converted into the open state opening the feedback loop. In this case, the measuring bridge is powered by direct current through the resistor R15.

2) Produce Zero Balance of the measuring bridge with a resistance box Rvar and dial gauge balance.

     3) Determine the resistance of the "cold" leg at room temperature according to the formula:

Rt=Rvar/10.                                                                                                         (2)

    4) Switch the TA to work, closing the key S1.

  5) Preset the value of overheating by using the tungsten temperature coefficient of resistance (TCR) and the resistance of the string sensor at a certain ambient temperature.

  6) Perform the correction of the frequency response of the TA. To do this the built-in square-wave generator is used, made on a microcontroller ATiny2313, deliver on the test input sensor in the form of a square waves with a frequency of about 2 kHz and a duty cycle of 2 [4]. Potentiometer Rf regulate the time constant of the instrument. The optimum setting when the peaks of the output signals are the most acute, but disruption of generation is not observed yet.

Achieved by means of field tests results allowed us to estimate the electrical time constant of the measuring circuit TA, which was about 20 microseconds, that ensures the required bandwidth from 0 to 20 kHz.

Static calibration of the TA was carried out in an aerodynamic stand ADS-200/250 with specialized information-measuring system calibration of hot-wire sensors [3]. Calibration was carried out in air flow velocity range from 1.5 to 10 m / s at temperatures from 20 to 45 0С. To enhance the reliability of measurement results for calibration was performed for three identical filar sensors (hereinafter – sensors number 1 – № 3) at three different filament overheating: Tw = 100, 140 and 180 0С.

The experiment for each sensor were determined according to the output voltage of TA from the flow velocity at four temperatures: Tg = 23, 30, 38 and 45 0С. (see Fig. 4).

4e

Figure 4 – A set of characteristic curves of TA output voltage from the flow velocity at different temperatures

The obtained results allowed to determine the ranges of output voltage of the TA and sensitivity of the speed at different temperatures strands (see Table 1). Based on the analysis of the results the requirements for the analog-to-digital converter (ADC) signal processing circuit TA are made:

- Operating voltage range from 0 to 2.5 V;

- Capacity to determine the static characteristics – at least 10;

- Capacity to determine the dynamic characteristics – at least 12.

 

Table 1 – The output range of the TA voltage and sensitivity over the speed at the minimum and maximum temperature thread overheating

Filament temperature

Tw, 0С

Output voltage ТА UТА, В

Sensitivity velocities S, В·с/м

100

1,05..1,65

0,03..0,085

180

1,55..2,25

0,08..0,125

 

Elaboration of the heat balance equation TA. As a basic heat balance equation for the approximation of obtained in the calibration of the experimental measurement data was chosen recommended in [5], the expression of the form

P / (Tw – Tg) = (A + B(ρv)n) · (Tw / Tg)m,                                                                                              (3)

 

where P – power supplied to the SE, W; Tw, Tg – SE temperature and flow respectively, K; ρυ – mass flow rate, kg/(m2*s); A, B, n and m – constants coefficients determined by individual calibration of the sensor.

While sensor calibration data processing and analyzing as the result of solving non-linear regression calibration coefficients was found that temperature correction power function

 (Tw / Tg)m ,                                                                                                          (4)

 
included into the basic equation (3) did not accurately describe the temperature dependence of the coefficients A and B. Figure 5 shows the typical form of the dependence of the coefficients A and B from the flow temperature Tg. It is specified that the coefficient A the experimental temperature dependence is well approximated by a quadratic function of the form

A(Tg) = a2Tg2 + a1Tg + a0 ,                                                                                                    (5)

 

and for the coefficient V – linear function

B(Tg)=b1Tg+b0.                                                                                                       (6)

 

Thus, improved heat balance equation takes the form

P / (Tw – Tg) = A(Tg) + B(Tg) · (ρv)n.                                                                                                  (7)

5

                                                                                       а)                                                               б)

Figure 5 – Type species of the coefficients dependence A and B from the flow temperature Tg:

a) A (Tg); b) B (Tg)

 

The proposed method of calculating the calibration coefficients for the improved equation (7) is as follows:

1) For the base equation (3) determine the value of the coefficient n.

2) At a fixed ratio of n = const determine the values ​​of the coefficients Ai and Bi of corresponding i-th stream temperature (i = 1 .. 4).

3) Approximates the dependence of B (Tg) function (6).

4) Clarify the meaning of Ai and approximate the dependence of A (Tg) function (5).

5) Estimate approximation errors of calibration experimental data according to specified equation (7).

For the evaluation of the calibration approximation error was selected relative standard deviation (SD).

f8e,                                                                                                (8)

 
where n – is the number of calibration points; ρυ apr and ρυ – the mass velocity, respectively found in the approximation and experimentally.

The calculation results of errors in the calibration while using fundamental and improved heat balance equations are summarized in Table 2 and graphically presented in Figure 6. From the results it follows that in all the cases for different sensors and different superheating improved TA thermal balance equation (7) provides calibration error reduction in comparison with fundamental equation (3) at an average 1.7 times.

6e
Figure6 – RMS approximation of the experimental calibration data fundamental (3) and improved equations (7)

Table 2 –Results of calculation errors of calibration by using a simple and improved heat balance equations

 

Sensor №1

Sensor №2

Sensor №3

Tw,0С

RMS_1,%

RMS_2,

RMS_1,%

RMS_2,%

RMS_1,%

RMS_2,%

100

3

1,5

6,7

5,8

2,5

1,7

140

2,6

1,3

4,9

3,1

1,9

1,3

180

2,4

1,2

4

1,7

1,9

1,1

Conclusion

In the course of the work it was found that the electrical time constant of the measuring system ,developed by the anemometer, is 20 microseconds, that provides the required bandwidth to 20 kHz obligatory for studies of the fine structure of turbulent flows.
    Based on the experimental results of the calibration the main metrological characteristics of the hot-wire anemometer identified and the requirements to the ADC signal processing are represented.
    Taking into account the temperature dependence of the coefficients A and B, the proposed refined heat anemometer balance equation provided the increase in the accuracy of the calibration approximately into 1.7 times.

 List of results

  1. Брэдшоу П. Введение в турбулентность и ее измерение – Москва, 1974. – 278 с. 
  2. ДонНУ СКТБ «Турбулентность» [Электронный ресурс]. – Режим доступа: http://turbulence.uaprom.net/
  3. Зори А.А., Кузнецов Д.Н. Методы и программно-аппаратные средства автоматизированной системы градуировки первичных измерительных преобразователей термоанемометров. Известия ТРТУ. Тематический выпуск: Материалы Всероссийской научно-технической конференции с международным участием «Компьютерные технологии в инженерной и управленч. деятельности». Таганрог: ТРТУ. – 2002. – №2 (25). – С.148-150.
  4. Кузнецов Д.Н., Зори А.А., Кочин А.Е. Измерительные микропроцессорные системы скорости и температуры потоков газа и жидкости – Донецк: ГВУЗ «ДонНТУ», 2012. – 226 с.: ил. 125, табл. 16, библиогр. 95.     
  5. Раннев Г.Г. Информационно-измерительная техника и электроника. – М.: Академия, 2006.– 512 с.
  6. Турбулентность – Дон [Электронный ресурс]. – Режим доступа: http://turbo-don.ru/
  7. Турбулентность [Электронный ресурс]. – Режим доступа: http://www.astronet.ru/db/msg/1188737
  8. Турбулентность в потоке жидкости или газа [Электронный ресурс]. – Режим доступа: http://www.physel.ru/mainmenu-4/eainmenu-15/198-s-193---.html 
  9. Ярин Л. П., Генкин А. Л., Кукес В. И. Термоанемометрия газовых потоков. – Л.: Машиностроение, 1983. – 198 с.
  10. Повх И.Л. Техническая гидромеханика. 2-е изд., доп. / И.Л. Повх. – Л.: Машиностроение, 1976. – 504 с.