Toward a Methodology for Testing Pneumatic Fans for Local Ventilation

B. M. Titov

(Presented by Prof., Doctor of Technical Sciences I. A. Balashev)

Statement of the Problem

Manufacturing plants, design organizations, and research institutes of the ministries of the ore-mining and coal industries are working on improving pneumatic fans. Among these organizations, especially those subordinate to different ministries, as a rule there is no proper exchange of information. Fan units are tested according to various individual (unpublished) methods, and the techno-economic characteristics are presented in such a form that they cannot be used even for operational calculations. For example, the characteristics of the VP-3 fan unit are given [11] as follows:

From this characteristic it is unclear: a) why, at different operating pressures, the air consumption by the motor remains constant; b) under what operating condition and pressure in the pneumatic network the fan has the indicated capacity and head; c) how the fan capacity can be determined when operating on a specific ventilation network. Characteristics of pneumatic fans are given in a similar manner in other literary sources [5, p. 188], [6, p. 306].

The absence of a unified testing methodology and the unsuccessful compilation of techno-economic characteristics make it difficult to compare and improve fan units.

In the present work, for the purposes of information and discussion, a methodology for testing pneumatic fans, developed by the author and adopted by the Tomsk Electromechanical Plant (TEMZ), is briefly set forth.^1

Fan tests, depending on the purpose set and the place where they are carried out, are divided into three groups.

1. Proving (laboratory) tests, the purpose of which is to obtain the complete technical characteristics of production fans, experimentally refine prototypes, and obtain initial data for designing new machines.

^1 In greater detail, with illustrations, a specific numerical example, and an assessment of the errors of the test results, this methodology is set forth in the author’s manuscript “Methodology for Testing Pneumatic Fans for Local Ventilation,” kept in the libraries of TEMZ and TPI.

1. Acceptance (factory) tests, the main purpose of which is to check the quality of the products manufactured by the plant.

2. Operational (mine) tests, in which the correspondence of the fan’s aerodynamic parameters to the characteristics of the ventilation network, the mechanical strength, wear resistance, and convenience of operation of the fan unit as a whole are checked.

In adjustment and acceptance tests, the characteristic of the fan unit is taken either completely or partially; in operational tests, the characteristic of the ventilation installation is taken.

A typical representative of modern local-ventilation fans with pneumatic drive is the VP-4 fan unit. The VP-4 unit consists (Fig. 1) of a fan with guide

Fig. 1. Section of a VP-4 fan unit: 1—guide apparatus; 2—fan impeller; 3—straightening apparatus; 4—turbine; 5—nozzle; 6—channel for discharging the air that has worked in the engine.

and straightening apparatuses and a pneumatic turbine motor. The blades of the turbine motor are arranged on the outer circumference of the fan impeller. Compressed air is supplied through the nozzle to the turbine; after doing work in it, it is discharged through the gap between the casing and the impeller into the pressure ventilation pipeline.

The arrangement of the experiments and the processing of the observational results are illustrated below using the results of tests of fan units of types VP-4 and VP-4M.

Adjustment Tests

The adjustment of experimental machines is most conveniently carried out by comparing the experimental characteristics of the original and the modernized turbine motor, fan, and fan unit as a whole. This method of investigating fans is widely used at TsAGI.

The aerodynamic characteristic of a fan, as is known, is taken in a test chamber or pipeline; moreover, with the experiments set up correctly, according to studies carried out by M. Ya. Gemberzhevskii [3], the fan characteristics obtained are identical.

For a long time it was believed that the magnitude of the pressure developed by a fan does not depend on whether it operates on suction or on delivery

[9, p. 147], and therefore, taking into account the more convenient conditions for measuring air flow rate and pressure, it was preferred to test fans while operating in suction. In reality, however, the value of the total head of a fan tested in discharge is obtained as somewhat greater than the same head when tested in suction [8]. At the same time, the power curves in both cases agree with one another quite well. The increment in head is explained by the fact that, in the discharge pipeline, the nonuniform velocity field is somewhat equalized, and the difference between the mean and actual dynamic pressure, which is not taken into account in suction tests, becomes an additional static head of the fan. To combine the positive aspects of both methods, fans should be tested in suction with a pipeline on the discharge side.

Equalization of the velocity field in the discharge pipeline downstream of centrifugal [3, p. 138] and axial (Fig. 2) fans occurs at a distance of no more than three pipe diameters ($D$). Further increase in the pipeline length does not produce an increase in static or total pressure.

Equalization of the velocity field and conversion of dynamic head into static head is, naturally, accompanied by a loss of pressure in the discharge nozzle—the diffuser. The procedure developed at Gipro-uglemash for testing the “Prokhodka-500” fans provides for analytical calculation of these losses and their addition to the measured fan head. It seems to us, however, that this should not be done.

First, the loss of head in the discharge nozzle, even with a length of $5D$ (according to the Gipro-uglemash procedure), is no more than 1% of the fan head and does not exceed the possible maximum experimental error.

Second, this loss of head can be regarded as usefully utilized only when the fan is operating in discharge.

Fig. 2. Influence of the length of the nozzle-diffuser on the characteristic of a pneumatic axial fan.

The air consumption by the turbo-engine averages 4–6% of the optimum fan capacity. However, when indicating the fan capacity, the air consumption by the turbo-engine should not be added to it, because: 1) this air is not used for ventilating the working when the fan operates in suction; 2) by means of the increased specific consumption of compressed air (low efficiency of the engine), the ...

lower the fan efficiency and 3) in the comparative evaluation of different machines lead to erroneous conclusions.

Figure 3 shows a stand for testing fan units with a diameter of the suction openings not exceeding \(D_h\). When throttling the fan by means of screens, the length of the pipeline from the throttling diaphragms to the point at which the static pressure is measured may be reduced to \(2D_h\) [3]. If the fan has no guide apparatus or ribs for fastening the front bearing and the fairing, then a straightening grid \(0.5D_h\) wide is installed between them and the measuring instruments; the pressure drop in this grid is added to the measured fan pressure [1]. The distance from the collector to a wall or to any object that might impede the entry of air into the pipe must be not less than one caliber \((D_c)\), and from the outlet section of the fan branch pipe to the wall not less than three calibers \((D_{\text{out}})\) of the pipeline. The inlet branch pipe of the stand is made replaceable, and its diameter is selected in accordance with the capacity of the fan being tested.

Fig. 3. Stand for testing a fan unit: 1—regulating slide gate; 2—starting valve of the turbo-blower; 3—regulating diaphragm; 4—diaphragm; 5—differential manometer; 6—manometer; 7—thermometer; 8—tachometer; 9—liquid manometer; 10—micromanometer; 11—barometer; 12—psychrometer.

To stabilize the temperature regime of the test installation, the fan unit must operate at a constant working pressure for not less than 30 min, and only after this, without stopping the unit, are the tests begun. The working pressure \(p_{\text{in}}\) is maintained constant by the regulating slide gate (Fig. 3); the starting valve of the turbo-blower is opened completely.

The operating regime of the fan unit is changed by means of regulating diaphragms. At each diaphragm, the air consumption of the turbo-blower, the shaft speed of the unit, and the capacity and pressure of the fan are measured. The compressed-air consumption is measured by means of a diaphragm and a differential manometer in accordance with “Rules No. 169” [10]. The shaft speed may be measured with a vibration tachometer, but more accurate results are obـ

are obtained if the end of the fan shaft¹ is brought outside and the measurement is made with a revolution counter and a stopwatch or with a tachometer. The capacity and pressure of the fan are determined according to the procedure adopted at TsAGI [1].

Measuring instruments are selected with allowance for the maximum permissible error of the experiments and are checked before the start of the tests in order to determine systematic errors and, after the tests, to verify serviceability. In the test report it is necessary to indicate the serial numbers of the instruments used.

In those cases where the fluctuation of the values of the measured quantities is less than the error of the measuring instruments used, readings on the instruments are taken only once; but, in order to avoid oversights, the experiments are carried out twice: the first time while decreasing the equivalent opening of the network, the second time while increasing it.

The characteristic of the fan unit is taken at a constant working pressure of the compressed air and is reduced to the normal test conditions:

temperature of compressed and ventilating air . . . . \(20^\circ\text{C}\),
atmospheric pressure . . . . . . . . . . . . . . . . . \(760\) mm Hg,
specific weight of atmospheric air . . . . . . . . . \(1.2\ \text{kg}/\text{m}^3\).

When reducing the characteristic of the fan unit to the normal test conditions, the number of revolutions of the fan shaft changes in direct proportion to the change in engine power and in inverse proportion to the change in fan load

\[ n = n_{op}\sqrt[3]{\frac{N_e}{N}}, \tag{1} \]

where \(n_{op}\) is the number of revolutions of the fan shaft during the experiment, rpm;
\(N_e, N\) are the powers on the shafts of the engine and the fan after reducing them to the normal test conditions, kW.

As is known, the power of a turbo-engine changes in direct proportion to the change in flow rate and in the specific work capacity of the compressed air

\[ \frac{N_e}{N'_e} = \frac{G_{dv.n}}{G_{dv}} \frac{L_{ad.n}}{L_{ad}}, \tag{2} \]

where \(N'_e\) is the power on the engine shaft under the test conditions, kW.

The change in the weight flow rate of the compressed air depends on its state before entering the turbo-engine [4, p. 328].

\[ \frac{G_{dv.n}}{G_{dv}} = \frac{p_{vx.n}}{p_{vx}} \sqrt{\frac{T_{vx}}{T_{vx.n}}}, \tag{3} \]

where \(p_{vx.n},\ p_{vx}\) are the normal and actual absolute pressure of the compressed air, kg/m²;
\(T_{vx.n},\ T_{vx}\) are the normal and actual absolute temperature of the compressed air, °K.

The specific adiabatic work of the air is calculated from the equation [4, p. 81]

\[ L_{ad} = \frac{k}{k-1}RT_{vx} \left[ 1-\left(\frac{p_{am}}{p_{vx}}\right)^{\frac{k-1}{k}} \right], \tag{4} \]

where \(k\) is the adiabatic exponent;
\(R\) is the gas constant, kgm/kgC°;
\(p_{am}\) is the atmospheric pressure, kg/m².

¹ The end of the shaft brought outside is subsequently, when testing the pneumatic engine, used for connection to the load generator.

Substituting the value of the flow rate and of the specific work of the air into formula (2), we obtain

\[ N_e = N'_e \frac{p_{\mathrm{in},n}}{p_{\mathrm{in}}} \sqrt{\frac{T_{\mathrm{in},n}}{T_{\mathrm{in}}}} \left[ \frac{ 1-\left(\dfrac{p_{\mathrm{atm},n}}{p_{\mathrm{in},n}}\right)^{\frac{k-1}{k}} }{ 1-\left(\dfrac{p_{\mathrm{atm}}}{p_{\mathrm{in}}}\right)^{\frac{k-1}{k}} } \right]. \tag{5} \]

The gauge pressure during tests is maintained constant, and the change in atmospheric pressure in comparison with the gauge pressure is insignificant; therefore the correction for the change in pressure is practically equal to unity and, consequently,

\[ N_e = N'_e \sqrt{\frac{T_{\mathrm{in},n}}{T_{\mathrm{in}}}} . \tag{6} \]

The power on the fan shaft changes in direct proportion to the change in the specific weight of the ventilating air,

\[ N = N' \frac{\gamma_n}{\gamma_{\mathrm{op}}}, \tag{7} \]

where \(N'\) is the power on the fan shaft under the test conditions, kW;
\(\gamma_n, \gamma_{\mathrm{op}}\) are the normal and actual specific weights of atmospheric air, kg/m\(^3\).

After substituting the power values into formula (1), we obtain

\[ n = n_{\mathrm{op}} \sqrt[6]{\frac{T_{\mathrm{in},n}}{T_{\mathrm{in}}}} \sqrt[3]{\frac{\gamma_{\mathrm{op}}}{\gamma_n}} . \tag{8} \]

The weight and volume flow rate of air through the engine is specified for normal test conditions \((t_{\mathrm{in}} = 20^\circ\mathrm{C},\ \gamma_n = 1.2\ \mathrm{kg}/\mathrm{m}^3)\)

\[ G_{\mathrm{dv},n} = G_{\mathrm{dv}} \frac{p_{\mathrm{in},n}}{p_{\mathrm{in}}} \sqrt{\frac{T_{\mathrm{in}}}{T_{\mathrm{in},n}}}, \tag{9} \]

\[ Q_{\mathrm{dv}} = \frac{G_{\mathrm{dv},n}}{\gamma_n}. \tag{10} \]

The characteristic of the ventilation unit, reduced to normal test conditions, is represented graphically (Fig. 4).

The fan characteristic is not recalculated to a constant number of shaft revolutions, since this does not correspond to the normal operating regime of the ventilation unit.

In order to separate the efficiency of the unit into the efficiency of the engine and that of the fan, an additional test of the turbo-engine is carried out. A specific feature of this test is the separation of the engine and fan impellers, which are structurally connected into a single whole. To separate the turbo-engine impeller, the fan blades are blanked off with disks made of roofing sheet iron (Fig. 5).

The engine may be loaded with a direct-current generator (preferably with independent excitation), whose power must exceed the rated power of the pneumatic engine by approximately 1.5 times. The electrical measuring instruments shown in the diagram of the stand for testing the engine (Fig. 6) are intended only for monitoring the operation of the generator.

Fig. 4. Characteristic of the VP-4 type fan unit.

Fig. 5. Impellers of a pneumatic fan unit:
1—in the normal version; 2—with the fan blades blanked off; 3—with the fan blades blanked off and without the turbine-engine blades.

The generator and the motor are connected directly. If a reducer is used, the latter must operate without slipping and must have constant ventilation and mechanical losses. As such a reducer one may use a chain drive operating without an oil bath.

Fig. 6. Test stand for testing a pneumatic motor.

The load generator is suspended on a balancing stand, and the power consumed by it is calculated from the results of measuring the load \(G\), which balances the balancing stand, and the number of revolutions of the pneumatic-motor shaft \(n_{\mathrm{dv}}\),

\[ N_{\Gamma}=\frac{G b n_{\mathrm{dv}}}{975\, i}\ \text{kW}, \tag{11} \]

where

\[ \begin{aligned} b &\text{— arm of the balancing stand, m;}\\ i &= \frac{n_{\mathrm{dv}}}{n_{\Gamma}}\text{ — gear ratio of the reducer.} \end{aligned} \]

Power is measured in \(\mathrm{kg\,m/sec}\), hp, or kW. The most widespread, in view of the broad use of electrical energy, is the measurement of power in kW; therefore it is logical to express the power of the generator and of the pneumatic motor in the same dimension, taking kW as the unit of measurement.

In testing the motor, the balancing stand does not measure the power expended in overcoming ventilation and mechanical losses in the part of the disk belonging to the fan impeller, the reducer, and the generator armature. Therefore the useful power of the motor is found by summing the generator load and the stand losses under the actual test conditions,

\[ N'_e = N_{\Gamma}+N_{tr.op}. \tag{12} \]

The ventilation and mechanical losses of the stand are determined on a special installation (Fig. 7). As the motor of this installation, the load generator suspended on the balancing stand is used, to which voltage is supplied from an external current source. The revolutions of the electric motor are varied by rheostats in the armature and field circuits, and the power developed by it is calculated from the torque and

...to the number of shaft revolutions. The voltmeter and ammeters are intended only for monitoring the operation of the electric motor.

Visible labels in Fig. 7 include: blade passages plugged; electric motor; \(B\); \(t_{cm}\); \(t_{am}\); \(G\).

Fig. 7. Installation for determining the ventilation and mechanical losses of the stand.

First the power expended by the motor on rotating the plugged fan impeller (Fig. 5) without the turbine-motor blades is measured (the weight of the load in these tests is \(G'_1\)); then the impeller is removed and the power expended on rotating the same installation, but without the impeller, is measured (the weight of the load in these tests is \(G'_2\)). The difference between the power consumed in the first and second tests corresponds to the ventilation losses in the disk of the turbine-motor and fan impeller. The part of the disk belonging to the turbine-motor impeller cannot be cut off, since there would then be nothing to fasten the plugs of the fan blades to; therefore the separation of the disk ventilation losses is carried out analytically:

\[ N_{tr.d}=\frac{b n_{ob}}{975}(G_1-G_2)\left(\frac{D_2}{D_1}\right)^5, \tag{13} \]

where

\[ N_{tr.d} \]
is the ventilation loss in the part of the disk belonging to the fan impeller, kW;

\[ G_1 \text{ and } G_2 \]
are the weights of the load on the pan of the balance stand, reduced to normal atmospheric conditions and to the revolutions of the motor shaft, kgf;

\[ D_1 \]
is the outside diameter of the disk tested, mm;

\[ D_2 \]
is the diameter of the fan impeller, mm.

The weight of the load balancing the balance stand is reduced to normal atmospheric conditions and, when testing with a reducer, to the revolutions of the pneumatic-engine shaft by the formula

\[ G=G'\frac{\gamma_n}{\gamma_{op}}\frac{1}{i}. \tag{14} \]

Summing the ventilation losses of the fan disk with the power expended on rotating the armature of the electric motor, the reducer, and the unit shaft, we obtain the losses of the stand reduced to the normal conditions of the test:

\[ N_{tr}=\frac{b n_{ob}}{975}\left[G_2+(G_1-G_2)\left(\frac{D_2}{D_1}\right)^5\right]. \tag{15} \]

The weight of the load \(G_1\) and \(G_2\), corresponding to the same number of shaft revolutions, is determined graphically from the experimental curves

\[ G_1=f_1(n_{\partial в}) \tag{16} \]

and

\[ G_2=f_2(n_{\partial в}). \tag{17} \]

Displacement of the stand due to vibration during the experiments is impermissible, since the additional expenditure of energy for this noticeably distorts the test results (Fig. 8).

Fig. 8. Weight of the load balancing the balance machine when measuring the ventilation and mechanical losses of the stand, reduced to \(\gamma_H = 1.2\ \mathrm{kg/m^3}\).

The dependence of the ventilation and mechanical losses of the stand on the number of revolutions of the motor shaft should be expressed graphically (Fig. 9).

Fig. 9. Ventilation and mechanical losses of the stand, not measured by the balance machine when testing a turbomotor.

To calculate the useful effective power of the motor by formula (12), the ventilation and mechanical losses of the stand, found from the curve in Fig. 9, are reduced to the actual test conditions:

\[ N_{tr.op}=N_{tr}\frac{\gamma_{op}}{\gamma_n}. \tag{18} \]

On the characteristic curve of the turbine motor (Fig. 10), the power and air consumption are indicated, reduced to normal test conditions by formulas (6), (9), and (10).

Fig. 10. Characteristic of the turbine motor of a VP-4 type fan unit.

From the power on the shaft of the pneumatic motor, its relative effective efficiency is calculated:

\[ \eta_{ce}=\frac{N_e}{N_{ad}} \tag{19} \]

and the fan efficiency is separated from the efficiency of the fan unit:

\[ \eta_v=\frac{\eta_{agr}}{\eta_{ce}}. \tag{20} \]

The fan efficiency is calculated with less error as the ratio of its useful (aerodynamic) power to the power on the shaft of the turbine motor, taken at the same number of revolutions of the unit shaft:

\[ \eta_v=\frac{N_p}{N_e}. \tag{21} \]

The maximum fan efficiency is found graphically (Fig. 11) from the curve \(\eta_v=f(Q)\).

Testing the fan at a constant number of revolutions leads to distortion of the actual mode of its operation, and therefore is not recommended.

Acceptance Tests

The manufacturing quality of the parts of the ventilator unit is monitored during production, while the technical and economic parameters are monitored at the testing station. The procedure for checking the dimensions, tightness, and strength of finished parts is outside our task; therefore, below we consider only the monitoring of the aerodynamic parameters of production ventilator units.

Fig. 11. Dependence of efficiency on the capacity of the VP-4 fan.

The impeller of the fan, the turbine blades, and the housing of the unit are cast from patterns; therefore it may be assumed that these parts are manufactured identically. The turbine nozzle is made detachable; its passages are milled and subsequently brought to the required dimensions by hand. The shaft of the unit rotates in rolling bearings. Thus, the most probable cause of discrepancies in the aerodynamic parameters of production ventilator units may be poor-quality manufacture of the nozzles or defects in assembly of the unit.

The air flow through the nozzle at a pressure drop above the critical value¹ does not depend on how it is tested—separately or in the assembled motor; therefore, it is recommended that the nozzles be tested for tightness and air flow before assembly of the ventilator unit.

According to our investigations, the power of the turbine motor depends significantly on the clearance between the oblique cut of the nozzle and the blades of the motor; therefore, when assembling the ventilator unit this clearance must be maintained with an error of not more than \(\pm 0.5\ \mathrm{mm}\).

A preliminary check of the accuracy of manufacture of the parts, the tightness of the joints, and the air flow through the nozzle makes it possible to simplify considerably the acceptance testing of ventilator units. In this case the acceptance test can be limited to comparing, under the same operating conditions, the shaft speed of the ventilator unit being tested with the shaft speed of a reference specimen. An example of the dependence of the shaft speed of the unit on the pressure of the air supplied to the motor is given in Fig. 12. The maximum correction to the shaft speed of the ventilator unit due to changes in the temperature of the compressed air and in atmospheric conditions usually does not exceed \(\pm 3.0\%\); therefore, to simplify the acceptance—

... tests it is neglected. The steady operating mode of the fan is most simply obtained by testing it without a ventilation pipeline.

Only a portion (not fewer than three) of arbitrarily selected production machines should be subjected to complete testing; the averaged characteristic of these machines is then indicated in the passport of the fan unit. Compiling the characteristic of the fan unit from the results of tests of experimental or carefully manufactured specimens leads to an understatement or overstatement of the actual technical and economic operating indices of production machines.

Operational tests

Under operating conditions, only those operating indices of local-ventilation fans should be checked which cannot be reliably studied in the laboratory, namely: the correctness of the choice of the design aerodynamic parameters of the fan unit, the strength and wear resistance of its construction, and the convenience of servicing and transporting it through mine workings. It is pointless to take the characteristic of a fan in a mine, since the aerodynamic conditions of its operation do not differ from laboratory conditions, while the production environment and the testing difficulties associated with it only increase the error of the results obtained.

Figure labels: fan-unit shaft speed, rpm; air pressure at engine inlet, atm; \(n = f(P_{\text{in}})\); with two compressed-air channels switched on; with three compressed-air channels switched on.

Fig. 12. Dependence of the shaft speed of the fan unit on the working pressure of the compressed air.

Under mine conditions, what is essentially tested is not the fan, but the ventilation installation as a whole, i.e., the fan unit in combination with the pipeline. The ventilation installation should be tested in the state in which it is found under actual operating conditions, i.e., taking into account the condition of the ventilation pipes and defects in the installation of the pipe string; and the true values of the quantities sought should be calculated as arithmetic means from the results of testing several identical ventilation installations.

The capacity of a fan operating in delivery is conveniently measured from the vacuum behind the collector, and the amount of air reaching the face—with a cup anemometer and a stopwatch. The method for measuring fan capacity from the vacuum behind the collector is described by M. Ya. Gemberzhevskii [3], and the magnitude of the discharge coefficients for various...

…collectors, taking into account the locations for sampling static pressure, was determined by V. I. Khonzhakov [12]. The method of measuring air flow with an anemometer is well known. These same methods of measuring air flow are used when the fan operates on suction with a nozzle-diffuser, and on discharge.

Measurement of static pressure in the pipeline is carried out with a liquid manometer or micromanometer through a tube fitted flush into the wall of the pipeline, or through a special attachment shown in Fig. 13.

Fig. 13. Static-pressure tube.

Having measured the capacity, pressure, and quantity of air delivered by the fan into the working face, one determines the leakage (or inflow) of air, the actual resistance of the pipeline, and compares them with the calculated values.

The quality of ventilation of a blind working is checked by analyzing air samples, which are taken in accordance with the instructions to the safety regulations in coal and shale mines [7, p. 92].

The fan installation must be tested under normal and forced operating conditions (after blasting the holes), with gradual extension of the pipeline to the maximum calculated length of the blind working. If it turns out that the aerodynamic parameters of the fan do not meet the requirements imposed on it, it is necessary to determine for what capacity and pressure the fan unit should be designed under the specified operating conditions.

The mechanical strength and wear resistance of the design of the fan unit, and the convenience of its maintenance and transportation through the workings, are determined in the course of operation (for not less than 3 months) by control disassembly, inspection, and questioning of the service personnel, with the drawing up of the appropriate reports. These reports also include the production workers’ comments on the design of the fan unit.

Instructions for Processing Observation Results

The accuracy of the experimental results is determined by the choice of the testing method, the quality of manufacture of the stand, the error of the measuring instruments used, and the experience of the experimenter. By determining the partial errors of individual measurements and summing them according to definite rules [2], one can establish the maximum possible and the most probable error of the result. In this connection, the principle of separately studying the sources of error is of substantial importance. It makes it possible to determine in advance the factors leading to the greatest error of the result, to pay special attention to them in the course of setting up the experiments, and thereby to reduce the error of the experiment.

The sensitivity of the instruments used for testing the fan unit is usually less than the change in the quantities being measured; therefore…

the readings of the latter remain constant in each experiment. Taking the above into account, the maximum possible error in reading the measured quantity is taken to be equal to one half of the value of a division of the instrument scale, and in accordance with this the errors of the quantities found from tables and graphs are subsequently established. To eliminate systematic errors of the instruments, their readings are corrected according to calibration curves. The measuring apparatus is checked against standard instruments (preferably of class 0.35). The accuracy of constructing the calibration curves must be no worse than \(\pm 0.5\%\) of the plotted quantity.

Mathematical operations with numbers are performed on a 25 cm logarithmic slide rule, and the results obtained are rounded according to the rules for operations with approximate numbers [13].

The choice of scales for the graphical representation of test results is made in accordance with the recommendations of M. E. Blanter [2, p. 108].

Conclusions

1. For a reliable comparison of different types of pneumatic fans, their testing and the processing of observational results should be carried out according to a unified methodology.

2. The aerodynamic characteristic of pneumatic fans must be given graphically for the most characteristic working pressures in the pneumatic network, with the maximum efficiency of the fan unit separated into the efficiency of the motor and of the fan.

3. The characteristic of the fan unit, taken at a constant working pressure in the pneumatic network, must be reduced to the normal parameters of ventilation and compressed air \((\gamma_n = 1.2\ \mathrm{kg}/\mathrm{m}^3;\ t_{vx} = 20^\circ\mathrm{C})\) according to the methodology proposed by us.

4. Presenting the aerodynamic characteristic of a fan unit in a table according to the “from and to” principle does not make it possible to judge correctly the technical and economic indicators of its operation; therefore such a method of specifying the characteristic should be considered unsuccessful.

References

1. Atlas of Fans and Deflectors. Aerodynamic Diagrams and Experimental Characteristics. Edited by K. A. Ushakov, TsAGI Proceedings, No. 172, 1934.

2. Blanter M. E. Methodology for the Study of Metals and the Processing of Experimental Data. Metallurgizdat, 1952.

3. Gembergzhevsky M. Ya. Testing of Centrifugal Fans in Pipelines. Industrial Aerodynamics. Collection No. 4. Publication of the Bureau of Scientific Information, 1953.

4. Zhiritsky G. S. Aircraft Gas Turbines. Oborongiz, 1950.

5. Zhumakhov I. M. Mine Fans. Ugletekhizdat, 1951.

6. Ilinchev A. S. Collected Works, Vol. 1. Mine Pneumatic Installations. Ugletekhizdat, 1953.

7. Instructions to the Safety Rules in Coal and Shale Mines. Ugletekhizdat, 1952.

8. Nevelson M. I. Centrifugal Fans. Gosenergoizdat, 1954.

9. Polikovsky V. I. Fans, Blowers, Compressors. Mashgiz, 1938.

10. Rules No. 169 for Measuring the Flow Rate of Liquids, Gas, and Steam by Means of Nozzles and Diaphragms, and Guiding Instructions to Rules No. 169. Main Administration of Measures and Weights, 1938.

11. Mine Axial Pneumatic Fans for Auxiliary Ventilation VP-3. Instructions for Operation and Maintenance. Krivoi Rog, 1949.

12. Khonzhakov V. I. Aerodynamic Characteristics of Collectors. Industrial Aerodynamics. Collection No. 4. Publication of the Bureau of Scientific Information, 1953.

13. Yakovlev K. P. Mathematical Processing of Measurement Results. State Publishing House of Technical-Theoretical Literature. Moscow, 1953.

1. For air, the critical pressure drop is
   \[ \frac{P_{\mathrm{ex}}}{P_{\mathrm{atm}}} > 0.528 . \] ↩