The present invention relates to a device for automatic-measurement of solid-volume conveyed by a gas through a pipe line.
The invention relates more particularly to an instrumentation system for measurement of mass flow rate of bulk solid material like sand, lime dust and coal dust along with a gas medium such as air through a pipe line of electrical nonconductive material like Perspex glass and flexi glass, without causing any obstruction to the flow of the said gas-solid by the sensing transducers used, in which the instantaneous velocity of the bulk solid is determined by using basically the instrumentation system disclosed in the granted Indian Patent Application No. 3004/DEL/1997 filed on 20th October 1997, and the instantaneous volumetric concentration of the bulk solid in the gas-solid medium flowing through the pipe line is measured by using capacitance based sensor along with processing circuit and related software and the mass flow rate of bulk solid is computed automatically from the signals produced in the said two measurements by means of a personal computer (PC) provided with softwares developed for the purpose and the results of computation are displayed suitably by means such as of CRO, tape recorder and teleprinter to the users of the instrumentation system developed.
The measurement of mass flow rate of bulk solid in the two-phase gas-solid medium conveyed through a pipeline
instaneously and precisely is required by the industries for control of processes. At present no such means is available for the industries like steel, coal, cement. There is therefore a need for developing an instrumentation system for the purpose to eliminate the time-consuming method followed at present by measuring directly the amount of bulk solid conveyed in a given time through the pipe line.
The object of the present invention is to provide an instrumentation system for displaying continuously the mass flow rate of bulk solid conveyed in two-phase gas-solid medium through a pipe line for process control in industries.
The mass flow rate of bulk solid conveyed in two-phase gas-solid medium conveyed through a pipe line is obtained from the following relation (1) :-
MS = A p α Vs ... (1)
where MS = Mass flow rate of the conveyed bulk solid in g/sec,
A = Internal cross sectional area of the pipe line
2 in cm ,
p = Bulk density of the conveyed bulk solid in
g/cm3,
α = Volumetric concentration of the conveyed bulk
solid in the gas-solid medium, being less than 1
and greater than 0,
VS = Velocity of the conveyed bulk solid in cm/sec.
In the invented instrumentation system, the velocity of the conveyed solid, VS, is measured by using two electro-
dynamic type transducers; the volumetric concentration of the conveyed solid, α is measured by using a transformer-ratio-arm-bridge capacitance type transducer; and the mass flow rate of the conveyed solid, MS, is computed by converting first the analog signals obtained for VS and α into corresponding digital signals and then applying the cross correlation technique for measurement of VS by means of a personal computer provided with software programs developed for the purpose for given values of the internal cross sectional area, A of the pipe line used and the bulk density p of the conveyed bulk solid. The invented instrumentation system is described fully and particularly in an unrestrictive manner with reference to the accompanying drawings in which -Figure 1 is a schematic layout of the signal flow in the
invented instrumentation system; Figure 2 shows the cross section of the pipe-line fitted with capacitance type transducer for measuring the volumetric concentration of the bulk solid conveyed through the same; Figure 3 is the electronic circuit used for measuring the
volumetric concentration of the conveyed bulk solid with a transformer-ratio-arm- bridge capacitance type transducer; Figure 4 is the schematic experimental set-up used for
calibration of the invented instrumentation system and comparison of the values of the mass flow rate of the conveyed solid obtained by using the invented
instrumentation system and those obtained by direct measurement of the mass of the bulk solid conveyed in a given time; and Figure 5 shows the comparative values of the mass flow rate of the conveyed bulk solid obtained by using the invented instrumentation system and those obtained by direct measurement of the mass of the solid conveyed in a given time in a number of experiments performed using the set-up shown in Fig. A. Referring to Fig. 1 one transformer-ratio-arm bridge capacitance type transducer (T) and two electrodynamic type transducers (E1 and E2) are fitted on the outside surface of a pipe line (P) of electrical non-conducting material such as perspex, glass and flexi glass conveying a gas-solid medium (M) through it, Transducer (T) is connected to unit (VC) for producing volumetric concentration signal (VOL) and transducers (E1 and E2) are connected respectively to units (ES1 and ES2) for producing signals (VEL1 and VEL2) corresponding to velocity of the bulk solid conveyed through pipe line (P) at the respective locations of E1 and E2 thereon. Signals (VEL1 and VEL2) are fed to cross correlator unit (CC) to produce the velocity signal (VEL) of the bulk solid flowing in the part of the pipe line (P) lying between locations of E1 and E2 thereon. Signals (VOL and VEL) are fed at the input of a personal computer (PC) along with the data relating to the internal cross sectional area (A) of the pipe line and bulk density (p ) of the bulk solid to produce signal (MFR) in the output of the (PC)
corresponding to the mass flow rate of the bulk solid, which is displayed in unit (DPY).
Referring to Fig. 2, two arcuates (18) of stainless steel strip, are attached to the outside wall of flexi glass pipe line (13) in diametrically opposite positions, subtending an angle (8) at the axis of the pipe line. The arcuates are wrapped with insulating material (16) such as paper, which in turn is wrapped with stainless steel strip (15). In a particular embodiment pipe line (13) is of internal radius 6.35 mm, of external radius 9.25 mm and angle (2 ) is 120°.
Referring to Fig. 3, transformer-ratio-arm-bridge capacitance type transducer (T) is excited by a 10 KHz, 20V AC supply (SI). The imbalance output signal of (T) is amplified by an operational pre-amplifier (Al). The ratio-arm transformer (TR) is a balanced torroid of ratio 1:1:1 wound on a ferrite core. The output of (Al) is amplified by bandpass amplifier (A2). The output of (A2) is supplied to the capacitance feedback amplifier (A3), buffer (A4), comparator amplifier (A5) and then to the phase sensitive detector comprising two MOSFETS (Ml and M2), such as MOSFET P(BS250) and MOSFET N(BS170) respectively. The outputs obtained at the source terminals (S) of the two MOSFETS are supplied at +Ve input terminal (2) and -Ve input terminal (1) of the Instrumentation Amplifier (IA) operating from a 12V DC supply. The output signal obtained between terminals (9) and (6) of the IA corresponds to the volumetric concentration (a) of the bulk solid, which is displayed for the users of the instrumentation system.
The functions performed by the PC is controlled by software programs developed in C-language. The main functions achieved by means of the software are s
(i) Data Acquisition, (ii) computation of velocity (VS) of Bulk Solid by cross Correlation Technique, (iii) computation of volumetric concentration (α) of the conveyed bulk solid and (iv) computation of mass flow rate (MS) of the conveyed bulk solid.
The software developed converts the all three analog signals which are acquired simultaneously into 12 bit digital data using a PCL-2Q8 type ADC card. The digital data are taken from the ports of the PCL-208 type ADC card to the RAM of the PC and transferred to a data file on hard disk for data storage. Commands for multiplexing and triggering of the ADC card and the data transfer are given by executing a program (adc.c) written in C language.
The algorithm used for data acquisition is presented below :-
i. Initialise the Base, Trigger, Multiplexing channel, End of conversion (EOC), Control and High Byte and Low Byte data registers as per the addresses given in the PCL-208 mannual.
ii. Take input file name (adc.c), number of readings to
be taken (user prescribed - 1000 in the present case) Note that the sampling time of the data - number of
readings x number of loops (set by operator=20 in the present case) x execution time. The number of readings and the number of loops are set in a file set.dat.
iii. Assign the multiplexer scan range to select the number of input channels to be used (3 in the present case), iv. Give the triggering command for starting A/D conversion for the sample of data, v. Check end of A/D conversion. If no, then check again after an interval, vi. If yes, then take the Low byte and High byte of (hex) data from respective registers to an array after converting to decimal and then increment the array add res s. vii. Assign the delay corresponding to the sampling rate, viii. Repeat step (iv) to (vii) for the total number of readings. ix. Transfer all the data taken from the array in RAM to a data file (avco.dat).
It is to be noted that the output gives the time in milliseconds, the readings (channel 1 and 2) for velocity measurement and the readings (channel 3) for volumetric concentration measurement. Although the sampling rate of the PCL-208 for a single channel is 60 KHz the speed of the data acquisition can be reduced to suit the required sampling rate by providing the loop delay in the program. The calibration of the A/D card has to be done to know the exact sampling rate corresponding to a loop delay and number of channels to be read.
For example, a square wave of frequency 0.5Hz and a loop delay of 100 gives an average of 328 number of sample in one half of the time period i.e. in 1000 ms (for a frequency of 0.5 Hz). These numbers are accurately found from the data file by counting the sample number between two consecutive sign changes. Now the number of samples are averaged for half of the time period from at least 5 to 6 time periods or more for accuracy of the measurement of time gap. Thus the time gap between two samples for a delay of 100 ms appeared to be (1000/328) ms = 3.049 ms. This time is essential in the cross-correlation program to find the velocity. Table 1 depicts the time between two samples for different loop delays as obtained experimentally. These experiments were performed by the data acquisition of square waves of magnitude 2 V from a precision signal generator (HP-331AA) with different frequencies and for different loop delays.
Calculation of velocity in the present case is done by normalised cross-correlation function. This is an improvement of the digital cross-correlation technique.
The expression for the digital cross-correlation
function is
(Equation Removed)
Where, N : number of sampled data for cross-correlation.
J : shift in number of samples (0 4000.0, correlation function<0.4 or VS<200.0. vi. Record the correlation function (Rxy) computed for
each value of time shift in the output file corr.dat and the velocity in the file result 1.dat. It is important to note that the sampling interval of the signal x and y determines the bandwidth as well as the time response of the overall system. The number of sampled data for cross-correlation (N) is also important. N has to be chosen as a trade-off between the noise rejection and sharpness of the
cross-correlation function. Theoretically, cross-correlation is immune to random noise. But that is true when N is very large. Again if we increase N very much the time response of the overall system will be slow because of calculation burden and the correlation function will also be relatively flatter. The increased flatness of the correlation function may reduce the resolution of the velocity measurement. Because, after calculating the correlation function, the peak of the correlation function is found out to compute the time delay between the electrodes.
The volumetric concentration measuring circuit suffers from the problem of constant voltage drift. So, the change in the root-mean-squared (r.m.s.) value of only the a.c. component of the volumetric concentration circuit output voltage is calculated. The change in d.c. level of the voltage due to the flow is not considered in order to neglect the effect of the d.c. voltage change due to the base line drift of the output voltage. This scheme, however, suffers from the limitation that it can be used only for dynamic measurement. The software that implements this is termed (avdata.c.)
The algorithm used for the program (av data.c) for volumetric concentration measurement is given below :
(i) Read the data of the concentration channel from the
file (avco.dat) in a two-dimensional array x [2] [i] where 2 denotes the concentration channel (0 and are used for the velocity channels) and i varies from 0 to the total number of readings (n).
(ii) Calculate the average of the concentration channel readings as
(Equation Removed)
It is to be noted that x [2] [0] is neglected in
average calculation since it is a faulty reading.
(iii) Subtract avg from each of the n readings and then
square them individually. Then add all these
readings that is
(Equation Removed)
(iv) Obtain concentration by multiplying value by a gain
factor of 10.
The values obtained for different known values of
volumetric concentration is then plotted during calibration to
obtain a power series relationship between the two from which
any unknown concentration can be predicted.
The algorithm used for program (con-cali.c) for
software (con-cali.c) which combines all the three said
programs for computing the mass flow rate comprises the following
steps :-
(i) Acquire the data using the program adc.c
(ii) Calculate the instantaneous velocity using the
program proj2.c
(iii) Calculate the volumetric concentration using the
program avdata.c Civ) Determine the mass flow rate.
For the ease of mass flow rate calculation, this programme is actually run in two parts.
During the first part, there is no solid flow. In this connection, a set of five data is taken where each data comprises of instantaneous velocity and instantaneous concentration readings. Since there is no solid flow, the instantaneous velocity is automatically zero and the instantaneous concentration is a value which corresponds to air-capacitance between the sensor plates.
The average of the first five concentration readings, taken under the condition of no solid flow, is assumed to be the initial concentration value.
After the initial set of five readings are taken, the second part of the programme is initiated. The solid flow is started and simultaneously keyboard is hit so that, the next set of five readings can be taken under the running condition.
Now set of five readings of velocity and volumetric concentration are obtained and the average of these five readings are computed, for both velocity and volumetric concentration.
The change in volumetric concentration value (i.e. new. concentration value - old concentration value) gives the volumetric concentration information.
With values of volumetric concentration (α), velocity (VS), the mass flow rate (MS) is found from :-
Ms = A p α Vs K (8)
Where K is the meter factor which is found by calibration of
the instrument.
The experimental set up for the measurement of mass flow rate of the bulk solid in a pneumatic conveying pipe line shown in Fig. A, consists of the following major subsystems :
Carrier air supply source (S), solid charge in bin (6),
and volumetric concentration of solid,,
flexi glass pipe line (P) for measuring velocity of solid and
flexi glass pipe line (Q) for collection of solid in collecting tank (12) for a given time of gas-solid flow through pipe line (P).
Compressed air is supplied to the set-up through air supply lines (11A and 11B). Air from the same source(S) is used as carrier for the solid particle as well as for pressurising the charging bin (6) to ensure smooth flow of solid in the system. Air supply rate is controlled with the help of a control valve (2) and a bypass valve (3). A rotameter (5) is incorporated in the carrier air line to measure the air flow rate. The carrier air pressure can be noted from a manometer connected in the down stream of the rotameter at (X). The air supply line is connected to one end of the T - mixer (T).
The charging bin (6) is a mild steel cylindrical vessel (having dimensions : length = 470 mm, diameter = 250 mm). At the bottom it is conical at an angle of 45 degree for easy flow of solid under the combined action of gravity and top pressure. The bin is provided with a removable top cover, a
pressure gauge (10), a vent cum filling line (11A') and a sight glass (10A) for inspection of the material level in the bin. The bottom end of the conical section is connected to T - mixer (T) through a short connecting pipe (11A") and a quick closing valve (7). Solid-air mixture is formed at the T - mixer which moves further downstream of the conveying line.
The horizontal conveying line (11A*'f) is connected to the vertical test section through a bent pipe (11), a flexible tube and a union joint (z). The pipe line (P) is 884 mm in length and 12.7 mm in diameter and connected between two quick closing valves (8 and 9). The handles of these two quick closing valves are attached to a single lever to actuate them simultaneously. For measurement of differential pressure across the test pipe line (?), pressure tappings are taken from the upstream and downstream of the test pipe line at X2 and X1 and connected to a manometer (MM). A rough estimate of the average solid concentration can be obtained from the pressure drop in the test pipe line between positions X1 and X2.
In the pipe line (P), three sensors (C, 1 and E2) are mounted at different axial locations. The lower one (C) is a capacitance sensor for measurement of volumetric concentration while the upper two ( and "E2), separated by 60 mm, are electro-dynamic sensors for velocity measurement using cross-correlation technique. All the sensors are connected to the measuring circuit via two - core shielded cables. The constructions and the dimensions of the sensors have been already described herein.
The solid collectingpipe line (Q) is 1105 mm long and 25.4 mm in diameter transparent flexi glass tube connected in the downstream of the test pipe line (P) by a 90° bent short pipe piece (11E) and solid particle collecting tank (12). The top of the pipe (Q) is closed by a wire mesh (w) with small perforations through which gas escapes to the atmosphere. The solid flows through pipe (Q) and is collected in the tank (12).
Compressed air at around 2-3 kg/cm2 pressure is
supplied to the air supply lines. The conveying line is purged
and air-flow is established by operating the control valves and
by-pass valve. Meanwhile the solid charging system is
pressurised with compressed air by operating control valve (l)
and a pressure regulator (A) to maintain the top pressure in
bin (6). The air-solid two phase flow is established by opening
quick closing valve (7) below the charging bin (6) which allows
the solid particles to flow to the T - mixer where air-solid
mixture is produced. This two-phase mixture flows downstream
through the conveying line and pipe (P) in the form of clouds.
When the tv/o-phase flow is established, simultaneous closing
of the two quick closing valves (8,9) across the pipe (P) traps
the solid-air mixture within pipe (P). The solid being heavier
settles down in the bottom of pipe (P), remaining portion of
which is filled with the air. The volumetric concentration of
the solid in pipe (p) can be estimated from the volume of the
solid collected in the test section. This can be used for the
calibration of the system. The solid collected in pipe (P) can
be removed by opening the union joint (z) and the flexible pipe
at the bottom thereof. Finally the solid collected in tank (12)
for a recorded period of time is weighed to calculate the mass flow rate. This can be compared with the mass flow rate measured by the system and used for calibration of the system.
The equation (1) provides the basis for the calibration of velocity and mass flow rate measurement. The volumetric concentration (α) of the pneumatically conveyed solid particles in pipe (P) at any given time is the ratio of volume of solid particles in the pipe to the total volume of the pipe (Vts). Mathematically α can be expressed as;
(Equation Removed)
where W_ = the weight of the solid (hold, up) in the test section at any given time,
p= the bulk density of the solid,
A = area of cross-section of the pipe,
Lts = length of the pipe between X1 and X2.
For the known density of solid and the hold up (Ws), the volumetric concentration can be estimated using equation (9). Now replacing in the equation (1) from the equation (9) and rearranging, the velocity of the solid particles VS can be written in terms of Ms and Vs as:
Thus the velocity of the solid can be estimated by the known mass flow rate (MS) and the hold up (Ws) in pipe (P) between X1 and X2. The average mass flow rate is estimated by weighing the solid collected at the collecting bin (12) over a given period of time and dividing the mass collected by time.
The experimental set-up (Fig. 4) is especially
designed and developed to facilitate the calibration of velocity
measurement, concentration measurement and mass flow measurement.
In this system the average mass flow rate is estimated
experimentally by collecting the solid particles over a suitable
period of time in a steady flow condition. The time is noted
by a stop watch. The collected solid particles are weighed in
a high accuracy weighing machine and an average mass flow rate
(MS) is calculated by dividing this weight by the duration of
the measurement. This mass flow rate (MS) is taken as reference
for calibrating the mass flow meter.
Calibration of the volumetric concentration
measurement by capacitance transducer is achieved by a reference
volumetric concentration calculated experimentally by known mass

flow rate (MS) and known hold up (WS) using equation (10).
The above procedure is repeated for different flow conditions by varying the air flow rate and the charging bin pressure for proper calibration of the instrumentation system.
The comparative values of mass flow rate as measured by using the invented instrumentation system and estimated from direct measurement on experimental set-up (Fig. 4) have been shown in a graphical form in Fig. 5, from which it is noted that there is a fairly close agreement between the two results.
Table : Calibration for loop delay
(Table Removed)

We claim:
1. A device for automatic-measurement of mass flow rate of solid materials like sand,
lime dust and coal dust conveyed by a gas through a pipe line, for process control in
industries, comprising (a) units (ESI and ES2) for measuring the velocity of the bulk
solid flowing through a pipe line formed of non-conductive material like flexi glass; (b) a
unit (VC) for measuring the volumetric concentration of the bulk solid flowing through
the pipe line, and (c) a personal computer (PC) as herein described for computing the
mass flow rate of the bulk solid through the pipe line, characterized in that the unit for
measuring volumetric concentration comprises
a capacitance type transducer (T) which consists of two articulates (18) attached to the outside wall of pipe line (13) in diametrically opposite positions, each subtending an angle 9 with the axis of the pipe line and wrapped with insulating material; and
an electronic circuit for measuring the volumetric concentration of the conveyed bulk solid.
2. The device as claimed in claim 1 wherein the electronic circuit for measuring the
volumetric concentration of the conveyed bulk solid comprises:
- operational amplifier (Al)
- ratio-arm transformer (TR)
- bandpass amplifier (A2)
- capacitance feed back amplifier ((A3)
- buffer (A4)
- comparator amplifier (A5)
- phase sensitive detector
- instrumentation amplifier (IA)
- capacitors (C1 to C11)
-resistors (R1 to R15)
3. The device as claimed in claim 1 and 2 wherein, the articulates (18) are wrapped with insulating material (16) such as paper, which in turn wrapped with conducting material such as stainless steel strip (15).
4. The device as claimed in claim 1 wherein the subtending angle 0 is 120°
5. The device as claimed in claim 1 and 2 wherein the ratio-arm transformer (TR) is a balanced torroid of ratio 1:1:1 wound on a ferrite core.

6. The device as claimed in claim 1 and 2 wherein the. phase sensitive detector comprises two MOSFETS (Ml and M2).
7. The device as claimed in claim 6 wherein MOSFETs are MOSFET P(BS250) and MOSFET N(BS170).
8. A device for automatic-measurement of mass flow rate of solid materials like sand, lime dust and coal dust conveyed by a gas through a pipe line, for process control in industries, substantially such as hereinbefore described in accordance with the accompanying drawings.