In an inductive flow meter, the magnetic field generator that generates
a magnetic field within a tube, is connected to a source of alternating
voltage via an electronic switch. The switch is controlled such
that it becomes conductive only upon a zero crossing of the alternating
voltage, respectively, and remains conductive until the next successive
zero crossing of the exciting current. The useful voltage taken
at the electrodes is integrated in an integrator, whereby the parasitic
voltage coupled into the electrode loop by self induction becomes
zero. In order to eliminate the influences of retentivity, each
current pulse, at which integration is performed, is preceded by
1. An inductive flow metering method, comprising:
providing a magnetic field generator,
connecting the magnetic field generator to an alternating voltage
via an electronic switch,
providing a pair of electrodes,
providing a control circuit having a zero crossing detector for
detecting zero crossings of the alternating voltage,
controlling the switch via the control circuit so that the switch
assumes a conductive state at a zero crossing of the alternating
voltage and is switched out at a zero crossing of an exciting current,
connecting an integrator to the electrodes,
integrating electrode voltage over an integration period equal
to at least one cycle of the alternating voltage, the integration
period starting substantially simultaneously with the switch assuming
a conductive state, and
generating a pre-pulse of the switch prior to the integration period,
during which the integrator is inactive and the switch carries current
of the same polarity as during the integration period.
BACKGROUND OF THE INVENTION
The invention relates to an inductive flow meter wherein a transversely
directed magnetic field is generated within a tube, the magnetic
field, together with an electrically conductive liquid flowing in
the tube, generating a voltage which is detected by electrodes arranged
at said tube and is evaluated. Such an inductive flow meter is known,
among others, from US Pat. No. 5018391 to Doll.
In the above mentioned known flow meter, positive and negative
exciting currents are generated alternatingly in the magnetic field
generator, the currents in turn generating a magnetic flux density
B within the tube. The exciting current flows for less than one
cycle of the applied alternating voltage, respectively. In an evaluating
circuit connected to the electrodes, the integral with respect to
the respective cycle of the alternating voltage is formed. Since
the forming of the integral is performed for exactly one cycle,
parasitic voltages coupled in from the mains and having the same
frequency as the exciting voltage are compensated. The value of
the integral at the end of the mains cycle is supplied to a differential
amplifier supplying the useful signal which may afterwards be freed
from further parasitic signals. The principle of the known flow
meter is based on the fact that the magnetic flow density B.sub.2
at the end of the cycle used for forming the integral is equal to
the magnetic flow density B.sub.1 at the beginning of the excitation.
However, in practice, this condition is met only approximately.
While the exciting current is zero at the beginning and the end
of the integration time, the flux density B is not. This is due
to the retentivity inherent to the iron of the magnetic field generator.
This retentivity has the effect that, at the beginning of the integration
interval, the flux density still has a value derived from the previous
exciting pulse of opposite polarity and falsifying the measuring
result, since the integral of the inductive parasitic voltage ##EQU1##
will be eliminated only if B.sub.2 =B.sub.1.
SUMMARY OF THE INVENTION
It is an object of the invention to provide a flow meter in which
integrals are formed over an integral cycle of the exciting voltage,
with no measuring errors being caused by retentivity or eddy currents.
The flow meter of the present invention corresponds in its general
structure and its effect to that of U.S. Pat. No. 5018391. Different
from this known flow meter, however, a pre-pulse is generated prior
to each integration time, the pre-pulse having the same polarity
as the pulse in the successive integration time in which the integral
is formed and the evaluation is made. The prepulse erases the "memory"
of a previous pulse of opposite polarity and the magnetic field
generator is prepared for the polarity of the next measuring pulse.
Thus, it is achieved that at the beginning and at the end of each
measuring pulse the values of the flux density are equal (although
they are other than zero) so that no difference falsifying the result
of the measure exists.
Although a single pre-pulse is sufficient before each integration
time, also a plurality of pre-pulses may be generated, e.g., from
one to four pre-pulses. On the other hand, a pre-pulse could also
be followed by a plurality of measuring pulses, e.g., 2 3 or 4
which would then be evaluated with a correspondingly longer integration
BRIEF DESCRIPTION OF THE DRAWINGS
The following is a detailed description of an embodiment of the
invention taken in conjunction with the accompanying drawings.
In the Figures
FIG. 1 is a schematic block diagram of the inductive flow meter,
FIG. 2 is a diagram of different signals occurring in the circuit
of FIG. 1 .
DESCRIPTION OF THE PREFERRED EMBODIMENT
In a manner known per se, the flow meter comprises a tube 10 of
a non-conductive material, in which flows the liquid, the flow rate
of which is to be measured. The tube 10 is provided with the magnetic
field generator 11 which in the present case comprises two coils
11a and 11b being arranged coaxially with respect to each other
at opposite sides of the tube and being electrically connected in
series. An exciting current flowing through both coils 11a, 11b,
generates a magnetic field inside the tube 10 that extends transversely
through the interior of the tube.
Connected in series to the electronic switch 13 the series connection
of the coils 11a and 11b is connected to the supply network through
the lines 14 and 15 so that the series connection of switch 13 and
coils 11a and 11b may be connected directly to the power supply
of, e.g., 220 V and 50 Hz.
At the inner wall of the tube 10 two electrodes 16a, 16b connected
to the evaluating circuit 17 are arranged opposite each other.
The evaluating circuit 17 comprises an amplifier 18 connected to
the electrodes 16a, 16b, the signal of which is supplied to an integrator
19. The output signal of the integrator 19 is supplied to two sample
and hold circuits 20a and 20b, the outputs of which are connected
to the two inputs of a differential amplifier 21. The output signal
23 of the differential amplifier 21 is proportional to the flow
rate of the liquid in the tube 10.
The amplitude of the alternating voltage on line 14 is supplied
to the control electronics 25 via the line 24. This control electronics
includes a zero crossing detector which generates a pulse upon each
zero crossing of the alternating voltage. The top line a) of FIG.
2 illustrates the course in time of the alternating voltage on lines
14 and 15 being designated as U. Line c) shows the needle pulses
26 generated by the zero crossing detector during the respective
zero crossings of the alternating voltage U.
In the top line a) of FIG. 2 the course of the exciting current
flowing through the magnetic field generator 11 is referenced as
i. The electronic switch 13 in the form of a triac is controlled
by the control electronics 25 through pulses Q.sub.0 supplied to
the gate thereof, such that it is controlled to the conductive state
during each respective zero crossing of the alternating voltage.
The triac remains in the conductive state until the current i, which
is phase shifted with respect to the alternating current U, has
become zero. Thyristors or triacs are known to be controlled to
the conductive state when the main current becomes zero without
there being a control voltage present at the gate. According to
FIG. 2d), the signal Q.sub.0 extends over one half-wave of the alternating
voltage U. Thus, the triac 13 may take the reversed state in the
next half-wave, when the current i becomes zero.
In the present embodiment, a second pulse Q.sub.0 is generated
after the first pulse Q.sub.0 during the second zero crossing of
the alternating voltage U, and three zero crossings after the beginning
thereof, a first pulse Q.sub.0 is generated again, etc. The first
pulse is a pre-pulse VP, respectively, which appears one cycle of
the alternating voltage before the main pulse HP.
In dependence on the zero crossing signals 26 the control electronics
25 generates the pulses Q.sub.1 illustrated in line e) of FIG.
2 which last for one cycle of the alternating voltage U, i.e.,
for one positive and one negative half-wave. Each pulse Q.sub.1
starts with the zero crossing of the alternating voltage U at which
the triac 13 is controlled to the conductive state by a main pulse
HP. These pulses Q.sub.1 control the integrator 19 such that it
performs integration only during the integration times TI defined
by the pulses Q.sub.1 and that its output signal is then reset obligatorily.
Since parasitic voltages occurring have the frequency of the alternating
voltage U (or a multiple thereof), the integral of this parasitic
voltage becomes zero over one full cycle of the alternating voltage.
In contrast thereto, the integral of the intensity of the magnetic
field within the respective cycle yields the value B, the magnitude
of which corresponds to the hatched area in FIG. 2b. The useful
voltage produced at the electrodes 16a and 16b is proportional to
the area B. At the end of each signal Q.sub.1 a useful signal is
present at the output of the integrator 19 the magnitude thereof
corresponding to the flow velocity and the integral B with respect
to time of the immediately preceding field pulse. This output value
of the integrator 19 is supplied to a sample and hold circuit 20a
or 20b while being clocked by the signal Q.sub.2 generated by the
control electronics. The signal Q.sub.2 is a short pulse generated
immediately after the end of signal Q.sub.1. The signals Q.sub.2
are timed such that a respective one of these signals alternatingly
activates the sample and hold circuit 20a, while the next signal
activates the sample and hold circuit 20b. Thus, the positive useful
signals, which in FIG. 2b) lie above the time axis t, are latched
into the sample and hold circuit 20a, whereas the negative time
integrals that lie below the time axis t are latched into the sample
and hold circuit 20b. It is the function of the sample and hold
circuits to accept and latch the respective output signal of the
integrator 19 upon a pulse Q.sub.2 and to keep it latched until
the next successive pulse Q.sub.2.
The positive input of the differential amplifier 21 is connected
with the sample and hold circuit 20a and its negative input is connected
with the sample and hold circuit 20b. Since the value contained
in the sample and hold circuit 20b is negative, the differential
amplifier 21 adds the two amounts of the contents of the sample
and hold circuits. At the output of the differential amplifier 21
a voltage arises which is proportional to the voltage between the
electrodes 16a and 16b and represents the measured result.
Line b) in FIG. 2 represents the course of the magnetic flux density
B of the magnetic field generator 11. While the signal Q.sub.0 performs
a pre-pulse VP, the retentivity is inverted in polarity, so to say,
and set to the polarity of the half-wave of the alternating voltage
U prevailing during the pre-pulse VP. In this phase, the integrator
19 is still inactive. The flux density B rises, e.g., from the negative
region, takes its positive maximum and falls with the exciting current
i. When the exciting current i has become zero, a retentivity value
of the flux density B larger than zero will remain. Upon the next
main pulse HP of the signal Q.sub.0 the flux density will rise
from this retentivity value, only to fall to the same value thereafter.
Thus, the integration time TI ends with the same value of the flux
density B it began with. It will be appreciated that the flux density
B rises to a lower value during the pre-pulse VP than it does during
the main pulse HP. In the successive polarity inversion, first,
a pre-pulse is performed with a negative direction of current and
then a main pulse is performed, also with a negative direction of
current. For the proper measuring interval, i.e., the integration
time TI, only the portion with the main pulse HP is evaluated.