Abstrict A magnetic flow meter nulling system wherein each electrode has
connected thereto an in-phase voltage adjusting network and a phase-shifted
voltage adjusting network. In particular, each electrode is connected
in a loop with a slide wire of a potentiometer. The slider of one
potentiometer is connected directly to the measuring circuit whereas
the other potentiometer's slide wire has the slide wire of a third
potentiometer connected across, the said other potentiometer having
its slides connected directly to the measuring circuit and the slider
of the third potentiometer being connected to the measuring circuit
via a capacitor.
Claims Having described my invention as required by the Statute, I Claim:
1. An electromagnetic flowmeter system, including in combination,
a. a primary element for producing an electrical potential difference
proportional to the flow rate of flowing electrically-conductive
material, and
b. a detecting and measuring circuit for measuring said electrical
potential difference such as to produce a measure of the rate of
flow of said material through said primary element;
c. said primary element having field means for producing an AC
magnetic field in the path of said material and for directing said
field transverse to the direction in which said material flows through
said primary element;
d. said primary element having a first electrode, the electrical
potential of which is due to voltage at a first location in said
material, due to flow of said material through and transverse to
the direction of said field;
e. said primary element having a second electrode, the electrical
potential of which is due to voltage at a second location in said
material, due to said flow of said material, said locations being
spaced one from the other along a line transverse to said flow;
f. said primary element having a first lead circuit connected to
said first electrode and a first output terminal for presenting
a first output voltage corresponding to the potential of said first
electrode;
g. said primary element having a second lead circuit connected
to said second electrode and having a second output terminal for
presenting a second output voltage corresponding to the potential
of said second electrode;
h. said first lead circuit having therein an adjusting element
for adjusting the amplitude of said first output voltage with respect
to the potential of said first electrode;
i. said second lead circuit having therein an adjusting element
for adjusting the amplitude of said second output voltage, with
respect to the potential of said second electrode;
j. said first lead circuit also having an adjusting element for
adjusting the phase of said first output voltage with respect to
the phase of said second output voltage; each said lead circuit
comprising a pick-up loop, said loop incorporating the corresponding
said electrode, and each said lead circuit including therein, remote
from the corresponding said electrode, a potentiometer of which
the slide wire is a series element of such loop, said slide wire
being connected to an input terminal of said detecting and measuring
circuit.
2. The electromagnetic flowmeter system of claim 1 wherein each
said lead circuit includes pick-up loop means for generating, in
said lead circuit, voltages due to alternating electromagnetic fields
adjacent to such lead circuit;
k. said second lead circuit further comprising a phase adjusting
element.
Description FIELD OF THE INVENTION
This invention pertains to a system for nulling (i.e., adjusting
to a minimum) the zero-or no-flow signal from a electromagnetic
flowmeter.
THE PRIOR ART
Many previous nulling systems have been purely mechanical in nature,
involving means of physically moving the electrode lead wire in
the magnetic field until a minimum no-flow output signal is reached.
A variation on this has been a mechanical device which instead of
moving the lead wires, moved conductive or magnetic elements to
induce eddy currents or otherwise distort the magnetic field in
the vicinity of the electrode leads. Again, these are positioned
for a minimum no-flow signal. With mechanical devices, the element
adjusted must be located in the magnetic field. Therefore, the meter
must either be nulled while in a state of partial disassembly or
remotely located means mechanically connected to the adjustable
elements must be provided for adjusting it in assembled state.
Other nulling systems have used electronic means, allowing the
adjustment to be made on conveniently located potentiometers.
The simplest of these, using potentiometers having electrode lead
loops, attempts to electrically adjust the "virtual center"
of the leads, or of the electrodes themselves, for a minimum null.
Although these systems work, they are not perfect because an amplitude
adjustment only is available, but not phase. Hence, at null there
generally is a slight phase difference between the electrode no-flow
signal and the signal induced in the lead loop. This makes complete
cancellation impossible.
Other electronic systems have included both amplitude and phase
adjustment. These, however, have been unduly complex, generally
requiring one or more transformers or at best, three leads to each
electrode (two leads being for a signal pick up loop and one lead
being a direct electrode connection).
SUMMARY OF THE INVENTION
The present invention, however, requires only one loop (two leads)
for each electrode, yet allows both amplitude and phase adjustment
thereby providing substantially complete nulling.
DETAILED DESCRIPTION OF THE DRAWINGS
FIG. 1 shows a magnetic flowmeter nulling system according to the
invention in the form of a box diagram.
FIG. 2 is like FIG. 1 except the boxes are replaced by actual circuitry.
In FIG. 1 reference numeral 1 denotes a conduit, through which
an at least slightly conductive liquid is flowing. Diametrically
disposed on the wall of conduit 1 are electrodes 2 and 3 which contact
the liquid. A winding 4 or similar suitable magnetic field producing
means, generates a magnetic field, normally alternating in sense,
and directed transverse both to a line adjoining electrodes 2 and
3 and to the direction of flow through the pipe. This is a conventional
so called electromagnetic flowmeter primary element. As is well
known, the electrodes are connected to the terminals of suitable
AC voltage detecting and measuring apparatus, which apparatus produces
an output in proportion to the voltage induced in the liquid flowing
between the electrodes. As is evident from its label, box 5 represents
such detection and measuring apparatus.
Electrode 2 is connected in a nulling loop by a pair of wire leads
6 and 7. As will be seen from the Figure, leads 6 and 7 are twisted
together except at their ends in the immediate vicinity of electrode
where they form a loop which is fixed in the rectangular configuration
shown, as, for example, by securing every point of the wire in the
loop directly to the outer surface of conduit 1. Preferably, the
electrode 2 is located at the center of the lower side of the rectangular
loops, and in any event the electrode 2 is in effect an electrically
conductive element of the loop formed by the ends of the leads 6
and 7. The twisted portion of the leads 6 and 7 runs to an in-phase
voltage adjusting network represented by box 8 having input terminals
9 and 10 and an output terminal 11. Leads 6 and 7 continue on to
a phase-shifted voltage adjusting network represented by box 12
having input terminals 13 and 14 and output terminal 15.
Electrode 3 is fitted out exactly like electrode 2 with leads 16
and 17 having their ends at the electrode forming a loop including
the electrode and having a twisted portion connected to an in-phase
voltage adjusting network 18 and thence to a phase-shifted voltage
adjusting network 19. The output terminals of networks 8 and 12
are connected to input terminal 20 of circuit 5 and the output of
networks 18 and 19 are connected to another input terminal 21 of
circuit 5. Networks 18 and 19 are provided with input and output
terminals exactly as are their respective counterparts, networks
8 and 12.
It is to be supposed that a voltage is induced in each electrode
lead loop by the alternating magnetic field and further, that each
network has the effective property of deriving from the induced
voltage an algebraic sum of two voltages, one in phase with the
induced voltage and one shifted in phase with respect to the induced
voltage, both derived voltages being adjustable.
Accordingly, an electromagnetic flowmeter may be manufactured in
completely assembled form regardless of whether or not completion
results in a device having substantially zero voltage at no-flow.
This is so, because if the finished primary element is found to
put out non-zero voltage at zero flow, it is then necessary only
to adjust one or more of the networks 8 12 13 and 19 in order
to null out unwanted zero flow voltages.
While the box diagram form of the circuit shown in FIG. 1 suggests
a somewhat complex system of networks, in practice, that turns out
not to be the case, as shown in FIG. 2 hereof. Thus, in FIG. 2
I not only take advantage of the fact that in general phase shifts
can be taken care of by using just one phase-shifted adjustable
source, but also that a source of in-phase voltage is right in the
electrode loops. Accordingly, the loop of the electrode formed by
leads 16 and 17 is completed by the slide wire 22 of a potentiometer
23 so as to allow adjusting the slider 24 of the potentiometer to
a point on slide wire 22 where there is zero loop voltage at no-flow.
The loop formed by leads 6 and 7 likewise is completed by potentiometer
25 having its slide wire 26 connected there across and a slider
27 which can be adjusted to a point on slide wire 26 where loop
voltage is zero at no-flow.
Two potentiometers 23 and 25 obviously can accommodate stray voltages
only if there is no phase shift. Therefore, to accommodate phase
problems, I connect the slide wire 28 of a potentiometer 29 across
the slide wire 26 and the slider 30 of potentiometer 29 I connect
via a capacitor 31 to terminal 20 and therefore to the slider 27
of potentiometer 25.
In use, the conduit 1 which may supposed to be of uniform diameter,
and a number of times its diameter in length, is stopped at both
ends and solidly filled with conductive liquid. Magnetic field producing
device 4 is energized and a suitable instrument (not shown) is connected
to slider 24 to measure the voltage between it and any suitable
reference point such as the fluid potential at the ends of the conduit.
Slider 24 is now adjusted until such voltage as exists between it
and the reference point is minimum. Because this adjustment does
not include a phase-shifted adjusting network, said minimum voltage
generally will not be zero. Likewise, the net voltage of slider
27 which includes the voltage of slider 30 shifted by capacitor
31 is measured, but this time in respect to the previously adjusted
voltage on slider 24. The two sliders 27 and 30 are now adjusted
for a minimum net signal, that is in practice, these two sliders
can be easily adjusted to substantially null out the no-flow signal
measured between terminals 21 and 20 substantially completely.
Although some harmonics of the field supply frequency may not be
thus reducible to zero, these will generally be rejected by the
measuring circuitry 5 which is customarily designed to reject harmonics
of the usual 60 hertz supply for the field.
The adjustment means that just described is capable of nulling
out, substantially completely, the differential no-flow signal,
that is, the no-flow signal measured between terminals 21 and 20.
This is possible because the inclusion of a phase-shifted voltage
adjusting network makes it possible to adjust the voltage on terminal
20 to be substantially equal to that on terminal 21. Because there
is no phase-shifted adjusting network connected to terminal 21
however, it is generally not possible to completely null out the
no-flow common mode voltage, that is, the no-flow voltage measured
between either terminal 21 or terminal 20 and the above mentioned
reference point. This generally is of no consequence, as the measuring
circuitry 5 is customarily designed to reject common mode voltages,
responding only to the differential voltage between terminals 21
and 20.
If it should be necessary to provide substantially complete nulling
of the no-flow common mode voltage a capacitor and second potentiometer
may be added to the nulling network connected to terminal 21 in
a manner identical to that connected to terminal 20. Each network
may then be independently adjusted for substantially zero no-flow
voltage between each terminal and the reference point.
The adjusting technique is not critical, and can only be described
as cut and try. In a typical system, slide wire 22 was 1 kilohm,
slide wire 26 was 1 kilohm, slide wire 28 was 10 kilohms, and capacitor
31 was 0.2 microfarad. With this arrangement, when 24 was adjusted
first, and sliders 27 and 30 were next adjusted in alternating fashion,
the no-flow voltage between terminals 20 and 21 could be gotten
down to 100 microvolts or less, said voltage consisting almost entirely
of harmonics of the 60 hertz supply. Under exactly the same circumstances,
using a standard commercial zero flow nulling technique, no-flow
signals ranged from 150 microvolts up, including a substantial portion
of the basic 60 herz supply. Furthermore, the aforesaid commercial
form for the nulling system, namely, that used in the present assignee's
1100L flow primaries often requires that the zero be readjusted
when changing the setting of the span dial on the detecting and
measuring apparatus (the assignee's 1100T flow transmitter). In
particular, on 0.1 inch and 0.2 inch diameter flow primaries having
the aforesaid commercial nulling system, when after nulling the
span is adjusted from 3 to 30 feet per second, there arises a 22
percent zero shift typically. Yet if the same transmitter is connected
to the same primary elements fitted with nulling circuitry according
to FIG. 2 of the present application, the same span change (i.e.,
from 3 ft/sec. to 30 ft/sec.) was found to produce less than 2 percent
of zero shift. |