Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
1175Sf~4
The present invention is directed primarily to indicating
the position of a valve for controlling the teeming or flow of
molten fluids, and more particularly the teeming of steel. It finds
particular utility in valves such as exemplified in U.S. patent
4,063,668 and Reissue patent 27,237, and more particularly as
modified for throttling of the subject exemplified valves as
exemplified in Canadian Patent 1,103,921 and Canadian Patent
1,136,829.
The patents and patent applications identified above are
directed to refractory-type sliding gate valves for controlling the
flow or teeming of steel. The patent applications which are
copending are directed to throttling in such valves where the
sliding gate and the fixed gate are initially in register for full
flow pour, and then the sliding gate is shifted in order to reduce
the effective size of the pouring orifice.
In most applications the throttling as well as the movement
of the sliding gate is accomplished by actuating hydraulic rams.
The operator normally monitors the flow of steel by watching the
stream. It will be appreciated that in the high temperature and
~0 corrosive environments involved, the accuracy of such visual
inspection of flow rates will depend heavily upon the skill and
experience of the operator, and even with the most skillful
operator, the accuracy and repeatability of the extent of throttling
is empirical.
It is therefore desirable, particularly for purposes of
repeatability and control, to have a means for accurately
determining the position of such a valve, open, closed, and
direction of open and close. In addition, where such a valve
involves means for throttling, it is important to measure the extent
of throttling, and to be able to repeat the same from pour-to-pour.
The present invention contemplates a method of monitoring
the position of a sliding gate valve where a hydraulic actuator is
:117~5~4
] emp~ ~d by measuring the fluid flow in and out of the hydraulic
actuator which drives the valve, either for throttling, or positioning.
Desirably the method includes inserting in series with the hydraulic
actuator a fluid motor which drives a rotary sensor. The rotary sensor
signal is read out remotely and preferably digitally, with intermediate
means translating the volume of fluid flow sensed into distance of
travel of the hydraulic actuator which, in turn, can be translated
into the movement of the sliding gate valve. Means for zero setting
upon actuation are provided within the method.
The apparatus contemplates primarily the combination of a
fluid motor in the hydraulic line which drives the hydraulic actuator
of a sliding gate valve. The rotation of the motor is then counted
and calculated to read out remote from the motor and in digital form
the position of the valve. A logic circuit is actuated by the read-out
to compute the hydraulic fluid displacement and thus indicate the
position of the valve. Means for zeroing the motor and counter for
each operation is further included in the combination.
It is a primary object of the present invention to read
accurately, and with repeatability, the position of the valve actuator
20 of a tundish or ladle gate valve thereby converting to the position of
the valve. This permits the operator to observe open, closed, or
extent of throttle.
A related object of the present invention looks to a method
of remotely positioning an electrical digital read-out for measuring
the position of a sliding gate valve and which will be in an environment
that is more conducive to long life and accurate operation.
Still another object of the present invention is to provide
a method and apparatus for reading the positions of a sliding gate
valve when it is throttled, whereby automatic controls can be developed
30 for controlling the level of steel in both a ladle and a tundish in
coordinated fashion for feeding a continuous caster.
Still another object of the present invention is to achieve
the above objectives within the bounds of an economically advantageous
unit.
1175564
l Further objects and advantages of the present invention will
become apparent as the following description of an illustrative method
and embodiment proceeds taken in conjunction with the accompanying
drawings, in which:
FIG. 1 shows in perspective, partially diagrammatic view the
environment of and read-out of the present indicator and method;
FIG. 2 is a partial block diagram showing the flow of hydraulic
fluid and coupling of the same to the electrical read-out and counter;
FIG. 3 shows how FIGS. 3A, 3B and 3C are to be arranged for
viewing;
FIGS. 3A, 3B and 3C together form a detailed schematic
diagram of the circuit which converts the output of the rotary sensor
to digital or analog display indicating valve position; and
FIG. 4 is a diagrammatic phase diagram of some of the pulse
relationships within the circuit.
The present method is based upon the proposition that where
the movement of a valve is a ~unction of hydraulic fluid displacement,
the valve being in a highly corrosive and high heat environment, the
best way to read the valve position is to determine the amount of
hydraulic fluid displaced in moving the valve. The method is well set
forth in FIG. 2 where it will be seen that a directional valve 60
controls the movement of the hydraulic actuator. In most commercial
applications a hydraulic cylinder or ram 61 is driven by a pump and
tank 67, but the present invention also finds utility where a hydraulic
motor is driven. All of the fluid displaced, whether the cylinder is
single acting or double acting and whether the valve is single acting
or double acting, passes through and drives a positive displacement
fluid motor 62. The positive displacement fluid motor, in turn,
drives a rotary sensor pulse generator 10 normally of the rotary type
with an optical shaft encoder. For a typical 3 1/4" bore hydraulic
cylinder which is stroked 3 1/2" in two seconds, the positive displacement
fluid motor will rotate at 1,130 revolutions per minute which approximates
20 revolutions per second.
117~5~i4
1 The positive displacement fluid motor 62 drives an optical
rotary shaft encoder 10 which generates two electrical signals, called
the "Clock~ 63 and the "Second Clock" 64, each of which consists of
200 square wave pulses per motor revolution, and which are generated
90 out of phase. Thus when the encoder rotates clockwise the Clock
lags the Second Clock and when the encoder rotates counterclockwise
the Clock leads the Second Clock. Consequently the direction of valve
65 movement may be detected from the relative phase of the Clock and
Second Clock by direction detector 68. Further, the 20 revolutions
10 per second times 2 seconds times 200 pulses per revolution delivers
8,000 pulses for 3~" of displacement. And any intermediate displacement
can be calculated from the number of Clock pulses generated by movement
to such intermediate displacement, including subtracting Clock pulses
when the movement is reversed.
To accommodate for the lost motion, wind-up, or mechanical
"slop~ in the valve upon change of direction of movement , an adjustable
skip delay 69 is interposed in the circuit after the direction detection
and before Clock pulse counting. The skip delay precludes the counting
of any preselected number of Clock pulses (the number is determined
20 empirically for each particular valve). Once the direction detection
and skip delay has been applied to the signal, it then proceeds into
the multiply, count and divide circuit 70. The purpose of this circuit
is to translate the number of pulses into the movement of the valve,
taking into account the diameter of the cylinder and the length of
stroke. Furthermore, the read-out can be made to be displayed on a
light emitting diode display 66 in either millimeters, or inches such
as shown in FIG. 1. In addition, a binary coded decimal output 72 is
available to be sent to a screen for a graphic display of the valve
position.
Separately, the method contemplates positioning a
digital-to-analog converter 73 also in parallel with the other displays
which can read the results of the position on a meter 74, as distinguished
from digitially.
S~4
l Finally, the method also contemplates a pulse output terminal
75 prior to the divide and multiply circuitry movement to feed into
automatic programming for further controlling the unit independently
of manual operation. For example, where the level of steel in the
mold for the continuous casting is constantly monitored, this information
can be passed to the pulse output receiver, and upon noting a lowering
in the level of the steel in the mold, the valve is told to open a
finite distance until the level is reestablished. While this involves
some electronic "hunting~ it takes place in so few milliseconds that
10 stabilization promptly occurs.
FIGS. 3, 3A, 3B, and 3C disclose the basic schematic diagram
for a logic circuit which performs the method as set forth above.
More specifically, pulse generator 10 outputs the Clock 63 and the
Second Clock 64 signals into the clock and data inputs, respectively,
of dual D flip-flop 11. Flip-flop 11 may be an integrated circuit of
type 7474. If the Second Clock is leading the Clock, then the Q
output 12 of flip-flop 11 will be high and the not-Q output 13 will be
low; whereas if the Second Clock is lagging the Clock, then output 12
will be low and output 13 will be high. Thus flip-flop 11 detects the
20 direction of movement of the pulse generator 10 (and thus also of the
valve being controlled), and this direction information is fed into
binary-coded decimal arithmetic units 14-20 as follows: If output 13
is low, then the arithmetic units are put into the add mode; whereas
if output 13 is high the arithmetic units are in the subtract mode.
The arithmetic units are integrated circuits of type 82S82, and the
overflow from unit 14 feeds into unit 15; the overflow of unit 15
feeds into unit 16, etc.; so seven-digit counts are within the capacity
of the circuit.
The Clock passes through OR-gate 51 for counting unless the
30skip delay prevents the passage through OR-gate 51. The skip delay is
only active just after the direction of movement of the pulse generator
changes, and is incorporated to account for lost motion, wind-up, or
mechanical slop in the valve. The operation of the skip delay will be
described later.
55~4
1 The counting of Clock pulses is as follows:
1. The Clock is fed into the load input of D flip-flops
21-27. The flip-flops may be integrated circuits of type 74175.
Because the output of the flip-flops is to be in engineering terms
(i.e. inches or centimeters) the Clock pulses are not counted directly
but rather are first multiplied by an appropriate factor, then counted.
For example, the previously computed generation of 8,000 pulses for
the total 3~N movement by the hydraulic cylinder would be analyzed as
follows: If arithmetic unit 20 and flip-flop 27 are to store the
lO number of inches of movement, then arithmetic unit 18 and flip-flop 25
would be storing the number of hundredths of an inch of movement.
Because one-hundredth of an inch of movement will correspond to approx-
imately 22.86 pulses (8,000 divided by 3.5 divided by 100) and because
these 22.86 pulses are to result in 10,000 counts (arithmetic unit 18
and flip-flop 25 are the ten thousands' digit), each Clock pulse
should produce 437 counts (10,000 divided by 22.86). These numbers
are approximations; however, if greater accuracy is desired, then one
has to add more arithmetic units and flip-flops so as to limit, in
essence, the round-off error on this factor of 437. The multiplication
20 by 437 iS obtained by the wire jumper patterns 28-34. The 437 is set
by connecting inputs 81, B2 and B4 of arithmetic unit 14 to the high
line and input B8 to the low line (this represents 7), the B1 and B2
inputs of arithmetic unit 15 to the high line and inputs B4 and B8 to
the low line (this represents 30), and input B4 of arithmetic unit 16
to the high line and inputs Bl, B2 and B8 to the low line (this represents
400), and all of the B inputs to arithmetic units 17-20 connected to
the low line.
2. The multiplication of 437 presumes that the output 13 is
low (so that the arithmetic units are in the addition mode) and that
30the flip-flops have been cleared (such as by the reset 36 or the
start-up reset 37), the various inputs are as follows: The arithmetic
units have Al,A2, A4 and A8 all low, and the Bl, B2, B4 and B8 as set
by the wire jumper pattern. The arithmetic unit outputs Fl, F2, F4
" li75S~4
l an~ and C are the binary sums of the A's and B's, so Fl is the sum
of A1 and Bl, F2 is the sum of A2 and B2 plus any overflow from F1,
etc. The flip-flops have inputs D1, D2, D4 and D8 equal to the F1,
F2, F4 and F8 outputs of the arithmetic units and the Q1, Q2, Q4 and
Q8 outputs have been cleared to low. When the Clock goes high (i.e. a
pulse) the flip-flops load the data at inputs D1, D2, D4 and D8 and it
is stored and made available at the outputs Q1, Q2, Q4 and Q8 until
the next time the Clock goes high (i.e. the next pulse). This stored
data is just that provided by the wire jumper pattern; in particular,
for arithmetic unit 14 the outputs prior to the first Clock pulse were
F1, F2 and F4 all high and F8 low, thus upon the pulse the data stored
and made available is Q1, Q2, Q4 all high and Q8 low. Because Q1, Q2,
Q4 and Q8 are connected to A1, A2, A4 and A8, the arithmetic unit adds
this to the wire jumper input at B1, B2, B4 and B8 and the sum appears
at F1, F2, F4 and F8; thus after the first pulse the outputs Q1, Q2,
Q4 and Q8 of the flip-flop are the same as the wire jumper input and
the data at D1, D2, D4 and D8 is twice the wire jumper input. At the
second time the Clock goes high (i.e. second pulse) this twice the
wire jumper input will be loaded into the flip-flop and made available
at the outputs Q1, Q2, Q4 and Q8, and additionally this output will be
inputted at A1, A2, A4 and A8 to be added to the wire jumper input at
B1, B2, B4 and B8, resulting in three times the wire jumper input
appearing at the arithmetic unit outputs F1, F2, F4 and F8 and at the
data inputs D1, D2, D4 and D8 of the flip-flop. In like fashion each
time the Clock goes high (i.e. a pulse impinges on the flip-flop load
input) another addition of the wire jumper input is made by loading
the previous sum at D1, D2, D4 and D8 inputs into the flip-flop and
thus inputting this sum to the arithmetic unit which adds another wire
jumper input to it and outputs it to D1, D2, D4 and D8. Whenever an
arithmetic unit overflows (i.e. contains more than nine) the overflow
is passed on to the next arithmetic unit as an input to sum with the
flip-flop stored and wire pattern inputs. The pin locations on arithmetic
units 15-20 are analogous to those shown in arithmetic unit 14, and
1~755~4
locations on flip-flops 22-27 are analogous to those shown in
flip-flop 21.
3. When the valve has reversed direction of movement, the
output 13 will have put the arithmetic units into the subtract mode
and for each count pulse the wire jumper input (i.e. 437) will be
subtracted from the count total stored in the flip-flops, but otherwise
the operation of the circuitry is analogous to add mode.
4. The data stored in, and the output of, flip-flops 25, 26
and 27 consists of a three-digit binary coded decimal which represents
the millions, hundred thousands, and ten thousands of counts and which
may be used to drive light-emitting diodes, analog devices, or any
other mode of display.
The skip delay is used to provide for a predetermined number
of Clock pulses to be ignored in the counting process each time the
valve reverses direction. Such pulses presumably reflect motion by the
hydraulic fluid without any movement by the valve itself and is termed
lost motion, wind-up, or mechanical slop in the valve. The skip delay
circuitry operates as follows:
1. Upon a change in direction of the pulse generator 10,
the Clock and the Second Clock change relative phase and this is
detected by flip-flop 11 and results in a change in the outputs 12 and
13. For example, if the pulse generator had been rotating clockwise,
then the Second Clock had been leading the Clock and the output 12
will be high; upon a reversal of rotation direction, the Clock will
now lead the Second Clock and the output 12 will drop to low.
Simultaneously the output 13 will change from low to high. Output 12
is the input for the one-shot multivibrator 37 which, upon a drop from
high to low by output 12, generates a single pulse output which is
inputted into OR-gate 39. Multivibrator 37 may be an integrated circuit
of type 74121. The output 13, which changes from low to high, is
inputted to one-shot multivibrator 38, which is identical to
multivibrator 37. Because a change from low to high has no effect
there is no output from multivibrator 38. If the change of the direction
~î~755~4
1 Of r~ Ition of the pulse generator 10 had been from counterclockwise
to clockwise then output 12 would have changed from low to high and
output 13 from high to low. In this event multivibrator 37 would have
no output and multivibrator 38 would output a pulse into OR-gate 39.
Thus if direction is changed either from clockwise to counterclockwise
or from counterclockwise to clockwise a pulse will be inputted to
OR-gate 39 and thus a pulse will be outputted by OR-gate 39 into
one-shot multivibrator 40. Multivibrator 40 is an integrated circuit
of the same type and connections as one-shot multivibrator 37 and
10 outputs a pulse to OR-gate 41 and to one-shot multivibrator 42, which
is also of the same type as multivibrator 37. The pulse into OR-gate
41 passes through and clears the three BCD up-down counters 43, 44 and
45. The BCD up-down counters may be integrated circuits of type
74192. The pulse from multivibrator 42 loads the preset digits in DIP
switches 46, 47 and 48 into the counters 43, 44 and 45. The counters'
outputs are combined and fed into OR-gate 51 and AND-gate 49. As long
as any of the counters 43, 44 and 45 contain any positive digit, the
combined outputs will be high.
2. The Clock is fed into AND-gate 49 together with the
20 combined outputs of the counters 43, 44 and 45, so as long as the
counters 43, 44 and 45 contain any positive digit the Clock will be
transmitted through the AND-gate 49 and into counter 43 at the down
input. Thus each pulse in the Clock will cause counter 43 to count
down by one. Of course, counter 43 borrows from counter 44 and counter
44 in turn borrows from counter 45; thus as the Clock pulses are
inputted into counter 43 the digits loaded from the DIP switches 46,
47 and 48 are counted down until counters 43, 44 and 45 all are reduced
to zero, at which time the combined outputs of these counters drops
from high to low. This drop from high to low turns off AND-gate 49
30 and also terminates the high inputted to OR-gate 51 continuously since
the pulse generator 10 had changed direction of rotation. Thus for
the first time since the change of direction the Clock passes through
OR-gate 51 unaffected; whereas prior to the counting down by counters
1175564
l 43 and 45, the output of OR-gate 15 had been a steady high and no
counting of the Clock had occurred in the arithmetic units 14-20 and
the flip-flops 21-27; this is illustrated in FIG. 4. Since the DIP
switches 46, 47 and 48 are adjustable, any predetermined number of
Clock pulses may be ignored prior to counting starts. The number of
such Clock pulses that are not to be counted is empirically determined,
and depends upon the particular characteristics of the valve and
hydraulic system involved.
Zero reset pushbutton 50 activates reset circuit 35 which
clears all of the stored data in flip-flops 21-27 to zero, and thus
also reduces to low the output which would be read on the light-emitting
diode display as zero displacement.
Start-up reset circuit 36 comprises a a one-shot multivibrator
which outputs one pulse when electrical power is first supplied to the
system. This pulse is used to: (1) set flip-flop 11 into the mode
with output 12 low and output 13 high; (2) clear BCD up-down counters
43, 44 and 45 to zero data and consequently low output; and (3) clear
flip-flops 21-27 to zero data and consequently low output. Thus when
the system is first activated the start-up reset circuit 36 aligns the
system for use.
In summary, the logic circuit multiplies the pulses outputted
from the pulse generator and counts the resultant multiplied pulses.
The count is outputted into display devices, such as light-emitting
diode digital display. The logic circuit is also provided with a reset
and start-up reset subcircuits to set the stored counts to zero for
the beginning of the counting of the pulse generators output. Further,
the logic CiLCUit is provided with a subcircuit which causes a preselected
number of pulses from the pulse generator to not be multiplied and
counted each time the pulse generator changes direction.
