Note: Descriptions are shown in the official language in which they were submitted.
107593Z
The invention described herein relates to an
improved vibrating wire stress meter and a method of
clamping the vibratory wire used.
Vibrating wire stress meters have been used for
some time to measure the stress in rocks or other structures
by noting variations in the vibrations of the tensioned wires.
Three examples of the prior art devices can be found in the
United States Patents 2,306,137 to W. Pabst et al, 2,969,677
to A. V. Lewis, and 3,675,474 to P~. D. Browne. In many of
these inventions the wire is tensioned in a longitudinal tube
lengthwise of the tube to measure compressive forces on the
tube ends. This type of arrangement along with the readout
` equipment needed to determine the frequency of vibration of
the wire is usually expensive to fabricate and requires a
knowledge of the modulus of the surrounding rock. This
improved stress meter tensions the wire generally perpendicular
of the tube 15 length to provide a unidirectional measurement
of the compressive forces on the tube's outer surface. This
construction allows the fabrication of a very stable, sensitive,
simple, inexpensive, and rugged gauge whose operation is
appreciably unaffected by changes in the surrounding borehole
rock modulus. It does this without relying on a voltage
source or value of the resistance from a connecting cable.
Another problem encountered in constructing vibrating
wire stress meters is firmly anchoring the tensioned wire at
its two ends. To allow for accurate readings in the vibrations
it is necessary to precisely tension the wire. This invention
not only provides for a precise tensioring of the wire but
does so with a minimum of machining of the cylinder body. It
also provides a watertight seal to the environment. ~s such it
2 ~
-` 1075932
eliminates the use of screws, and the like, as anchoring
devices which suffer from lack of accuracy and watertightness,
as well as requiring space for screw clamps.
According to the invention, a wire is tensioned
and transversely clamped across the diameter of a tubular
housing of a vibrating wire stress meter. Within the
housing a magnetic coil attached to a readout meter at
one end is used to pluck and thereby vibrate the wire at
a known frequency of vibration. A wedging apparatus can
be used to fix the housing within a borehole drilled in
rock. To clamp the ends of the tensioned wire within two
aligned holes of the housing the wire is first placed to
extend through both holes and then tensioned. When in such
a position, two small tubes are placed over the two ends of
the wire outside of the housing and each are accurately
punched into recessed portions of countersunk housing holes
to become extruded therein. This controlled punching force
can, as a result, be used to precisely tension the wire.
One embodiment of the invention will now be
described, by way of example, with-reference to the
accompanying drawings, of which:-
FIG. 1 illustrates a vibrating wire stress meter
in situ in a mine borehole,
FIG. 2 is a cross-sectional exploded view of the
meter,
FIG. 3 is an assembled view of the stress meter, and
FIG. 4 is an enlarged cross-sectional view of one
clamped end, immediately before punching and extruding takes
place.
In FIG. 1, the vibrating wire meter 1 is shown in
~ .
7~93~
situ in a mine. It is connected by electrical cable 3 to
a readout meter 5. The stress meter is rigidly fixed in
the borehole 7 of the surrounding rock 9 by a wedge assembly 11.
Presently tools allow the meter to be set into boreholes to
depths ranging from 1-1/2 inches to about 100 feet. An
operator, located in the mined-out portion 13, is able
to connect and/or disconnect a readout meter 5 via cable 3
to the meter. Readings on the readout meter are given in
terms of the inverse frequency or period of the unknown
signal and are displayed to four place accuracy to within
- one part in 105 (the number of cycles of a high frequency
oscillator within the readout unit). This readout meter
per se forms no part of this invention.
; In an exploded cross-sectional view taken along
the longitudinal center line of the meter (FIG. 2), the
hollow cylindrical steel gauge body 15 houses the
vibrating wire 17 that is rigidly held therein by two
capillary hollow tube clamps 19 and 21. In the present
preferred embodiment, the gauge cylinder is made from
4140 steel, heat treated to Rockwell 45, having a yield
strength of 160,000 pounds per square inch (psi) and an
ultimate strength of 200,000 psi, the wire of steel
having an ultimate strength of 420,000 psi; and the tubes
from stainless steel 316 series having a yield strength
between 38,000 and 55,000 psi. The wire extends through
two aligned countersunk holes in the gauge body such that it is
generally perpendicular to the longitudinal extent of the
body and free to vibrate within the body's hollow interior. The top
. .
107~932
portion 27 of the housing is flattened such that it can
complimentarily engage the flattened undersurface 29 of the
wedge 30 forming part of the wedge assembly 11. The upper
platen 31 and the hollow shear rivet 33 that joins the platen
and wedge form the rest of the assembly. When it is desired
to insert the meter in a borehole as in FIG. 1, the thin edge
of the wedge is located so as to face towards the borehole
- opening and the wedge's flat undersurface 29 touches the top
27 of the gauge body. The platen is riveted to the upper
surface of the wedge. A setting tool head (not shown) is
used to carry the meter, wedge, and platen down the borehole
and also to grip the wedge and pull it between the meter and
platen. Front wedge hole 34 is used in this operation. This
pulling motion breaks the hollow rivet 33 to firmly wedge the
meter and wedge assemhly in the borehole.
Inserted within the hollowed out cylindrical portion
35 of the body 15 is an actuating mechanism 37 to pluck and
sense the vibration of the vibrating wire 17. This mechanism's
rear end cap portion 39, when inserted in the hollow gauge body,
is slightly indented from the rear surface 41 of the body and
its forward end has a hole 42 to receive the vibrating wire 17.
The basic components of the actuating mechanism include its
housing, the cable 3, the two neoprene sealing O-rings 43,
cable cl~l~ 45, potting compound 47, thermistors 49, wire
coil 51, magnet 53, yoke 55, end cap 39 of the assembly's body,
roll pin 57 to secure the assembly to the gauge body 15 when in
hole 44, and the adhesive foil disc 60. The coil in conjunction
with its magnet enables the wire to be vibrated. The two end
legs of the U-~haped carbon steel yoke surround the coil and
magnet. The magnet and coil with the yoke act as a three-pole
1075932
electromagnet and complete the magnetic path from the wire to
the magnet. Cable 3 transmits a modulated on and off power
signal Erom an external source (readout meter) to the coil to
generate a magnetic field in the magnet to thereby pull the
wire towards the magnet and then release it.
The functions of the other components of the actuating
mechanism are straight forward. The O-rings seal the assemblv
in the gauge or meter body 15; the potting compounds encapsulate
the electrical components for protection and easy assembly into
the body; the end cap protects one end of the electrical
package; the foil disc protects the other end of the same
package and allows the placement of indicia for identification
of the unit. The therrnistors 49 are used to measure tempera-
ture to enable the meter readings to be corrected for
differential thermal effects between the rock and meter. The
particular type of thermistor used in the preferred embodiment
was manufactured by Fenwall Electronics, Framingham, Massachu-
setts, U~S.A., and had Model No. UUA32J3.
The vibrating wire 17 is a high tensile steel music
wire sometimes referred to as piano wire. Its ultimate
tensile strength i5 usually around 420,000 psi and its
vibratory length about .78 inch and its cross-sectional
diameter about 0.009 inch. Variations in dimensions and
strength are, of course, possible. Initially, the wire before
it is clarnped at its ends in the extrusion process to be
described hereafter is subjected to about 200,000 psi--the
peak tensile stress it is ever under.
When stress changes occur in the surrounding rock,
they cause a small change in the diameter of gauge cylinder
body 15. Since the wire is anchored in this body, these
107~93Z
.
changes will be transmitted to the preloaded wire. The
wire's natural frequency (f) or period T of vibration and its
; stress and strain are given by:
T ~ ~ pg = ~ E g (1)
where Q, E, and p are the vibratory length, modulus of
elasticity and density of the wire, respectively. The letter
g represents the gravitational constant and ~ and ~ the wire
stress and strain. A more meaningful expression than (1) for
our purposes is one that gives the wire deformation ~. This
is given by:
4 3 f2 (2)
Eg
One working embodiment would result in ~ for equation (2)
being 4 636 x 10 11 in. , when E = 30 x 10 psi, ~ = 0.78 inch,
T
g = 386 in./sec.2, and p = 0.286 lb/in.3. As will be dis-
cussed hereafter, T is the period of wire vibration and also
the four digit meter reading which is displayed.
The basic theory of operation of vibrative wire
gauges is well known. The coil has current flowing through it
to cause its associated magnet to send out a magnetic flux
field which cuts into the adjacent tensioned steel wire. To
achieve the desired result, the wire, coil and magnet geometry
are all very important. To excite the wire at its natural
frequency, the magnetic field pulls the tensioned wire
; approximately at its midpoint towards the magnet and then, due
to the modulated current in the coil windings, returns it to
its prior unplucked state. The change in inductive reactance
of the coil is directly related to the work (force times
deflection) done by the flux linkage on the wire. Generally,
. .
107~93Z
this is maximized when the flux linkage is as great as
possible ( shaped pole pieces, small wire-magnet distance,and
large diameter wire) and the wire deflection is as great as
possible.
The same driving coil and magnet can be used to
sense or pick up the vibrations of the wire. This occurs when
the magnet induces a lGcal magnetic field in the wire. As the
wire moves, it appears as a changing magnetic field to the
nearby coil, thereby induciny a voltage in it. This usable
induced voltage ranges from 0.1 to 20 millivolts depending upon
the vibration amplitude of the wire, the streng~h of the magnet
and the coil, and the wire stand-off distance.
To use the gauge, the wire is plucked electro-
magnetically for a number of cycles at its natural resonance
frequency. These pluck signals are generated by a counter
` circuit and solid state driving circuit in the readout unit
that is connected via cable 3 to the coil 51. The frequency of
these plucking signals is determined by a readout voltage
; controlled oscillator (VCO) whose nominal free runing frequency
varies as the operator turns a dial on the readout unit's face.
Generally,the VCO running frequencies range from 2,000 to 8,000
hertz, the detection range of the instrument. Then an amplifier
; listens for the return signal for a period of time. Because
the return signal appears as a damped sinusoid followin~ the
resonance excitation, the signal will eventually damp out to a
point where it falls below the detection threshold. In one
working embodiment of the stress meter, the detectable cycles
range up to about 700; however, only the first 300 or so were
free enough from noise to be usable.
In the previously mentioned embodiment of the readout
107~i932
meter, digitzl logic is used to count and display the unknown
frequency (in terms of a period) for a known period of time
but the count is limited to a discrete number of cycles. For
example, say 6667 cycles are counted in one second, the
frequency of the signal is 6667 herz accurate to approximately
one point in 105. Had the known count time been only 0.1
second, then the count would have been 66 cycles which gives
an accuracy of the frequency measurement of only one part in
103. Thus, to count cycles and display then accurately in
- 10 terms of frequency, the counting time should extend over at
least one second. Rather than do this and still achieve four
place accuracy when only 250 cycles are available to be counted,
an alternative system has been used. This alternate is to
count exactly 100 cycles of the unknown frequency received
from the gauge and at the same time count the n~Imber of cycles
from a stable 105 hertz high frequency oscillator. The number
of counts from the oscillator is then displayed as the digital
readout. For example, if it takes 0.15 second to count 100
cycles at 6667 hertz, then during this same time frame 1500
cycles of the reference time base oscillator are counted since
it is counting 100 times as fast. Thus, the displaced number
1500 represents inverse frequency (l~f) or the period of the
unknown signal. With an accuracy of one part in 10 in the
oscillator,four place accuracy can be obtained for the wire
period. This reasoning explains why the vibrating wire readout
meter displays its digital readings as a period rather than as
straight frequency.
FIG. 3 illustrates the assembled FIG. 2 stress meter
with the wedge assembly above it. The end cap 39 is inserted in
the hollow opening 35 (FIG. 2~ such that it is slightly
107:'
vertically recessed from the surface 41. Power and sensing
cable 3 extends from the center portion of the actuating
mechanism to the readout meter. When placed in a borehole, the
two flattened surfaces--surface 27 of the meter and under
surface 29 of the wedge--engage each other to fixedly secure the
assembly therein. The upper surface of the platen and the
surface 27, as well as the hollow hole 35, would all be gen-
erally parallel in such a state. The dimensions of the
assembled unit of FIG. 3 are about 1.5 inches long by 1.125
inches in diameter.
;~ FIG. 4 is an enlarged cross-sectional view of a
portion of the gauge body 15 where it clamps the vibrating
wire 17. The lower hole in the gauge body is similarly con-
figured and the method to be described to extrude tube 21 into
the upper countersunk hole is equally applicable to extruding
tube 19 into the lower countersunk hole. Each of these holes
has a smaller and larger diameter outwardly facing portion 23
into which the stainless steel tube clamp 21 of approximately
the same diameter is seated. The wire extends through the
larger diametex hole portion and its encircling tube and also
through the smaller hole section 61 into the hollow gauge body
volume 35. The lower end of the same wire extends into smaller
hole section 63 (see E'IG. 2) and then out the larger hole
portion 25 into which tube 19 is seated. Thus, the two counter-
sunk holes are used as seats for the tubes by allowing the
tubes to be seated in their larger diameter portion but pre-
venting their passage in the smaller diameter inner sections.
To anchor these tukes to the wire, the wire is clamped above
an hydraulic cylinder attacned to punch 59 to first hold the
wire virtically. About 12-1/2 pounds are hung from the lower
107593Z
end of the wire to stress it to roughly 200,000 psi. The upper
hollow punch 59 is pumped do~m to contact the tube's upper
surface with around 1550 pounds of force for 30 seconds. This
results in the relatively soft steel material (38 - 55,000 psi
yield) of the tube 21 being extruded into upper hole portion 23
and gripping the wire. Next, the upper clamp is removed so
that the wire and the 12-1/2 pound weight are supported solely
by the extruded tube clamp. A lower hydraulic cylinder pump
and a second tube 19 now are utilized to extrude this lower
tube into larger hole portion 25. As this second lower tube
grips the wire and supports the weight, the tension in the
length of the wire inside the hollow gauge volume 35 is reduced.
When the tension inside the gauge body drops to around 150,000
psi, the readout inside the gauge body drops to around 150,000
psi and the readout meter 5 starts to display the period
readings. From this point on, the tube clamping pressure can
be controlled in conjunction with the meter readings. Actual
settings have been made to within ~ 10 units of the required
meter readings by exercising care and experience. After the
proper setting is obtained, the 12-1/2 pound weight is increased
to around 25 pounds, which breaks the wire at its lower tube 19.
After a few days when the tension has initially been
set, the meter readings for some gauges show a slight decrease
(i.e., the wire tension increases). This is thought to be due
to viscoelastic recovery of the clamping tubes after the high
extrusion stresses. Not all gauges respond in this way.
Further readings a few days after this noted decrease and
show a settling down to a steady state condition which remains
unchanged.
It is important to se]ect the strengths of the
1075932
materials making up the tubes, gauge cylinder body and wire
with care. It is important that the wire have very high
strength to avoid stress relaxation under high stresses. The
tubes should have a much lower yield strength than the body
or wire so that they can be extruded into the gauge body holes
as the punch compresses them. One additional consideration for
the gauge body is that it be sufficiently elastic over its
working range. The Rockwell 45 hardness 4140 steel insures
this type of elastic behavior.
~Iha~ has been achieved by the foregoing disclosure
is to construct a simple, durable, inexpensive, sensitive, wire
stress meter that is impervious to most of the normal environ-
mental conditions it would usually be used under, including
`~ water-filled boreholes. The unique method of clamping its
vibrating wire insures a constant and accurate readout of the
detected vibrations.
The stress meter described in this disclosure is
unidirectional in its measuring capabilities. Generally, to
get an accurate picture of what is occurring,three meters whose
wires are set 120 degrees apart when viewed directly down the
borehole, are used. In this way when stresses act on the gauge
bodies from different directions, changes in the resultant
stresses can be detected with a high degree of accuracy.
12
