Note: Descriptions are shown in the official language in which they were submitted.
The invention rela-~es to the measurement of axial positions
and displacements for rotating bodies, especially but not
exclusively, to the measurement and control of the size of the
disc clearance in a chip refiner.
In refining devices of this type there is a clearance
be.ween two refining plates mounted on discs. Either one of the
discs is stationary and the other rotates around a shaft
perpendicular to the plate, or both of the discs rotate in
opposite directions. The plates are provided with suitable
patterns on their sides facing one another and the material
which is to be ground up is usually fed in through a hole in the
middle of one of the plates and after refining is taken out at
the periphery of the concentrically rotating plates. The size of
the clearance between the two plates is of course crucial to the
refining results, and therefore there is as a rule some ~orm of
device for controlling the same. The problem is complicated by
the great forces involved, both in the form of centrifugal
forces and in the form of pressure on the material which is
being refined. One must also consider the effect of varying
temperature on the unit, making it di~ficult to maintain a set
clearance. Depending on the device~ the usual desired clearance
is from one or two millimeters down to some tenths of a milli-
meter. Because o F the forces involved, it can happen that the
operating position for the plates corresponds to what is called
"negative clearance", meaning tha-t the plates have been
displaced so far towards one another that the positions of the
rotating shafts have passed -the point at which the plates at
rest would make metallic contact with one another.
1-
: . :
: -` ' , ', ' ~ ,.
.. . . .
,, ', ~ ~
One purpose of the invention is now to ~chieve a device
with wllich the clearance can be measured during operation.
Another problem encountered in clearances of this type
ls wllat is usually called "out of tram", meaning that the plates
are not parallel, so that the clearance varies along the periphery.
This is usually caused by one or both of the rotating shafts
changing orientation, e.g. because of temperature gradients in
the frame in which the shafts are journalled. If such an error,
whiGll c.an be seen from the refining results, is corrected by
movltlg tlle reining plates closer to one another, the plates can
easily be brought into metal]ic contact so that they are damaged.
Therefore it is another purpose of the invention to
achieve a device for measurement of clearance during operation,
W]liC]l can be arranged so that the clearance can be measured at
several points along the periphery.
According to the present invention there is provided
a device for use in making out-of-tram and clearance measurements
in C]lip refiners having first and second normally parallel and
spaced apart coaxial discs~ at least said first disc being
~0 rotatable with a coaxial shaft having a longitudinally extending
- a~is of rotation, comprising a pair of linear converging and
cliverging indicia located on the periphery of said first disc
d ~ositioned so that a plane normal to said axis of rotation
intersects both indicia of said pair of indicia~ first sensing
means positioned at a first stationary location substantially
adjacent to the periphery of said first disc for sensing the
passage of said pair of indicia past said first sensing means
and generating a signal whenever one of said pair of indicia
- 2 -
., .. :-. -. : .;. ,
; ;, ,; - .,,, : .~ ,,
.. .
- . . - .: .. , :
,. . . .
, , . .. ,, , ., .
: , :~ , , :. . .
passes thereby and second sensing means positioned at a second
stationary location, different from said first stationary loca-
tion, substantially adjacent to the periphery of said first disc
for sensing the passage of said pair of indicia past said second
sensing means and generating a signal whenever one of said pair
of indicia passes thereby, whereby the time interval between
; successive signals produced by said first and second sensing
mcans can be used in making the out-of-tram and clearance measure-
mcnts .
t0 Although the invention will now be described in connec-
`tiOIl with an application with two discs rotating in opposite
directions, it will be obvious to the skilled art worker that
the same inventive idea is also applicable without major modifi-
cations to measurement of a clearanceJ where one of the plates
is stationary~ and the measurement according to the
- 2a -
`~
.j .,, ~ ,,
" , : .
- . , i , . ~ :
: `~ . , ' , . ,
'~ '`" ` ` ` ~ . ' ' ". ' ` ' ' `
I
invention is done on the rotating plate. The skilled art worker
will see that the posi-tion of the second plate can in this case
be determined in a number of different ways, and that the
problem is thus greatly simplified since the stationary plate,
although it can be deformed~ is at least standing still.
A non-limi-ting embodiment of -the invention will now be
described with reference to the drawings.
Fig. 1 shows a portion of a chip refiner, showing the
placement of a pair of sensors.
Fig. 2 is a sketch showing portions o~ the peripheral
surfaces ofthe two discs, which are provided with line members
to be sensed, preferably by magnetic sensors.
Fig. 3 shows an example of a simple circuit for determina-
tion of clearance.
The example shown is based on a reiner of known type, e.g.
the model RSA 1300 manufactured b~ Sunds AB. Although the
following description relates particularly to such a machine, it
is obvious that the invention can be applied to many different
types of units where de-termination of the axial position ~f
rotating objects is desired. Although the inven-tion was
developed in connection wi-th work relating to refining machines
in the cellulose industry, it is obvious -that the principle of
the invention is applicable to many other machine parts, in
turbines for example.
Fig. 1 shows an outline of the basic arrangement of a chip
refiner with two discs 3 and 4~ mounted on coaxial shafts 5
journa]led in bearings 6. Material to be refined is fed in
through a screw device 7. Since this belongs to prior art~ it
. , "
:
: . . : ~, . - ~ :
will not be dealt with in more detail here. The present
invention rela-tes to the measurement of the clearance between
the plates on the discs 3 and 4. Two sensors 1 and 2 are
arranged for this measurement, and they are disposed to sense
special line members arranged on the peripheral edges of the
discs.
Fig. 2 shows said line members on the peripheral edges of
the discs. The lines consist of ferromagnetic bars which are
attached to the peripheral casing or to the discs of non-
magnetic material, possibly sunk into the same, at angles to theaxis of rotation, preferably about 4S~ although other angles
are possible. T~e sensors are indicated as small circles in this
figure which only shows portions of the peripheral casing
surfaces of the discs. The directions of rotation are indicated
lS by arrows. When the discs rotate, they will be sensed by the
sensors 1 and 2 along the dash-do-t lines 20 and 21 respectively.
If, for example, disc 4 is moved axially a certain distance a,
the sensing would now be done along the line 22.
Now observation of -this displacement is possible according
to the invention, in spite of the fact that the disc is rotating,
by virtue of the fact tha-t, as is shown in the figureg -the
distance along the sensed peripheral line between the bars 12
and 13 is in this case smaller. ~The bars can of cour6e be
placed so that the distance becomes greater ins-tead.) In the
example shown, with undiminished speed, the -time interval
between the respective detections of the two bars will be
reduced, so that the displacement a i6 immediately calculable,
if the angles of the bars in relation to the axis of ro-tation
.
--4--
~ ~ ,
and the peripheral speed of the disc are known,
Since the discs are usually made of austenitic and non-
~agnetic stainless steel, it is easy to sense the passage of
the ferromagnetic bars with magnetic sensors. Even lf the discs
were magnetic it would of course be possible to sense the
passage of the bars, if they are raised and the distance-
dependence of the sensors is sufficiently great. As a rule, this
does not present any problem, since most sensors of this type
have characteristics which decrease by a factor lying somewhat
between the square and the cube of the distance. In the
embodiment which has been tested, a Phillips PR 9262 electro-
magnetic sensor was used~ which worked excellently.
In the presently preferred embodiment there are two bars on
each disc, which are screwed fast to the peripheral s~rfaces.
These surfaces are cylindrical with their axes coinciding with
the axis of rotation, In view of the high rotational speed of
the discs, it is of course necessary after screwing the bars
down to rebalance the discs. The bars are rectangular and have
the dimensions 5 x 10 x 75 mm. The circumference o the discs is
approximately 4,~ mm, and the bars have a space between them of
about 150 mm, measured from their centers. It is also possible
to mount more than two bars around the circumference if desired,
In order to be able to measure the clearance at several
different poin-ts along the periphery, pairs of sensors 1,2 are
2S arranged at three different places, namely a-t the location shown
in Fig. 1~ and at 90 therefrom in either direction. This is in
order to be able to measure "out of tram" (not shown in the
figure~, Other placements are of course conceivable, according
,, " ., ,:
, , : ~, ; ,
:; , ,
to the suitability in the refiner which is to be equipped.
Pairs of sensors are arranged in holes in covers screwed
onto -the casing around the discs (not shown). By vir~ue of the
fact that the sensors are screwed into threaded holes in the
covers, it is possible by screwing in or screwing out to adjust
their position as to depth and thereby their distance to the
respective rotating disc. The sensors are adjusted in this way
to lie several millimeters above the bars (with the above-
mentioned sensor from Phillips) a-t most 8-10 mm *rom the bars.
The bars should be protected from corrosion in a suitable
manner, e.g. by treatment with an appropriate plastic lacquer.
One should also ~alvanically insulate between the disc and the
bar, since galvanic corrosion can occur in this use.
The functional principle of the invention can best be seen
from Fig. 2 partially showing the peripheral surfaces of the
discs 3 and 4, with ferromagnetic bars 10 and 11, and 12 and 13
respectively. When the discs rotate, the sensors 1 and 2 will
sense the bars along the circles 20 and 21, drawn with dash-dot
lines. It will be assumed that the discs ro-tate at a constant
r.p.m. and that the time intervals between the sensing of the
two pairs of bars are tl and t2 respectively. It is obvious that
if the right-hand disc 4 in the figur~ were to be moved to the
left the distance a~ -the sensor 2 would instead sense along the
line 22. It i5 immediately evident from the figure -tnat -the
distance between the pair of bars along this line is shorter
than the distance along line 21, so that the time interval for
their passage in front of the sensor 2 is reduced. It is also
immediately evident that the time difference is a linear
:' :
:;' , ~ . ;
'.f-
fl~
function of the displacement a. If the new time interval between
the passage of the bars is now t4 instead of t2, it is easily
seen that the dis-tance a can be calculated from the equation
a = k(t2 - t4), where the constant k is only dependen-t on
s constant geometric conditions.
I~ in the case aceording to Fig. 2 there is a cer-tain
clearance Mdo, for which the time intervals tl and t2
respectively are measured, and then the time intervals t3 and
t4 respectively~ the new clearance can be computed from the
equation
Md MdO ~ (kl(tl - t3) + k2( ~ 4 (1)
or
~ d K + klt3 + k2t4 t2)
if the constant K is determined from the computed values ~orthe
case where the clearance is zero. If the geome-try is otherwise,
for example if the angles ~etween the bars are in the opposite
directions, then the signs in the equations will be reversed.
A relatively simple device for treating the signals from
the sensors 1 and 2 is shown in Fig. 3. The sensors are each
coupled to an individual means for measuring the time interval
between the successive passages of the bars and giving off
analog voltage signals which are proportional to the respective
time intervals. These voltages are coupled to a precision
resistor in such a manner that the two analog voltages are
summed. The resistance is a potentiome-ter and one end of the
resistance and the slidable contact are coupled to a digital
voltmeter. By adjusting the setting of the slidable contact of
the potentiometer it is possible to produce a value which
:`
--7--
::: .,
.
~o~
~~
differs from the magnitude of the clearance by a constant ~ac-tor.
This presupposes that the constants kl and k2 are equal, which
is the case if in Fig. 2 the con~igurations of the bars are
mirror images of one another. If this is not the case, the units
15 and 16 must be made adjustable to include the appropriate
calibration constan-ts, Even the constant K in Equation (2) can
be adjustable either in units lS and 16 or in uni-t 18.
The skilled art worker will see that this trea-tment of -the
signals can be done in many different ways, either digitally or
by analog. In the embodiment shown there was used ~or units 15
and 16 the Hewlett Packard 5300 A measurement system. This
comprises a crystal-controlled oscillator and a scaler, which is
turned on at the sensing o~ he first bar and is shut off at the
sensîng of the second. During the intervening time interval,
lS pulses are counted from the oscillator~ and the resulting
counted number is presented at the ou-tput in the form of an
analog voltage signal. The potentiometer 17 is a wire-wound
precision potentiometer of Helipot manufacture, which provides
an accurate setting of a calibration constant, e.g. so that the
~ readings on the voltmeter 18 are in millimeter units for
variations in the clearance.
In the embodiment shown here the refiner motors are
synchronous motors, as is usually the case to assure that ~e
power factor in the load on -the electricity supply system will
remain constant and not be too reactive. This means tha-t after
the motors have been started and brought in-to synchronous
operation with the mains frequency~ -then no special correction
is required for varying r.p.m.'s. O-therwise it is necessary to
.
:;;
make a special correction for varying rotational speeds,
suitably with a tachome~er on the motor shafts. The skilled art
worker will see how such a correction can be carried ou-t.
As disclosed previously, in the arrangement described there
are three different pairs of sensors with accompanying measure-
ment equipment. This makes it possible to determine ~he size of
the clearance at three different locations around the periphery
and to thereby diagnose any obliqueness or "out of tram".
The described measuring device provides a measurement of
the clearance during operation. This makes it possible, in
addition to manual control, to automatically control the size of
the clearance by return couplin~ to the means, which are known
per se, by which the clearance is normally adjusted during
operation. Previously it was necessary to control -the clearance
based on the sensing of the axial positions of the shafts in
relation to the machine frame. This causes obvious problems if
one notes that the clearance should be set with an accuracy of
better than one tenth of a mîllimeter~ that -the distance be-tween
the points of measurement is on the order of two meters, and
that the temperatures Oll the order of 100C are present in
certain parts of the refiner at the same -time as other parts are
possibly a-t room temperature. Therefore the -temperature
gradients can cause considerable difficultiesj even if a-ttempts
are made to solve these problems by careful temperature control
of the machine frame. Measuring the clearance in situ elimina-tes
most of these problems. Even a distortion of the machine frame,
causing the disc shafts to be out of parallel, can be brought
under control by arranging several pairs of sensors around the
" : "
: ~: : . :
- . . ; . .. . .
, : . - ~ :
, .. .
,, : : .
9~
discs in the manner described above.
It was mentioned in ~he introduction tha-t ~he plates~
primarily due to the ef~ect of the centrifugal force and the
pressure o~ the ma-terial between the plates, do not have the
same form as in the machine at rest. ~owever refining discs oi
the type intended here usually have concave surfaces so that the
clearance is least at t~le periphery. This means that chan~es are
re~lected in a relevant manner when the clearance is measured
according to the principle of the invention, since wha-t is
measured .is the clearance at the periphery. This means tha-t -the
problem occurring in many refiners according to the prior art,
namely that one must work with a so-called negative clearance,
is eliminated. ~ change in the pressure on the fed-in material
or in its composition, which af~ects the size of the clearance
in spite of the fact that the motor shafts do not move axially,
.is thus measurable and controllable.
The invention has been described with re~erence to an
application within the cellulose indus-tryO It can also be
applied generally in order to solve problems relating to similar
axial displacemen-ts o~ moving machine parts where it is not
suf~icien-t to read the positions at res-t. The sk~led worker in
the ~ield in question will see that the application in a
particula~ case will require modifications of the application
descr:ibed here. It is intended that -the following claims will
also cover applica-tions which are similar and obvious to the
skilled art worker. For example, it is not necessary that the
line members be ferromagne-tic, with magnetic sensors. In many
cases it is possible to use optical means instead, as the
--10--
,:.
.
.
skilled art worker will see. Thus the sensing can be done for
example by directing a light from a light-emitting diode against
a line member and is reflected against a phototransistor, which
then gives off a pulse every time a line member passes. Thus
there are qui-te a number of conceivable detecting principles,
for example capacitive sensing, sensing of the Foucault effect
in line members with lower resistivity -than the machine parts on
which they are mounted, etc~ Thus innumerable variations of the
present inventive idea are possible, and it is impossible to
mention more than a few ill~strative examples here.
~11~
; .
.
