U.S. patent application number 15/664853 was filed with the patent office on 2017-11-16 for effort modulation for process control of friction stir operations.
The applicant listed for this patent is Brigham Young University (BYU). Invention is credited to Kenneth Alec Ross, Carl D Sorensen.
Application Number | 20170326679 15/664853 |
Document ID | / |
Family ID | 49261306 |
Filed Date | 2017-11-16 |
United States Patent
Application |
20170326679 |
Kind Code |
A1 |
Ross; Kenneth Alec ; et
al. |
November 16, 2017 |
EFFORT MODULATION FOR PROCESS CONTROL OF FRICTION STIR
OPERATIONS
Abstract
A system and method for making adjustments to output effort from
a spindle driver using a multi-stage nested control loop of an
active controller to provide constant power to a friction stir zone
during a friction stir operation. Providing constant power
facilitates temperature control within the friction stir zone and
thereby improves the result of the operation
Inventors: |
Ross; Kenneth Alec;
(Mishawaka, IN) ; Sorensen; Carl D; (Provo,
UT) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Brigham Young University (BYU) |
Provo |
UT |
US |
|
|
Family ID: |
49261306 |
Appl. No.: |
15/664853 |
Filed: |
July 31, 2017 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
13853798 |
Mar 29, 2013 |
9751153 |
|
|
15664853 |
|
|
|
|
61769920 |
Feb 27, 2013 |
|
|
|
61618266 |
Mar 30, 2012 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
B23K 20/1235 20130101;
G05B 19/188 20130101; G05B 2219/49077 20130101; B23K 20/125
20130101 |
International
Class: |
B23K 20/12 20060101
B23K020/12; B23K 20/12 20060101 B23K020/12; G05B 19/18 20060101
G05B019/18 |
Claims
1. A method for using an active controller to control a friction
stir zone, said method comprising: providing an active controller
for controlling operation of a friction stir tool; and controlling,
to a power setpoint, power provided by the friction stir tool to a
friction stir zone using the active controller, the power provided
to the friction stir zone is controlled by making adjustments to an
output effort of a spindle driver that is used to drive the
friction stir tool.
2. The method as defined in claim 1, wherein controlling the power
provided to the friction stir zone further comprises using the
active controller to maintain a power input level.
3. The method as defined in claim 1, further comprising
implementing a multi-stage nested control loop.
4. The method as defined in claim 3, wherein the implementing the
multi-stage nested control loop further comprises: 1) implementing
an inner control loop to thereby maintain a power input level
through control of the output effort of the spindle driver; and 2)
implementing an outer control loop that adjusts the power provided
to thereby maintain a temperature for the friction stir zone.
5. The method as defined in claim 1, wherein controlling the output
effort of the spindle driver further comprises controlling an input
flow to the spindle driver.
6. The method as defined in claim 1, wherein controlling the output
effort of the spindle driver further comprises maintaining an
effort level.
7. The method as defined in claim 1, wherein controlling the output
effort of the spindle driver further comprises stabilizing control
of the spindle driver.
8. The method as defined in claim 7, wherein stabilizing control of
the spindle driver further comprises using spindle flow
feedback.
9. The method as defined in claim 1, wherein maintaining the power
input level through control of the output effort of the spindle
driver further comprises increasing the output effort of the
spindle driver in response to a decrease in the spindle flow.
10. The method as defined in claim 1, wherein maintaining the power
input level through control of the output effort of the spindle
driver further comprises decreasing the output effort of the
spindle driver in response to an increase in the spindle flow.
11. The method of claim 1, wherein the friction stir tool is a
welding tool.
12. A friction stir system, comprising: a friction stir tool; a
spindle coupled to the friction stir tool; a spindle driver coupled
to the spindle to thereby cause the friction stir tool to rotate;
and an active controller for controlling, to a power setpoint,
operation of the friction stir tool, the active controller
configured to adjust an output effort of the spindle driver that is
used to drive the friction stir tool to thereby control power
provided to a friction stir zone.
13. The system of claim 12, wherein the active controller is
further comprises means for receiving spindle flow feedback.
14. The system of claim 12, wherein the active controller is
further configured to approach and maintain a temperature for the
friction stir zone.
15. The system of claim 12, wherein the active controller comprises
a multi-stage nested control loop.
16. The system of claim 15, wherein the multi-stage nested control
loop comprises an inner control loop configured to maintain a power
input level through control of the spindle driver effort, and an
outer control loop that adjusts the power provided to thereby
maintain a temperature in the friction stir zone.
17. The system of claim 12, wherein the friction stir tool is a
welding tool.
18. A method, comprising: controlling, to a power setpoint, power
provided by a friction stir tool to a friction stir zone using an
active controller, the power provided to the friction stir zone is
controlled by making adjustments to an output effort of a spindle
driver that is used to drive the friction stir tool.
19. The method of claim 18, wherein controlling the power provided
to the friction stir zone includes using the active controller to
maintain a power input level.
20. The method of claim 18, further comprising: using a multi-stage
nested control loop.
Description
RELATED APPLICATIONS
[0001] This application is a continuation of U.S. patent
application Ser. No. 13/853,798, filed on Mar. 29, 2013, entitled
"EFFORT MODULATION FOR PROCESS CONTROL OF FRICTION STIR
OPERATIONS," which claims priority to and the benefit of U.S.
Provisional Application 61/618,266 entitled "TORQUE MODULATION FOR
PROCESS CONTROL OF FRICTION STIR PROCESSING" filed on 30 Mar. 2012
and to U.S. Provisional Application 61/769,920 entitled "EFFORT
MODULATION FOR PROCESS CONTROL OF FRICTION STIR PROCESSING" filed
on 27 Feb. 2013. Each of the foregoing applications are
incorporated herein by reference in their entirety.
BACKGROUND
Field
[0002] This invention relates generally to friction stir operations
such as friction stir welding (FSW), friction stir processing
(FSP), friction stir mixing (FSM), friction surfacing, friction
hydro pillar processing, friction stir forming, friction extrusion,
and friction stir spot welding (FSSW) (and hereinafter referred to
collectively as "friction stir operations"). More specifically, the
present invention relates to the use of control algorithms that
utilize control of effort (e.g. torque) in order to improve control
of the friction stir operations.
Description of Related Art
[0003] Friction stir welding is a technology that has been
developed for welding metals and metal alloys. The FSW process
often involves engaging the material of two adjoining workpieces on
either side of a joint by a rotating stir pin. Force is exerted to
urge the pin and the workpieces together and frictional heating
caused by the interaction between the pin, shoulder and the
workpieces results in plasticization of the material on either side
of the joint. The pin and shoulder combination or "FSW tip" is
traversed along the joint, plasticizing material as it advances,
and the plasticized material left in the wake of the advancing FSW
tip cools to form a weld. The FSW tip can also be a tool without a
pin but only a shoulder for processing of another material through
FSP.
[0004] FIG. 1 is a perspective view of a tool being used for
friction stir welding that is characterized by a generally
cylindrical tool 10 having a shank 8, a shoulder 12 and a pin 14
extending outward from the shoulder. The pin 14 is rotated against
a workpiece 16 until sufficient heat is generated, at which point
the pin of the tool is plunged into the plasticized workpiece
material. Typically, the pin 14 is plunged into the workpiece 16
until reaching the shoulder 12 which prevents further penetration
into the workpiece. The workpiece 16 is often two sheets or plates
of material that are butted together at a joint line 18. In this
example, the pin 14 is plunged into the workpiece 16 at the joint
line 18.
[0005] FIG. 2 is a cross-sectional view of a typical tool 10, but
should not be considered as limiting. A collar 22 is shown gripping
both the shank 8 and the FSW tip 24, wherein the FSW tip is
comprised of the shoulder 12 and the pin 14.
As the tool 10 is rotated, torque (i.e. effort) is transmitted from
the rotating shank 8 to the collar 22 and then to the FSW tip
24.
[0006] Referring to FIG. 1, the frictional heat caused by
rotational motion of the pin 14 against the workpiece material 16
causes the workpiece material to soften without reaching a melting
point. The tool 10 is moved transversely along the joint line 18,
thereby creating a weld as the plasticized material flows around
the pin 14 from a leading edge to a trailing edge. The result is a
solid phase bond 20 at the joint line 18 that may be generally
indistinguishable from the workpiece material 16 itself, in
contrast to welds using other conventional technologies. It is also
possible that the solid phase bond 20 is superior to the original
workpiece material 16 because of the mixing that occurs.
Furthermore, if the workpiece material is comprised of different
materials, the resulting mixed material may also be superior to
either of the two original workpiece materials.
[0007] It is observed that when the shoulder 12 contacts the
surface of the workpieces, its rotation creates additional
frictional heat that plasticizes a larger cylindrical column of
material around the inserted pin 14. The shoulder 12 provides a
forging force that contains the upward metal flow caused by the
rotating tool pin 14.
[0008] During friction stir welding, the area to be welded and the
tool 10 are moved relative to each other such that the tool
traverses a desired length of the weld joint. The rotating friction
stir welding tool 10 provides a continual hot working action,
plasticizing metal within a narrow zone as it moves transversely
along the workpiece materials 16, while transporting metal from the
leading edge of the pin 14 to its trailing edge. As a weld zone
cools, there is typically no solidification as no liquid is created
as the tool 10 passes. It is often the case, but not always, that
the resulting weld is a defect-free, recrystallized, fine grain
microstructure formed in the area of the weld.
[0009] Travel speeds of friction stir tools change depending upon
the specific type of friction stir operation being performed, the
application and the material being processed. Some examples of
travel speeds are over 1 m/min with rotation rates of 200 to 3000
rpm. These rates are only examples and should not be considered to
be limiting the operation of the present invention. Temperatures
reached are usually close to, but below, solidus temperatures.
Friction stir welding parameters are a function of a material's
thermal properties, high temperature flow stress and penetration
depth.
[0010] Friction stir welding has several advantages over fusion
welding because 1) there is no filler metal, 2) the process can be
fully automated requiring a relatively low operator skill level, 3)
the energy input is efficient as all heating occurs at the
tool/workpiece interface, 4) minimum post-weld inspection is
required due to the solid state nature and extreme repeatability of
FSW, 5) FSW is tolerant to interface gaps and as such little
pre-weld preparation is required, 6) there is typically no weld
spatter to remove, 7) the post-weld surface finish can be
exceptionally smooth with very little to no flash, 8) there is
often no porosity and oxygen contamination, 9) there is little or
no distortion of surrounding material, 10) no operator protection
is required as there are no harmful emissions, and 11) weld
properties are often improved. Throughout this document, friction
stir operations will be considered to include all processes that
can be performed using a friction stir tool, including but not
limited to friction stir welding, friction stir processing,
friction stir spot welding and friction stir mixing.
[0011] In friction stir welding, the temperature of the process
zone or friction stir zone affects the properties of the resulting
weld and has a dramatic effect on tool life such as in PCBN
(polycrystalline cubic boron nitride) tools. This tool example
should not be considered to be limiting. An active control system
that changes process parameters to control weld temperature is
desirable.
[0012] Controlling the weld temperature throughout the length of
the weld is an important undertaking because weld properties, such
as fracture toughness and corrosion resistance, vary with weld
temperature. If specified properties are desired throughout the
weld, the weld temperature must be adjustable and in control
throughout the length of the weld.
[0013] The short tool life of PCBN tools limits the application of
friction stir processing (FSP) in steels and other high softening
temperature (or high melting temperature) materials. Controlling
tool temperature should increase the tool life of PCBN tools
because some temperature problems can cause damage. For example, if
the temperature is too low, the tool is overstressed by forces that
increase as tool temperature decreases. In contrast, if the tool
temperature is too high, PCBN tools fail quickly because of
chemical wear. High temperature can also cause creep in the locking
collar allowing the PCBN insert to rotate. Point stresses will then
likely exist at the corners of the insert on cooling that may lead
to cracking and failure.
[0014] The first efforts to control weld temperature used passive
control techniques. Researchers considered that the weld
temperature was proportional to weld pitch (spindle speed)/(travel
speed) or various "pseudo heat indexes" which are functions of
spindle speed and travel speed. Passive control techniques assume
that the process has reached a self-limiting equilibrium
condition.
[0015] Passive control techniques are not adequate for temperature
control because equilibrium conditions may not exist along the
length of a weld. Causes of temperature changes along the length of
the weld include: inadequate cooling of the tool or backing plate;
changes in the thermal boundary condition; and insufficient time to
reach equilibrium. Passive control techniques are not versatile
because they do not adjust for process disturbances.
[0016] FIG. 3 is a block diagram of an active control system for a
friction stir welding machine. A friction stir welding machine
includes a friction stir welding tool that is coupled to a spindle,
which in turn is coupled to a spindle motor. The friction stir
welding machine also includes a surface for clamping or supporting
workpieces to be friction stir welded. A friction stir welding
machine can be controlled by the active control system to thereby
perform friction stir welding.
[0017] FIG. 4 shows that the prior art also teaches a two-stage
control model that contains an inner loop that controls the spindle
speed to keep power constant and an outer loop for setting the
desired power based on weld temperature. FIG. 4 is a block diagram
showing a close-up of the inner loop of FIG. 3.
[0018] FIG. 3 shows a temperature control algorithm where T is
temperature, w is spindle speed, M is torque and P is power.
[0019] FIG. 5 is provided as a block diagram of a close-up view of
a stir zone from FIGS. 3 and 4. A plant is the combination of the
spindle motor and the stir zone. For the outer loop the reference
is desired temperature, the controlled variable is temperature and
the manipulated variable is power. For the inner loop the reference
is commanded power and the controlled variable is power.
[0020] The prior art shown in FIG. 3 teaches that the inner loop
adjusts spindle speed to maintain constant power. It is critical to
the present invention to understand this aspect of the prior art.
The purpose of the inner loop is to maintain a desired power
provided to the stir zone.
[0021] The relationship between power and spindle speed is given
by:
P=.omega.M Equation (1)
where P is power, .omega. is spindle speed in radians/s and M is
torque. Power control by adjusting spindle speed uses torque
feedback and Equation (1) to solve for the spindle speed required
to produce the desired power. A block diagram for power control by
adjusting spindle speed is shown in FIG. 4 where M(filtered) is the
filtered value of the reported torque. This control scheme assumes
that torque is constant during each PLC time step. Slew limits
define the maximum acceleration of the spindle. Slew limits are
used to prevent the system from going unstable due to noise in
torque feedback. If slew limits are set too high, the system will
amplify noise in the torque feedback and become unstable.
[0022] Results for power control by adjusting spindle speed with a
slew rate of 0.83 RPM/s are shown in FIG. 6C. When power control is
enabled, the torque is high and the RPM is low. As the weld
progresses the plate heats and softens causing a decrease in
torque. As torque decreases, the RPM increases, thereby maintaining
constant power. Large spikes in power persist throughout the weld
as shown. Power spikes occur because the spindle motor attempts to
accelerate the spindle instantaneously to achieve the commanded
RPM. Adjusting spindle speed to control power results in large
power spikes throughout the weld. The average of these power spikes
is the desired power value.
[0023] It would be an advantage over the prior art to provide an
improved active control system that does not use spindle speed to
control power.
BRIEF SUMMARY
[0024] The present invention is a system and method for making
adjustments to effort from a spindle driver in order to provide
constant power to a friction stir zone during friction stir
operations and control the temperature within the friction stir
zone and thereby improve the weld or other result of the friction
stir operations. In certain embodiments, a commanded or requested
power is obtained by modulating the effort variable in response to
the flow variable feedback. The flow feedback variable may be
measured or estimated.
[0025] Furthermore, as disclosed herein, a friction stir system for
controlling power provided to a friction stir zone during friction
stir operations may include a friction stir tool, a spindle that is
coupled to the friction stir tool, a spindle driver that is coupled
to the spindle to thereby cause the friction stir tool to rotate
and an active controller for controlling operation of the friction
stir tool, wherein the active controller adjusts an output effort
of the spindle driver that is used to drive the friction stir tool
to thereby control power provided to a friction stir zone.
[0026] A corresponding method for using an active controller to
control the temperature of a weld during friction stir operations
may include providing an active controller for controlling
operation of a friction stir tool, controlling power input by the
friction stir tool to a friction stir zone using the active
controller to control a temperature of the friction stir zone,
wherein the power input is controlled by making adjustments to an
output effort of a spindle driver that is used to drive the
friction stir tool.
[0027] These and other objects, features, advantages and
alternative aspects of the present invention will become apparent
to those skilled in the art from a consideration of the following
detailed description taken in combination with the accompanying
drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0028] FIG. 1 is a perspective view of a tool as taught in the
prior art for friction stir welding;
[0029] FIG. 2 is a cut-away profile view of an FSW tip, a locking
collar and a shank from the prior art;
[0030] FIG. 3 is simplified block diagram for an active control
system for temperature control as taught in the prior art;
[0031] FIG. 4 is a block diagram for a spindle speed power control
system as taught in FIG. 3;
[0032] FIG. 5 is a block diagram for a close-up of a plant for a
control model of the prior art;
[0033] FIGS. 6A and 6B are system block diagrams depicting power
control via modulation of effort;
[0034] FIG. 6C is a graph showing power control by adjusting
spindle speed;
[0035] FIG. 7 shows that constant spindle speed welds contain
variation in power and temperature;
[0036] FIG. 8 is a block diagram of the inner loop of an active
controller;
[0037] FIG. 9 is a graph showing the results when torque (i.e.
effort) is adjusted to keep power constant;
[0038] FIG. 10 is a graph showing that constant spindle speed welds
contain variation in power and temperature;
[0039] FIG. 11 is a graph of temperature response;
[0040] FIG. 12 is a graph showing that poor reported torque
resolution results in unreported torque dynamics;
[0041] FIG. 13 is a graph showing linear torque-spindle speed
assumption used for comparing spindle speed control to torque (i.e.
effort) control;
[0042] FIG. 14a is a graph showing a step increase in power
adjusting spindle speed;
[0043] FIG. 14b is a graph showing a step increase in power
adjusting torque; and
[0044] FIG. 15 is a block diagram showing the inner loop and the
outer loop of an active controller.
DETAILED DESCRIPTION
[0045] Reference will now be made to the drawings in which the
various elements of the present invention will be given numerical
designations and in which the invention will be discussed so as to
enable one skilled in the art to make and use the invention. It is
to be understood that the following description is only exemplary
of the principles of the present invention, and should not be
viewed as narrowing the claims which follow.
[0046] As disclosed herein, the effort of a friction stir system is
modulated based on flow feedback. Based on the principle of system
similarity in system dynamics a power, P, is defined as an effort,
e, multiplied by a flow, f. This relationship is shown in Equation
(2) as:
P=e*f (2)
[0047] For example, power in the mechanical rotation domain is
defined by torque (effort) multiplied by angular velocity (flow).
Examples of various power domains, efforts and flows are shown in
Table 1 below.
TABLE-US-00001 TABLE 1 Power domains Domain Effort (e) Flow (f)
Mechanical (translation) Force F [N] Velocity v [m/s] Mechanical
(rotation) TorqueM[Nm] Angular velocity .omega. [rad/s] Pneumatic
Pressure p [Pa] Volume Flow Q [m.sup.3/s] Hydraulic Pressure p [Pa]
Volume Flow Q [m.sup.3/s] Electric Voltage u [V] Current i [A]
Magnetic Magnetomotive Flux rate force
[0048] One method of modulating output effort of the spindle driver
includes modulating the input flow or effort of the spindle driver.
If the spindle driver is a transformer, output effort can be
effectively modulated by modulating input effort. If the spindle
driver is a gyrator, output effort can be effectively modulated by
modulating input flow. For example, an electric motor driver is a
gyrator. The input flow (e.g. current) can be used to modulate the
output effort (torque).
[0049] This detailed description discloses specific examples that
demonstrate controlling an electric spindle motor. Electric motors
perform a transformation from the electrical domain, through the
magnetic domain to the rotational mechanical domain. In particular,
for an electric motor the output torque is controlled by means of
adjusting the input current. A device that transforms effort in one
domain to flow in another domain, or vice versa, is known as a
gyrator. Many devices that provide spindle energy are gyrators.
Although the electric spindle motor is a representative embodiment,
the invention is not limited to such embodiments. For example, the
power of a friction stir system could be controlled by monitoring
any form of flow (including those presented in Table 1) and
modulating any form of effort (including those presented in Table
1).
[0050] As used herein, the "temperature" of the friction stir zone
refers to any direct measurements or estimated measurements and may
include any useful method of measuring or approximating weld (i.e.
friction stir zone) temperature. For example the friction stir zone
temperature may be approximated by measured tool temperature or
backing plate temperature. Methods to measure the internal
temperature of the friction stir operation zone, such as ultrasonic
technology, may also be used.
[0051] System diagrams depicting power control via modulation of
effort by an active controller 610 are shown in FIG. 6A and FIG. 6B
where P.sub.cmd is commanded power, f.sub.rep is reported flow,
f.sub.real is real flow and e.sub.cmd is commanded effort. It is
important to note the reported flow may be a measured value or an
estimated value. FIG. 6A shows reported flow as a measured value
while FIG. 6B shows reported flow as an estimated value. One of
skill in the art will appreciate that there are a variety methods
of providing reported flow and that FIG. 6A and FIG. 6B are
examples for reference only.
[0052] The present invention is an improved control system for
friction stir operations. Research shows that power provided to a
friction stir zone leads tool temperature. Due to the inertia
associated with the spindle, power control is best achieved by
commanding torque (i.e. effort) rather than spindle speed (i.e.
flow) as will be explained.
[0053] The active control system disclosed herein was initially
developed for a dual loop control system for FSW where the inner
loop maintains constant power and the outer loop adjusts power to
maintain constant temperature. Although much of the operation of
the present invention as described herein refers to a dual loop
control system for controlling welding operations, the invention is
not limited to a dual loop control system nor to welding
operations.
[0054] FIG. 7 is provided to show the hardware elements 700 of a
friction stir welding system that are relevant to an understanding
of the present invention. These elements include a spindle motor
710 that is part of a friction stir welding machine (not shown).
The spindle motor 710 is coupled to a spindle 720. The spindle in
turn is coupled to a friction stir welding tool 730 that is used to
perform friction stir welding in all its various forms and on both
low melting temperature and high melting temperature materials.
There are various configurations of friction stir welding tools and
this document should not be seen as limiting the variety that can
be used with the present invention.
[0055] The present invention uses the relationship between torque
(i.e. effort) and power as expressed in Equation (2) to control the
FSW process. In this first embodiment, the spindle motor (i.e.
driver) 710 is a variable frequency AC induction motor. Although an
induction motor is shown, any appropriate driving mechanism may be
used including mechanical, pneumatic, hydraulic, electric and
magnetic drivers. When running under torque control with an
induction motor, the motor interface control software commands and
maintains a constant torque using flux vector control as directed
by an active controller (not shown). The desired torque is obtained
from the spindle motor by controlling the current fed to the
induction motor. This control diagram is shown in FIG. 8.
[0056] It should be noted that torque (i.e. effort) control without
spindle speed (i.e. flow) feedback is unstable. A constant torque
can only be maintained for a short time. The torque can only be
controlled when the load supports the torque. Torque that is
greater than the natural process torque leads to greatly decreasing
loads causing an exponential increase in spindle speed. This is
because when spindle speed and torque are increasing, the power
increases and the material softens. Conversely, if the torque is
lower than the natural process torque, the spindle speed will
decrease exponentially as the material cools and hardens. Process
variation causes the commanded torque to be either too low or too
high to maintain equilibrium causing the spindle to rapidly
decelerate until it stops or accelerate until machine safety limits
are triggered.
[0057] It has been determined that controlling power provided to a
friction stir zone by adjusting torque (i.e. effort) in response to
changes in spindle speed (i.e. flow) is a stable process. Torque
increases in response to decreasing spindle speed to maintain a
constant power. Torque decreases in response to increasing spindle
speed to maintain a constant power.
[0058] FIG. 9 shows the results from a weld where torque (i.e.
effort) is adjusted to maintain constant power. As the workpiece is
heated, the material softens. The torque and RPM signals are mirror
images of each other. The spindle speed increases as the material
softens. The torque (i.e. effort) decreases with increasing spindle
speed to maintain the constant power. FIG. 9 shows that power
control achieved by torque control is a stable process.
[0059] FIG. 10 shows that constant spindle speed welds contain
variation in power and temperature. The measured power contains
power spikes that persist throughout the weld. The filtered power
value varies throughout the weld in response the changes in the
process.
[0060] Previously, it was shown in FIG. 6C that power control
achieved by adjusting spindle speed results in a constant average
power. However, the actual power contains torque spikes throughout
the weld. The temperature response shows that the tool gradually
rises in temperature until thermal equilibrium is reached and the
temperature is constant.
[0061] FIG. 9 showed that power control achieved a torque (i.e.
effort) control to maintain constant power produces a constant
power with negligible variation. The temperature response is
linear. This indicates that the tool is still heating and will
reach a steady state-temperature.
[0062] A demonstration of the principles of the present invention
was performed and the results are shown in FIG. 11. A power control
weld adjusting torque to keep power constant was run for 406.4 mm
(16 in) at 2.238 kW then 609.6 mm (24 in) at 2.536 kW in AA 7075.
The temperature response is shown in FIG. 11. Before the change in
power is commanded, temperature is constant. After the change in
power is commanded, the temperature increases logarithmically until
a new steady-state temperature is reached. These results show that
changes in power lead changes in temperature.
[0063] In the prior art when spindle speed is adjusted to control
power, the commanded spindle speed is determined by the torque that
is reported by the spindle controller. When torque is adjusted to
control spindle speed, a command that is transmitted to control
torque is sent to the spindle controller. The difference in
resolution between reported and commanded torque affects the design
of the power control loop.
[0064] Torque measurements used in this embodiment are reported by
the spindle motor controller. However, the reported torque fails to
capture torque dynamics. FIG. 12 shows the torque reported by the
controller and the torque calculated using load cells on the FSW
machine in response to a sinusoidal torque command. The sinusoid is
reported as a square wave due to limited resolution in reported
torque.
[0065] The sinusoid can be seen in the torque calculated from
forces measured by load cells. Torque calculated using load cells
has poor signal quality due to cross talk. Because changes in
torque during temperature and power control are small spindle
torque will be assumed to be equal to commanded torque.
[0066] In developing the present invention, a dynamics analysis was
performed to compare adjusting spindle speed and adjusting torque
to maintain constant power. Returning to FIG. 7, M(mtr) is the
motor torque, M(spn) is spindle torque, D(mtr) is the diameter of
the motor pulley, D(spn) is the diameter of the spindle pulley,
M(b) is torque lost due to bearings, .omega.(spn) is the rotational
velocity of the spindle, and J is the mass moment of inertia of
associated with the spindle. The gear ratio, R, is given by:
R=D(spn)ID(mtr) Equation (3)
and has a value of 2.5.
[0067] The relationship between torque and spindle speed is derived
using FIG. 7 and is given by:
RM(mtr)=M(spn)+M(b)+J.omega.(spn) Equation (4)
The effective motor torque, M(mtr_eff), is given by:
M(mtr_eff),=RM(mtr) Equation (5)
Reported torque in experimental data is M(mtr_eff). Substituting
Equation (4) into Equation (5) yields:
M(mtr_eff)=M(spn)+M(b)+J.omega.(spn) Equation (6)
For the purpose of comparing spindle speed control to torque
control, it is assumed that .omega. (spn) and M(spn) have a linear
relationship as shown in FIG. 13 and given by:
.omega.(spn)=M(spn)/B Equation (7)
where it is assumed that J, M(b) and B are constant.
[0068] Because the spindle motor's maximum torque is finite,
changes in commanded spindle speed result in large torque spikes,
shown in FIG. 14a, in attempt to instantaneously accelerate the
spindle to the desired rotational velocity.
[0069] The spindle speed response to a step change in effective
motor torque was found to be exponential. However, the spindle
speed response to changes in torque contains no
discontinuities.
[0070] Experimental results for step changes in spindle speed and
torque (i.e. effort) validate that torque (i.e. effort) has a
derivative relationship with spindle speed, and spindle speed has
an integral relationship with torque (i.e. effort). FIG. 14a shows
the effective motor torque response to a step change in RPM. At the
instant the change in desired spindle speed is made, the motor
attempts to instantaneously accelerate the spindle in order to
achieve the desired spindle speed. This results in a large spike in
motor torque.
[0071] FIG. 14b shows the spindle speed response to a step increase
in power where power control was obtained by adjusting torque (i.e.
effort). When the change in power is commanded, the torque
instantly increases slightly to obtain the desired power value. Due
to the increased power the material begins to soften causing the
spindle speed (i.e. flow) to increase and the motor torque or
effort to decrease. The spindle speed response to torque contains
no discontinuities.
[0072] Experimental data shown in FIGS. 14a and 14b indicate that
higher spindle speeds correspond to lower torques and lower spindle
speeds correspond to higher torques. These results suggest that the
slope shown in FIG. 13 should be negative.
[0073] The control methods presented assume M(mtr_eff) is
approximately equal to M(spn). Torque (i.e. effort) has a
derivative relationship with spindle speed (i.e. flow). When a
change is RPM is commanded, the spindle motor attempts to
instantaneously accelerate the spindle to a new RPM causing a spike
in motor torque. A near instantaneous acceleration of the spindle
motor would cause a large difference between motor torque and
spindle torque. The motor torque is not approximately equal to
spindle torque when a change in RPM is commanded.
[0074] Experimental data shows that when torque is adjusted to keep
power constant, the difference between motor torque and spindle
torque is much smaller than when spindle speed is used to keep
power constant.
[0075] When a step change in power is commanded as shown in FIG.
14a, the torque error during the spindle spike is 203.0 Nm (149.7
ft-lb) which corresponds to a 406% error in power at the high power
level and a 565% error at the low power level. In contrast, for
power control by adjusting torque as shown in FIG. 14b, the error
associated with the spindle acceleration after the change in power,
where acceleration is greatest, is 7.4 Nm (5.46 ft-lb) which
corresponds to 10% error m power.
[0076] Adjusting spindle speed to keep power constant is
undesirable because reported torque is a poor control signal.
Torque spikes cause the difference between motor and spindle torque
to be as high or even higher than 400%. The reported torque signal
has low resolution, adding to the error.
[0077] The main advantage of using torque to control power is the
avoidance of artificial torque spikes caused by attempting to
change the RPM instantaneously. Using torque control to control
power results in a smooth power signal with low uncertainty. The
difference between the motor torque and actual spindle torque is
proportional to the acceleration of the spindle. Low uncertainty
exists because under torque control, power and spindle speed change
in a controlled fashion.
[0078] Having addressed the inner control loop, attention is now
directed to the outer control loop. FIG. 15 is a block diagram
showing both the inner and outer loops of the first embodiment of
the present invention. Proportional-integral-derivative (PID)
controllers are the most commonly used type of closed-loop feedback
controller. The outer loop uses PID control to adjust power to
maintain a commanded temperature. The implementation of PID control
in the outer loop is shown in FIG. 15. P(init) is a constant power
value. The PID controller provides a change in power that is added
to the P(init) term. The P(init) term is the spindle power at the
moment the controller switches from commanding a constant spindle
speed (i.e. flow) to temperature control. Once temperature control
is engaged P(init) is constant until temperature control is
disengaged.
[0079] It is to be understood that the above-described arrangements
are only illustrative of the application of the principles of the
present invention. Numerous modifications and alternative
arrangements may be devised by those skilled in the art without
departing from the spirit and scope of the present invention. The
appended claims are intended to cover such modifications and
arrangements.
* * * * *