FIELD OF THE INVENTION
-
The invention relates to a method of reducing the
cycle time for metal forming and more particularly to a
method of reducing the part-to-part cycle time of quick
plastic forming and superplastic forming of a metallic
sheet alloy into an automotive sheet metal component by
locally modifying a die surface to control friction.
BACKGROUND OF THE INVENTION
-
Typically, an automobile sheet metal component is
made by stamping or shaping a low carbon steel or an
aluminum alloy sheet stock into a desired shape. Automobile
sheet metal components are formed and welded or
otherwise joined to form vehicle body or closure panels.
It is a goal to make the panels from as few parts as
possible in order to minimize manufacturing cost and the
overall weight of the vehicle. It is another goal to
make the sheet metal components as quickly as possible
to minimize manufacturing cost. In order to accomplish
these goals, there is an incentive to devise more formable
metal alloys and better forming processes so that
fewer automobile body panels having a more complex shape
can be made and joined to form either the vehicle body
or closure panels rather than welding or bolting together
a myriad of smaller, simpler pieces.
-
Lubrication is a critical aspect of the forming
process. Typically, a lubricant with a low coefficient
of friction is selected to enhance material flow in a
die. By minimizing friction, sticking between a blank
and the die is likewise minimized, and part removal
without distortion is facilitated.
-
Superplasticity is the capability of a material to
develop unusually high tensile elongation with a reduced
tendency toward local necking during deformation at
elevated temperatures. Necking is a defect that results
from excessive local thinning during forming and can
ultimately lead to failure during or after forming.
Alloys which exhibit superplasticity are capable of
being subjected to superplastic forming, wherein portions
of a preform are expanded by the application of
fluid pressure against the surface of a forming member.
The forming member is usually in the form of a die
which produces structures of predetermined shapes. The
expansion of the preform occurs through an increase in
the surface area of the preform produced by an elongation
in the length and a reduction in thickness of individual
material elements.
-
The above process is typically called superplastic
forming or SPF. Recent advances have resulted in a
mixed-mode deformation process termed quick plastic
forming or QPF.
-
In superplastic forming and quick plastic forming
operations, the preform is clamped firmly at its periphery,
thus ideally allowing for material to be stretched
from the area inside of the clamped periphery only.
Thinning of a preform as a result of stretching is
highly uniform except in areas coming in contact with
the forming surface. A unique area that comes in early
contact with the forming surface is the forming member
cavity entrance radius area, i.e., the intermediate
region between the peripheral portion of the preform and
the part expanding into the cavity. As the preform
drapes over the radius area there is a tendency to increase
the local rate of material elongation at sharp
features which, in turn, may produce localized thinning
or necking in this area. The necking can ultimately
lead to splits or tears during subsequent forming. If
the forming cycle is too aggressive at a given temperature,
the blank will neck and/or split just below the
entry radius. Necking makes it difficult to obtain
uniform thickness profiles in the structure and can lead
to failure during forming. To reduce necking, and obtain
uniform thickness profiles, a release coating which
is capable of producing a high coefficient of friction
has been used. The use of the release coatings in specific
areas results in an increase in the frictional
force and a lower net force causing material expansion,
and, in turn, reduced necking at the radius area. The
alternative to preventing necking is to form components
at very slow cycle times, which is prohibitive for high
volume production.
-
While conventional thinking in metal forming has
been that increasing the amount of lubricant enhances
forming, it has been shown that in specific cases, using
lubricants having different coefficients of friction
actually improves formability. Varying lubricant types
across a blank may be plausible for low volume applications.
However, such a method is difficult to use in
high volume automobile production. In addition, forming
lubricants are costly and can lead to surface blemishes
on automobile outer body components.
-
It would be desirable to produce an automobile
sheet metal component using a superplastic for quick
plastic forming process where forming time, necking,
excessive thinning, and splitting are minimized by
locally modifying a die surface to control friction.
SUMMARY OF THE INVENTION
-
Consistent and consonant with the present
invention, a method of producing an automobile sheet
metal component where forming time, necking, thinning,
and splitting are minimized by locally modifying a die
surface to control friction has surprisingly been
discovered.
-
The method of producing an automobile sheet metal
component comprises the steps of:
- providing a metal blank;
- providing a metal forming die having a die surface
to effect forming of the metal blank thereon,
the die surface having at least one coefficient
of friction; and
- forming the metal structure by applying pressure to
the metal blank with the metal forming die;
- forming the metal structure at an elevated
temperature using one of a superplastic forming
and a quick plastic forming process;
wherein at least a portion of the die surface is
modified to change the at least one coefficient
of friction thereon to minimize a cycle time
for the forming and militate against at least
one of necking, localized thinning, and
splitting of the metal blank during the
forming.-
DESCRIPTION OF THE DRAWINGS
-
The above, as well as other advantages of the
present invention, will become readily apparent to
those skilled in the art from the following detailed
description of a preferred embodiment when considered
in the light of the accompanying drawings in which:
-
Fig. 1 is a top view of a panel formed by a die
in accordance with the method of the present
invention;
-
Fig. 2 a sectional view of the panel illustrated
in Fig. 1 taken along line 2-2;
-
Fig. 3 is a table showing pressure-time cycle
data for blanks having various friction
characteristics obtained during experimentation to
arrive at the method of the present invention;
-
Fig. 4 is a table showing pressure-time cycle
data for die surfaces having various friction
characteristics obtained during experimentation to
arrive at the method of the present invention;
-
Fig. 5 is a graphical representation of the
effect of die lubricant on a license plate pocket
panel wall thickness distribution;
-
Fig. 6 is a graphical representation of the
effect of die lubricant on a license plate pocket
panel wall thickness distribution shown in Fig. 5,
compared with a graphical representation of the
effect of die lubricant on a license plate pocket
panel wall thickness where a die has a higher
coefficient of friction at a die entry radius;
-
Fig. 7 is a partial cross-sectional view of the
panel of Figs. 1 and 2 showing a die having a surface
thereof modified to control the die friction; and
-
Fig. 8 is a partial cross-sectional view of the
panel of Figs. 1 and 2 showing a die having a metal
insert to control the die friction.
DESCRIPTION OF THE PREFERRED EMBODIMENT
-
Referring now to the drawings, and particularly
Fig. 1, there is shown generally at 10 a license
plate pocket panel formed by a die in accordance with
the method of the present invention. The license plate
pocket panel 10 was used in the development of the
present invention since there exists therein a very
aggressive die entry radius 12. A cross-section of
the license plate pocket panel 10 is illustrated in
Fig. 2. All testing was performed on a 1.2 mm thick
aluminum alloy 5083-H18 such as that produced by
Pechiney Rolled Products, Ravenswood, WV. Three
lubricants were also used in the testing, a Boron
Nitride (BN) lubricant sold under the trademark
LUBRICOAT, Milk of Magnesia or Mg(OH)2, and a
lubricant sold under the trademark SEALMET. The BN
lubricant and the SEALMET lubricant were supplied by
ZYP coatings, Oak Ridge, TN. The SEALMET lubricant
contained an unspecified mixture of metal oxides. The
BN lubricant was provided in two forms: (a) as an
aerosol spray which contained 97% hexagonal BN and 3%
magnesium silicate with an alcohol/acetone carrier;
and (b) as a paint which consisted of a suspension of
hexagonal BN (25 wt%) and Al2O3 (4 wt%) in a water
carrier. The Milk of Magnesia consisted of an aqueous
suspension of Mg(OH)2 at a concentration of 80 mg of
Mg(OH)2 per ml of water. Forming was conducted on an
up-acting, 320 ton hydraulic press designed for
superplastic forming.
-
Two types of experiments were performed. The
first type involved forming lubricated blanks in a
bare or uncoated die to show the effect on cycle time
by varying lubricity. The blanks were lubricated
with one of two lubricants: (1) BN LUBRICOAT
lubricant or (2) Milk of Magnesia. The license plate
pocket panel 10 was formed with the two different
lubricant conditions at incrementally faster cycle
times ranging from 6 minutes down to 13 seconds, to
determine when necking and splitting occurred. The
difference in cycle time was produced by changing the
pressurization rate of the forming operation, as well
as by increasing the dwelling time at the peak
pressure. For the sake of clarity, results are
compared using cycle time, but could also be easily
compared using pressurization rate. All blanks were
formed at a temperature of 450 degrees Celsius with a
5 minute preheat to ensure proper blank temperature.
The pressure-time cycles for each trial using the BN
LUBRICOAT lubricant or Milk of Magnesia are shown in
Fig. 3. After forming, each of the license plate
pocket panels 10 was evaluated for necks or splits at
or near the die entry radius 12.
-
The second type of experiment consisted of
forming bare or unlubricated blanks in a selectively
lubricated die to show the effects on necking and
splitting by varying die surface friction. The
friction condition was varied across the die by one
of two methods. The first method, called Pattern 1,
involved spraying lubricants with extremely different
lubricity on the two halves of the die. This was
accomplished by masking one half of the die with tape
and spraying a thick layer (approx. 0.001") of BN
LUBRICOAT lubricant on the die using the BN lubricant
spray. The surface was burnished by rubbing with wax
paper to create a very slippery surface. The second
half of the die was then sprayed with the SEALMET
coating without any subsequent polishing or
burnishing. This provided a very rough surface with
high friction. The second method for varying friction
in the die, called Patterns 2, 3, 4, and 5, involved
masking off selected regions of the die and then
spraying with the BN lubricant aerosol spray. The
die surface was then burnished as described
previously. After burnishing, the tape was removed
leaving areas of bare tool surrounded by highly
lubricious BN lubricant. The license plate pocket
panel 10 was formed under similar conditions to the
previously described lubricated blank trials to
determine the onset of necking and splitting. The
pressure-time cycles for each trial are shown in Fig.
4.
Bare Die Trials
-
Blanks lubricated with either BN lubricant or Milk
of Magnesia were formed in an unlubricated license
pocket die at successively faster pressurization rates
(i.e. faster cycle times) until necking or splitting
occurred. The goal of these trials was to establish the
effect of lubricity on cycle time. BN lubricant and Milk
of Magnesia have been previously shown to exhibit
different lubricity during elevated temperature friction
testing, with BN lubricant giving a significantly lower
coefficient of friction.
-
Referring now to Fig. 3, blanks coated with Milk
of Magnesia could be formed using cycles as fast as
13 seconds without evidence of necking or splitting.
However, the license plate pocket panels 10 formed
using the BN lubricant coated blanks exhibited
necking, at cycles of 4 minutes or less, and
splitting at cycles of 2 minutes or less. Thus, the
"poorer" lubricant, Milk of Magnesia, actually
produced a lower cycle time than the "better"
lubricant, BN. These results clearly demonstrate that
frictional modifications can significantly affect
cycle time. It should be noted, however, that the
forming of parts with a poor lubricant often leads to
sticking during part release, which affects the
dimensional accuracy of the part. In addition, a poor
lubricant can also cause formability problems in very
deep sections, as the material will be unable to flow
into the critical feature areas without splitting.
Lubricated Die Trials
-
The goal of the second set of experiments was to
determine whether locally tailored coefficients of
friction on the surface of the die could produce the
same cycle time reductions that were observed for
blanks lubricated with BN lubricant and Milk of
Magnesia. Four different lubricant test patterns were
produced. Each test pattern will be described below,
followed by the forming results for experiments using
the described pattern. The pressure-time cycles and
trial results are summarized in Fig. 4.
Pattern 1
-
Pattern 1 was made by coating one-half of the
die with the lubricious BN lubricant and the other
half with the non-lubricious SEALMET lubricant. The
purpose for testing this first pattern was to
determine whether drastically different friction
conditions in only a portion of the die produced the
same effect as when the entire die or blank consisted
of the same frictional condition. The license plate
pocket panels 10 were formed using different
pressurization rates to determine the onset of
necking and splitting.
-
For cycle times greater than 4 minutes, no
significant difference in necking behavior was
observed between the two sides of the die. Both
halves of the die successfully formed without
necking. For cycle times between 2 and 4 minutes,
necking at the die entry radius 12 was observed on
the side of the license plate pocket panel 10 that
was provided with the "good" lubricant (BN). The
SEALMET lubricant side of the die showed no evidence
of necking. The difference in friction or roughness
between the two coatings is demonstrated by examining
the differences on the die side of the formed license
plate pocket panels 10. The BN lubricant side of the
license plate pocket panel 10 was very clean with no
evidence of galling or scratching, indicating low
friction. The SEALMET lubricant side of the license
plate pocket panel 10 showed evidence of galling
marks, scratches, and other blemishes indicating
significant interaction between the tool and the
blank during forming.
-
For cycle times at or below 2 minutes, splitting
occurred, but only on the side of the license plate
pocket panel 10 which was formed in the BN lubricant
side of the die. These results indicate that the
presence of BN lubricant on half of the tool
controlled the ability to form the part. Necking and
splitting were observed at similar cycle times to the
tests with BN lubricant coated blanks in the bare die
trial section above.
-
The thickness profile for each "side" of a
license plate pocket panel 10 formed in 4 minutes was
also measured, as shown in Fig. 5. The side of the
license plate pocket panel 10 formed with the SEALMET
lubricant coating showed more overall variation in
thickness than the side of the license plate pocket
panel 10 formed with BN lubricant. The die entry
radius 12 regions were thicker on the SEALMET
lubricant side of the license plate pocket panel 10,
but the bottom corners were thinner. The BN
lubricant side showed a more uniform thickness across
the entire bottom of the license plate pocket panel
10. The thickness profile indicates that while the
SEALMET lubricant side of the license plate pocket
panel 10 did not exhibit necking, it exhibits higher
strain values at the bottom of the pocket and a less
uniform strain distribution than the BN lubricant
side of the panel 10. This would become an increased
issue of concern if the depth of the license plate
pocket panel 10 were increased.
Pattern 2
-
Pattern 2 was produce by masking off specific
regions of the die to produce unlubricated areas,
while the remainder of the die was coated with BN
lubricant. The goal of evaluating Pattern 2 was to
determine whether locally increasing friction in the
vicinity of the die entry radius 12 could prevent
necking and to determine the optimal location for
increasing friction. Regions were masked off either
slightly above the die entry radius 12 on a plateau
14, on the die entry radius 12, or slightly below the
die entry radius 12 on a wall 16, as illustrated in
Fig. 1. Some of the die entry radius 12 regions were
not masked off for comparison. The trials were
performed at a temperature of 510 degrees Celsius to
help exaggerate the necking phenomena. The gas
pressure time cycles for these trials are summarized
in Fig. 4.
-
Initial trials were conducted using a
seven-minute cycle time with pressurization rates of
either 90 psi/min or 30 psi/min. In both trials, the
license plate pocket panels 10 split catastrophically
in the areas where there was full BN lubricant
coverage. A subsequent trial was performed at a
slower cycle of 12 minutes with a pressurization rate
of 15 psi/min. In that trial, the license plate
pocket panel 10 split at the die entry radius 12
along the entire side, except the region where the
die entry radius 12 was void of BN lubricant. The
split occurred right up to the point where the
lubricant was removed from the die, at which point it
arrested. The split traveled into the region where
the lubricant was removed from the plateau 14. Some
slight necking was observed at the die entry radius
12. No necking was observed in the region where the
die entry radius 12 was void of lubricant. However,
removing the lubricant either above or below the die
entry radius 12 did not prevent necking, indicating
that the best pattern for minimizing necking is to
remove the lubricant from the region right at the die
entry radius 12.
Pattern 3
-
The results from lubricant Pattern 2 indicated
that locally increasing friction at the die entry
radius 12 could prevent necking. To evaluate this, a
third pattern was studied where the die entry radius
12 on half of the die did not have lubricant while
the other half of the die was completely coated with
BN lubricant. Blanks were formed using three
different ramp rates. 200 psi/min, 100 psi/min, and
25 psi/min at a temperature of 450 degrees Celsius.
In all three cases, the license plate pocket panels
10 split during forming on the side of the die where
the lubricant was present on the die entry radius 12.
No splitting or necking was observed on the side of
the die where the lubricant was removed from the die
entry radius 12. In fact, the split that initiated on
the side of the die where the die entry radius 12 was
lubricated did not spread into the region of the die
where the die entry radius 12 was bare. The gas
pressure time cycles for these trials are summarized
in Fig. 4.
Pattern 4
-
The success in preventing necking demonstrated
by locally removing the lubricant at the die entry
radius 12 led to testing of a fourth pattern to see
whether the locally tailored friction could produce
an improvement in cycle time. The die entry radius
12 along both sides of the die was masked off to
eliminate lubricant. Bare blanks were formed at an
aggressive ramp rate of 400 psi/min. During testing
for Pattern 4, it was discovered that the region of
the unlubricated die entry radius 12 did not extend
far enough along the side of the die. The license
plate pocket panels 10 split along the edge in the
area where the die entry radius 12 was lubricated.
None of the other regions showed evidence of necking.
While this test was unsuccessful in reducing cycle
time, it clearly demonstrated the effect of friction
at the die entry radius 12 on necking and splitting.
The gas pressure time cycles for these trials are
summarized in Fig. 4.
Pattern 5
-
Pattern 5 involved removing lubrication from the
die entry radius 12 portion of the entire die.
Blanks were formed at a temperature of 450 degrees
Celsius using pressurization rates from 400 to 2000
psi/min. No splitting or necking at the die entry
radius 12 was observed in any of the trials for
Pattern 5. The fastest cycle time was 23 seconds.
-
The thickness distribution in the license plate
pocket panel 10 formed using a cycle of 2.5 minutes
with pattern 5 is shown in Fig. 6. The license plate
pocket panel 10 formed with the tailored BN lubricant
die coating showed a more gradual change in thickness
across the bottom of the license plate pocket panel
10 than the license plate pocket panel 10 shown
previously with the SEALMET lubricant coating
illustrated in Fig. 5. This clearly demonstrates that
the use of dies having tailored friction areas can
significantly reduce cycle time while also preventing
localized splitting and necking. The gas pressure
time cycles for these trials are summarized in Fig.
4.
-
The testing described above demonstrates that
metal forming cycle time can be significantly reduced
by locally modifying the friction characteristics of
a die, i.e. the coefficient of friction. In addition
to cycle time reduction, local control of friction
characteristics in metal forming tools has other
benefits including: control of as-formed panel
thickness distribution; the prevention of localized
thinning; part release is facilitated, thereby
improving the dimensional accuracy of parts; and
improved surface quality of formed parts by
minimizing lubricant buildup on die entry radii.
-
Tailoring the friction characteristics of a die
by surface modification has proven to be a critical
enabler for cycle time reduction. While the methods
used in the present study were excellent for
characterizing the phenomenon, they are not practical
for high speed production. The lubricant sprayed on
the tooling was almost completely removed after only
forming a few of the license plate pocket panels 10
and would require reapplication. It is also difficult
and time consuming to apply the spray coating
uniformly to create a smooth surface that would be
acceptable for an exterior body panel. Thus the
method of the present invention for directly
modifying the die to control friction solves this
problem. Thus, referring to Fig. 7, by directly
modifying a surface 18 of a die 20 to control die
friction, production can proceed without lengthy
interruptions for lubricant application and the like.
Methods for modifying the surface 18 of the die 20
may include chemical etching, laser surface dimpling,
scribing, sand blasting, laser particle injection,
laser ablation, local oxidation, and combinations
thereof, for example. Referring to Fig. 8, a
dissimilar metal insert 22 can also be used to
control friction as desired for the die 20. It is
understood that other methods of controlling die
friction can be used without departing from the scope
and spirit of the invention.
-
From the foregoing description, one ordinarily
skilled in the art can easily ascertain the essential
characteristics of this invention and, without
departing from the spirit and scope thereof, can make
various changes and modifications to the invention to
adapt it to various usages and conditions.
