Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
. CA 02221076 1997-11-14
Back~round of the In~ention
1. Field of ~e Invention
The present invention relates to techniques for superplastic forming of parts, and more
particularly, for control of thinout in such parts.
2. Background of the Invention
Superplastic forming hereinafter (SPF) is a metal forming process used throughout the
aerospace industry for manufacturing detailed parts and built-up structures. The design
flexibility that is offered by SPF has resulted in substantial cost savings in the fabrication of
detail parts and assemblies. Further savings have been apparent in the reduc~ion of
weight in aircraft. The prior art SPF process for manufacturing parts consists of several
steps. ~hese steps are illustrated in FIGIJRES lA to lD and can be ~ 7.e~ as
follows: heating a die to an al~plopliate temperature for a particular metal alloy; placing
a metal sheet, also referred to as a blank, in the die; closing a lid to the die; applying
restrair~ing forces to hold the die and lid together, applying a forming gas pressure to the
blank in order to push the blank into the die cavity; completing the time required in the
forming cycle; and removing the fini~hed part from the die.
FIGURE 2 shows a sçhem~tic plan view of the die with the lid removed for illustration
purposes. The blank or sheet 10 is supported in the regions 12 surrounding the sealed
area 14 by the lower die 16, as shown in FIGURE lA. The double lines 18 outline the
seal area, within which a part will be formed. The reason that the material does not
thinout uniformly is that once the lid is closed on the SPF die, the periphery of the m~t~ri~l
is restrained such that the m~t~ l is not allowed to "draw-in" the material outside of the
seal area.
CA 02221076 1997-11-14
FIGURE 3 shows a schematic cross section view of a part formed by SPF. The dotted
lines show where the part will be cut or trimmed. The run-out is in the die region outside
of the net part area. A correctly designed die will optirnize the run-out configuration so
that thinout is rni.nimized in the part area and maxirnized in the run-out material.
FIGURE 4 is a side elevation cross-section illustrating the thinout problem. For example,
the part thicknesses at 20 and 22 are very tllin, and could potentially be below the
thicknesses specified by the Engineering drawing.
Summary of the Invention
The present invention defines a method of increasing part thickness in specific areas of
SPF details by preferentially cutting out one or more areas of the starting material '~blank".
The cut-out area of the blank becomes stretched so that minimal thinning results in the
area near the periphery of the cutout. The process utilizes a second sheet of material to
push the cutout blank, with cutout(s), onto the die.
Brief Description of the Drawin~s
The foregoing aspects and many of the attendant advantages of this invention will become
more readily appreciated by reference to the following detailed description, when taken in
conjunction with the accompanying drawings, wherein:
FIGURES lA to lD are a side elevation cross-section drawing of a prior art SPF in four
consecutive modes of operation.
FIGURE 2 is a plan view of a die with lid removed.
FIGURF 3 shows a schematic cross section of a part formed by SPF.
FIGURE 4 is a schematic side elevation cross sectional view of a die and part
manufactured by SPF.
, CA 02221076 1997-11-14
FIGURES SA to 5 ~show a cross section of a die and part in six consecutive stages of
the SPF process of this invention.
FIGURES 6A and 6B show how the secondary sheet of this invention can entrap the
blank after forming.
FIGURE 7 is a schematic showing the axial and biaxial stresses around the cutout of this
invention.
~IGURE 8 compares thickness data at various locations on SPF parts made using prior
art techniques and using the techniques of this ~nvention.
Figure 9A shows an orthogonal projection of one-half of a SPF part.
Figure 9B shows a method for manipulatingdie surfaces to arrive at the size of a cutout of
this invention.
FIGURE 9C illustrates a plan view of the blank after manipulating and projecting the die
surfaces of Figure 9B.
Description of the Preferred Embodiment
One of the greatest chaUenges associated with Superplastic Forming is predicting m~t~,ri~l
thinout and then achieving that thinout duling part f~hric~tion. The m~teri~l thinout
challenge is inherent to the SPF process and stems from varying m~te,ri~l thickness across
the part area after SPF. Engineering drawings typically call out a minimllm allowed
m~t~ l thickness across the entire part or in specific regions of the part.
The reason that material does not thinout uniformly is that once the lid is closed on the
SPF die, the periphery of the material is restrained such that the material is not allowed to
CA 02221076 1997-11-14
"draw-in" from the edges. As a result, the m~t~ri~l that will become the part area must be
stretched from the material inside of the seal area.
During forming, the stretching of the material within the seal area progresses until the
m~t~ l eventually contacts the die sur~ace. Upon contact, the mz~t~ l sticks to the die
sur~ace. The remaining m~teri~l that has not yet contacted the die continues to stretch
until it too contacts the die surface and sticks. Once the material is completely formed, the
thinnest regions are generally those that are the last to form. These regions equate to the
deepest areas of the die and radii, in particular spherical radii (corners).
Since material thinout is dependent on the die geometry, the die design is ~ritical in
achieving the proper material thinout. Of specific importance is the die "run-out", which
- is the die region outside the net part area. A correctly designed die will optimize the "run-
out" configuration so that thinout is minimi7~d in the part area and maximized in the "run-
out" material.
Once the part area and "run-out" of the die have been machined and the first SPF part is
formed, there are only a few options for recouTse if the part is too thin according to
engineering drawing requirements. The two most common options are: (1) Start with a
thicker gauge of blank m~t~,ri~l, (2) Preform the blank prior to forming it into the final
part configuration.
The former option is the easier of the two options to irnplement and provides relatively
quick results for thickness analysis. However, it is not a ~ua~ ee for achieving the
correct minimnm thickness since adding thickness to the starting blank does not equate to
a sufficient thickness increase in the thinnest areas of the part. Furthermore, an increase in
the starting m~t~,ri7l1 gauge adversely effects the part weight.
. CA 02221076 1997-11-14
The latter option, designing a prefor n for the blank, carries a fair amount of risk.
Designing a successful preform geometry potentially requires several iterations, an
expensive and tinne consuming process. As with increasing the material gauge, preforming
will not guarantee a successful part.
With this invention a third option becomes available for selectively increasing the material
thickness in specific regions of the part. This option too, does not guarantee that the
minimum material thinout will be attained. However, through a combination of increasing
the starting gauge and utilizing this third option, the odds of attaining a successful part are
significantly increased.
Depending on the part configuration, it is possible to minimi7e the material thinout by
placing a strategically located cutout(s) in the starting blank. The typical applicable part
configuration is one that has an area of the net trim that is in~f~rn~l to the part itself (i.e. a
pocket or "bowl"). A simplified example would be a pan-shaped part that has the bottom
of the pan cut away, resulting in a ring-shaped part.
Cutting out a hole(s) in the material allows for the hole(s) to enlarge as the m~t~ ri~l is
stretched onto the tool surface. This enlarging takes the place of stretching and thinning
the material if the holes were not present. The basic concept is that the hole enables more
axial stretching of the material and minimi7e~ the biaxial stretching (ref.Figure ~. The end
result is minimi7ed material thinout in the axially stretched material. The thinout in areas
of the hole that are stretched biaxially is also minimized (relative to not using the
CA 02221076 1997-11-14
cutout(s), but to a lesser extent than the axially stretched regions (ref. Figure 6 and Data
Table I).
Since the SPF process uses gas pressure to form the m~t~n~l, it is imperative that the
sheet being formed does not have any holes through it. This requirement is in direct
opposition to the process of this invention. Subsequently, a second sheet of material that
does not contain any cutout locations, is required to form the blank (material with the
cutout(s). This second sheet is placed between the blank and the die lid and becomes the
membrane which can be pressurized and formed onto the tool geometry. While forming,
this secondary sheet also forms the blank with the cutout(s). Once the blank is fully
formed (die surface is in intimate contact with the entire blank), the blank and secondary
sheet are separated and the secondary sheet is discarded
There are three critical factors-that must be dealt with to successfully utilize the disclosed
process. Those factors are: (1) l_ocation of cutout(s) on the blank, (2) Shape and Size of
the cutout(s) on the blank, and (3) Indexing the blank to the tool.
Cutout Location: The location of the cutout(s) is op~ ized when the periphery of the
cutout(s) is located as close as possible to the net trim of the part after forming. Locating
the cutout(s) as such will maximize the m~t~ thickness at the trim line.
Shape and Size of the Cutout(s): The cutout(s) shape and size are critical in that an
un~ler~i7ed cutout will result in unnecessary thinout. Conversely, an oversized cutout will
result in undercutting the trim line of the part. While both sizing and locating the
cutout(s), caution must be taken so that the formed blank does not become entrapped by
the secondary sheet during the separation of the two sheets (ref. Figure 6B).
. CA 02221076 1997-11-14
There are several methods for deter~rlining the proper location and size of the cutout(s) on
the blanks. Besides trial and error, a highly accurate "best guess" can be made utilizing a
model of the die surface. Among the most easily manipulated mode is a Computer Aided
Drafting (CAD) model. Once generated, the tool surfaces can be projected or rotated to
one plane so that the trim line of the part can be seen on that plane - the equivalent of a
forming blank (ref. Figure 9). ~his planar trim line de~lnes a preliminary location, si~, and
shape of the cutout(s). The ~mal size and shape of the cutout can then be obtained by
applying a reduction factor anad corner radii to the preliminary size and shape.
Indexing the Blanlc to the Tool: Once the size, shape, and location of the cutout(s) have
been detennined, it is paramount that the blank be located to the die accurately. Without
accurate, repeatable alignment, it is not possible to produce a consistent part. One method
of locating the blank to the die is through the use of "pins~' or "posts" that extend from the
die sealing surface. Holes can then be cut into the blank and secondary sheet tocorrespond to the pins in the die.
Description of Figures:
Figure 5: Illustrates the process steps for the Disclosed Invention
Prior to Figure 5A, the region(s) of the blank that are to be cutout must be
removed. The location, size and geometry of the cutout(s) is of critical importance
for the successful forming of the part. In addition, it is also critical that the blank
be indexed to the die surface so that the cutout area(s) form into the desired areas
of the die.
Figure 5A: This figure illustrates the die, die lid, material blank and the secondary
sheet. Once the die is heated to the forming temperature for the particular material
. . CA 02221076 1997-11-14
alloy, the blank and secondaIy sheet are placed on the surface of the die. The
location of the secondary sheet is between the die lid and the blank.
Figure 5B: The blank and secondary sheet are "sandwiched" between the die and
lid by means of a force applied to the lid.
Figure 5C: As gas pressure is introduced to the top side of the secondary sheet,the secondary sheet and blank are formed into the die cavity.
Figure SD: Forming continues as the gas pressure is increased.
Figure 51~: Upon completion of the forrning cycle, the blank is in full contact with
the die surfaces.
Figure 5F: The gas pressure is vented and the lid is removed. The blank and
secondaly sheet are separated and removed from the die.
Figure 6A: This figure illustrates the result of correctly calculating and locating
the cutout(s) on the blank. In this instance, the secondary sheet does not entrap the
blank when the two are separated.
Figure 6B: This figure illustrates the potential problem of material entrapment
caused by incorrectly calculating the size and location of the cutout(s) on the
blank. In this instance after forming, the edge of the cutout(s) became located on a
near-vertical surface, creating entrapment of the blank by the secondary sheet.
This entrapment does not allow for separation of the blank and secondary sheet
without cutting the two apart~
. CA 02221076 1997-11-14
Figure 7: This figure illustrates the types of stretching of the cutout periphery that
take place during forming of the blank. Any straight line regions of the cutout
periphery will undergo stretching in one direction ~axial). (~urved segments of the
cutout will stretch in two directions (biaxial). In general, the axial stretching that
takes place will result in less thinout' than in the regions that are biaxially
stretched.
Figure 8: This figure illustrates data obtained from fabricated test parts. All parts
started with the same material thickness and were measured in the same locations.
From the data it is possible to see that by cutting hole(s) in the blank, it is possible
to achieve a 69% increase in as-formed material thickness, in comparison to parts
formed without the invention process.
Figure 9: This ~lgure illustrates how the basic geometry and location of the
cutout(s) can be obtained. This key information be obtained through several
methods. However, the easiest method is through the manipulation of Computer
Aided Drafting (CAD) data of the die geometry. Use of such data is illustrated in
this ~lgure.
Figure 9A: This figure illustrates half of a symmetrical part. The dashed lines
indicate surfaces of the die.
Figure 9B. This figure shows the axis about which the surfaces are rotated. Oncethe surfaces are rotated to the starting plane, a preliminary outline of the cutout
can be determined. The edge of the surface rotated is determined by the net trim of the
part.
