0% found this document useful (0 votes)
64 views15 pages

Stamping Dieter

This document discusses various aspects of stamping processes including draw beads, strain distribution, loose metal and wrinkling. It describes how draw beads are used to control material flow and explains how properties like strain hardening exponent affect strain distribution and formability. The document also covers issues like loose metal, wrinkling and controlling blank holder pressure and draw beads to avoid defects.

Uploaded by

Geovanny Muñoz
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
64 views15 pages

Stamping Dieter

This document discusses various aspects of stamping processes including draw beads, strain distribution, loose metal and wrinkling. It describes how draw beads are used to control material flow and explains how properties like strain hardening exponent affect strain distribution and formability. The document also covers issues like loose metal, wrinkling and controlling blank holder pressure and draw beads to avoid defects.

Uploaded by

Geovanny Muñoz
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

P1: KAE

0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

16 Stamping

16.1 STAMPING
Operations called stamping, pressing, and sometimes drawing involve clamping a sheet
at its edges and forcing it into a die cavity with a punch, as shown in Figure 16.1. The
sheet is stretched rather than squeezed between the tools. Pressure on the draw beads
controls how much additional material is drawn into the die cavity. In some cases there
is a die, which reverses the movement of material after it is stretched over the punch.

16.2 DRAW BEADS


Draw beads (Figure 16.2) are used to create tension in the sheet being formed by
preventing excessive drawing. As a sheet moves through a draw bead, it is bent three
times and unbent three times. Each bend and each unbend there requires plastic work.
Over each radius there is friction. This create resistance to movement of the sheet.
If the resistance is sufficiently high, the sheet will be locked by the draw bead. The
restraining force of the draw bead can be controlled by the height of the insert.
The restraining force has two components. One is caused by the work necessary to
bend and unbend the sheet as it flows over the draw bead and the other is the work to
overcome friction. The restraining force F per length resulting from the bending and
unbending can be crudely estimated with the following simplifying assumptions: Work

DRAW BLANK-
BEAD HOLDER

PUNCH

16.1. Schematic punch-and-die set with


blank holder.
DIE BLANK
RING
DIE

255
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

256 STAMPING

16.2. Sketch of a draw bead. The strains on the top surface are indicated. There is friction wherever
the sheet contacts the tooling.

hardening, elastic core, movement of the neutral plane, and the difference between
engineering strain and true strain are neglected (ε = e). The strain is given by e = z/r,
where z is the distance from the neutral plane. In this case the work per volume for an
element at z is Y (z/r ), where Y is the yield strength and r isthe radius of the bend at the
t/2
neutral axis. The work for all elements then is dW/dL = 2 0 (Y/r ) zdz = Y t 2 /(4 r ).
The restraining force cause by bending three times and unbending three times as the
sheet moves through the draw bead is then

Fb = 6dW/dL = 1.5Y t 2 /r. (16.1)

This treatment neglects the Bauschinger effect. A more accurate determination of Fb


can be obtained by pulling a strip through a fixture in which freely rotating cylinders
simulate the draw beads (Figure 16.3).

16.3. Determining the bending contribution to the draw bead restraining force.
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

16.3. STRAIN DISTRIBUTION 257

The position of draw beads is important. They should be placed perpendicular to the
direction of metal flow. If they are too close to the trim line, material being drawn over
them may become part of the finished product and create a surface defect. If they are
too far from the trim line, their effectiveness in preventing drawing will be diminished.

16.3 STRAIN DISTRIBUTION


Forming limit diagrams do not completely describe the forming behavior of different
metals. Two materials may have nearly the same forming limits ε1∗ , but differ substan-
tially in formability. For example, consider a symmetric part (Figure 16.4) with a line
of length L 0 parallel to the 1-direction. The stress is not uniform along the line so ε1
will vary from place to place. The total length of line will be given by
 L0
L = L0 + ex dx. (16.2)
0

The height h max of a hemispherical cup at failure depends on L/L 0 when ε1 max =
ε1∗ . In other words, h max depends strongly on the distribution of ε1 on x. The ratio of
ε1 in lightly loaded regions to that in heavily loaded regions depends primarily on the
strain-hardening exponent n, but also to some extent on the strain-rate sensitivity, m.
Increasing values of n and m distribute the strain so permit deeper parts to be formed.
Figure 16.5 illustrates this.
When a sheet that is locked by draw beads is stretched over a hemispherical punch,
the failure site depends on the friction between the sheet and the punch. With high
friction, the failure usually occurs in the unsupported material near contact with the
punch. Just outside of the contact region, the deformation is in plane strain because of
the hoop direction constraint from the neighboring material in contact with the punch.
With lower friction the failure site moves toward the dome and the depth of cup at
failure increases.

16.4. Increase of line length, L 0 .


P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

258 STAMPING

16.5. Distribution of strains during punch stretching of sheets of several metals. The punch was
stopped at the same peak strain in each case. Materials with higher values of n had more widely
distributed strains and formed deeper cups. From S. P. Keeler and W. A. Backofen, Trans. Q. ASM, 56
(1963), pp. 25–48.

16.4 LOOSE METAL AND WRINKLING


When flat sections are not stretched enough, they become loose or floppy. If a potential
car buyer puts his hand on the roof of a car and finds that it isn’t stiff, he will perceive
poor quality. The same is true if the hood of a car vibrates at high speed. Loose metal
can be voided by ensuring that all parts of a stamping are deformed to some minimum
effective strain (perhaps 4%).
Wrinkling occurs when the stresses in the sheet are compressive. With thin sheets,
wrinkling will occur under very low amounts of compressive stress. Figure 16.6 shows
the region of wrinkling on a forming limit diagram.
Wrinkling is a potential problem in drawing of many parts. Consider the forming
of a flat-bottom cup with conical walls, as shown Figure 16.7. As the punch descends,
the shaded element is drawn closer to the punch so its circumference must shrink.
Stretching of the wall in tension will reduce the circumference, but if the wall isn’t
stretched enough it will wrinkle. The amount of stretching can be increased by raising
the blank-holder pressure. However, too much restraint by excessive blank-holder
pressure or draw beads prevents flange material drawing in and may cause the wall to
fail in tension. Therefore the blank-holder pressure must be controlled.∗
There is a window of permissible levels of blank-holder force. Too little causes
wrinkling; too much force results in wall failure. This is shown in Figure 16.8. Since
more circumferential contraction occurs when material with high R-values is stretched,
there is more contraction in the plane of the sheet so an increased R widens the left
side of the window. With an increased value of n, more stretching can occur before
failure. For any given material and required depth of draw, there is a window of
permissible blank-holder force. Increasing the depth of draw beads has the same effect
as increasing blank-holder force. It is possible to change the blank-holder force during


J. Havranek, in Sheet Metal Forming and Energy Conservation, Proc. 9th Bienn. Cong. Int. Deep
Drawing Res. Group, ASM, Ann Arbor, MI (1976), pp. 245–63.
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

16.5. FLANGING 259

ε1

16.6. A forming limit diagram showing the uniaxial


path of a uniaxial tension test (dashed line). tension
To the left of this line, the stress σ 2 is
compressive so wrinkling is possible.
wrinkling
possible

ε2

blank holder
16.7. In drawing of a conical cup, wall wrin-
kling may occur if there is not enough tensile element
die
stretching of the wall.
punch

60
Depth of draw, mm

wrinkling increasing R

increasing n
16.8. Forming window for a conical cup. For wall failure
a draw depth of 55 mm, the blank-holder 40
force must be between 40 and 75 kN. window for
forming

0 50 100 120
Blankholder force, kN

the draw. Greater depths of draw are possible if the force is low in the early stages of
the draw and increases in the later stages.∗

16.5 FLANGING
Often edges of sheets are flanged. If the flange is concave, cracking may occur at
sheared edges where the edges are elongated in tension. This occurs when a hole is

D. E. Hardt and R. C. Fenn, Proc. ASM Symposium on Monitoring and Control of Manufacturing
Processes (1990), pp. 349–72.
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

260 STAMPING

edge cracks

wrinkles

16.9. Edge cracks on a concave flange and wrinkling on a convex flange.

expanded or a concave edge is flared as shown in Figure 16.9. The tensile strain in such
a case is
ε1 = ln(d0 /d), (16.3)

where the diameters before and after flanging are d0 and d. The tendency is aggravated
if shearing has left a burred edge. On the other hand, if the flange is convex, the
contraction of the flanged edge may lead to wrinkling.

16.6 SPRINGBACK
Control of springback is very important, especially in lightly curved sections like
the bottom of the stamping in Figure 16.10. As indicated in Chapter 12, springback
is minimized if tension is sufficient to cause tensile yielding across the whole cross
section. Yet the wall force F2 must not cause a tensile stress that exceeds the tensile
strength. The relative magnitudes of F1 and F2 can be calculated approximately as
follows: A radial force balance on an element in the bend gives the normal force, dN
= F dθ, so the frictional force dF on the element is µ dN = µF dθ, where µ is the
coefficient of friction.
From a force balance in the circumferential direction,

F + dF = F + Fµ dθ,
 F2  θ
dF/F = µ dθ so F2 = F1 exp(µθ). (16.4)
F1 0

Neglecting differences between the plane-strain and uniaxial tensile and yield strengths,
F2 < (Su )wt and F1 > (Y )wt, where w is the dimension parallel to the bend, t is the
sheet thickness, Su is the tensile strength, and Y is the yield strength. Therefore

Su /Y > exp(µθ). (16.5)


P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

16.8. TAILOR-WELDED BLANKS 261

16.10. Sheet being bent over a punch.

Thus the tensile-to-yield strength ratio must exceed a value that depends on both
the coefficient of friction and the bend angle. For example, if µ = 0.2 and θ = π/2,
Su /Y > 1.37.
In springback calculations it should be recognized that when a metal is stretched in
tension and unloaded, the unloading is not linear. This is called the Bauschinger effect.
Figure 16.11 shows typical behavior. Springback is much greater than that calculated
by assuming linear elastic unloading.

16.7 STRAIN SIGNATURES


The strain path varies from one place to another in a stamping. The strain path of
a particular point is called its strain signature. Figures 16.12 and 16.13 show stamp-
ings of a hemispherical cup and a shallow pan with the strain signatures of several
locations.

16.8 TAILOR-WELDED BLANKS


In recent years parts are being stamped from blanks made by welding two or more
sheets of different thickness or different base materials. The purpose is to save weight by
using thinner gauge material where its strength is sufficient and using thicker or stronger
material only where necessary. Some difficulties are encountered during forming. Offset
blank-holder surfaces are required to assure adequate hold-down. Welding hardens the
weld zone, which may reduce the formability and cause cross-weld failures if the
direction of major strain is parallel to the weld. The best blank orientation is with the
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

262 STAMPING

stress

slope = E* < E

slope = E

strain linear
recovey

total recovery

16.11. Nonlinear unloading results in greater springback than calculated assuming linear elasticity.

B
ε1
A
A
C C D
B
E

E ε2
16.12. Strain signature in drawing of a hemispherical cup.

weld perpendicular to the major strain axis. In this case failure is likely to occur by
splitting of the thinner material parallel to the weld. This problem can be alleviated to
some extent by decreasing the hold-down pressure on the thicker material and allowing
more of the thicker metal to flow into the die. The frequency of failures parallel to the
weld is reduced by decreasing the movement of the weld in the die.

16.9 DIE DESIGN∗


At the center of most sheet stampings a certain effective strain level is necessary to
prevent “slack” or “loose” metal and to minimize springback. Otherwise the panel

This section is a result of private discussions with Edmund Herman, Creative Concepts Company,
Shelby Township, MI (2005).
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

16.9. DIE DESIGN 263

A
B

ε1 D ε1 CÕ
BÕ DÕ
E

C D

E
B

A
ε2 AÕ ε2
16.13. Strain signatures in a stamping of a shallow pan. The strain path at B is likely to cause wrinkling.

may not retain the desired shape or it may feel floppy to the customer. A target strain
at the center of the panel will also ensure enough work hardening to strengthen the
panel. To achieve this target strain, there must be sufficient tension in the sheet at that
point. Because of friction and bending, the tension at the binder must be much higher.
Figure 16.14 illustrates a part being stamped. The final part is actually from A to C. The
material between C and the edge of the binder must be stretched in the die to achieve
the required stress at point A, but is later trimmed off and discarded. The example

Region I
A

punch Region 2
C

y
neutral line
Stretch wall
D
Binder
wrap Region 3
E

16.14. Schematic of a stamping.


P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

264 STAMPING

below is in plane strain but real parts are usually stretched in the normal direction as
well. The highest stress will occur in the stretch wall between C and D. To determine
whether this stamping can be made, it is necessary to fix the strain level required at A
and then calculate the stress in this region to determine whether it exceeds the tensile
strength of the material.

EXAMPLE 16.1: Determine whether the stamping in Figure 16.14 can be made assum-
ing a target strain at A of εA = 0.035. Other data are: the sheet thickness = 1 mm;
the stress–strain relation is σ = 520ε0.18 MPa; the radii are r1 = 8 m, r2 = 0.10 m,
and r3 = 0.025 m; the angles are θ1 = 0.25, θ2 = 1.05, and θ3 = 1.30 radians; and
the friction coefficient is µ = 0.10.
For simplicity, assume that this is a plane-strain stamping.

SOLUTION: The thickness at A, tA = t0 exp(−εA ) = (1 mm) exp(−0.035) =


0.9656 mm. The tensile force per length at A is FA = 520tA εAn = 520(0.9656)
(0.035)0.18 = 274.6 kN/m.
The tensile force at B must be enough larger than this to overcome the friction in
all the bends.

FB = FA exp(µθ) = 274.6 exp[(0.10(0.25)] = 275.3 kN/m,


FC = FA exp(µθ) = 275.3 exp[(0.10)(1.05)] = 305.8 kN/m = FD .

The maximum force, FD , corresponds to the tensile strength time the original thickness
= t0 K n n exp(−n) = (0.001)(520)(0.180.18 ) exp(−0.18) = 319.0 kN/m. The value of
FE is less than this so the stamping can be made.

EXAMPLE 16.2: Find the required length of the stretch wall.

SOLUTION: Guessing that the horizontal component of the stretch wall is 0.04 m, the
length of the wrap is 8 × 0.25 + 0.1 × 1.05 + 1.3 × 0.025 + 8 × 0.25 + 0.04 = 2.178.
The force per length at B, FB = FA exp(µθ) = 274.6 kN/m [(0.10)(0.25)] =
274.6, but FB also equals tB σB = t0 exp(−εB )K εB0.18 so εB0.18 exp(−εB ) = FA /(K t 0 ) =
274.6/520 = 0.5281.
Solving by trial and error, εB = 0.0521.
The force per length at C is 305.8 kN/m but FC also equals tC σC = t0 exp(−εC )
K εC0.18 so εC0.18 exp(−εC ) = FC /(K t0 ) = 305.8/(520) = 0.5881.
Solving by trial and error, εC = 0.083.
Taking the average strain between A and B as (0.035 + 0.0521)/2 = 0.0436, the
increase in the length of line is (8 × 0.25)[exp(0.0436) − 1] = 0.089 m.
Taking the average strain between B and C as (0.0521 + 0.083)/2 = 0.0676, the
increase in the length of line is (0.1 × 1.105)[exp(0.0673) − 1] = 0.0077 m.
The change of length of the stretch wall is 0.04[exp(εC ) − 1] = 0.0035 m.
The total change of length is then 0.089 + 0.009 + 0.0035 = 0.10 m or
100 mm. To this must be added any length of material that comes out of the
binder.
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

16.10. TOUGHNESS AND SHEET TEARING 265

16.15. Sketch of a sheet failure by through-


thickness necking.

16.16. Plastic zone in a deeply notched


tensile specimen.

16.10 TOUGHNESS AND SHEET TEARING


Figure 16.15 is a sketch of a plane-strain neck in a sheet. The volume of the plastic
region is proportional to t 2 L where t is the specimen thickness and L is the length of
the crack. Since the area of the fracture is t L, the plastic work per √
area is proportional
to the thickness t, and the fracture toughness K C is proportional to t. Very thin sheets
tear suddenly at surprisingly low stresses.
A fracture toughness can be associated with the process of localized necking and
tearing. A simple way of measuring the toughness is to test a series of deeply edge-
notched tensile specimens∗ like that shown in Figure 16.16. The deep notch confines


B. Cotterell and J. K. Reddel, Int. J. Fract., 13 (1977), pp. 267–77.
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

266 STAMPING

16.17. Schematic plot of the work per area, W/Lt0 , as a function of the distance between notches. G
is the value at L = 0.

16.18. Shape of plastic zone during local necking of an ideally plastic material.

the plastic deformation to a circular region whose diameter is equal to the distance L
between notches. The work is composed of two terms. One is the plastic work in the
circular region. The other is the work to propagate the local neck. The total work is
W = αL 2 t0 + G Lt0 . (16.6)
Here G is the fracture toughness or energy per area to cause the local neck. The constant
α reflects the strain distribution in the circular area. G can be determined by plotting
W/(Lt0 ) as a function of L as shown in Figure 16.17. The intercept at L = 0 is G.
At any instant during the necking of an ideally plastic material, plastic deforma-
tion is restricted to a region of length t that is equal to the current thickness.∗ See
Figure 16.18. In this case it can be shown that
 εf
G = t0 K ps εn exp (−2ε) dε. (16.7)
0

Here K ps is the coefficient in the power-law expression for plane strain, σ = K ps εps
n
,
and is related to K in σ̄ = K ε̄ through the yield criterion. For von Mises,
n

K ps /K = (4/3)(n+1)/2 , whereas for equations 13.22 and 13.23 with R = P, K ps /K =


(4/3)(n+1)/a . The value of the integral is not very dependent on εf as long as
εf ≥ 1 (R A = 63%). The main problem with this analysis is that the shapes of the
necks vary with the material, whereas this analysis predicts θ = 26.6◦ . Necks are often

W. F. Hosford and A. G. Atkins, J. Material Shaping Technol., 8 (1990), pp. 107–11.
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

NOTES OF INTEREST 267

16.19. Dependence of press performance for three parts on n and R̄.

more gradual than this prediction, so the plastic zone is longer than t. This is probably
the result of a positive rate sensitivity, m, and makes the experimental values of G
larger than those predicted by equation 16.7.

16.11 GENERAL OBSERVATIONS


The relative amount of stretching and drawing varies from on part to another. Where
stretching predominates, formability depends mainly on n and m, whereas if drawing
predominates, R̄ is most important. Figure 16.19 relates actual press performance to n
and R̄.

NOTES OF INTEREST
Johann Bauschinger (1833–1893) was the director of the Polytechnical Institute of
Munich. In this position, he installed the largest testing machine of the period and
instrumented it with a mirror extensometer so that he could make very accurate strain
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

268 STAMPING

measurements. He discovered that prior straining in tension lowered the compressive


yield strength and vice versa. Bauschinger also did the original research on the yielding
and strain aging of mild steel.
The aluminum pie dish is one of the very few stamped items with wrinkles that
consumers accept willingly. The wrinkled sidewalls actually strengthen the pan.

REFERENCES

Z. Marciniak, J. L. Duncan, and S. J. Hu, Mechanics of Sheet Metal Forming, Butter-


worth and Heinemann, 2002.
E. M. Meilnik, Metalworking Science and Technology, McGraw-Hill, 1991.
John H. Schey, Metal Deformation Processes: Friction & Lubrication, M. Decker,
1970.

PROBLEMS

16.1. Referring to Figure 16.5, determine the values of L/L0 for brass and aluminum.
16.2. Plot the contour of ε̄ = 0.04 on a plot of ε1 vs. ε2 (i.e., on a forming limit
diagram).
16.3. Show how the dashed line in Figure 16.6 would change for a material having
an R-value of 2.
16.4. A sheet of HSLA steel having tensile strength of 450 MPa and yield strength
of 350 MPa is to be drawn over a 90◦ bend. What is the maximum permissible
coefficient of friction?
16.5. Consider the deep drawing of a flat-bottom cylindrical cup. Sketch the strain
path for an element initially on the flange halfway between the periphery and
die lip.
16.6. Calculate the total drag force per length of a draw bead attributable to plastic
bending if the bend angle entering and leaving the draw bead is 45◦ and the
bend angle in the middle of the draw bead is 90◦ . Assume the bend radii are
10 mm and the sheet thickness is 1 mm.
16.7. Calculate the drag force attributable to friction in Problem 16.6, assuming a
friction coefficient of 0.10.
16.8. Carefully sketch the strain signature for the conical cup drawing in Figure
16.20.

A
16.20. Drawing of a conical
cup.
B

C
D
P1: KAE
0521881210c16 CUFX149/Hosford 0 521 88121 8 printer: Sheridan October 4, 2007 18:15

PROBLEMS 269

16.9. Why isn’t the bending included in the drag in regions A–B and B–C in Example
16.1?
16.10. It has been noted that when a cup is drawn with a hemispherical punch, greater
depths can be achieved before failure when the punch spins. Explain why this
might be so.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy