LOCKHEED M A R T I N_ :;_

Friction Stir Process

Mapping Methodology

Gerry Bjorkman

Alex Kooney

Lockheed Martin Space Systems Company

Carolyn Russell

Marshall Space Flight Center

m

('3

o

(o

p..=

¢..vo,o

0

0I

0C_

O0

https://ntrs.nasa.gov/search.jsp?R=20020094338 2020-06-22T03:56:03+00:00Z

Process Overview ,°,,,,,,

Friction Stir Welding (FSW)- FSW is a solid-state process using

a rotating tool with a shoulder anda projecting pin.

- The pin tool is rotated and plungedinto the joint until the shouldercontacts the top surface.

- The frictional heating between thepin tool and the joint plasticizes thematerial in the local region nearthe pin.

- The material at the weld centerline

is joined through a combination offorging processes that occurs inthe local region of the pin tool.

- Three significant parameters:spindle speed, travel speed,plunge load or plunge position

Heel

Plunge

Travel

Rotation

Penetration

Ligament

9/18/2002 Friction Stir Process Mapping Methodology 2

FSW Process L@©KllEED MARTIN_

A process map summarizes the weld process performance for a given

pin tool geometry and joint configuration.

Targeting a consistent penetration ligament, the process is simplified

into two parameters: travel speed and rotation speed.

Other parameters, such as plunge force, traverse force, weld nugget

geometry, NDE response, and mechanical properties are assumed to

be dependent variables.

I . |#.

4,4 ,j r..,,,

i I/ i _

wP • IP"

YELLOW: Unusual flow

patterns, unstable position andprocess loads, excessive flash,poor mechanical properties

GREEN: Symmetric flowpatterns, stable position andprocess loads, good strength

RED: NDE rejections,volumetric defects, poorstrength, excessive processloads

IPM

Friction Stir Process Mapping Methodology9/18/2002 3

FSW Process LO¢I(ilEED HARTIN_

A selected rpm/ipm combination (weld schedule) provides a specificnugget geometry, heat input, and mechanical strength.The selected nominal weld schedule, or sweet spot, is the bestcompromise between process stability, mechanical strength, NDEresponse, and machine capability.Once the nominal schedule is selected, process loads and heelpositions are explored to determine their acceptable operatingwindows.

Statistical process control in conjunction with the process map dataprovides quality control and grounds for reduced NDE requirements.

a.

Process envelope

(acceptable weld schedules)

Nominal weld schedule

(sweet spot)

Operating Window(acceptable variation about

nominal weld schedule)

IPM

Friction Stir Process Mapping Methodology9/18/2002 4

(_ FSW Process L@¢KHEED MARTIN_

Methodology Overview

Determine joint configuration, pin tool

design, anvil and clamping system

Phase I Quick Look

Phase II Testing

Select weldschedu

Characterize process withnominal weld schedule

9/18/2002 Friction Stir Process Mapping Methodology 5

Phase I Quick Look L@¢KilEED MABTIN_

The "quick look" provides a general overview of the process map- Three weld schedules are performed on a 24 inch long test panel

- Weld schedules are performed "hot" to "cold" by changing the travel

speed (constant rotation speed)

- Metallographic samples are excised near the end of each weldschedule

I

-f- Sc.edu,e.1iI

4 - 6.0"i

4 - 6.0" Macro 1i,I

___ 'II

24"

Schedule #2

Macro 3

Weld Start 4.0"

I

: Schedule #3III

II

III

Macro 2 :IIIIIi

6.0" 11.0" 14.0" 16.0"

9/18/2002 Friction Stir Process Mapping Methodology 6

Z _00_/8 L/6/,6olopoq;alAl 6u!ddelhi $£eOOJd J!],cGuop,o!J-!

Bldl _dl

•spleM >lOOq_io!no eq_ woJ; e;ep o!qdeJ6olle;ew

eq_ uo peseq pe_eeu!lep aJe suo_6eJ peel pue 'ueeJE) 'MOlle,k

_R I I U V FI 8 | | ll J g O "I

Y )!oo'1 >io!no I eseLld __

Phase I Quick Look LOCKllEED MARTIN_

Thin gauge similar alloy configuration- Low (A), medium (B), and high (C) heat input

C

IPM

Friction Stir Process Mapping Methodology9/18/2002 8

Phase I Quick Look

• Thin gauge similar alloy configuration

- Heel plunge vs. travel rate

/.@¢/(HEED

IPMFriction Stir Process Mapping Methodology9/18/2002 9

Phase I Quick Look L@¢KllEED MARTIN_

Thin gauge similar alloy configuration

- Scaling/Galling

IPMFriction Stir Process Mapping Methodology9/18/2002 10

LL _00_/8 L/6/,6olopoq;e_ 6u!ddelhl ssaooJd J!;S uo!lo!J-I

_dl

"U

_N I L I V iVY

Olt/IJO01

qsel-I e^!sseox3 pue ;e66nN pleM e6Jeq -

uo!_eJn6!juoo _Olle Jel!w!s e6ne6 u!qi

)!oo'7 >lo!no I aseqd

Phase I Quick Look ,,=,,,,° ,,,,,,_

• Thin gauge similar alloy configuration- Root voids and "worm holes"

1'111

I:L

IPMFriction Stir Process Mapping Methodology9/18/2O02 12

Phase I Quick Look• Thick gauge similar alloy configuration

- Low (A), medium (B), and high (C) heat input

LOCKHEED MABTIN_

C

IPM

Friction Stir Process Mapping Methodology9/18/2002 13

17L _00_/9 L/6_6olopoq;elhl 6u!ddelhl sseooJcl J!;S uo!;o!J-I

0

_NIIlVlJ 4II/IJOOTY

;ndu! ;eeq (0) q6!q pue '(8) wn!pew '(V) Mo-I -

uo!_eJn6!_uoo _Olle Jel!w!ss!p e6ne6 u!q_L

_!oo-I _lo!no I aseqd

£ I. _00_/81./6_6olopoqJ, e_ 6u!dde_ sseOOJd J!),S uop, ou_-I

_dl

,,eloq LUJOM,,pue p!o^ _ool9 -

uo!;ejn6!juoo _Olle Jel!LU!ss!pe6ne6 u!q/

_" " ""v ""'°°' )!oo-I >io!no I eseLld

Phase I Quick Look L@¢KilEED MARTIN_

Thin gauge dissimilar alloy configurationIrregular nugget flow

Location of particular alloy influences flow within the nugget

Reversed alloy locations

IPM

Friction Stir Process Mapping Methodology9/18/2002 16

L L gOOg/8 L/6_6olopoqlelhl 6u!ddelhi sseooJd J!;S uo!lop=l

INdl

O

:12"12

;ndu! ;eeq (9) qB!q pue '(8) Lun!peLu'(V) Moq -

uo!;eJnB!juoo _Olle Jel!UJ!SS!pe6ne6 _lo!q/ °

_,,,,,, ,,,,,°" )!oo-I )lo!no I eseLld

8 L _00_/8 _/6/;6olopoq_elhl 6u!ddelhl ss8OOJd JP,S uo!:lO!J-I

_dl

"10

,,SalOq LUJOM,,ql!M le66nu PleM esdellOO/pleM lndu! ;eeq q6!H -

uo!;eJn6!_uoo f,Olle Jel!w!ss!pa6ne6 >io!q_L •

_,,,,,, ,,,,,°., _!oo-I _lo!no I eseqdy

Phase I Quick Look

• Thick gauge dissimilar alloy configuration- Low Heat Input Weld with "Worm Holes"

LO©KHEED FI A R T I N_

G.

L

IPM

Friction Stir Process Mapping Methodology9/18/20O2 19

Phase II AL@©KllEED MARTIN_

Weld schedules that provide acceptable metallographic profiles

from Phase I are performed on 24" long test panels.

- The longer weld provides adequate time for weld to reach stability

• More reliable NDE response and tensile tests

- Process load data becomes more consistent

Tensile tests are conducted at the expected service

temperatures of the weld

These tests define the process envelope and begin to focus in on

the "sweet spot"

W! | _IIMIWm | m t _J __J

12"x24" Test Panel

IPM

Friction Stir Process Mapping Methodology9/18/20O2 20

Phase II TestinThe process envelope is delineated using the Phase II test data

- Mechanical strength, NDE, and tool performance are factors to consider

IPM

0

0

IPM

9/18/2002 Friction Stir Process Mapping Methodology

IPM

IPM

21

Phase II Testin LO©KHEED

Tensile strength increases with faster travel speeds

- Cryogenic strength is more sensitive than room temperature strength to

heat input

Process loads, especially traverse loads, increase with travel speed

The ability to perform cold welds depends on the machine's control

system response

m

a. ==

IPM

IPM

Friction Stir Process Mapping Methodology

V

9/18/2002 22

Weld Schedule Selection LOCKHEED

Schedule selection is the best compromise between thefollowing factors:

- Process stability

- Mechanical strength

- NDE response

- Machine capability

The nominal scheduleshould be near the center of

the process envelope toensure robust performanceto variations in the

manufacturing environment a.

MAHTIN_

IPM

9/18/2002 Friction Stir Process Mapping Methodology 23

Process Characterization ,,,,.,,, ,,,,,.

Multiple welds are performed with varied setup conditions

Process information, such as plunge load, is collected and

acceptable bounds are established

I

.... I .... t i ' '10 15 20

Weld Position (in)

,,D

8

,qma.

• Average 1

Minimum

Maximum

-3sig

+3sig

i : i

5 10 15 20 25

We_Pos_n(_)

9/18/2002 Friction Stir Process Mapping Methodology 24

The weld process performance for a given weld jointconfiguration and tool setup is summarized on a 2-D plot ofRPM vs. IPM

A process envelope is drawn within the map to identify the

range of acceptable welds

The sweet spot is selected as the nominal weld schedule

The nominal weld schedule is characterized in the expected

manufacturing environment

The nominal weld schedule in conjunction with process control

ensures a consistent and predictable weld performance

9/18/2002 Friction Stir Process Mapping Methodology 25