Hoja de datos de 63800-0029 Crimp Quality Handbook de Molex

Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 1 of 24
Revision: D Revision Date: 12-23-09
QUALITY CRIMPING
HANDBOOK
Order No. 63800-0029
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 2 of 24
Revision: D Revision Date: 12-23-09
Table of Contents
SECTION
1 Introduction to Crimp Technology
2 Purpose
3 Scope
4 Definitions
5 Associated Materials
6 Procedures
7 Measurement
8 Crimp Process Control
9 Troubleshooting
10 Wire Gauge Chart
Terminal TooHng
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 3 of 24
Revision: D Revision Date: 12-23-09
SECTION 1
INTRODUCTION TO CRIMP TECHNOLOGY
Developed to replace the need to solder terminations, crimping technology provides a high quality connection
between a terminal and a wire at a relatively low applied cost. The methods for applying crimp terminations
depend on the application and volume, and range from hand-held devices to fully automated systems.
The application methods include a basic hand tool, a press and die set, a stripper crimper, or a fully automatic
wire processing system. However, no matter what method is used, the setup of each tool is critical for
achieving a quality crimp.
Today, many OEM companies are using Statistical Process Control (SPC) to continuously improve their crimp
terminations. Crimp termination is a complex process and to ensure consistent quality it is necessary to
understand the variability and inter-relational interactions that the technology involves.
Without a thorough understanding of the crimping process, and all the factors that can affect it, the result may
not meet expectations. The three key elements in the crimping process are the terminal, the wire, and the
tooling.
Terminal
For most applications, it is not economically practical for connector manufacturers to design a terminal to
accept one wire size, one wire stranding, and one insulation diameter (UL type). Most terminals accommodate
many wire sizes, stranding, and a range of insulation diameters, and the terminals are designed to meet
acceptable levels over this entire range.
Wire
The wire stranding and insulation type can vary widely within one wire size. For example, there is more than
18% more material in an 18 AWG by 19-strand wire than an 18 AWG by 16-strand wire. The insulation
diameter of an 18 AWG wire can range from 1.78mm (.070") to over 4.57mm (.180"). Wire strands can be
copper, tinned, over coated, or top coated. Wire insulation materials, thickness, and durometers vary from
application to application.
Tooling
What type of tooling does the application require? Does the application require hand stripping of the wire or
does the volume dictate an automatic wire-stripping machine? Does the application and volume require hand
tools, press and die, or fully automatic wire process machines? Crimping with a manual hand tool, semi-
automatic press and die, or fully automatic wire processor, all involve different levels of variability. The
terminal, wire, and type of application tooling all affect the quality of the completed terminations.
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 4 of 24
Revision: D Revision Date: 12-23-09
SECTION 2
PURPOSE
This handbook provides general guidelines and procedures for understanding and achieving acceptable crimp
terminations. A glossary in Section 4 lists common terms and definitions. Section 5 lists the tools that are
necessary to take accurate measurements and evaluate the crimp's acceptability.
The tooling setup is critical in determining the quality of the finished crimp. The attributes that need to be
considered include crimp height, conductor brush, bell mouth, and cut-off tab and strip length and insulation
position. Variability in one or more of these attributes can reduce the measured pull force. It can be difficult to
establish acceptable variability limits because the attributes all interact with one another.
For example, a track adjustment for bell mouth also will change the cut-off tab length and the insulation wire
position while strip length and wire locations affect the conductor brush and insulation position. Adjusting the
insulation crimp height may result in a slight change to the conductor crimp height measurement. It may be
necessary for the setup person to make multiple adjustments before establishing an optimal setup.
The order the setup is done may help reduce the number of repetitions required for an optimum setup.
Section 6 has a flowchart for a process setup while Section 9 is a troubleshooting guide for common problems.
Using Statistical Process Control (SPC) during the crimping process can help minimize the amount of defects.
Section 8 provides a general explanation of the benefits of using SPC.
This handbook is structured so that parts, or all, of its contents can be used as a procedural guide for ISO
requirements.
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 5 of 24
Revision: D Revision Date: 12-23-09
SECTION 3
SCOPE
This handbook is intended for Molex customers who are crimping Molex open barrel crimp terminals
and are using Molex tooling, primarily in semiautomatic or automatic wire processing termination methods.
The handbook's contents may slightly differ from other connector manufacturers' guidelines or individual
company procedures.
This handbook provides a basic overview of what to look for in an acceptable crimp. It is not intended to
replace individual product and/or tooling specifications.
Individual terminals or applications may have special requirements. Tooling limitations also may not permit an
attribute to be adjusted to meet optimum requirements.
SEAM
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 6 of 24
Revision: D Revision Date: 12-23-09
INSULATION
CRIMP
BRUSH
STRIP
LENGTH
BELL
MOUTH
CRIMP
HEIGHT
CONDUCTOR
CRIMP
CUT-OFF
TAB
SEAM
EXTRUSIONS
EXTRUSION EXTRUSION
TERMINAL CROSS
SECTION VIEW
Figure 4-1
BEND
DOWN
BEND UP
ROLLING
TWISTING
SECTION 4
DEFINITIONS
Anatomy of a Crimp Termination (Figure 4-1)
Bell Mouth (Flare)
The flare that is formed on the edge of the
conductor crimp acts as a funnel for the wire
strands. This funnel reduces the possibility that a
sharp edge on the conductor crimp will cut or nick
the wire strands. As a general guideline, the
conductor bell mouth needs to be approximately 1
to 2 times the thickness of the terminal material. *
Bend Test
One way to test the insulation crimp is by bending
the wire several times and then evaluating the
movement of the insulation and wire strands. As a
rule, the insulation crimp should withstand the wire
being bent 60 to 90 degrees in any direction,
several times. Use care when working with small
wire sizes so the wire at the back of the insulation
crimp does not shear.
Conductor Brush
The conductor brush is made up of the wire strands
that extend past the conductor crimp on the contact
side of the terminal. This helps ensure that
mechanical compression occurs over the full length of
the conductor crimp. The conductor brush should not
extend into the contact area.
Conductor Crimp
This is the metallurgical compression of a terminal
around the wire's conductor. This connection creates a
common electrical path with low resistance and high
current carrying capabilities.
Conductor Crimp Height
The conductor crimp height is measured from the top
surface of the formed crimp to the bottom radial surface.
Do not include the extrusion points in this measurement
(See Figure 4-1). Measuring crimp height is a quick,
non-destructive way to help ensure the correct
metallurgical compression of a terminal around the
wire's conductor and is an excellent attribute for process
control. The crimp height specification is typically set as
a balance between electrical and mechanical
performance over the complete range of wire stranding
and coatings, and terminal materials and plating.
Although it is possible to optimize a crimp height to
individual wire stranding and terminal plating, one crimp
height specification is normally created.
*Consult individual terminal specification requirements
**Ht
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 7 of 24
Revision: D Revision Date: 12-23-09
28.5
29.5
30.5
31.5
Thousandths
Measurements
Crimp Height
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Sample
Contol Limit
Control Limit
Upper Specification
Lower Specification
Example Control Chart
Figure 4-3
Figure 4-2
Cut-off Tab Length
This material protrudes outside the insulation crimp
after the terminal is separated from the carrier strip.
As a rule, the cut-off tab is approximately 1.0 to 1.5
times terminal material thickness. * A cut-off tab
that is too long may expose a terminal outside the
housing or it may fail electrical spacing
requirements. In most situations, a tool is set-up to
provide a cut-off tab that is flush to one material
thickness.
Extrusions (Flash)
These small flares form on the bottom of the
conductor crimp resulting from the clearance
between the punch and anvil tooling. If the anvil is
worn or the terminal is over-crimped, excessive
extrusion is the results. An uneven extrusion may
also result if the punch and anvil alignment is not
correct, if the feed adjustment is off, or if there is
insufficient/excessive terminal drag.
Insulation Crimp (Strain Relief)
This is the part of the terminal that provides wire
support for insertion into the housing. It also allows
the terminal to withstand shock and vibration. The
terminal needs to hold the wire as firmly as possible
without cutting through to the conductor strands.
The acceptability of an insulation crimp is subjective
and depends on the application. A bend test is
recommended to determine whether the strain relief
is acceptable for each particular application.
Insulation Crimp Height
Molex does not specify insulation crimp heights
because of the wide variety of insulation thickness,
material, and hardness. Most terminals are
designed to accommodate multiple wire ranges.
Within the terminal’s range, the strain relief may or
may not completely surround the diameter of the
wire. This condition will still provide an acceptable
insulation crimp for most applications.
1. A large strain relief should firmly grip at least
88% of the wire.
2. A smaller strain relief should firmly grip at least
50% of the wire and firmly hold the top of the
wire.
To evaluate the insulation crimp section, cut the wire
flush with the back of the terminal. Once the optimum
setting for the application is determined, it is important to
document the insulation crimp height. Then, as part of
the setup procedure, the operator can check the crimp
height.
Insulation Position
This is the location of the insulation in relation to the
transition area between the conductor and insulation
crimps. Equal amounts of the conductor strands and
insulation needs to be visible in the transition area. The
insulation position ensures that the insulation is crimped
along the full length of the insulation crimp, and that no
insulation is crimped under the conductor crimp. The
insulation position is set by the wire stop and strip length
for bench applications. For automatic wire processing
applications, the insulation position is set by the in/out
press adjustment.
Strip Length
The strip length is determined by measuring the
exposed conductor strands after the insulation is
removed. The strip length determines the conductor
brush length when the insulation position is centered.
*Consult individual terminal specification requirements
Process
The combination of people, equipment, tooling,
materials, methods, and procedures needed to produce
a crimp termination. Process Control is used to track
attributes over time to aid in the detection of change to
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 8 of 24
Revision: D Revision Date: 12-23-09
Figure 4-4
Figure 4-6
PUNCHES
WIRE
ANVILS TERMINALS
Figure 4-5
RAM
PRESS BASE PLATE
SHUT HEIGHT
GAUGE
the process. Detecting a process change when it
happens helps prevent many thousands of bad
crimps.
Pull Force Testing
Pull force testing is a quick, destructive way to
evaluate the mechanical properties of a crimp
termination.
Pull force testing results out of allowed range are
good indicators of problems in the process. Cut or
nicked strands in the stripping operation, lack of bell
mouth or conductor brush, or incorrect crimp height
or tooling will reduce pull force tests results. Wire
properties and stranding, and terminal design
(material thickness and serration design), also can
increase or decrease the value of the results of a
pull force levels.
If results of a pull force tests are within an allowed
range, it assures that proper crimp force has been
applied during crimping. It is crucial as, when
making a crimp, enough force must be applied to
break down the layer of non-conductive oxides that
may build up on the stripped conductor and the tin-
plating on the inside of the terminal grip. This is
necessary to provide a good metal-to-metal contact. If
this does not occur, resistance can increase. Over-
crimping a crimp termination will reduce the circular area
of the conductor and increase resistance.
Shut Height
This is the distance (at bottom dead center on a press)
from the tooling mounting base plate to the tooling
connection point on the ram of the press.
Terminal Position
The terminal position is set by the alignment of the
terminal to the forming punch and anvils, and the carrier
strip cut-off tooling. The tool set-up determines
conductor bell mouth, cut-off tab length, and terminal
extrusions
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 9 of 24
Revision: D Revision Date: 12-23-09
SECTION 5
ASSOCIATED MATERIALS
Caliper
A gauge, consisting of two opposing blades. It used
for measuring linear dimensional attributes.
Eye Loupe
This is an image
magnification tool,
normally 10 times
power or greater, which
is used in the aid visual
evaluation of a crimp
termination.
Crimp Micrometer
This is a micrometer specifically designed to
measure crimp height. The measurement is taken in
the center of the crimp so the conductor bell mouth
does not influence it. It has a thin blade that supports
the top of the crimp while a pointed section
determines the bottom radial surface.
Ruler (Pocket Scale)
This is used to measure the length of bell mouth, cut-off
tab, conductor brush, strip length and estimate wire
position. The recommended maximum resolution is
0.50mm (.020”).
Pull Tester
A device used to determine the mechanical strength of a
crimp termination. Most pull testing is done with a
device that clamps the wire, pulls at a set speed, and
measures force by means of a load cell. A pull tester
also can be as simple as hanging fixed weights on the
wire for a minimum of one minute.
Toolmaker’s Microscope
This is used for close visual evaluation and statistical
measurement of bell mouth, cut-off tab, conductor
brush, wire position and strip length.
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 10 of 24
Revision: D Revision Date: 12-23-09
SECTION 6
PROCEDURES
Tool Setup (Reference Procedures Flow Chart)
1. Check that tooling is clean and not worn. If
necessary, clean and replace worn tooling.
2. Disconnect power to the press and remove
guarding devices.
3. Install the appropriate tooling into the press.
4. Load terminals into the tooling so that the first
terminal is located over the anvil.
5. Manually cycle the press to help ensure a complete
cycle can be made without interference. If it
cannot, remove tooling and check press shut
height. Go to procedure 3.
6. Check that the tooling is aligned. Check the
impression on the bottom of the crimp that was
made by the anvil tooling. Check that the extrusions
and crimp form are centered. If not, align tooling
and go to procedure 5.
7. Check that the terminal feed locates the next
terminal over the center of the anvil tooling. If not,
adjust terminal feed and feed finger and go to
procedure 5.
8. Re-install all safety devices that were removed
during the set-up. (Follow all safety
requirements listed in individual press and/or
tooling manuals.)
9. Crimp sample terminals under power.
10. Evaluate cut-off tab length and conductor bell
mouth. If adjustment is necessary, disconnect
power to the press and remove guarding. Adjust track
position. Manually cycle the press and check the feed
finger for feed location, go to procedure 7.
11. Evaluate conductor brush. If adjustment is necessary,
disconnect power to the press and remove guarding.
Adjust wire stop for bench applications or press position
on automatic wire processing equipment. Go to
procedure 8.
12. Evaluate insulation position. If necessary, adjust strip
length, crimp new samples, and go to procedure 11.
13. Loosen insulation crimp height.
14. Crimp sample terminals.
15. Measure conductor crimp height and compare to
specification. If necessary, disconnect power and
remove guarding. Adjust conductor crimp height, install
guards, connect power, and go to procedure 14.
16. Perform a pull force test. Refer to troubleshooting
(Section 9) if this test fails.
17. Adjust insulation crimp.
18. Crimp sample terminals.
19. Evaluate insulation crimp. If necessary, disconnect
power and remove guarding. Adjust insulation crimp
height, install guards, connect power, and go to
procedure 18.
20. Measure crimp height and compare to specification. If
necessary, disconnect power and remove guarding.
Adjust conductor crimp height, install guards, connect
power, and go to procedure 18.
21. Document measurements.
Please Work Safely At All Times.
PROCEDURES FLOW CHART
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 11 of 24
Revision: D Revision Date: 12-23-09
PROCEDURES FLOW CHART
Disconnect power and
remove necessary guards
Install tooling into
the press
Load terminals
into tooling
Manually cycle
the press
Evaluate tooling
to ensure it is
lean and not worn
Replace
tooling / clean
Set shut height
of the press
Remove tooling
from the press
Start
Complete
cycle?
Tooling
aligned?
Terminal feed
OK? Terminal
centered over
anvil?
Evaluate cut-
off
tab and
bell mouth
Install safety guarding
and connect power
Crimp
samples
B
Adjust terminal feed
and/or feed finger
Alig
n
tooling
A
Adjust track
position
Manually cycle
the press
Disconnect power and
remove necessary
guarding
Crimp
samples
Adjust press
position
Evaluate
conductor
brush
Dis
connect power and
remove necessary guards
Adjust
wire stop
Loosen insulation
crimp
A
Bench press
or wire
processing
B
Evaluate
insulation
position
Crimp
samples
Adjust
strip length
Install safety guarding
and connect power
Adjust conductor
crimp height
Adjust
insulation
Trouble shooting
(See Section 9)
Disconnect power and
remove necessary
guarding
Conductor
crimp height
OK?
Pull
Force test
Crimp
samples
Insulation
OK?
Conductor
crimp
OK?
Install safety guarding
and connect power
Adjust conductor
crimp height
Disconnect power and
remove necessary
guarding
Disconnect power and
remove necessary
guarding
Adjust insulation
crimp height
Install safety guarding
and connect power
End
Wire
processing
Bench press
No
No
Yes
Yes
Yes
No
No
Yes
Good
Not
Good
No
No
Yes
Good
Not
Pass
Yes
Pass
Good
Not
Good
Not
Good
No
Yes
Size of Conductor Pullout Force’ 28 0.08 2 5.9 24 0.20 5 22.3 20 0 519 13 57.9 14 2.08 50 222 6 10 5 261 80 356 0
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 12 of 24
Revision: D Revision Date: 12-23-09
SECTION 7
MEASUREMENT
Pull Force Testing
1. Cut wire length to approximately 150mm (6”) long.
2. Strip one end to 13mm (.50”), or long enough so no
wire insulation is under the insulation grip, or
loosen the insulation crimp so it has no grip on the
insulation of the wire.
3. Terminate the appropriate terminal to the wire to
the nominal crimp height.
4. Visually inspect the termination for bell mouth, wire
brush and cut strands.
5. Set pull tester to 254mm per minute (1.00" per
minute). For most applications, a higher rate will
not have a significant impact on the data. The
slower rate prevents a sudden application of force
or jerking that snaps strands. Verify higher pull
rates with data taken at 1.00” per minute.
6. If necessary, knot the un-terminated end of the wire
(If insulation slips on wire).
7. Regardless of pull tester type, both wire and
terminated end must be securely clamped. (Note:
Clamp terminal contact interface, do not clamp
conductor crimp)
8. Activate pull test.
a. Record pull force readings. A minimum of five
pull force measurements should be done to
confirm each set-up. A minimum of 25 readings
should be taken for determining process
capability.
b. Compare lowest reading to minimum pull force
specification.
Note: High variability and lower value of Cpk (see
section 8 for explanation of Cpk) are common
when two wires are crimped together. The
variability is due to more variation in conductor
brush, conductor bell mouth and fewer strands of
one wire being in contact with the serrations on the
terminal barrel. A double wire crimp is considered
no better than the smallest wire crimped. Higher
pull force readings can be seen if both wires are
gripped and pulled exactly together. Pulling each
wire individually will result in a much lower pull
force reading. If both wires are of the same size,
the top wire will normally result in a lower reading than
the bottom wire due to the effects of the terminal
serrations.
Wire Chart
Note: Pull force has only a minimum specification. For Cpk
calculations, the average reading is assumed nominal and
the upper specification limit is set so Cp and Cpk are equal.
High pull force readings that increase the standard
deviation can lower Cpk even if the mean and lowest
reading are increased.
Test Values for Pull Force Test
UL486A
Size of Conductor Pullout Force*
AWG mm2 Lbf N
30 0.05 1.5 6.7
28 0.08 2 8.9
26 0.13 3 13.4
24 0.20 5 22.3
22 0.324 8 35.6
20 0.519 13 57.9
18 0.823 20 89.0
16 1.31 30 133.5
14 2.08 50 222.6
12 3.31 70 311.5
10 5.261 80 356.0
8 8.367 90 400.5
*Consult individual specifications
Crimp Height Testing
1. Complete tool set-up procedure.
2. Crimp a minimum of five samples.
3. Place the flat blade of the crimp micrometer across the
center of the dual radii of the conductor crimp. Do not
take the measurement near the conductor bell mouth.
4. Rotate the micrometer dial until the point contacts the
bottom radial surface. If using a caliper, be certain not
to measure the extrusion points of the crimp.
5. Record crimp height readings. A minimum of five crimp
height readings is necessary to confirm each set-up. A
minimum of 25 readings is necessary to determine
capability.
6. Check crimp height every 250 to 500 parts throughout
the run.
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 13 of 24
Revision: D Revision Date: 12-23-09
Note: Crimp height is usually control charted
because it is a quick, nondestructive measurement
and is critical for the termination's electrical and
mechanical reliability. There are three primary
purposes for control charting. One, the number of
setup samples is normally small, and its statistical
value is limited. Two, since special cause/effects
on a process are irregular and unpredictable; it is
necessary to have a means of catching changes in
the process as soon as they occur. This prevents
having to scrap thousands of terminations after the
run is over. Three and this is most important, the
data is necessary to assess and improve the crimp
process.
|||||||l||||||I|||| W3
Quality Crimping Handbook
Order No 63800-0029 Release Date: 09-04-03 UNCONTROLLED COPY Page 14 of 24
Revision: A Revision Date: 09-04-03
Figure 7-3
CALIPER
Figure 7-2
CRIMP MICROMETER
Figure 7-1
PULL TESTING
Capability 1 99.73 2,699 1.67 9999+ 0.57 To‘erance 7 USL - Mean 7 7 Mean - LSL 7
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 15 of 24
Revision: D Revision Date: 12-23-09
SECTION 8
CRIMP PROCESS CONTROL
The crimp process is the interaction of a terminal, wire,
tooling, personnel, methods and procedures, and
environmental attributes. When this process is
controlled, it will produce a quality termination. Quality
control is an important part of quality crimping. It
should not take excessive setup or inspection time to
do, and it can save a harness manufacturer thousands
of dollars in potential rework or re-manufacturing.
Variability is the slight change that occurs from crimp to
crimp. There are two types of variability, common or
special. Common causes of variation affect the
process uniformly and are the result of many small
sources. Common variability is inherent tolerances
within a reel of wire or terminals. Common variability
also is created by the natural tolerances of the stripping
and crimping machines.
Reducing variability at the common level typically has
to come from changes to the wire, terminal, and tooling
manufacturer.
Special causes of variation occur irregularly and
unpredictably. Without checks throughout a run, having
a tool become loose after the first hundred crimps or a
jam resulting from a damaged tool may be undetected
until thousands of crimps are made.
Process Capability
Before putting a new crimping tool in production, Molex
recommends that each customer do a capability study,
using the specific wire that will be used in its process.
A capability study, which is based on the assumption of
a normal distribution (bell-type curve), estimates the
probability of a measurement being outside of
specification.
Capability
Cpk +/- Sigma % Yield PPM*
0.67 2 95.45 45,500
1 3 99.73 2,699
1.33 4 99.99 63
1.67 5 99.99+ 0.57
2 6 99.99++ 0
* PPM - Parts per million potential defects.
A 25 piece minimum sample needs to be taken from
the crimping process. Calculate the average and
standard deviation for the sample. A capability index is
defined by the formula below. Cp may range in value
from zero to infinity, with a larger value indicating a
more capable process. A value greater than 1.33 is
considered acceptable for most applications. Cp is
calculated with the following formula.
___Tolerance___
6*Standard Deviation
The Cpk index indicates whether the process will
produce units within the tolerance limits. Cpk has a
value equal to Cp if the process is centered on the
mean of specification; if Cpk is negative, the process
mean is outside the specification limits; if Cpk is
between 0 and 1 then some of the 6 sigma spread falls
outside the tolerance limits. If Cpk is larger than one,
the 6-sigma spread is completely within the tolerance
limits. Cpk is calculated with the lesser of the following
formulas:
__(USL - Mean)__ __ (Mean - LSL)__
3*Standard Deviation 3*Standard Deviation
USL = Upper Specification Limit, LSL = Lower
Specification Limit
Six sigma is a goal of many companies because it
represents virtually zero defects. The ability of a
company to achieve a six-sigma level depends on the
amount of common variability in its process. For
example, hand stripping the wire produces more
variability than a stripping machine; crimping hand tools
produce more variability than a press and die set, and
bench terminations produce more variability than a
wire-processing machine.
A part of the variability in crimping will result from the
type of instruments that are used to measure the parts
and the operator's ability to repeat the measurement. A
crimp micrometer will measure more accurately than a
dial caliper. An automatic pull force system will
measure better than a hook type scale. It is important
that the measurement gauge has enough resolution.
Example Control Chart W man" m -gm ,mmm 7mm." wisp-m Dun-Wm . 1 ‘ ‘ 5. 1 ‘ mummmmuwmuu ”emu;
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 16 of 24
Revision: D Revision Date: 12-23-09
Two operators may measure the same part differently,
or the same operator may measure the part differently
when using two types of gauges. Molex recommends a
gauge capability study to identify what part of the
variability is coming from measurement error. Micro-
terminals crimped to small wire sizes need a tight crimp
height range to maintain pull force. The variability from
measurement error can keep Cpk values low.
The capability of the crimping tools needs to be re-
confirmed if the production data is significantly different
from the capability study.
Production
Before the tool is ready for production, the level of
capability needs to be established. Many harness
manufacturers run only a few hundred or few thousand
wires at one time. In this case, it is not practical or
economical to run a twenty-five-piece capability with
every set-up.
Visual Inspection
It needs to be standard operating procedure for the
operator to manually fan each bundle of crimped wires
and visually check bell mouth, conductor brush,
insulation position, cut-off tab length, and insulation
crimp.
Control Charting
Crimp height is typically control charted because it is a
quick nondestructive measurement and is critical for the
termination's electrical and mechanical reliability.
There are three primary purposes for control charting.
One, the number of setup samples is usually small, with
limited statistical value. Two, since special cause
effects on a process are irregular and unpredictable; it
is necessary to be able to catch changes in the process
as soon as they occur. This prevents thousands of
terminations from being scrapped after the run is over.
Three, and most important, this data is necessary to
assess and improve the crimp process.
Once the tooling process is setup and the wire size
does not change, keep one control chart for wire color
changes, wire length changes, terminal material
changes, or setup adjustments. Record the data point
on the chart before making a crimp height adjustment.
If data is recorded after each adjustment, the process is
likely to assume control and provide little data for
improving the process. The operator needs to make as
many notes as possible on the chart. The only truly
effective and economically sensible way to manage a
manufacturing process is to understand, monitor and
reduce sources of variability that are inherent to the
process itself. Every minute required for setup or
adjustments is unproductive.
What does this sample chart tell us?
X and R Chart
Control limit for sample of 5 = Avg
(Avg of 5 readings) + .577 x Avg (Ranges)
It indicates that a process shift occurred between
measurement 12 and 13. This type of shift could occur
due to a change in wire, a change in terminal lots, a
jam in the machine that damaged the tooling, a change
in operators, or an adjustment to the insulation crimp.
Since the measurements are still within specification,
would you stop production to adjust crimp height?
A shift in the process due to a change in material may
warrant a crimp height adjustment. A shift after a jam
would not indicate an adjustment, but a close
evaluation of the tooling. A shift in the process
between operators would not indicate an adjustment,
but an evaluation of measurement capability. The
purpose of a control chart is to identify what caused the
shift in process to determine if an adjustment to the
process is needed.
Symptom Cause Solution Worn tooling Replace tooling Wire cut depth too shallow Adiust cut depth Damaged tooling Replace tooling Cut depth too deep Adiust cut depth Conductor not on wire center Contact wire supplier Worn tooling Replace tooling Wire cut depth too shallow Adiust cut depth Wire drive rollers/belts worn Replace belts/rollers Insulation dummeter too hard Increase drive pressure Wire straightener too loose or tight Adiust Wire straightener Wrong strip length (Figure 9-4) incorrect setup Re-setup tooling
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 17 of 24
Revision: D Revision Date: 12-23-09
SECTION 9
TROUBLESHOOTING
Wire Preparation
Symptom Cause Solution
Worn tooling Replace tooling Irregular insulation cut
(Figure 9-1) Wire cut depth too shallow Adjust cut depth
Damaged tooling Replace tooling
Cut depth too deep Adjust cut depth
Cut or nicked strands
(Figure 9-2) Conductor not on wire center Contact wire supplier
Worn tooling Replace tooling Irregular conductor cut-pulled strands
(Figure 9-3) Wire cut depth too shallow Adjust cut depth
Wire drive rollers/belts worn Replace belts/rollers
Insulation durometer too hard Increase drive pressure
Wire length variability too high
(Figure 9-4) Wire straightener too loose or tight Adjust wire straightener
Wrong strip length (Figure 9-4) Incorrect setup Re-setup tooling
Figure 9-1
IRREGULAR INSULATION CUT
Figure 9-2
CUT STRANDS
Figure 9-3
PULLED STRANDS
Figure 9-4
WIRE LENGTH VARIABILITY OR WRONG STRIP LENGTH
Bell Mouth and Cut-off Tab Length Symptom Cause Solution Excesslve bell mouth, no cut-ell tab Adjust track position for small cut-off tab Adjust track position Check for worn cut-off and replace if necessary
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 18 of 24
Revision: D Revision Date: 12-23-09
Bell Mouth and Cut-off Tab Length
Symptom Cause Solution
Excessive bell mouth, no cut-off tab Adjust track position for small cut-off tab Low pull force
(Figure 9-6 and 9-7) Excessive bell mouth, cut-off tab alright Check for worn or incorrect punch tooling and replace
Adjust track position Cut or nicked strands
(Figure 9-8) No bell mouth and/or excessive cut-off tab Check for camber in terminal strip
Check for worn cut-off and replace if necessary
Long cut-off tab (Figure 9-9) Good bell mouth and excessive cut-off tab Check for worn punch tooling, replace, and re-adjust track
Figure 9-5
OPTIMAL CRIMP
Figure 9-6
EXCESSIVE BELL MOUTH
Figure 9-7
EXCESSIVE BELL MOUTH
NO CUT-OFF TAB
Figure 9-8
NO BELL MOUTH EXCESSIVE CUT-OFF TAB
Figure 9-9
EXCESSIVE CUT-OFF TAB GOOD BELL MOUTH
Conductor Brush and Insulation Position 5 m Cause Solukion conducmr brush F' ure 9-10 Bench «2 cri n - ere st ' inoorrem Ad ust wire 5 to center of transition area 3 iticnforwire ssi a \cations on ad' (Sm \ lo 3 iticnforwire ssi a \cations V o ratorswire cememabil O tortramin,reducecnm' ra‘e IMP, ‘
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 19 of 24
Revision: D Revision Date: 12-23-09
Conductor Brush and Insulation Position
Symptom Cause Solution
Insulation under conductor crimp, good
conductor brush (Figure 9-10) Strip length too short Check specification, adjust strip length longer
Bench top crimping - Wire stop position incorrect Adjust wire stop to center of transition area Insulation under conductor crimp, long
conductor brush length (Figure 9-11) Wire Processing – Press position incorrect Adjust press position away from wire
Check specification, adjust strip length longer
Insulation under conductor crimp,
short or no conductor brush (Figure 9-12) Strip length too short Re-adjust wire stop position for bench top applications OR
re-adjust press position for wire processing applications
Check specification, adjust strip length shorter
Strip length too long Re-adjust wire stop position for bench top applications OR
re-adjust press position for wire processing applications
Insulation edge centered in transition area,
conductor brush too long (Figure 9-13) Irregular wire cut-off or strands pulled
from insulation bundle Check for worn stripping tooling
Check specification, adjust strip length longer
Insulation edge centered in transition area,
conductor brush too short (Figure 9-14) Strip length too short Re-adjust wire stop position for bench top applications OR
re-adjust press position for wire processing applications
Check specification, adjust strip length shorter
Insulation edge under insulation crimp,
Good or long conductor brush (Figure 9-15) Strip length too long Re-adjust wire stop position for bench top applications OR
re-adjust press position for wire processing applications
Bench top crimping - Wire stop position incorrect Adjust wire stop to center of transition area
Wire processing - Press position incorrect Adjust press position away from wire
Insulation edge under insulation crimp,
short or no conductor brush (Figure 9-16) Verify operators wire placement ability Operator training, reduce crimping rate
Figure 9-10
INSULATION UNDER CONDUCTOR
CRIMP, GOOD CONDUCTOR BRUSH
Figure 9-11
INSULATION UNDER CONDUCTOR CRIMP,
CONDUCTOR BRUSH TOO LONG
Figure 9-12
INSULATION UNDER CONDUCTOR CRIMP,
SHORT OR NO CONDUCTOR BRUSH
Figure 9-13
CONDUCTOR BRUSH TOO LONG
Figure 9-14
CONDUCTOR BRUSH TOO SHORT
Figure 9-15
INSULATION UNDER INSULATION CRIMP,
CONDUCTOR BRUSH TOO LONG
Figure 9-16
INSULATION UNDER INSULATION CRIMP,
CONDUCTOR BRUSH TOO SHORT
Insulation Crimg Tighten insuiation crimp height Evaluate terminal diameterwire (Figure 9-22) into conductor strands (Figure 9-23) fails bend test (Figure 9-24) INSULATION CRIMP INSULATION CRIMP INSULATION CRIMP ACCEPTABLE INSULATION CRIMP MARGINAL INSULATION CRIMP IMF
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 20 of 24
Revision: D Revision Date: 12-23-09
Insulation Crimp
Symptom Cause Solution
Tighten insulation crimp height Terminal surrounds less than 88% of a large
diameter wire (Figure 9-21)
Crimp too loose, not enough
terminal insulation barrel Evaluate terminal
Terminal contacts less than 50% of a small
diameter wire (Figure 9-22) Too much terminal insulation barrel Evaluate terminal
Insulation crimp barrels cut through insulation
into conductor strands (Figure 9-23) Crimp too tight Adjust insulation crimp height*
Insulation not firmly gripping insulation,
fails bend test (Figure 9-24) Crimp too loose Adjust insulation crimp height tighter
* Inexpensive hand tools provide no adjustment for the insulation crimp. A hand tool is intended for low volume
applications. Although you are not able to adjust the insulation crimp on a hand tool, an insulation crimp, which
pierces the insulation, may still be considered acceptable for many applications. This criterion only applies to hand
tools due to their low speed crimp cycle. If the insulation crimp pierces the insulation, the wire strands tend to move
aside without damage.
Figure 9-17
PREFFERRED
INSULATION CRIMP
Figure 9-18
PREFFERRED
INSULATION CRIMP
Figure 9-19
ACCEPTABLE
INSULATION CRIMP
Figure 9-20
ACCEPTABLE INSULATION CRIMP
Figure 9-21
MARGINAL INSULATION CRIMP
Figure 9-22
MARGINAL INSULATION CRIMP
Figure 9-23
MARGINAL INSULATION CRIMP
Figure 9-24
MARGINAL INSULATION CRIMP
Crimg Height Symptom Cause Solution Changed insuiaiion coior or duromeier Changed crimp press (shut neigni) Changed tooling set-up Wire variabilii Damaged, ioose or worn iooling Tooling replacement or iigniening cm or missing Wire strands Siripping process adiusimeni anul bnuificuiion Nominfl u m7 4 x E a g a i 4 2 , 0.0440 0.0450 WNW 0.!M7li 0.04M mus Muss «ms aims Crimp Height Luwel L .r , sumac-nu.- “‘"“‘"" Swag-mm suntafiikun Wm“ sucti'n’i'm u .2 ‘0 w E x 2 x 2 ii i 2 J JH mi ”J W ‘mii 0W .. “‘J J H'“ ' ' ‘ ‘ ‘ ' 03““ 0,0150 mum (Lllflli Mum) mus ““55 mm“ mm! Mus IMIISS Midi-5 M1475 (‘nmp might cm... new IGH
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 21 of 24
Revision: D Revision Date: 12-23-09
Figure 9-25
OPTIMAL CRIMP HEIGHT CHART
Figure 9-26
CRIMP HEIGHT OFF TARGET
Figure 9-27
CRIMP HEIGHT VARIABILITY TOO HIGH
Crimp Height
Symptom Cause Solution
Changed wire type vendor or stranding
Changed insulation color or durometer
Changed crimp tooling
Changed crimp press (shut height)
Changed press type (manufacturer)
Changed terminal reel (lot code)
Changed tooling set-up
Adjust tooling back to target
Crimp height off target
(Figure 9-26)
Damaged or worn tooling Replace tooling
Wire variability
Terminal variability Inspect incoming product
Damaged, loose or worn tooling Tooling replacement or tightening
Measurement error Gauge capability analysis
Terminal spring-back too great, over crimping Crimp height adjustment
Crimp height variability too high
(Figure 9-27)
Cut or missing wire strands Stripping process adjustment
Pull Force 10 Musul emznh Symptom Cause Solution Crimp height too low Adjust crimp height insulation crimp through insulation wall Raise insulation crimp height Conductor bell mouth too big Adrust tooling track Terminal material thickness too small Light serrations on terminal Contact your local sales engineer Lower Spuil‘iuliun l2 20.0 25.0 30.0 15.0 40.0 A51: 50.0 Pull um: um Murmur-cub aner Specificnimi :rll 2M zs.n 30.0 55.0 40.0 45.0 50.0 Pull Force (In) it . - ..
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 22 of 24
Revision: D Revision Date: 12-23-09
Pull Force
Symptom Cause Solution
Cut or nicked strands Check the stripping process
Crimp height too low Adjust crimp height
Small or no bell mouth Adjust tooling track
Wire breaks before conductor
crimp - low pull force
(Figure 9-29) Insulation crimp through insulation wall Raise insulation crimp height
Crimp height too high Adjust crimp height
Small or no conductor brush Increase strip length
Conductor bell mouth too big Adjust tooling track
Gold terminal application
Terminal material thickness too small Evaluate the terminal application
Wire pulls out of conductor
grip - low pull force
(Figure 9-29)
Light serrations on terminal Contact your local sales engineer
Figure 9-28
OPTIMAL PULL FORCE CHART
Figure 9-29
LOW PULL FORCE CHART
8 8.302 .01287 1 .1280 3.25 .128 16384 2175.00 489.0 8 7.820 .01212 19 .0285 3.68 .145 15433 2048.72 460.6 10 4.740 .00735 37 .0159 2.92 .115 9354 1241.75 279.2 10 5.006 .00776 49 .0142 2.95 .116 9880 1311.63 294.9 10 5.320 .00825 105 .0100 2.95 .116 10500 1393.89 313.4 12 3.308 .00513 1 .080 2.05 .081 6529 866.69 194.8 14 2.270 .00352 7 .0253 1.85 .073 4481 594.81 133.7 14 1.941 .00301 19 .0142 1.85 .073 3831 508.59 114.3 14 2.078 .00322 41 .0100 1.85 .073 4100 544.28 122.4 16 1.307 .00203 65 .0063 1.50 .059 2580 342.48 77.0 16 1.330 .00206 105 .0050 1.47 .058 2525 348.47 78.3 18 .823 .00128 1 .0403 1.02 .040 1524 215.60 48.5 20 .519 .00080 1 .0320 .81 .032 1024 135.94 30.6 20 .563 .00087 7 .0126 .97 .038 1111 147.53 33.2 20 .507 .00079 10 .0100 .89 .035 1000 132.75 29.8 20 .616 .00096 19 .0080 .94 .037 1215 161.43 36.3
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 23 of 24
Revision: D Revision Date: 12-23-09
SECTION 10
Wire Gauge Chart
Wire Area Stranding Wire Diameter Circular Wire Break
AWG Sq. mm Sq. inch No. Dia. mm In. Mills N Lbf.
8 8.302 .01287 1 .1280 3.25 .128 16384 2175.00 489.0
8 7.820 .01212 19 .0285 3.68 .145 15433 2048.72 460.6
8 7.955 .01233 49 .0179 3.73 .147 15700 2084.21 468.6
8 8.605 .01334 133 .0113 3.73 .147 16983 2254.49 506.9
8 8.513 .01319 168 .0100 3.73 .147 16800 2230.22 501.4
8 8.424 .01306 665 .0020 3.73 .147 16625 2206.99 496.2
10 5.261 .00816 1 .1019 2.59 .102 10384 1378.44 309.9
10 4.740 .00735 37 .0159 2.92 .115 9354 1241.75 279.2
10 5.006 .00776 49 .0142 2.95 .116 9880 1311.63 294.9
10 5.320 .00825 105 .0100 2.95 .116 10500 1393.89 313.4
12 3.308 .00513 1 .080 2.05 .081 6529 866.69 194.8
12 3.632 .00563 7 .0320 2.44 .096 7168 951.56 213.9
12 3.085 .00478 19 .0179 2.36 .093 6088 808.16 181.7
12 3.294 .00511 65 .0100 2.41 .095 6500 862.88 194.0
12 3.3118 .00514 165 .0063 2.41 .095 6549 869.37 195.5
14 2.082 .00323 1 .0641 1.63 .064 4109 545.45 122.6
14 2.270 .00352 7 .0253 1.85 .073 4481 594.81 133.7
14 1.941 .00301 19 .0142 1.85 .073 3831 508.59 114.3
14 2.078 .00322 41 .0100 1.85 .073 4100 544.28 122.4
14 2.112 .00327 105 .0063 1.85 .073 4167 553.24 124.4
16 1.308 .00203 1 .0508 1.30 .051 2581 342.58 77.0
16 1.433 .00222 7 .0201 1.52 .060 2828 375.43 84.4
16 1.229 .00191 19 .0113 1.47 .058 2426 322.07 72.4
16 1.317 .00204 26 .0100 1.50 .059 2600 345.15 77.6
16 1.307 .00203 65 .0063 1.50 .059 2580 342.48 77.0
16 1.330 .00206 105 .0050 1.47 .058 2625 348.47 78.3
18 .823 .00128 1 .0403 1.02 .040 1624 215.60 48.5
18 .897 .00139 7 .0159 1.22 .048 1770 234.93 52.8
18 .811 .00126 16 .0100 1.19 .047 1600 212.40 47.8
18 .963 .00149 19 .0100 1.24 .049 1900 252.23 56.7
18 .825 .00128 41 .0063 1.19 .047 1627 216.03 48.6
18 .823 .00128 65 .0050 1.19 .047 1625 215.72 48.5
20 .519 .00080 1 .0320 .81 .032 1024 135.94 30.6
20 .563 .00087 7 .0126 .97 .038 1111 147.53 33.2
20 .507 .00079 10 .0100 .89 .035 1000 132.75 29.8
20 .616 .00096 19 .0080 .94 .037 1216 161.43 36.3
20 .523 .00081 26 .0063 .91 .036 1032 136.99 30.8
20 .519 .00081 41 .0050 .91 .036 1025 136.07 30.6
22 .324 .00050 1 .0253 .64 .025 640 84.97 19.1
22 .355 .00055 7 .0100 .76 .030 700 92.93 20.9
24 .201 .00031 10 .0063 .61 .024 397 52.69 11.8 24 .241 .00037 19 .0050 .58 .023 475 53.06 14.2 24 .200 .00031 41 .0031 .58 .023 394 52.31 11.8 28 .089 .00014 7 .0050 .38 .015 175 23.23 5.2 28 .093 .00014 19 .0031 .41 .016 183 24.24 5.4 30 .051 .00008 1 .0100 .25 .010 100 13.28 3.0 30 .057 .00009 7 .0040 .30 .012 112 14.87 3.3
Quality Crimp Handbook
Order No: TM-638000029 Release Date: 09-04-03 UNCONTROLLED COPY Page 24 of 24
Revision: D Revision Date: 12-23-09
Wire Area Stranding Wire Diameter Circular Wire Break
AWG Sq. mm Sq. inch No. Dia. mm In. Mills N Lbf.
22 .382 .00059 19 .0063 .79 .031 754 100.11 22.5
22 .329 .00051 26 .0050 .76 .030 650 86.29 19.4
24 .205 .00032 1 .0201 .61 .024 404 53.63 12.1
24 .227 .00035 7 .0080 .58 .023 448 59.47 13.4
24 .201 .00031 10 .0063 .61 .024 397 52.69 11.8
24 .241 .00037 19 .0050 .58 .023 475 63.06 14.2
24 .200 .00031 41 .0031 .58 .023 394 52.31 11.8
26 .128 .00020 1 .0159 .40 .016 253 33.56 7.5
26 .141 .00022 7 .0063 .53 .021 278 36.88 8.3
26 .127 .00020 10 .0050 .51 .020 250 33.19 7.5
26 .154 .00024 19 .0040 .48 .019 304 40.36 9.1
28 .080 .00012 1 .0126 .32 .013 159 21.08 4.7
28 .089 .00014 7 .0050 .38 .015 175 23.23 5.2
28 .093 .00014 19 .0031 .41 .016 183 24.24 5.4
30 .051 .00008 1 .0100 .25 .010 100 13.28 3.0
30 .057 .00009 7 .0040 .30 .012 112 14.87 3.3
30 .060 .00009 19 .0025 .30 .012 118 15.64 3.5
32 .032 .00005 1 .0080 .20 .008 64 8.50 1.9
32 .034 .00005 7 .0031 .20 .008 67 8.93 2.0
32 .039 .00006 19 .0020 .23 .009 76 10.09 2.3
Americas Headquarters
Lisle, Illinois 60532 U.S.A.
1-800-78MOLEX
amerinfo@molex.com
Far East North Headquarters
Yamato, Kanagawa, Japan
81-462-65-2324
feninfo@molex.com
Far East South Headquarters
Jurong, Singapore
65-6-268-6868
fesinfo@molex.com
European Headquarters
Munich, Germany
49-89-413092-0
eurinfo@molex.com
Corporate Headquarters
2222 Wellington Ct.
Lisle, IL 60532 U.S.A.
630-969-4550
Fax: 630-969-1352
Visit our Web site at http://www.molex.com