IOSR Journal of Engineering (IOSRJEN)

ISSN (e): 2250-3021, ISSN (p): 2278-8719

Vol. 04, Issue 12 (December 2014), ||V1|| PP 11-17

www.iosrjen.org

Reduction of Bending Defects in Hard Disk Drive Suspensions

Khwanchai Huailuk1, Napassavong Rojanarowan2
12

Department of Industrial Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, Thailand

Abstract: - The objective of this research was to reduce the rate of bending defects in hard disk drive
suspensions by applying the Six Sigma method. The study first defines the problem, gives the project objective
and scope. Then, the Attribute Agreement Analysis was performed to ensure the reliability of the inspection
results. Process Mapping then lists the Process Input Variables that lead to bending defects. Then, the effects of
various variables on bending defects were analyzed using the Cause and Effect Matrix. The criteria of the Failure
Mode and Effects Analysis technique were then used to reduce the number of variables and prioritize the Key
Process Input Variable (KPIV) for further analysis. Following this, process factors were improved by changing
tray design in Laser Welding, Cutting, Forming, Final Trim, Damper Attachment and Backend processes. The
tooling die was modified in the forming process. The handling tool was modified in the Final trim and Damper
Attachment processes. Finally, the creation or revision of standard work instructions (Do/ Dont) were done and
the operators were trained. After that, a control plan was set up to control production processes. The results show
that the bend defect rate decreased from 15649 PPM to 9263 PPM.

Keywords: - Bend Defect, Hard Disk Drive Suspension, Six Sigma, Hypothesis Testing, Defect Reduction
I.

INTRODUCTION

A suspension is a component of a hard disk drive designed to record and write data on a hard disk.
Suspension manufacturers need to focus on the costs of production as well as the performance, reliability and
stability of the system. As a result, manufacturers must find ways and means to improve quality in order to
remain competitive in terms of cost and quality in a highly competitive industry.
This research was designed to reduce production costs by reducing waste generated from the
production of defective suspensions and to produce quality products to meet the needs of consumers. Bending is
one of the top three defects and generates the highest scrap costs in suspension system production. Bend defects
in suspension systems occur from improper assembly processes. Thus, there is a need to understand each factor
in the process that creates bending defects in order to reduce defects. To determine this relationship, effective
experiments have to be performed to obtain experimental results. Hypothesis testing is a technique used to test
the significance of factors analyzed and post-process improvement.
Hard disk drive suspensions consist of five main parts: a mount plate, load beam, hinge, damper and
flexible circuit (flexure). The suspension assembly process starts with cleaning, cutting and forming raw
material (load beam and flexure), then laser welding assembly of components (formed load beam, formed
flexure, mount plate), cutting suspension, pre-gram load and statistic forming suspension, final trim suspensions,
cleaning, annealing, damper attachment, gram load and static attitude adjustment, final visual inspection, quality
audit and packing. In the assembly process there are many factors that may affect disk bending such as machine
tooling, the die contacting the suspension directly and incorrect operating procedures. There will be a reduction
of defects if these factors are corrected.

II.

LITERATURE REVIEW AND METHODOLOGY

This research aims to study and improve all types of factors affecting bend defects in hard disk drive
suspensions. Thus, a systematic method was utilized to prioritize, select, and improve the most important factors
leading to such defects. The principle objective was to decrease defects, reduce production costs, and create
value [1]. The Six Sigma approach consists of five steps known as DMAIC which represents Define, Measure,
Analyze, Improve, and Control. This paper implemented the DMAIC procedure of the Six Sigma approach. In
the Define phase, a project charter and process map were written to describe the problem and current processes.
In the Measure phase, the measurement system and process capability were analyzed to ensure that the current
performance of the process was correctly understood. In the Analyze phase, key process input variables (KPIVs)
were identified and reduced for further investigation in the next phase using the Cause-and-Effect Matrix and
the criteria from the Failure Mode and Effects Analysis. In the Improve phase, the design of experiment
technique was used to test for the significance of the selected factors and to determine optimal levels of
significant factors. In the Control phase, a control plan and other control tools including control charts and check
list, were set up to maintain performance after improvement [2].

International organization of Scientific Research

11 | P a g e

Reduction of Bending Defects in Hard Disk Drive Suspensions

Many studies have applied the Six Sigma approach and shown that it can help companies worldwide
save costs and effectively achieve targets for the reduction of defect rates in production [3],[4],[5]. Also, the Six
Sigma method is an effective way to perform quality improvement in various industries as has been shown by
Ruthaiputpong and Rojanarowan [6] who used the Six Sigma approach to increase the read/write area of hard
disk drives. Sonphuak and Rojanarowan [7] applied DMAIC steps in a case study to improve the strength of
modulus of rupture (MOR). Artharn and Rojanarowan [8] presented methods to reduce dent defects in the
flexible printed circuit manufacturing process by applying the Six Sigma quality improvement method. Finally,
Kaewon and Rojanarowan, [9] applied the Six Sigma method to study the effect of assembly process factors on
seal strength of a VRLA battery.

III.

RESULTS

3.1 Define Phase

In the Define Phase, the project charter was written. The project charter consists of the Problem statement,
Objective Statement, Project Scope, Project Metrics, Project constraints, Project assumptions, Team members
and Timeline. Table 1 shows the project charter. The rate of bend defects in hard disk drive suspensions was
reported at 15649ppm. This defect rate amounts to 9,944,224 THB/year in losses. The research goal is thus to
decrease the defect rate to 13800ppm (70% reduction from entitlement 13000ppm) by December 2013.
Table 1 Project Charter
Project Title:
Reduction of bend defects in hard disk drive suspensions
Business Case:
Suspension product defects cause a significant scrap cost and customer claims.
Problem Statement:

Objective Statement:

Project Scope:
Project Metrics:

Project Assumption:
Project Constraints:
Team Member:

Based on the hard disk drive suspension yield report of product S+M, there was a
high bend defect rate of 15649 ppm from January-June 2013. This led to losses of
suspension products on average of 9,944,224 THB/year.
To reduce bend defects from 15649 ppm to 13800 ppm (70% from entitlement
13000 ppm) of product S+M by end of December 2013 as measured by a hard
disk drive suspension yield report.
Study of product S+M
Business metric: Customer satisfaction Primary metric: Bend Defect rate
Secondary metric: machine UPH Consequential metric: Investment in resources
Financial Metric: Cost of scrap and rework
The champion supports the project and team has authority to do experiments and
implement solutions.
Team meets two hours/week with information updates by email.

Director/ Manager/ Engineers: Manufacturing, Tooling, Equipment,

Product and Quality

3.2 Measure Phase

In the Measure Phase, the Measurement System Analysis (MSA) of bend defects is performed to make sure that
measurements are reliable. The measurement system is tested by visual inspection for Attribute data. Then, an
Attribute Agreement Analysis was performed to study 42 operators and two replicates. In this study the
Repeatability results within Appraiser for each operator was more than 95.0% while the value between
Appraisers (Reproducibility) was 90.0%. The Kappa was >0.90 which shows that the measurement system had
sufficient reliability from inspection agreement to measure the process [10].
In order to understand the process before the Analyze phase, a study of problems and preliminary data
analysis was performed. Bend defects were defined then a failure analysis was performed by characteristics and
Zone.
The pie chart in Fig. 1 shows major defect characteristics by zone. The contribution on defects from the four
zones was as follows: BE1 (48%), BE2 (13%), BE3 (16%), and BE4 (23%). Thus, BE1, BE3, and BE4 zones
were chosen to be improved further since they contributed 87% of total defects.
Following that, the defect analysis by operation was done to find out which main operations created the
defect by visual mapping before and after each process. A sample size (n) of greater than 708 pieces was needed
to ensure the power of the test was >80% [11].

International organization of Scientific Research

12 | P a g e

Reduction of Bending Defects in Hard Disk Drive Suspensions

Fig. 1 Pie Chart showing Contribution on Bending Defects

Table 2 Defect Data by Process

The results shown in Table 2 found six processes created bend defects including: laser welding, cutting,
forming, final trim, damper attachment and backend. Next, process mapping was performed to list Process Input
Variables (PIV) that affected bend defects. Then each factor was prioritized using the Cause & Effect Matrix
and the score rating from FMEA [12], [13] led to six process variables for analysis of significance in the
analysis phase.
3.3 Analyze Phase
In this phase, the KPIVs which affected the bend defect of each process were identified and prioritized.
The variables are listed according to 5M categorization (man, machine, method, material, measurement). The
six process factors were selected to be analyzed further to see if they statistically affected bend defects of
suspension systems. Hypothesis testing was done to test the significance of those factors. The details of
experiments from visual inspection of each KPIV with a sample size of 1000 pcs (power of test>80%) was
noted [11] and then analyzed using a proportion test to establish whether and which KPIVs created bend defects
or not [14]. See the results in the Hypothesis of list KPIVs in Table3.
: Proportion of bend defects
H0 : = 0,
Ha : > 0

International organization of Scientific Research

13 | P a g e

Reduction of Bending Defects in Hard Disk Drive Suspensions

Process
Laser
Weld

Cutting

Forming

Final
Trim

Damper
attach

Backend

Table3 Hypothesis Testing Results for KPIVs with = 0.05

KPIVs
Hypothesis
Gage
KPOVs
Result
Test
Position to lift the TG
Proportion Visual
Bend
Significant
out of tray
inspection
Defect

Revise WI
Do Dont

Clearance between part

and tray

Proportion

Visual
inspection

Bend
Defect

Significant

Design
Tray

Clearance between part

and hand tool

Proportion

Visual
inspection

Bend
Defect

Not
Significant

Clearance between part

and hand tool
Clearance between part
and tray (strip level)
Clearance between part
and die
Clearance between part
and hand tool
Clearance between part
and tray (strip level)
Clearance between part
and die
Position to hand carry
the tray
Die condition

Proportion

Proportion

Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect

Not
Significant
Not
Significant
Significant

Clearance between part

and die/pin
Clearance between part
and tray (strip level)
Clearance between part
and tray (piece part
level)
Clearance between part
and hand tool
Position of part in tray

Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection
Visual
inspection

Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect
Bend
Defect

Significant

Proportion
Proportion
Proportion
Proportion
Proportion
Proportion
Proportion

Proportion
Proportion

Proportion
Proportion

Clearance between part

and tray
Handling

Proportion

Clearance between part

and hand tool
Position of part in tray

Proportion

Position to hand carry

the boat
Clearance between part
and tray
Handling

Proportion

Proportion

Proportion

Proportion
Proportion

International organization of Scientific Research

Not
Significant
Significant
Significant
Not
Significant
Not
Significant
Not
Significant
Significant
Significant

Significant
Significant
Significant
Significant
Significant
Significant
Not
Significant
Not
Significant

Action Plan

New

Modify Die
Design New
Tray
Modify Die
Design New
Tray
Design New
Tray
Design New
hand tool
Design New
hand tool
Design New
Tray
Design New
Tray
Design New
hand tool
Design New
hand tool
Revise WI
Do Dont
-

14 | P a g e

Reduction of Bending Defects in Hard Disk Drive Suspensions

3.4 Improve Phase
In the improvement phase, KPIVs which displayed significant differences were included in
improvement plans to reduce bend defects which have four actions in table 4. Action 1- Design a new tray to
solve the problem of clearance between parts and the tray where the upper tray directly presses on the parts; a
small lump makes the part bend. Action 2 - Design a new hand tool to solve the clearance problem between
parts and the hand tool which tends to get stuck with the upper tray during unloading from the final trim tray by
the strip below. This either presses the part into its correct position or makes it drop outside its correct position
which creates bends. Action3 - Modify the die to solve the problem of clearance between parts and the die by
tooling the die to operate with direct contact with the work piece. Action4 - Revise work instructions to train the
operator in the correct operating method.
Table 4 Improvement Plan
Concept
Improvement
1. The lump on the
Design new tray to
center bar presses on
prevent lumps on the
the suspension make
center bar pressing
the bend defect
on the suspension
2. The upper tray
and to keep the upper
sticks to the strip of
tray from sticking to
the lower tray,
the strip of lower tray
causing a bend defect. and allocate the
3. Location of pin at
locating pin to
tail causes bend
correct position.
defect.
Locating cavity is
Design locating
oval so the position is cavity alignment by
not so accurate. This
circle hole.
causes the part to fall
outside its cavity.

No.

Action Plan

Cause of Problem

Design New
Tray

Design New
hand tool

Modify Die

Clearance between
the part and die is
close which creates
opportunities for bend
defects while the
tooling die operates.

Revise
Work
Instruction

The problem caused

by operator error such
as mistake in load-in
and re-loading.

Implement
Process
Laser,
Cutting,
Forming,
Final Trim,
Damper
and
Backend

Hypothesis
Test
Proportion

Significant

Final Trim,
Damper
Attachment

Proportion

Significant

Relief tooling
material to avoid
direct contact while
die operates to
prevent bend defect.

Forming

Proportion

Significant

To reduce this
problem, Do/Dont
WI is created.

Laser
Weld,
Backend

Proportion

Significant

3.5 Control Phase

In the control phase, the control plan was created as shown in Table 5.
Table 5 Control Plan
Process
Checking Method
Sampling Plan
Equip
ment

Product
Characte
ristics
Bend
defect

Specification
/ Disposed
by
Refer to
visual
inspection
document

Sample
Size/
Frequency
10Pcs./
2 Lots

Laser Weld,
Micro
Cutting,
scope
Forming, Final
10X
Trim, Damper
Attach and
FVMI
Legend: Action: 1 = Stop Machine, 2 = Adjust / Fix/ Change

International organization of Scientific Research

Result

Triggering point

Res
ponse

Trigger
Limit

Action

Response

QC

Out of
spec

1,2

Process/
Maintenance
Engineer

15 | P a g e

Reduction of Bending Defects in Hard Disk Drive Suspensions

seen in

The result after implementation on the production line is the Bend defect decreased to 9263 ppm as
Fig. 2 Bend defect trends.

Fig.2 Bend Defect trends

Table 6 shows the comparative results before and after improvement. It shows that the defective rate
from bend defect decreased from 15649 ppm to 9263 ppm, which achieved the target of 13800 ppm. The cost
reduction of 9,944,224 THB/year to 4,583,970 THB/year (5,360,254 THB/year saving) was realized.

Item
Input (Piece/ Year)
%Bend Defect
Bend defect (Pieces)
Price / piece (Baht)
Loss / Year (Baht)

Table 6 Bend Defect Before-After Improvement

Before Improve
After Improve
(Forecast Input)

Improvement

38,797,932
1.56
605,248
16.43

30,000,000
0.93
279,000
16.43

0.63
326,248
-

9,944,224

4,583,970

5,360,254

IV.

CONCLUSION

This paper presented a method to reduce bend defects in suspension systems in hard disk drives. The
Six Sigma approach with DMAIC steps was applied. In the Define Phase the problem, the objective, and the
scope of project were identified. In the Measure Phase, the attribute agreement analysis was performed. Then,
the Process Input Variable (KPIV) of bend defects were prioritized by using the Cause and Effect Matrix and
the criteria of FMEA. Next in the Analyze phase, the Hypothesis Testing was used to test for significant causes
of bend defects. In the Improve phase, steps were taken to design a new tray, a new hand-tool, modify the die
and revise the work instructions to improve the bend defect. Finally, in the Control phase, the control plan was
set up by document control and responsible technicians assigned. The results after implementation is bend
defects were reduced from 15649 ppm to 9263 ppm (better than target 13800ppm).
REFERENCES

[1]
[2]
[3]

[4]

B. Klefsj, H. Wiklund, and R.L. Edgeman, Six Sigma seen as a methodology for total Management,
Measuring Business, 5, 2001, 31-35.
T. Pyzdek, The Six Sigma handbook revised and expanded (McGraw-Hill, 2003)
J. Antony, M. Kumar, and M.K. Tiwari, An application of Six Sigma methodology to reduce the engine
overheating problem in an automotive company, Journal of Engineering Manufacture, 209, 2005, 633646
R.B. Anand, S.K. Shukla, A. Ghorpade, M.K. Tiwari and R. Shankare, Six Sigma-based approach to
optimize deep drawing operation variables, International Journal of Production Research, 45, 2007,
2365-2385.

International organization of Scientific Research

16 | P a g e

Reduction of Bending Defects in Hard Disk Drive Suspensions

[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]

S. Coleman, Six Sigma: an opportunity for statistics and for statisticians, Significance, 5, 2008, 94-96.
P. Ruthaiputpong and N. Rojanarowan, Improvement of track zero to increase read/write area in hard
disk drive assembly process, Uncertain Supply Chain Management, 2013, 165-176.
W. Sonphuak and N. Rojanarowan, Strength improvement of fibre cement product, International Journal
of Industrial Engineering Computations, 2013, 505-516.
P. Artharn and N. Rojanarowan, Defective Reduction on Dent defects in flexible printed circuit
manufacturing process, IOSR Journal of Engineering, 3, 2013, 23-28.
W. Kaewon and N. Rojanarowan, Seal strength improvement of valve-regulated lead-acid (VRLA)
battery, IOSR Journal of Engineering, 3, 2013, 39-43.
Automotive Industry Action Group (AIAG), Measurement System Analysis (MSA) (Michigan, USA.,
2002).
Breyfogle, F. W. III, Implementing SIX SIGMA Smarter Solutions Using Statistical Methods (New
York: John Wiley & Sons, 2001).
A. Shahin, Integration of FMEA and the Kano model: An exploratory examination, International Journal
of Quality& Reliability Management, 21, 2004, 731-46.
Q. Feng, X. Wu, J. Luo, and B. Qi, Research and Application of Special Vehicle Design Based on the
FMEA, Applied Mechanics and Materials, 404, 2013, 171-176.
D.C. Montgomery, Applied Statistical and Probability for Engineers (New York: John Wiley & Sons,
2011).

International organization of Scientific Research

17 | P a g e