■ If Takt Time Cannot Be Achieved, Volume Production Cannot Be Said to Work
It could be automated.
But applying one piece takes 30 seconds.
The takt time the customer needs is 3 seconds.
This cannot be used in volume production.
In other words, in evaluating automation, it is necessary to look not only at
whether it can or cannot be done,
but
whether it can be done stably at the required speed.
■ Raising Speed Brings Other Problems
Peeling faster.
Transporting faster.
Applying faster.
Then
component deformation,
static electricity,
air bubbles,
and misalignment
may increase.
In other words, in automation, it is necessary to find
not the maximum speed,
but a stable speed at which quality can be maintained.
■ The Equipment Stoppage Rate Is Also Important
It fails to pick up a component once in 100 times.
In prototyping, this may not seem like a big problem.
But equipment that runs tens of thousands of times a day stops frequently.
The equipment stops.
Workers restore it.
Productivity drops.
In other words,
even 99% success
may be insufficient in volume production.
Automation requires extremely high reproducibility.
■ Equipment Stoppage Costs Can Be Greater Than Material Costs
A material a few yen cheaper per sheet was adopted.
But peeling defects increased.
Equipment stoppages increased.
Workers restore it.
As a result, the total cost went up.
In other words, in volume production, it is necessary to look at
total cost including equipment utilization, not just component unit price.
This cannot be judged by purchase price alone.
■ In Automation, "Yield" Alone Is Not Enough
For example, product yield is 99.5%.
As a number, it looks high.
But if the automated application process fails 0.5% of the time, 50 defects arise in 10,000 pieces.
Furthermore, equipment stoppages and rework occur.
That is why, in mass production, even small defect rates become big problems.
■ Changing the Component Design Slightly Can Make Automation Easier
For example,
making rounded corners larger,
making narrow sections slightly wider,
adding recognition holes,
changing tab positions,
and changing the shape considering the peeling direction.
Such small changes can greatly improve automation stability.
In other words, the option of
reviewing the component shape before upgrading the equipment
is also important.
■ Materials Are Sometimes Changed to Suit the Equipment
In thermal performance, material A is the best.
But material A is too soft to be handled by automated machines.
Material B has slightly lower thermal performance.
But automated application is very stable.
Looking at the actual device as a whole, material B can give better volume production quality.
In other words,
the best material performance and the best product performance are not the same.
It is necessary to judge including volume production capability.
■ Failure Example: Trying to Automate Prototype Materials as Is
In prototyping, a soft, high-performance TIM was applied by hand.
Thermal performance was good.
In volume production, an attempt was made to supply the same TIM on reels.
However,
it could not be peeled.
It stretched when picked up.
The application position was unstable.
In the end, major modification of the equipment side was needed.
This is a failure caused by
deciding on the prototype material first and thinking about automation afterwards.
■ Failure Example: Processed per the Drawing, but the Equipment Could Not Recognize It
A transparent film was processed with high precision per the drawing.
The dimensions were fine.
But the automated machine’s camera could not recognize the outline.
As a result, the application position was not stable.
In this case, processing quality is high.
But it does not work as a volume-produced component.
What was needed was design that considered
recognition holes,
marks,
and contrast with the background.
■ Failure Example: Prioritizing Minimum Material Width Made Transport Unstable
To improve material yield, the carrier width was made as narrow as possible.
Material cost decreased.
But it meandered during reel transport.
Component positions shifted.
The amount of image correction increased.
Equipment stoppages increased.
As a result, productivity worsened.
In other words, this is an example where
local optimization of material yield worsened the volume production process as a whole.
■ Failure Example: Changing the Component Design After the Automated Equipment Is Completed
The equipment manufacturer completed the machine.
After that, the component shape was changed.
Then
it did not fit the suction head,
the recognition mark position changed,
and the peeling angle did not match.
Equipment modification became necessary.
In other words, developing components and equipment separately tends to cause rework.
■ The Researcher's Perspective: In Automation, Look at the "Dynamic Properties" of Components
In material catalogs, one often looks at static values such as
thickness,
hardness,
thermal conductivity,
and adhesive strength.
In automated processes, however, what matters is how it behaves
when peeled at high speed,
when bent,
when picked up,
and when accelerated.
In other words, it is necessary to look not only at
the performance of the material at rest, but its performance when moved.
■ The Shop-Floor Perspective: Look Where the Equipment Stops
When considering automation problems, looking only at finished products does not reveal the cause.
Actually look at the equipment.
Look at
where the component remains,
where it shifts,
where pick-up failures occur,
and at what timing it stops.
Then improvement points on the component side become visible, such as
shape,
peeling force,
carrier,
and static electricity.
■ OTIS's Perspective: Designing Components on the Premise of Automation
What OTIS should aim for is not only
“receiving a drawing and cutting it.”
It is checking how it will be handled in volume production equipment,
and, if necessary, proposing as far as
● Shape changes
● Adding tabs
● Partial adhesion
● Half-cutting
● Carrier design
● Pitch design
● Reel conversion
In other words,
not making automation solely the equipment manufacturer’s job.
Making automation work from the component side too.
This is where a processing company’s value lies.
■ Thinking of the Reel Product Itself as a Single Product
Rather than looking only at the finished component,
the component,
liner,
carrier,
pitch,
winding direction,
and reel
are all considered as a single supply product.
It can be used stably from the moment it is fed into the customer’s equipment.
If it can be made to this state, it is differentiated from simple die-cutting.
■ Thinking of Components and Equipment as a Set
OTIS designs and processes the components.
The equipment manufacturer designs the peeling and application equipment.
The customer presents takt time and quality requirements.
These three parties are connected at an early stage.
This can reduce rework such as
“the component was completed, but cannot be used in the equipment,”
or “the equipment was completed, but cannot pick up the component.”
In other words, it is important
to view everything from material selection to automation as a single development.
■ What OTIS Can Do
At OTIS, through
● High-precision die-cutting
● Thin-film processing
● Microfabrication
● TIM processing
● Graphite processing
● Insulating film processing
● Adhesive processing
● Partial adhesion
● Half-cutting
● Carrier design
● Pitch design
● Reel-to-reel processing
● Supply formats for automated application
● Proposals for easily recognizable component shapes
● Image inspection
● Positional accuracy control in long-run processing
● Stabilization of volume production conditions
we can contribute to processing thermal management components into forms that are easy to feed into automated processes.
■ Outside OTIS's Expertise
OTIS is not a company that specializes in
● Design and manufacture of automated application equipment itself
● Robot body design
● Servo control
● Development of image processing algorithms themselves
As needed, we work with equipment manufacturers.
What OTIS should take on is
“automation design on the component side” between materials and equipment.
■ Summary: Components That Cannot Be Automated Cannot Be Used in Mass Production
High thermal conductivity.
Good insulation performance.
Thin.
Light.
These are important.
But in volume production, they alone are not enough.
It must be possible to
peel it,
hold it,
find it,
orient it correctly,
apply it in the correct position,
repeat it at high speed,
and not stop the equipment.
In other words, components in the automation era are required to have, in addition to material performance,
equipment compatibility.
And equipment compatibility is not something to consider after the machine is built.
Materials.
Shape.
Adhesive.
Liner.
Carrier.
Reel.
Equipment.
These must be connected from the design stage.
Automation is not just replacing human hands with machines.
It is also redesigning the component itself so that machines are less likely to fail.
That is why what is required of processing companies from now on is not only
“we can cut it into this shape.”
It is being able to say “With this component, it can be fed directly into the automated process.”
The last barrier for thermal management components moving into volume production is sometimes not thermal conductivity.
It is whether automated machines can handle that component the same way every time.
Only when this is achieved does it become a volume-produced component.
Column Supervisor: Koji Kakumoto (President, OTIS Co., Ltd.)
After studying abroad and working in planning and development at a trading company, he joined OTIS Co., Ltd. in 2011. While primarily working in the Corporate Planning Department, he has also served concurrently in manufacturing and technical divisions, and since 2018 has served as President, working to drive business growth and strengthen the organization.
This article is a general technical explanation intended for educational purposes and does not refer to any specific company, product, or technology.
