TECH COLUMN

9-7. Prototype Conditions and Volume Production Conditions Differ (Part 2) ― Success in Prototyping Does Not Guarantee the Same Results in Volume Production ―

Heat

release:
9-7. Prototype Conditions and Volume Production Conditions Differ (Part 2) ― Success in Prototyping Does Not Guarantee the Same Results in Volume Production ―

In Part 1, we looked at variable factors that did not exist in prototyping.
Material lots. Tolerances. Equipment wear. Speed. Temperature and humidity. Storage. Operators. Differences between machines. Overseas transfer. Transport.
These cannot be reduced to zero.
Even with automation, variation remains.
Then the only option is to design on the premise of variation.
An average temperature of 70°C in prototyping.
In volume production, it may be distributed from 65°C to 85°C.
If the customer specification is 80°C or below, rejects will appear even if the average is good.
99°C in prototyping.
The specification upper limit is 100°C.
It passes on paper, but considering volume production variation, it will soon exceed the limit.
In other words, what is asked in volume production is
not the highest performance of a single unit,
but never falling outside the specification, no matter how many tens of thousands are made.
In Part 2, we look at
the perspective of worst-case values and process capability,
why design margins are not overdesign,
the premise that inspection cannot save everything,
the decision to choose stable conditions rather than optimal conditions,
and the DFM approach of considering manufacturing from the design stage.
We also look at actual failure cases:
a case where full-surface contact was achieved in prototyping, but part lifted in volume production;
a case where peeling defects occurred with the next material lot;
a case where it was good in Japan, but warping increased in overseas volume production;
a case where the component was good at shipment, but could not be applied after arriving at the customer.
Finally, we organize the work of turning prototype performance into volume production performance.

Table of Contents

■ Passing Only the Thermal Evaluation Does Not Mean Volume Production Is Established

■ Initial Performance and Performance over Time Also Differ

■ In Volume Production, the "Worst Case" Matters More Than the "Average"

■ Absorbing Variation with Design Margin

■ Design Margin Is Not Waste

■ The Concept of Process Capability

■ Inspection Cannot Save Everything

■ Be Especially Careful with Defects That Become Invisible in Later Processes

■ Traceability Also Becomes Important in Volume Production

■ Keeping Data Makes Process Changes Visible

■ Volume Production Defects Do Not Necessarily Have a Single Cause

■ Why "Everything Is Within Spec, Yet It Is Defective" Happens

■ The Processing Method Used in Prototyping Sometimes Cannot Be Carried into Volume Production as Is

■ Decide the Volume Production Method at an Early Stage

■ Easy to Process in Prototyping, but the Cost Does Not Work in Volume Production

■ Cutting Processes for Cost Reduction Can Change Performance

■ A Supplier Change Can Also Be a Performance Change

■ Failure Example: Full-Surface Contact in Prototyping, Partial Lifting in Volume Production

■ Failure Example: No Problem with One Material Lot, Peeling Defects with the Next

■ Failure Example: Good in Japan, Increased Warping in Overseas Volume Production

■ Failure Example: Good at Shipment, but Cannot Be Applied After Arriving at the Customer

■ The Researcher's Perspective: Volume Production Is "the Science of Variation"

■ The Shop-Floor Perspective: Line Up Defective and Good Units and Look at the Differences

■ Record the Prototype Conditions

■ Think of Volume Production Conditions as a "Window"

■ Choose Stable Conditions over Optimal Conditions

■ The DFM Approach Becomes Important

■ OTIS's Perspective: Not a Company That Makes Prototypes, but One That Transfers Prototype Performance to Volume Production

■ The Significance of OTIS Getting Involved Early

■ "As per Drawing" Alone Is Not a Volume Production Guarantee

■ What OTIS Can Do

■ Outside OTIS's Expertise

■ Summary: Prototype Success Is Not the Goal, but the Start of Volume Production Design

■ Passing Only the Thermal Evaluation Does Not Mean Volume Production Is Established

For thermal management components, temperature is looked at first.

But volume production products also need

insulation,

waterproofing,

peeling,

durability,

dimensions,

and appearance.

Even if thermal performance is good, other functions may fail due to volume production variation.

In other words, volume production evaluation must look at multiple functions.

■ Initial Performance and Performance over Time Also Differ

Early volume production units are all good.

But six months later,

the adhesive shifts.

The TIM hardens.

The interface lifts.

Then thermal performance changes.

In other words, establishing volume production requires considering not only the state right after manufacturing, but the state after use.

■ In Volume Production, the "Worst Case" Matters More Than the "Average"

In prototyping, the average temperature was 70°C.

Very good.

In volume production, it may be distributed as

65°C,

70°C,

78°C,

85°C.

If the customer specification is 80°C or below, rejects will appear even if the average is good.

In other words, in volume production, what matters more than the highest performance is staying within the specification even under worst-case conditions.

■ Absorbing Variation with Design Margin

For example, the specification upper limit is 100°C.

99°C in prototyping.

“It passed.”

That is the judgment.

But considering volume production variation, it will soon exceed 100°C.

In other words, a design right at the edge of the specification in prototyping is dangerous.

What is needed is margin to absorb volume production variation.

■ Design Margin Is Not Waste

When pursuing higher performance, one wants to push right up to the design limits.

But in volume production, there are

material tolerances,

equipment tolerances,

environmental changes,

and degradation.

That is why a certain safety margin is built in.

This is not overdesign, but design to absorb volume production variation.

■ The Concept of Process Capability

In volume production, it is not only that

the average is within specification,

but also

whether the variation is sufficiently small

that is looked at.

For

positional accuracy,

thickness,

application accuracy,

and so on, it is checked whether the process is stable.

In other words, what is looked at is not whether good units can be made, but whether the process can continue to make good units stably.

■ Inspection Cannot Save Everything

If there is variation, just inspect every unit.

One could think that way.

But some things cannot easily be seen after completion, such as

internal air bubbles,

contact pressure,

and micro-delamination.

Furthermore, adding inspection processes increases cost and time.

That is why what matters is not only inspection, but creating processes where defects are less likely to occur.

■ Be Especially Careful with Defects That Become Invisible in Later Processes

The TIM is applied.

A heat sink is attached on top of it.

After completion, the TIM cannot be seen.

In other words, even if there is an application defect, it is hard to confirm.

In such processes,

the material condition before application,

application position,

and pressing conditions

are controlled.

In other words, process assurance can be more important than finished-product inspection.

■ Traceability Also Becomes Important in Volume Production

A defect occurred.

Which material lot was it?

Which equipment?

On what date was it processed?

Under what conditions?

If these are known, the cause is easier to trace.

In volume production, it is important not only to deliver performance but also to keep records.

■ Keeping Data Makes Process Changes Visible

For example,

application position,

dimensions,

burrs,

and peel force

are recorded continuously.

Then trends become visible, such as

gradually worsening,

changing after a material lot change,

and only a specific machine being different.

In other words, inspection data is not only for pass/fail judgment, but becomes information for improving the process.

■ Volume Production Defects Do Not Necessarily Have a Single Cause

A product with high temperature appeared.

On investigation,

the TIM thickness was slightly thick.

The application position was also slightly off.

The pressing pressure was also slightly low.

Each alone was within specification.

But the three combined to raise the temperature.

Such things happen.

In other words, with volume production defects, it is necessary to look at combinations of small variations.

■ Why "Everything Is Within Spec, Yet It Is Defective" Happens

Each part is within spec.

The equipment is within spec.

The material is within spec.

Even so, the finished product fails.

This may be because the respective tolerances stacked up in the bad direction at the same time.

In other words, in volume production, tolerance design as a finished product is necessary, not just specifications for individual parts.

■ The Processing Method Used in Prototyping Sometimes Cannot Be Carried into Volume Production as Is

Prototyping uses laser processing.

Volume production uses die processing.

Prototyping uses manual application.

Volume production uses automatic reel application.

Prototyping uses manual lamination.

Volume production uses roll lamination.

If the processing method changes,

edges,

pressure,

dimensions,

and material stress

also change.

In other words, if the volume production method differs, it is not necessarily the same component even with the same drawing.

■ Decide the Volume Production Method at an Early Stage

If, after prototyping ends, one thinks

“How shall we make it in volume production?”

rework increases.

If possible, from the prototyping stage, consider:

Will volume production use dies?

Reels?

Automatic application?

Which equipment will be used?

In other words, it is important not to treat prototyping and volume production as separate projects.

■ Easy to Process in Prototyping, but the Cost Does Not Work in Volume Production

For a single prototype, even complex processing can be handled.

But as volume production quantities increase, processing time becomes cost.

For example,

multiple laminations,

detailed manual work,

and many inspections.

If these remain, volume production costs rise.

In other words, establishing volume production includes not only performance, but also takt time and cost.

■ Cutting Processes for Cost Reduction Can Change Performance

After volume production starts,

the adhesive is changed to a cheaper one.

Pressing time is shortened.

Inspection is reduced.

Such cost reductions are carried out.

Then problems that did not exist in the initial design may appear.

In other words, process changes may require re-evaluation including thermal performance.

■ A Supplier Change Can Also Be a Performance Change

Because it is an equivalent product, the material manufacturer is changed.

Same thickness.

Same thermal conductivity.

But

hardness,

surface condition,

peel force,

and tolerances

differ.

Then processability and contact state may change.

In other words, equivalent specifications do not necessarily mean equivalent volume production performance.

■ Failure Example: Full-Surface Contact in Prototyping, Partial Lifting in Volume Production

During prototyping, an engineer applied it by hand.

No air bubbles, full-surface contact.

In volume production, it was changed to automatic application.

Application speed increased, and pressing time also became shorter.

In some products, the TIM did not make full contact, and temperature variation occurred.

The material was the same.

The design was the same.

What differed were the process conditions.

■ Failure Example: No Problem with One Material Lot, Peeling Defects with the Next

Automation evaluation was performed with prototype material.

It peeled stably.

After volume production started, with a different lot, the peel force rose slightly.

Because the margin of the equipment conditions was small, component residue occurred.

In other words, the prototype conditions had not included material variation.

■ Failure Example: Good in Japan, Increased Warping in Overseas Volume Production

There was no problem on the prototyping line in Japan.

It was transferred to an overseas volume production factory.

The local temperature and humidity conditions are different.

The material storage conditions are also different.

As a result, warping of the thin-film laminate increased.

In other words, even though the drawings and materials were the same, the environmental conditions differed.

■ Failure Example: Good at Shipment, but Cannot Be Applied After Arriving at the Customer

Reel products were shipped.

There were no problems in in-factory inspection.

They were exposed to high temperatures during summer transport.

On arrival at the customer, the liner peel force had changed, making it hard to peel on the automatic application machine.

In other words, the volume production conditions included not only the manufacturing process but also logistics.

■ The Researcher's Perspective: Volume Production Is "the Science of Variation"

In prototyping, one looks at

representative values

and the highest performance.

In volume production, one looks at

distribution,

tolerances,

worst-case conditions,

and changes over time.

In other words, volume production technology is not only about raising average performance.

It is about understanding variation and confining it within the specification.

■ The Shop-Floor Perspective: Line Up Defective and Good Units and Look at the Differences

A temperature defect appeared in volume production.

Rather than looking only at the material catalog,

good units

and defective units

are lined up.

TIM compression marks,

position,

thickness,

air bubbles,

material lot,

and equipment history

are compared.

Then what is different may become visible.

In volume production problems, it is important to search for the differences between good and defective units in the actual items.

■ Record the Prototype Conditions

When a prototype succeeds,

recording only the material part number is not enough.

Which material lot was it?

What pressure?

What temperature?

How was it applied?

What thickness?

These are recorded.

That makes it easier to create reproduction conditions during volume production.

■ Think of Volume Production Conditions as a "Window"

For example,

rather than

“it succeeded at a pressure of 100 N,”

it is stronger to know

“it succeeds stably between 80 and 120 N.”

Rather than a single optimal point, check

how far conditions can change while still maintaining performance.

In volume production, the wider this stable range, the easier it is to handle.

■ Choose Stable Conditions over Optimal Conditions

For example,

thermal performance is best at a pressure of 120 N.

But straying slightly outside 110–130 N causes performance to deteriorate sharply.

On the other hand, at 100 N the highest performance is slightly lower, but it is stable from 80 to 120 N.

In volume production, the latter may be better.

In other words, it is also important to prioritize process margin over the highest performance.

■ The DFM Approach Becomes Important

DFM is the approach of designing with ease of manufacturing in mind.

Even for thermal management components, consider from the design stage:

Is this shape easy to die-cut?

Can this narrowness be processed stably?

Can this adhesive composition be peeled automatically?

Will this lamination not warp?

In other words, volume production is not a problem for the manufacturing department to solve at the end, but begins at the design stage.

■ OTIS's Perspective: Not a Company That Makes Prototypes, but One That Transfers Prototype Performance to Volume Production

For OTIS, 9-7 is a very important theme.

Processing prototype components.

This can be done.

But the real value lies beyond that.

The performance obtained in prototyping is

reproduced in the same way

even when material lots change,

even when equipment changes,

even when production speed increases.

In other words, it is turning prototype performance into volume production performance.

■ The Significance of OTIS Getting Involved Early

A processing request is received after the customer has decided everything about the material and shape.

Processing can still be done.

But volume production problems may be found later.

If, from the prototyping stage,

material tolerances,

shape,

application method,

automation,

and reel supply

are considered, rework at the transition to volume production can be reduced.

In other words, the earlier the processing manufacturer gets involved, the easier it is to approach the establishment of volume production.

■ "As per Drawing" Alone Is Not a Volume Production Guarantee

All drawing dimensions pass.

But it is hard to apply.

It warps.

Air bubbles get in.

Automatic machines cannot handle it.

Then it is not sufficient as a volume production component.

In other words, what is needed is not only drawing quality, but

quality that can be used in the process.

■ What OTIS Can Do

At OTIS, through

● High-precision die-cutting

● Microfabrication

● Thin-film processing

● TIM processing

● Graphite processing

● Insulating film processing

● Metal foil processing

● Adhesive processing

● High-precision lamination

● Lamination of dissimilar materials

● Partial lamination

● Half-cutting

● Carrier design

● Reel-to-reel processing

● Supply formats for automatic application

● Image inspection

● Processing condition control

● Traceability

● Process design from prototyping to volume production

we can contribute to transferring functional components established in prototyping into volume production processes.

■ Outside OTIS's Expertise

OTIS is not a company that specializes in

● Design of customers’ entire products

● Developing thermal materials themselves

● Design and manufacture of automation equipment itself

● Dedicated CFD analysis

As needed, we work with material manufacturers, equipment manufacturers and customers’ design departments.

Within this, what OTIS should take on is bringing materials, processing and supply formats down into volume production conditions.

■ Summary: Prototype Success Is Not the Goal, but the Start of Volume Production Design

In prototyping, performance can be confirmed by

using good materials,

applying carefully,

and creating optimal conditions.

But in volume production,

materials vary.

Equipment changes.

Temperature and humidity change.

Speed increases.

Operators change.

Transport conditions are added.

Amid all of this, the same performance must continue to be delivered.

That is why

“it cooled in prototyping” and “it keeps cooling in volume production” are different things.

What is needed in volume production is

not the highest performance of a single unit.

It is never falling outside the specification, no matter how many tens of thousands are made.

And having the process margin to maintain performance even if material lots change or equipment changes somewhat.

Thermal management is truly complete not at the moment the temperature drops in prototype evaluation.

It is when that temperature performance can be reproduced in the same way in volume production.

In other words, thermal management is not complete just by choosing good materials.

Incorporating it in a good state.

Defining that state as a process.

Managing variation.

Reproducing it in volume production.

Only when all of this is done is it established as a product.

And this work of “turning prototype performance into volume production performance” is precisely the major role of processing companies in thermal management.

 

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.

Latest Articles

Related Articles

  • Chapter 9: Failure Cases in Heat Problems ― Why Does Thermal Management Fail?<br>9-1. Failing by Choosing Based on Material Specifications Alone ― Why Doesn’t It Cool Even Though a Material with High Thermal Conductivity Was Chosen? ―

    Chapter 9: Failure Cases in Heat Problems ― Why Does Thermal Management Fail?
    9-1. Failing by Choosing Based on Material Specifications Alone ― Why Doesn’t It Cool Even Though a Material with High Thermal Conductivity Was Chosen? ―

    When choosing thermal management materials, one of... Read More
  • 9-2. Failing by Overlooking Contact Resistance (Part 2) ― Even with High-Performance TIM, Heat Does Not Flow “If It Is Not Touching” ―

    9-2. Failing by Overlooking Contact Resistance (Part 2) ― Even with High-Performance TIM, Heat Does Not Flow “If It Is Not Touching” ―

    In Part 1, we looked at how contact thermal resist... Read More
  • 9-4. Not Fitting into Automated Processes (Part 2) ― Even High-Performance Components Cannot Be Used in Volume Production If Automated Machines Cannot Handle Them ―

    9-4. Not Fitting into Automated Processes (Part 2) ― Even High-Performance Components Cannot Be Used in Volume Production If Automated Machines Cannot Handle Them ―

    In Part 1, we looked at why automated machines bec... Read More

CONTACT

Okayama

+81-867-42-3690

Tokyo

+813-6810-4830

Contact