.png)
.png)
What is Design for Manufacture (DFM) and When Should It Happen?
Rules and Costs
Someone has told you your design needs a DFM review, or a manufacturer has come back with a list of changes, and you are not sure whether they are reasonable.
Most of them probably are. The rules behind them are specific and mostly predictable once you know what they are.
This article covers what design for manufacture involves, the main rules for moulded and fabricated parts, when it should happen, and what a change costs at each stage.
What is Design for Manufacture?
Design for manufacture, usually shortened to DFM, is the process of adapting a product design so it can be produced reliably and economically by a specific manufacturing method.
It covers decisions about:
Wall thickness and how evenly material is distributed
Draft angles that let a part release from a mould
Tolerances, and how tight they genuinely need to be
Material selection for the process as well as the application
The number of separate parts in the product
The order in which those parts go together
Where parting lines, gate marks and ejector pin marks will appear
When done correctly, these decisions are often unnoticeable to the untrained eye, but getting them wrong can cause flaws in the appearance and/or performance, make a product expensive to manufacture, or in some cases impossible.
Why DFM Matters
A product can be designed to look great and function well but can be uneconomical to build.
DFM is what closes that gap. The value of doing it early saves on costly design modifications, or even worse tooling modifications.
A geometry problem found during concept development costs some design time
The same problem found during detailed CAD costs redesign and re-documentation
The same problem found after a tool has been cut costs tooling rework, typically in thousands of dollars.
The same problem found after production has started adds scrap, delay and sometimes recall
A Note on Terminology
Three related terms get used interchangeably and it helps to know the difference when reading a quote.
DFM, design for manufacture, concerns making each individual part producible by the chosen process.
DFA, design for assembly, concerns how easily the parts go together. Reducing part count, designing components that can only be fitted one way, avoiding fasteners that need two hands.
DFMA covers both, and is what most people actually mean when they say DFM.
In practice the two are difficult to separate, because combining two parts into one is both a DFM decision and a DFA decision. This article treats them together.
The Six DFM Decisions That Matter Most
The specifics change by process. The examples below focus on injection moulding, which is the most common process for consumer and light industrial products and the one where the financial consequences are largest.
1. Wall Thickness
What it is. How thick the material is, and how consistent that thickness stays across the part.
The general rule. Uniform walls between 1.5 mm and 4 mm for most engineering polymers. Thick sections should be cored out rather than left solid.
What can go wrong. Thick and thin sections cool at different rates. That causes sink marks on visible surfaces, warping across large panels, and internal stress that shows up as cracking later. Thick sections also increase cycle time, which raises unit cost on every part made.
Cost of fixing it late. Wall thickness is built into the core geometry, so changing it after tooling usually means significant re-working or a new tool.
2. Draft Angles
What it is. A slight taper applied to vertical faces so the part releases cleanly when the mould opens.
The general rule. One to two degrees on smooth surfaces. Three to five degrees or more on textured surfaces, with deeper textures needing more.
What goes wrong without it. Parts stick in the tool, drag marks appear on visible faces, and ejector pins push through or distort the part. Insufficient draft is one of the most common reasons a first-off sample is rejected.
Cost of fixing it late. Adding draft changes the visible geometry, so it usually requires both a design change and tool modification. This is also the change designers resist most, because it alters how the product looks from certain angles. Designing the form around the draft from the start avoids the argument entirely.
3. Tolerances
What it is. How much variation is permitted on a dimension.
The general rule. The loosest tolerance that still allows the product to function and assemble correctly.
What can go wrong. Tight tolerances require better tooling, slower cycle times and more frequent tool maintenance, and they produce more rejected parts. Specifying tighter tolerances than the application needs is easy to do and expensive to live with, because the cost repeats on every part.
Cost of fixing it late. Loosening a tolerance can save money. Tightening one after tooling can require a new tool.
What is Design for Manufacture (DFM) and When Should It Happen?
4. Part Count
What it is. How many separate components make up the product.
The general rule. Fewer is better, provided the combined part can still be moulded. Every part removed takes a tool, an assembly step and a source of error with it.
What can go wrong. Each additional component adds tooling cost, assembly labour, inventory, and something that can be fitted incorrectly. The costs can compound across the life of the product rather than just at initial production.
Cost of fixing it late. Combining parts after tooling means scrapping tools. This is one of the clearest reasons to review part count during concept development.
5. Part Lines and Gate Location
What it is. The part line is where the two halves of the mould meet. The gate is where molten plastic enters the cavity. Both leave a visible mark.
The general rule. Place them where they matter least, on a hidden surface, along an existing edge, or on a face the user does not touch. Decide this during design rather than leaving it to the toolmaker.
What can go wrong. A part line across a highly visible curved surface, or a gate mark in the middle of a display face. Both can be unsightly and/or make a surface rough to touch.
Cost of fixing it late. Moving a gate or parting line after tooling generally means significant tool rework.
6. Ribs, Bosses and Undercuts
What it is. Ribs stiffen a part without adding wall thickness. Bosses accept screws and fixings. Undercuts are features that prevent the part from releasing straight out of the tool.
The general rule. Ribs at roughly 50 to 60 per cent of the adjoining wall thickness, to add stiffness without causing sink marks on the opposite face. Bosses supported by ribs rather than thickened walls. Undercuts avoided where possible, since each one needs a side action or lifter in the tool.
What can go wrong. Overly-thick ribs cause visible sink marks. Unsupported bosses crack when a screw is tightened. Each undercut adds mechanism to the tool, which adds cost, cycle time and something that can fail in production.
Cost of fixing it late. Undercuts are the expensive one. Designing them out during concept development can remove several thousand dollars from tooling cost.
What Goes Wrong When DFM Happens Late
The problems that surface after a design is considered finished are often more technical than cosmetic.
Geometry requiring undercuts the tool cannot produce without adding sliders
Wall sections that will warp regardless of how the tool is gated
Assembly sequences that look sensible in CAD and take a factory worker four times as long as planned
Fixings that cannot be reached with a normal driver once the product is assembled
Tolerances stacked across an assembly so that parts at the extremes of tolerance do not fit
These are not adjustments. Resolving them often means rebuilding sections of the CAD model, and by that point the design has usually been approved by people who now need to approve it again.
The Gap Between Looking Right and Being Manufacturable
A design developed with only form in mind produces geometry that looks great in renders. The render has no draft, no parting line, and no gate mark. It looks exactly like the product that was imagined.
A design developed with the process in mind looks slightly different from the beginning. The geometry carries its draft. Parting lines sit where they matter least. Surface transitions are shaped to work with the tooling.
The second is not a compromised version of the first. It is a different starting point, and it is the one that reaches production without being redesigned along the way.
How DFM Works at Concept Evolution
Design for manufacture is not a separate stage at Concept Evolution. It runs alongside design from the point the manufacturing method is confirmed.
Our industrial designers work closely with a range of manufacturing partners, which means manufacturing questions get raised during concept development rather than after a design direction has been approved. Our workshop, including in-house 3D printing, lets us build and check parts as the design develops.
For specific processes, we work with our network of manufacturing partners including metal fabrication, injection moulding, thermoforming and extrusion. Their input on tooling and process constraints feeds back into the design while it is still easy to act on.
This also applies to work on existing products. Evaluating a part that is already in production, to find why it is causing problems in manufacture, assembly or installation, is a common piece of work and often produces the significant cost savings.
The Concept to Production Pathway
At Concept Evolution, projects typically move through four general stages. Each one resolves details or issues in the design before the next stage begins.
1. Design Review
We start by reviewing your product idea from several perspectives, to refine the design brief and guide concept development.
The questions we work through include:
What is the need for this product?
Who will use it, and how?
What is the market opportunity?
What materials and processes could be used to make it?
What technical or regulatory requirements apply?
The question about materials and processes is where DFM begins. Confirming the manufacturing method early is what allows the design to be shaped around it.
2. Concept Development
We take the refined brief and explore multiple design directions in both form and function, using:
Concept sketches
3D CAD modelling
Detailed 3D product visualisations
We work closely with you to refine the design and prepare it for physical development and evaluation.
3. Prototype and Test
Prototypes range from basic form studies, which assess size and shape, through to high-fidelity working prototypes.
Our workshop is equipped to assemble and test prototypes, including in-house 3D printing for rapid prototyping, which makes evaluating a design fast and economical. For specific materials, processes or treatments, we use our network of manufacturing partners.
Some products advance to production after one prototype. Others require revision and further rounds of prototyping, testing and refining.
4. Production
When you approve the final prototype, the product is manufactured and goes to market.
It is optimised for efficient mass production through design for manufacturing, and to comply with the relevant regulations and standards.
We can manage production for you through our network of trusted manufacturing partners in Australia and overseas, or work with your own manufacturing network.
DFM and Manufacturing in Australia
Sydney has a good base of toolmaking, moulding, machining and fabrication capability. Having tooling made locally generally means faster correction cycles compared to offshore production, which can expedite the time to market for a new product.
Concept Evolution is based at Taren Point in southern Sydney, with studio and workshop space on the same site. We work with clients across Australia and internationally, in three broad groups:
Manufacturers extending into design without building an in-house design team
Brands and businesses improving or extending an existing product line
Startups and inventors developing a product for the first time
Key Takeaways
Design for manufacture is adapting a design so a specific process can produce it reliably and economically.
For injection moulding, aim for uniform walls of 1.5 mm to 4 mm, one to two degrees of draft on smooth faces, and more on textured faces.
Ribs at roughly 50 to 60 per cent of the adjoining wall thickness avoid sink marks.
Specify the loosest tolerance that still works. Tighter-than-required tolerances can add unnecessary cost.
Reducing part count removes tooling, assembly labour and failure points.
Tooling corrections typically cost $5,000 to $30,000 each, which is why it is important to conduct DFM analysis prior to tooling production.
Every process has its own rules. Sheet metal, machining, casting, thermoforming and extrusion.
Frequently Asked Questions
What does DFM mean in product design?
DFM stands for design for manufacture. It means designing a product with the requirements of the chosen manufacturing process in mind, so the parts can be produced reliably and economically at scale. For injection moulded products this covers wall thickness, draft angles, parting lines, gate location, tolerances and part count. These decisions should be addressed in the design phase rather than being added after the design is approved.
When should design for manufacture happen?
As soon as the manufacturing method is confirmed, part design should be developed with DFM in mind. Waiting until detailed CAD is complete means decisions have already been locked in that DFM would change. A change at concept stage costs design time. The same change after tooling costs tooling rework.
Does DFM only apply to injection moulding?
No, although moulding is where the financial consequences are largest because of tooling cost. Sheet metal fabrication has minimum bend radii, bend relief and grain direction requirements. CNC machining is limited by tool access and internal corner radii. Die casting, thermoforming and extrusion each have their own constraints. The principle is the same in every case.
How much does DFM add to the design cost?
It shifts cost earlier rather than adding it. A DFM review during design typically adds two to five days of engineering time. Finding the same problems after tooling costs ten to fifty times that in rework alone, before accounting for the delay.
My product looks right in the CAD renders. Does it still need DFM?
Yes. Renders usually do not show draft angles, do not simulate how material cools and shrinks, and do not flag parting line or gate mark problems. A part can look great in a render and still carry several issues that will require tooling corrections after the first sample. A render is a communication tool, not a manufacturing check.
What is a good wall thickness for an injection moulded part?
For most engineering polymers, a uniform wall between 1.5 mm and 4 mm. Consistency matters more than the exact figure, because uneven sections cool at different rates and cause sink marks, warping and internal stress. Thick sections are usually cored out rather than left solid.
What draft angle does an injection moulded part need?
One to two degrees on smooth vertical faces is typical. Textured surfaces need more, commonly three to five degrees, increasing with texture depth. Without adequate draft, parts stick in the tool and show drag marks on visible surfaces.
Can DFM be done on a product that is already in production?
Yes, and it is often worthwhile. Reviewing an existing part to find why it causes problems in manufacture, assembly or installation frequently produces significant cost savings, particularly where part count can be reduced or a tolerance loosened. The constraint is that changes to an existing tool cost more than changes to a design, so the review usually focuses on what can be improved without scrapping tooling.
Final Thoughts
Most DFM problems are not difficult engineering. Wall thickness, draft and tolerance are well understood, and the rules have been established for decades.
The problems can be expensive when they are discovered late, due to potential design changes and re-working of tooling.
If you have a design in progress, or a part already in production that is causing trouble at the factory, send it through. A review usually takes a few days and may highlight some changes that can save significant costs.
Talk to the Concept Evolution team or call +61 2 9521 5502.
Concept Evolution is an industrial design, engineering and manufacturing company at Taren Point in Sydney. Over 20 years, more than 200 projects delivered, and over 70 clients across Australia and internationally. Our services span new product development, design and engineering, and prototyping and manufacturing. Related reading: what a product prototype is, 3D printing versus traditional prototyping, and how long product development takes in Australia.