The Secrets to Designing for Metal 3D Printing Revealed

The Secrets to Designing for Metal 3D Printing Revealed

3Diligent Among Experts Featured in Design for Metal 3D Printing Article

A short while ago, our CEO Cullen Hilkene was interviewed by Ann Thryft of Design News about a topic seemingly on every designer’s mind these days: Metal 3D Printing.

In particular, the intention of the article was to speak with experts across the digital manufacturing industry with expertise in metal additive about how designers need to think differently about the process.  With a clearer understanding of design rules, designers stand not only to have more successful builds, but also can take fuller advantage of the broader design possibilities of metal additive to deliver entirely new and better designs.

The article provides 1) General Design Guidelines, 2) Specific Rules of Thumb, and 3) a List of Differences between metal and polymer AM design.  3Diligent is featured alongside industry leaders such as GE Additive and 3D Systems in the article.

We are pleased to report that the 22-page article was just published, and you can read it here.


Design for Metal 3D Printing Guidelines

Below, we’ve recreated the 10 Rules of Thumb, which drew significantly from 3Diligent’s design for additive manufacturing experience:

Design Rules of Thumb for Metal AM Design

These Top 10 design rules are aimed at powder bed fusion processes for AM— primarily direct metal laser sintering (DMLS) or selective laser sintering (SLS), which are the most commonly used metal
AM technologies.

1. Avoid supports: post-processing hassles.

“When considering laser powder bed 3D printing with metals, the most important thing design engineers need to be aware of is supports. The most dollars lost in 3D printing with metals occurs because of iterative design: too many iterations are needed just to get the supports to work or to get
rid of them completely. In the laser powder bed, the part cools unevenly. So residual stresses are
built into the metal as it hardens, which makes it ‘potato chip’ while it is being built and when you
take the part out.”

“To cope with this, there are some things engineers need to do that drive design. The frst is
designing supports to hold the part down to the metal plate so it doesn’t potato chip. The other
reason for supports is to conduct heat away from the top layers of the part.”

“But now, you have to cut out those supports with a tool when the part is done. This really becomes
a problem when your part has internal passages; you have to get a tool in there to cut them out,
since it’s not like plastic, where they just break away or can be dissolved in water. And you need
enough room to remove them. This is real design-for-manufacturing. You must either design the part
to not need internal supports or design it so you can get a tool in there if needed.” —PADT’s Miller

2. Avoid supports: the 45-degree rule and overhangs.

“Consider the process mentally as you’re designing. With powder bed, thermal stresses are created
by the build, so you may need to use supports. As a general rule, you want to design to avoid
supports. To do that, typically you want to orient the part in the build chamber to avoid overhangs of
more than 45 degrees. Generally speaking, angles more vertical than 45 degrees are self-supporting,
but angles under 45 degrees will need supports. Because these supports are made of the same
metal material as the part, removing them can be time consuming and costly.” —3Diligent’s Hilkene

3. Avoid thin-walled features.

“As a general rule, you don’t want walls less than 500 microns thick and we recommend a minimum
of 1mm. That’s not the thinnest achievable with 3D printing—we’ve done 150 micron thick walls—but
they’re not tall. A 40 to 1 ratio for walls is also a good rule: keep to a 40mm height for a wall
thickness of 1mm. Operate at a ratio bigger than that and you’re putting that feature at risk.” —
3Diligent’s Hilkene

“Design engineers need to know what wall thicknesses work with a manufacturing process. For
example, with additive manufacturing, if it’s a big block of metal, you’ll get a lot of shrinkage. And the
transition from thin to thick can cause shrinkage problems. There are certain minimums for diferent
additive metals processes, for certain manufacturers of 3D printers for each of those processes, and
even for diferent models within a manufacturer. For each manufacturer of laser powder bed 3D
printers, the specifc laser and its settings often determine wall thickness. There are similar issues for
other metal 3D printing technologies.” —PADT’s Miller

4. Stair-stepping has post-processing consequences.

“Engineers must be aware of the fact that because AM with metals is a layered manufacturing
process, the main consequence is stair-stepping. How does this impact design? Engineers must be
aware of the typically rough surface that results, and they must specify the desired surface fnish.
Can you and your customer live with that roughness? If not, you must specify the surface fnish you
can live with. That means knowing the machine’s parameters and how to tweak part parameters if
you’re the operator. Design engineers aren’t taught in school about manufacturing issues, so they
often don’t know this.”—PADT’s Miller

5. Beware additive’s orthotropic planes.

“The other main consequence [of stair-stepping] is that the microstructure is unusual compared to
other metal manufacturing processes. In the X-Y plane, the crystal may be quite large, but it’s also
really thin. In the Z plane, of course, the metal crystallography is diferent and the material properties
are diferent from those in the X-Y plane. In other metal manufacturing processes, the properties are
isotropic in all planes. But with additive metal, they are orthotropic.”
“Stair-stepping is the most visible aspect of the diferences between diferent printer manufacturers’
product lines. It constrains some of the geometry, since the minimum feature size in the Z direction is
layer height. You can’t make features smaller than that height, and the feature has to be positioned at
the beginning of that layer.” —PADT’s Miller

6. Orientation matters.

“Orientation matters: the design engineer should specify part orientation relative to build direction.
Stair-stepping, supports, and not making features thinner than layer thickness—these constraints are
all dependent on part orientation in the machine. For example, if you build a cylinder-shaped object
standing upright in the powder bed, no supports are needed and you won’t get much stair-stepping.
But if you build it on its side, you will need supports and stair-stepping will be more pronounced.
When you design machined metal parts, thinking about setup steps is critical, since you want to
minimize them, and it’s the same for additive.” —PADT’s Miller
“Which direction the part is made in—oriented in either the X-Y plane or in the Z axis—afects the
area of the cross-section and therefore how much material shrink occurs in a given layer. Design
engineers should try to have gradual changes in the cross-section to avoid drastic changes in
material shrink.” —Protolabs’ Utley

7. Design at the system level, not the part level.

“A trap we see lots of people fall into is thinking about part-for-part replacement of the current
product: today I’m stamping a part and tomorrow I’m going to print it. But from a productivity
perspective, it’s important to fnd the opportunity at the product level and not look for only a piecepart
replacement. In the GE Catalyst engine, GE Aviation reduced 855 parts to 12 printed assemblies
using 3D printing and the redesign it made possible. The innovation was not a piece-part
replacement, but in thinking about this product diferently from the beginning—by looking at the
system, at parts consolidation, at soft costs, at optimizing the business case.” —GE Additive’s

8. Combine subassemblies.

“You can combine subassemblies that don’t need to be separate into a single assembly. You can
also make new features like captured features, where elements of a design are built inside or
interlocked with other features. When doing this, you’ll just need to be mindful of how you’ll remove
powder and whether it can be printed without supports.” —3Diligent’s Hilkene

9. Design for post-processing removal of excess material.

“Since 3D printing is good for small quantities, some customers want to give us a design they’re
already machining and see if we can print it more afordably. But you only occasionally see a cost
advantage in doing that, because the parts were designed with a diferent process in mind—
especially if you’ve already invested in tooling to support that process. To unlock the potential of
additive, you should really design with the process [of 3D printing] in mind. One thing it can do is
remove excess material where it’s not contributing meaningfully to the part’s performance. This is
commonly done in applications like aerospace and high-end cars. There is a class of software
referred to as topological optimization or generative design that is focused on this area.” —
3Diligent’s Hilkene

10. Design for post-processing removal of loose powder.

“Loose powder in internal chambers will need to be removed, so you may want to design with
drainage holes for it to escape. You also need to consider clearances between features for removing
lightly sintered powder. For example, there may be lightly sintered powder near a wall feature that’s
even harder to get out of crevices than loose powder.”
—3Diligent’s Hilkene