Spot the difference

As additive parts are rolled out across industry, they’re increasingly put in to actual service. This is especially true for metal parts, where processes such as selective laser sintering (SLM) have usurped more traditional subtractive fabrication and machining methods.

This is still small volume stuff, of course, and it is primarily limited to aerospace. Prototyping is still the mainstay of 3D printing, more widely, but it is increasingly being explored for things such as tooling as well as serviceable parts, not just models for marketing departments. And here, it is not limited to a few unmanned satellites, Boeing and Airbus are both putting printed parts on production aircraft, today.

The Airbus A350 XWB aircraft has a variety of plastic and metal brackets, with material and structural properties that have been well tested and validated. However, using additive processes and materials is not straightforward.

Firstly, there is the design challenge. Design tools like topological optimisation are still not largely available. And even when they are, usability is somewhat specialist. Mainstream CAD packages are yet to really offer AM specific features and functions to make a design optimised for printing. But much of this is not the fault of CAD firms, the fact is, design rules for AM are not fully established and everyone is still learning.

Then there is the challenge of manufacture, of course. Raw materials such as titanium alloy powders, do not always have the same rigid controls as bulk billets of metals. Traceability can be an issue, particularly if a part uses powder from different suppliers, but to the same specification. The trouble is, there can be deviation.

And then there is the printing process itself. Two machines from the same manufacturer will likely produce slightly different results of the same part. There are so many variables involved from laser power, deposition speed, layer thickness. Setting up machines to produce the same exact parts goes well beyond being a skill. It’s an art. There is always likely to be a slight change to porosity, the internal structure of the metal, or surface finish. It all comes back to the age old issue of quality assurance.

“Although you might have the material, it is the robustness of the process that’s essential,” says Kevin Cummings, chief executive of GKN Aerospace. “We have parts flying on the secondary structure of aircraft now. And while everyone wants to make sure we don’t move too quickly, there are already primary structures being tested and qualified.

“3D printing is still in its infancy. It is where composites were in the 70s. So there is a long way to go. It will be slower than we want, but in the end it will probably be more comprehensive than we can possibly imagine.”

GKN has gained much experience working with materials that have so many variables to control right the way through production – notably the vast amount of work it has done with composites. Despite its 3D printing effort being focussed around titanium, much like a composite, properties are created as the part is made.

“You have a series of ingredients in a recipe, and changing any of those ingredients – or how long or how hot the oven is – will change the output substantially,” says Ian Chatting, vice president of technology at GKN Aerospace.

“We are building up the knowledge base so you have to look at the properties of the materials coming in, so take titanium powder for example, you have to validate it at every stage of the production process.”

The result is that there is a drawn out process required for validation, and flight certification. Unlike bulk materials, or series production, testing material samples does not offer the same safe guards. Instead GKN has had to build up its database of processes and powders to show statistical proof that a process is valid.

“We have made thousands of test coupons and these are static tested and fatigue tested in a very traditional way,” adds Chatting. “The difference is there are so many more variables, hundreds. So what we have to do is narrow those down and understand them.

“It is a challenge in the aerospace industry. Anyone can 3D print parts, you can buy a machine and start doing it yourself. But producing qualified parts for flight, that’s the challenge.”

The database that GKN has built up shows that the properties of the titanium have largely exceeded initial expectations. They are reported to be very favourable in many instances to forgings, and exceed castings in almost every instance.

Testing Fatigue

There is substantial interest in how to test and validate additively manufactured parts, not just for flight, but perhaps for the track or road.

A seminar organised by test and analysis manufacturer Shimadzu at the Warwick Manufacturing Group (WMG) set up to specifically address the question of fatigue testing of 3D printed parts. With a focus on aluminium and titanium alloys additively manufactured parts, the event drew experts from a number of leading UK engineering firms including Augusta Westland, Jaguar Land Rover, Red Bull Racing, Rolls Royce, and GKN Aerospace.

One of the findings according to Shafaqat Siddique, a researcher at The Department of Materials Test Engineering at TU Dortmund University, is that fatigue strength of a SLM was less than a bulk material due to the inherent surface roughness of a finished part. However, this can be improved during post processing using a number of different techniques.

Another interesting finding is that parts made by SLM will fail below its fatigue limit. It means that stress applied below the commonly agreed fatigue limit of a material can still lead to failure.

Using very high cycle fatigue (VHCF) testing techniques has led to Siddique claiming that these material do not in fact have a fatigue limit at all.

He has tested samples made from a variety of alloys, some of which show a peculiar change to how and where cracks begin to propagate.

Between high cycle fatigue (104) to VHCF (109) cracks start to appear in the subsurface of a part, as opposed to on the surface.

The test samples were produced using a commercially available SLM material and were made in AlSi12 alloy with the VHCF tests carried out on a Shimadzu USF-2000 ultrasonic fatigue test system at 20kHz.

Here, the vibrations are transmitted through the solid body of the test part so the longitudinal waves resonate.

The specimens were clamped only at one end so the maximum stress occurs in the middle of the part, and maximum displacement therefore at the free end. To eliminate temperature effects the specimens were cooled with pulse-pause bursts of compressed air.

“The experiments were performed using the staircase method,” said Siddique. “If a specimen at ultrasonic frequency failed at less than 109, the stress amplitude is decreased by 5MPa for the next experiment. If the specimen did not fail, the stress amplitude is increased by 5MPa.”

Experiments showed that fatigue fracture occurs beyond the high cycle fatigue region in both batches.

“New developments in testing machines enables going beyond the previously known limits of knowledge,” concludes Siddique. “This opens the door for more intensive testing in the most realistic conditions.”

Author
Justin Cunningham

This material is protected by MA Business copyright
See Terms and Conditions.
One-off usage is permitted but bulk copying is not.
For multiple copies contact the sales team.

 

Supporting Information
Do you have any comments about this article?
Name
 
Email
 
Comments
 

Your comments/feedback may be edited prior to publishing. Not all entries will be published.
Please view our Terms and Conditions before leaving a comment.

© MA Business Ltd (a Mark Allen Group Company) 2018