• High Speed Imaging 
    of LPBF

Author: Jochen Philippi

What do trying to catch a fly, the hunting ability of an eagle and typical process development within LPBF today have in common?

Trying to catch a fly with your bare hand is tricky, for a simple reason. The facet eyes of flies have an almost 360° view and are quite a bit faster than the human eye or our motion. If you try to catch it, the moment you started moving your hand towards it the fly has seen the threat already and typically made its way. By using a fly swatter, the movement of the swat is again faster than the facet eye and …well… squishy fly is the end of the story…

 

Well what does this have to do with AM process development?

AM process development is carried out by people who have excellent knowledge of the processes. The problem is, since our human eyes are not made for process development, we need tools to gather valuable information. The process itself is happening typically at scan speeds between 500-2000 mm/s and with focal diameters of around 100µm where the action within a machine happens. So, too fast and the same time too small to observe of what`s really happening.

How can an eagle find its prey from up to a 1000 m distance? Specialized eyes that are super sensitive and have an integrated “zoom” lens are nice to have and would make us humans capable of reading a newspaper from 100 m distance…

But what if one could combine the positive characteristics of the fly and the eagle and use them to equip the process developer?

Typically a process development starts with experimental exposure lines on substrates to figure out basics of the system. These tests are examined with a magnification glass and require interrupted workflows like preparing a job, waiting for the process conditions being established, firing a few experiments, taking a job out, examining the results, planning the next test series, and so on. At a later stage powder is filled into the machine and the process developer is running series of experiments observing the process with his eyes. Based on his experience of observations he adjusts the process in an iterative way until he is satisfied with the result, goes into laboratory test and loops on and on.

High speed cameras with specialized optics and optimized illumination make both possible. Not only does it lead to an improved understanding of the interaction between laser and material, but it also helps optimizing process variables on the fly with only few vector exposures. This can be accomplished without the time consuming and expensive lab-work. 

Often it is possible to compare variying process parameters on only a few exposed vectors. In order to mak this reality motion capture, with high magnification as well as high frame rates for observing a small section of the available building envelope is required

 

Are these systems integratable in existing PLBF machines?

Typically, these kinds of investigations are executed on specially designed test-rigs or massively modified machines not representative of the environmental conditions within a series system.

With EOS machines you can exchange the process chamber door with an integrated high speed camera, in a working distance of down to 80 mm resulting in a smallest possible picture frame of down to 1,2x 0,8mm. The camera used in this example can create up to 50,000 pictures per second if requested, at a high image detail resolution. The illumination of the scene requires a special additional laser. Having that technology available within the AM system you can gain very interesting insights of the cause and effects of parameter modifications.

Spatter reducing exposure parameters

The following video shows a hatch vector return, whereas with every start of an exposure vector spatters are visible. Such spatter can disturb a homogeneous process, through e.g. absorption/scattering of laser light, affect other parts of a job, lay down on e.g. protective glasses of optical components, or lead to higher material consumption. Avoiding spatter formation during the exposure process can be one of many targets for an optimization. This can be achieved with some fairly sophisticated control mechanisms. While I am unfortunately still not at liberty to share details on how it works, I can provide a glimpse at its capabilities.

In the following video you can see the exposure of different optimizations onto an Inconel Substrate. The left and 1st exposure is carried out using the standard up to date exposure process. The second exposure in the middle ejects spatters only towards the left side of the video. The third exposure to the right has spatter surpression active and shows a very visible reduction of spatters.

AM Users strive for flexible and at the same time reliable process parameters for various applications and materials

  • An optimization of new process parameter (e.g. laser-power, scan-speed or layer thickness,…) is considered standard today
  • Adding the possibility to working with variable focal diameters (https://lnkd.in/gabtW_D) can increase the buildspeed additionally
  • Beamshaping technology (https://lnkd.in/g5ygAzr) opens up an additional complex solutions space of variable focus shapes

But how to develop the ideal balance on this extended field of variables, without many, mostly powder contaminated test samples, preparing endlessly machines and jobs and without external investigations? At EOS we do have the key in our hands.

Please enjoy this last video as much as we do showing an exposure in the heat conduction welding regime, using technologies such as beam shaping at very high powers and layer thickness as an outlook into the not so distant future.

EOS Innovation Management is bringing two beam shaping technologies to powder bed additive manufacturing, opening up new and amazing possibilities. At EOS we strive for perfection in helping our customers with best in class available technology. The EOS Innovation Management dedicated itself to developing the products for tomorrow and the day after tomorrow. The maturity level after the handover to series product development lies typically within a technology readiness level from 3 to 5. This said, the technology is today commercially not available for the wider market but can potentially be made accessible through a cooperation agreement.