Embedding sensors during additive manufacturing

Author: 
Alan Turner - Senior Concept Developer, Lloyd's Register
 
isl-2016-AM-specimens_630

When we talk about the benefits of additive manufacturing (AM), we tend to focus on the ability to produce complex parts in a shorter timeframe, faster time to manufacture, and reduced material usage. One aspect Lloyd’s Register is exploring is the ability to embed sensors within a printed part while the structure is created around it. Imagine getting a data feed from a static asset? Conceivably, we could monitor critical factors along the way, such as in the machine operating conditions, the laser-feedstock interaction zone (known as the “melt pool” for metal-based processes) or the in-service life of the final component.
 
As part of our focus on advanced manufacturing and the internet of things, Lloyd’s Register ran a low-fidelity experiment that embedded fibre Bragg grating (FBG) sensors in the thermoplastic layers during their printing by a MakerBot (Replicator 2). This experiment demonstrated the ability to embed FBG sensors for potential testing of material qualities during the certification process. This objective was pursued with consideration to previous results achieved by Maier et al. Another goal was to assess the viability of embedded sensors in subscale engineering models for testing (in environments such as wave pools, wind tunnels and environmental chambers etc.).
 
FBGs were chosen because they have been extensively studied for use in the monitoring of civil structures (highways, bridges, buildings, dams, etc.), smart manufacturing and non-destructive testing (composites, laminates, etc.), remote sensing (oil wells, power cables, pipelines, space stations, etc.), smart structures (airplane wings, ship hulls, buildings, sports equipment, etc.), as well as traditional strain, pressure and temperature sensing. 
 
A key advantage of FBGs for mechanical sensing is that these devices perform a direct transformation of the sensed parameter to optical wavelength, independent of light levels, connector losses, fibre losses, or other FBGs at different wavelengths. 
 
Two objects were selected to be printed:
  • Cantilever beam – The sensor was to measure static and dynamic response of the beam. Representation of a subscale model.
  • Dog Bone (ASTM D638 Engineering Specimen) – This was meant to represent a configuration for testing the material properties of a part during the product certification process.
isl-2016-AM dog bone specimen_306 isl-2016-AM cantilever beam_306
 
The methodology:
  1. Identify the challenges for prepping and mounting of sensor in fixture prior to printing so as to ensure proper placement of sensor in-situ. This included proper egress and protection at the exit point from the object.
  2. The measurement of strain and temperature during manufacture.
  3. Demonstration of sensor response after manufacture.
isl-2016-AM sensor experiment setup_630
 
For the “dog bone” ASTM specimen test, a single fibre was embedded in order to assess how best to place the fibre during the printing process. A lesson learned was we found that manually placing the fibre was difficult due to the machine structure/layout, and could be improved through more detailed work instructions or automation of the sensor placement. Also, changing the printer resolution may allow for more clearance for the fibre placement and less interaction with the printer head.  
 
Two cantilever beams were also printed and FGB sensors embedded during the printing of the parts. The sensors were read during the printing process to gauge temperature and compression of the fibre. After the parts were printed, the parts were bent slightly to gauge the response of the sensors.
 
Numerous R&D studies have helped establish fibre optic sensor embedment during AM as a feasible concept, which is ready to transition into an industry-driven proof of concept. With the marine and offshore sectors beginning to ramp up their focus and adoption of AM, we would like to work with the industry on some key next steps:
  • Mounting a sensor in a dog bone to demonstrate ability to certify material properties including the Instron tensile test machine. 
  • Embedding a sensor in a subscale model as demonstrated for evaluation of structures in a wave pool, wind tunnel or environmental chambers. 
  • Selection of a “Smart Structure” candidate to enable monitoring of a structure for reliability and\or degradation or to utilize the sensor information as feedback for control of the structure (also known as Structural Health Monitoring). 
The full methodology and the experiment's results are available. To obtain a copy, or to discuss a potential collaboration around sensor embedment and AM, please contact Alan Turner, Senior Concept Developer for Lloyd’s Register, at alan.turner@lr.org or (+1) 832-295-5957.