#AMneeds: From the Physical to the Digital Asset

2020-06-10

Aya Bentur  

Gas and liquid separator 3D Printed with HP 3D High Reusability PP enabled by BASF

After a part is additively manufactured and before it’s passed on to the next phases of post-processing, there is a certain gap raising the question, how can the physical product be linked to its digital process? How can we ensure the right process is applied throughout the entire production until reaching the right end customer? In this installment of our #AMneeds series, we look at the need to manage the process as a whole, specifically fingerprinting technologies as a way to link the digital process to the physical outcome. Where does the need stem from and how can we address it in additive manufacturing (AM)?

Embedding Information

As AM matures as an industrial production alternative, post-processing methods are becoming more complex, therefore hard to manage without tracking. In traditional manufacturing tracking a part can be done through creating a batch identification number. Since the parts are mass-produced there is no need for individual part identification but rather batch identification. It doesn’t require individual treatment such as imprinting each part separately therefore it can even just be a barcode on a box signifying the specific batch. These codes can hold information from the model and serial numbers to factory information, date, and time of production, up to location, and performance. During the production process, the code can indicate its location in the process, what phases it went through, and what the next phases are until it exits the production line. In any manufacturing method, in any industry, there is a need for part identification, for recall purposes, quality control, tracking expiration dates but most of all in order to track and follow the different phases of production.

Between the Manufacturing Stages

The most important aspect of such tracking is knowing what is the next manufacturing stage, what post-processing steps are needed to finish the part or product. While additive manufacturing allows for a digitized and automated process and post-processing methods have also become digitized, there is a gap left in the in-between, where human error can disturb the entire operation. In medium and large production batches, the challenge is in taking advantage of the AM versatility and producing variations or different parts all together in one production batch. The number of parts produced can make it harder to keep track and maintain an accurate and repeatable production through all stages of manufacturing. For instance, you can have in the same bed one part that requires heat treatment post-processing, another requires coloring, while another can stay as is (below before and after of a post-processed automotive duct 3D printed using HP Nylon 12 – Photo via AMT). SwipeGuide, for example, offers a technological yet manual service, including training and real-time guidance of the worker throughout the process. One of their customers, a Maintenance Manager at a European Beverage Manufacturer stated that “One of the challenges we were facing on the shop floor is low visibility of instructions and standards. There is a clear need for a tool like SwipeGuide to drive shop floor excellence.”

Before and after of a post-processed automotive duct 3D printed using HP Nylon 12 - Photo via AMT

One and Many

The need for tracking in the manufacturing process doesn’t really apply to parts that are unique and expensive, the need comes up with bigger quantities and parts that are in fact identical in their geometry but have unseen differences. For example, the same part printed on different machines. Even if these machines employ the same technology, they can require different handling and post-processing such as using a different curing material for SLA (below NEOS 800 RPS SLA 3D printer). Post-processing can also differ according to the conditions when and where the part is printed, such as humidity, this can change by room or by day. Once the item reaches its final destination, if a problem occurs the identification can help pinpoint when and where the part was produced. This will minimize the items requiring recall as well as identifying the problem itself and quickly implementing the needed solution.

NEOS 800 RPS SLA 3D Printer

Another instance where in-process tracking is required is connecting the item through its location in the print bed to the right box, the right shipment, and the right end customer. The production of AM dental aligners is one example. Dental aligners are manufactured in rather large quantities, at Invisalign, for example, aligners are produced at a rate of  410,000 per day. Multiple units can be printed together in the same bed – these can be aligners for different customers or aligners for the same customer but for different stages of aligning. The location within the bed indicates which belongs to whom, but identifying an item according to print bed location is risky. One option can be using a robot to dismantle the bed, dispersing the items to marked bins, but this also isn’t bulletproof as items can accidentally fall out of the bin.

A Unique Identification of a Unique Part

Using additional technologies after the printing process such as imprinting or using color, changes the part itself. Yet there are other solutions when it comes to AM such as marking the material itself. This can help differentiate between the use of authorized and unauthorized material but it can’t help identify a specific batch. There are a few companies developing solutions specific to additive manufacturing needs such as NEC as well as HP that have a few related patents that were recently published: Data-bearing medium, Locating a region of interest on an object, and Digital signature authentication (up top gas and liquid separator 3D printed with HP 3D High Reusability PP enabled by BASF).

One solution that caught our attention (and we hope it and others like it succeed) is FeaturePrint developed by Alitheon, a machine vision software company that recently raised $14.9 million to develop its technology. FeaturePrint doesn’t require any additional steps in production in order to create identifying features, the part is additively manufactured as is and the technology uses the existing identifying features – the unique surface structure to identify and authenticate the part. “Alitheon’s ability to identify an item using only the object itself—independent of markings, tags or other modifications—is groundbreaking. FeaturePrint technology has the potential to bring a new level of trust to supply chains that does not currently exist”, said Marcus Behrendt, Partner at BMW i Ventures. Alitheon uses off-the-shelf cameras to capture unique surface characteristics, without added technologies, codes, materials, or production costs, just an image of the surface itself holds the metadata (below).

Turbine blade Alitheon featurePrint

There are other advantages brought by such solutions, for example maintaining an audit trail after the part has left the factory and is in the hands of the customer, making sure the right part gets to the right customer and in the case of a malfunction the faulty part can be tracked all the way through the different stages, allowing the company to identify the root of the failure while recalling only one part. But the heart of the matter is within the manufacturing process answering the need for a seamless connection through and through, from the digital file to the physical product. For us at LEO Lane, this is an excellent enabler, which will allow us to provide the same level of enforcement that we currently provide in the AM process and extended it to the post-processing stages. We are excited to see the opportunities within tracking the physical item back to the digital process. Going forward, this could answer many of the needs around AM production from the audit trail to certification.

What are the unmet needs you see in the additive manufacturing ecosystem? Tell us about your #AMneeds. For more insights and information follow us on LinkedIn or subscribe to our newsletter for weekly updates.

Subscribe to Our Newsletter

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>