CPU company wants to automate 3D printing

Like many other leading tech firms, CPU manufacturer Intel regularly ventures outside its comfort zone in pursuit of new revenue opportunities. The chip maker maintains a dedicated investment fund for backing startups in promising markets, while its engineering division usually runs at least several projects that fall under Google’s definition of a moonshot. One such initiative appears to be aimed at making 3D printing more competitive against traditional manufacturing methods.

Intel is approaching the challenge by trying to reduce the amount of manual work involved in the fabrication process. According to a recent patent filing, the company is placing the emphasis on automating printer calibration, which can be a major challenge due to the numerous factors that need to be taken into account. Operators must  adjust the configuration settings based not only on the specifications of a given project but also environmental conditions such as room temperature, the pressure inside the printing chapter and sometimes even brightness.

The result is that the setting up a printer often represents a disproportionately large part of the work. This problem is exacerbated by the fact that it’s often difficult to address every operational parameter the first time around, which leads to fabrication errors. Operators then have to sink yet more time into troubleshooting the issue before they can try again. The automation system outlined in Intel’s patent application aims to provide a solution using a novel combination of analytics software and depth-sensing cameras.

According to the filing, the cameras are there to provide visibility into what’s happening inside a printer while objects are undergoing fabrication. Intel’s system uses the video feed as part of an elaborate testing procedure that involves creating a concept model to check if everything is in order. The chip maker specifies that the sample item can be “of any shape and design”, which should enable users to match its physical characteristics to their project requirements. Once everything is in place and the printer starts working, the software monitors the fabrication process for deviations from the blueprint.

A positive hit can trigger one of two reactions. If an error has a limited impact on the sample item, the system will automatically adjust printing settings to compensate for the issue and proceed to complete the test run. Users can then move on with their projects knowing that everything has been properly calibrated according to their requirements. A more complex problem, on the other hand, will lead Intel’s system to halt fabrication and produce a diagnostics report with the information necessary to perform troubleshooting.

The software can identify not only environmental factors that are interfering with printing but also wear and tear, which is usually much difficult to anticipate before a project. Intel doesn’t specify how accurately its system will be able to diagnose such issues, but its engineers should be able to cover a lot of ground with the right data. The company could collect diagnostics information from real-world deployments to assemble a database of the errors that affect different kinds of printers and projects.

If it ever leaves the drawing board, the system could score Intel a lot of points with the maker community while advancing 3D printing as a whole. Reducing the amount of manual work involved in the fabrication process is one of the most important steps that the industry will need to take if the technology is to become viable for large-scale commercial use.