In Search of the Plane Truth

Measuring flatness has always been an essential part of engineering and manufacturing processes, and with the advancement in technology, various methods have been developed to achieve accurate measurements. These methods include optical flatness measurement, profilometry, interferometry, and mechanical flatness measurement.

Optical flatness measurement is achieved using interferometry, while profilometry uses a stylus to scan the surface and provide a 3D representation of the surface topography. Mechanical flatness measurement involves using a straight edge or a set of gauge blocks to measure the flatness of the surface.

In the aerospace industry, flatness is crucial in the design and manufacture of aircraft components such as wings, fuselages, and engine parts. The flatness of these components can impact the aerodynamics of the aircraft and, therefore, its performance. The automotive and semiconductor industries also have strict flatness requirements in their manufacturing processes to ensure that the final products function as intended.

When extreme precision is required, the tools needed to make those micrometer-level measurements start to get quite expensive. For industry, tools that cost several thousand dollars might not be too big of a concern — but if a weekend hacker wants one of these professional-grade tools, that sort of price tag is generally a bit too much to stomach.

A Toronto-based engineer by the name of Bryan Howard wanted a highly accurate tool to measure flatness, but did not want to spring for a high-priced piece of equipment. Instead, he decided to come up with a plan to build his own, using inexpensive components that he already had on hand.

With modern image sensors packing high pixel counts and low price tags, Howard saw the humble webcam as a perfect sensing device to use in the design of his device. After removing the lens from in front of the webcam’s image sensor to eliminate distortion, a horizontal line laser, from an off-the-shelf laser level, was shined directly at the camera from a distance.

With the laser beam staying in the same position, an assembly containing the camera is moved across the surface of the object to measure for flatness in a series of steps. The level at which the horizontal laser beam intersects with the image sensor’s pixels can then be measured at each of these steps.

Howard modified and improved an existing software tool that calculates the exact position of the laser beam in the image frame and tracks the change in position over time. With a series of measurements, it is able to tell the user exactly how flat a surface is, or is not, to within about 0.5 to 2 micrometers.

The system is extremely sensitive — walking around, or even standing up, while the measurements are being collected can cause a wobble that leads to inaccuracies. Howard also notes that the system should be turned on for at least five minutes before use to bring the image sensor up to full operating temperature. Before that time, there will be some amount of drift.

This project is by no means a plug and play solution, but with the instructions provided, anyone that has a need for such a device is likely to also have the skills needed to be able to put one together. And while there are a few caveats that you have to be aware of to get accurate results, taking a seat while the device is in operation is well worth saving a few thousand dollars.