Allen Face

As discussed in last month’s column, my father Sam and I were outsiders to the slab installation business; concrete floors just happened to be the surfaces we coated and topped with plastic (thin polymer overlays). However, we asked ourselves: what’s missing? Cement masons had been placing and finishing concrete floors for a century, so why hadn’t someone already figured out how to make them flat, no less superflat?

The answer turned out to be a simple case of reverse mater artium necessitas (“necessity is the mother of invention”). Because there had never been a general need for extraordinary concrete floor flatness, no one had been motivated to tackle the problem. To make matters worse, the decades of indifference had allowed a specious but formidable obstacle to become institutionalized: the venerable old straightedge tolerance.

As long as you don’t think about it too much, the straightedge test looks plausible enough. Just lay a straight stick on the slab, and let the gaps beneath it indicate the surface flatness. Of course, it really depends on your definition of flat. Most folks agree in regard to slabs, flatness is a property that gauges the relative severity of the top surface’s waves. For more than two centuries, it has been known that any finite waveform—no matter how jagged—can be decomposed into a collection of simple sine waves. Because every sine wave’s relative severity is determined by two parameters—its amplitude and its wavelength—it follows directly that flatness is a property involving the specification of two characteristics. All straightedge tolerances, however, only look at a single dimension: the maximum allowable gap. But it is impossible to define any two-dimensional object using only one parameter.

Rule No. 15a: All maximum-allowable-gap type straightedge tolerances, no matter what their wording, are technically incompetent.

By only gauging—and very crudely at that—the profile’s local amplitudes while ignoring the wavelengths, the straightedge test can’t determine flatness. Because the straightedge is allowed to freestand, levelness is disregarded completely. The test also suffers from a disquieting anomaly: its results vary according to the side of the profile (top or bottom) to which the straightedge is applied. After nearly a century of use, it is no coincidence the straightedge test has yet to make its way into ASTM—though E6 is now developing such a standard (notwithstanding its inevitable uselessness). In short: the straightedge can’t determine flatness, and it can’t determine levelness. All it can really do is what it’s always done: inanely cause the concrete industry a lot of trouble.

In an effort to improve the flatness and levelness of the slabs placed in new narrow-aisle warehouses and realizing the straightedge test’s deficiencies, it became obvious that the missing element was an effective specification and measurement system. Trying to improve floor quality without a profiler was like trying to do astronomy without a telescope. Because the ultimate goal was to gain control over hard-axle vehicle motions, the idea of looking at the floor from the truck’s perspective suggested itself.

Over the winter of 1976, we built the first self-propelled, electronic, continuous recording profileograph. Engaging a leveled taut wire with a floating column, this machine provided an extremely accurate, continuous paper tape record of the floor’s elevation along each individual wheel track. The separate track profiles were differentiated manually on a light table to yield the image of the surface seen by the truck.

Soon the machine was fitted with transverse and longitudinal accelerometers to measure the side-to-side and front-to-back differences between all the tracks—effectively turning it into a true vehicle simulator. By generating the differential profiles directly, the laborious task of correlating the individual traces was eliminated. Christened the Face Floor Profileograph, this instrument is still used daily to test for F-min.

To learn more about F-min, read “F is for Flatness (or Face).”