When training Lean Six Sigma Black Belts, I often emphasize that figuring out how to measure something is often the hardest but most important breakthrough one can achieve. This article from The Economist is focused on rheology, the study and measurement of material flow.
If you have ever given a bottle of tomato ketchup a good shake to make it pour more easily, then you have experimented with rheology. This is the study of how materials flow, and it considers the many elements which give a liquid its overall viscosity. Shaking ketchup invokes one of those elements: shear thinning. This reduces a liquid’s reluctance to flow, or viscosity, by forcing the layers within it to separate, suddenly making it runny. If you want to make the perfect ketchup, therefore, rheology is important. Measuring what is going on in a ketchup factory can, however, be hard.
As a result, manufacturing difficulties may not be detected quickly. But Julia Rees, Will Zimmerman and Hemaka Bandulasena of the University of Sheffield in England, are riding to the rescue. They have invented a small, cheap device which—when combined with some clever mathematics—can measure viscosity-changing phenomena such as shearing. Moreover, it can do so on the fly, rather than requiring samples to be taken off to a laboratory. If the sauce coming out of a factory does not have the requisite gloopiness, that can then be detected immediately.
The new rheometer—described in a recent edition of Measurement Science and Technology—contains a channel through which some of the ketchup (or any other material of interest) passes. This channel has a corner in it, and when the fluid turns this corner its velocity, pressure and shearing rate all change in ways that give away its rheological secrets. The changes in question are mapped using a technique called micron-resolution particle-image velocimetry, which seeds the fluid with tiny particles and follows their progress with a digital camera. All of which is not too hard, using modern technology.
The trick is how you interpret what you see. What Dr Rees, Dr Zimmerman and Dr Bandulasena (or, rather, their computers) do is to take the equations that describe rheology, feed in the observed behaviour and then crunch through the possibilities to determine what combination of rheological properties would actually result in that behaviour. This, in turn, reveals the probable physical characteristics of the fluid, including those that are not immediately obvious—and thus whether it is being turned out according to spec.
It seems to work. So far, results from experiments using test liquids such as ketchup and mayonnaise have correlated well with those obtained using conventional equipment in the university’s chemistry department. Moreover, if the new device were deployed in a real factory, it would always be dealing with the same substance. There would then be no need for a general-purpose algorithm to do the calculation, and the task of working out what was going on could instead be handled by a simple processor built into the rheometer itself. Nor is the technique restricted to industrial applications.
One rheometer made by the three researchers had a test channel the width of a human hair. This has attracted interest from biologists who study fluids such as blood and lymph. The researchers are now looking for industrial partners to carry out further tests. What their invention is not yet able to offer, though, is a way to get the last bit of ketchup out of the bottle.