A Chilling Experience

Students Tunnel Their Way To Revised Wind Chill Table
 

Lowering their wind tunnel into a simulated Antarctic environment- a conventional freezer - are Candace Gildner and Matther Grubar

Gildner, Grabski, & Gruber. Although these names hold promise for an onomatopoeically-inclined barbershop trio, they actually belong to three mechanical engineering seniors from the University of Rochester, N.Y., who have a better chance of being remembered for their compelling research revising the currently accepted wind chill table.

Current, however, may not be the best word to describe a table based on a single experiment conducted 60 years ago in less-than-ideal conditions. That is why Candace Gildner, Dan Grabski, and Matthew Gruber decided to revisit this old chart during their Solids and Materials Lab last fall. Not coincidentally, they happen to live in a place where windchill matters. As Grabski puts it, “We have more than enough experience in Rochester with cold weather and a lot of wind.”

What they found is that the wind chill values—which provide an alternative temperature reading based on the cooling effects of wind—are largely overblown, often by more than 10 degrees Fahrenheit. That’s a significant difference, but one consistent with recent theories on the subject, and with subjective observations that it frequently doesn’t feel as cold as the wind chill table indicates.

A Tiny Tunnel

The team came to its results through a combination of ingenuity, experimentation, and nail-biting statistical analysis.

Using a miniature wind tunnel of their own creation allowed the students to gather data in a more controlled environment than the original experiment, which had been conducted in the 1940s with containers of water on the rooftop of a scientific research building in the Antarctic. Specifically, their wind tunnel eliminated two of the initial experiment’s flaws: fluctuations in outdoor wind speed and wind speed measurements taken several meters above the water samples.

According to Gildner, however, building the wind tunnel wasn’t the hard part; finding a steady source of cold air was. Dismissing their initial ideas of bringing cold air to the wind tunnel via liquid nitrogen or compressed air, they decided instead to bring their wind tunnel to the cold air. “We racked our brains, and the best thing we came up with was a freezer,” says Gildner.

Home-Grown Lab

Shaping a three-foot-wide and one-foot-long oval out of ventilation ductwork, the team assembled a wind tunnel small enough to fit inside a conventional freezer unit. Complete with variable speed fan and a chamber for water samples, their home-grown Antarctic lab could produce temperatures from 25 to -10 degrees Fahrenheit and wind speeds ranging from 0 to 15 mph.

They began their experiments by cooling the entire contraption to a desired temperature, selecting a wind speed, and inserting a soda can filled with hot water into the tunnel. As soon as the water temperature cooled to body temperature, two thermocouples—one measuring the wind tunnel’s ambient air temperature, another measuring water temperature—were used to track the rate of heat loss from the water sample, giving the students enough data to plot a curve of the cooling trend.

Water was used because it is a convenient medium to work with, and because its thermal properties resemble those of skin. The metal soda can, with its negligible thermal gradient, eliminated another potential variable. The original researchers, says Gildner, ignored the effects of their thermal gradients, which, in all likelihood, were not negligible.
 
A Breezy Concept

Once the team had collected data, their work began in earnest. As one might imagine, the process of turning raw data into useful wind chill figures is complicated by several factors.

ME students at the University of Rochester derived a new windchill table (top) after conducting experiments inside a small-scale wind tunnel. Their results indicate that the commonly accepted table (below), based on a 1940s experiment in the Antarctic, overestimates the cooling effects of wind on the body.

First, although the rates at which skin and water lose heat are similar, they are not identical, and, even when talking skin, differences in blood flow and surface area cause various parts of the body to lose heat at different rates. Furthermore, unlike water which forms a thin layer of ice in cold temperatures, the human body’s circulatory system continually replaces heat lost from the skin, yet the surface temperature of skin does drop somewhat when exposed to cold. Both factors influence the rate of heat loss.

“Wind chill is a number of importance to people working outside, but it’s also a number that is hard to get a good scientific meaning for,” says course professor Paul Funkenbusch. “What do you mean by wind chill? Do you mean the wind chill on your head, or the wind chill on your fingers, or on your arm?” The best one can do, he says, is “define what you mean and hope that you’re picking something that is practical.”

That’s exactly what the team did, researching the ways other scientists have derived wind chill numbers, before settling on their own approach. During their research they even came across an obscure alternative to wind chill—watts-per-meter2 of heat lost to wind—which Canada tried to implement some years ago. “It was more accurate,” says Grabski, “but nobody understood what watts-per-meter2 was.” Since the public had a hard time making sense of the figure, it was doomed, and Canada reverted to the wind chill.

Theoretical Quagmire

The Rochester team eventually decided to navigate this theoretical quagmire by focusing their attention on the body’s thermal resistance, or its initial rate of heat loss—an approach that eliminates the variable of changing skin temperature. “We were only concerned with the rate of loss the instant you walk out the door,” explains Gildner.

To obtain this measurement, they relied on a decay-time constant, which they calculated by plotting the cooling data for a given wind speed and ambient temperature, then matching the exponential curve using a best-fit line analysis on Excel.

“When we performed the data analysis, all of our work began to bear fruit,” says Gruber. “We knew that our results were in the general region we had hoped, so we knew our experiment would be a success.”

Although, on average, the Rochester team’s results show the old wind chill table to be overvalued (showing colder wind chill temperatures), in some cases by as much as 12 degrees Fahrenheit, the relationship between the two charts is not strictly linear. For instance, those unlucky enough to find themselves in a -10 degrees Fahrenheit environment with 15 mph winds, will find that the Rochester chart shows a wind chill temperature of -50 degrees Fahrenheit, which is five degrees below the old value of -45.

As excited as the team is about their work, publishing it would require extending their testing range to colder temperatures and faster wind speeds, a difficult task considering they have already maxed out the capabilities of their fan and freezer.

Yet even if the team finds time to overcome these hurdles, they hold little hope for the nation’s quick acceptance of an updated table. Gruber draws a parallel with the English measuring system.

“Although the metric system is more practical and easier to use, convincing people to use it has proven very difficult. [If we continue our research], I hope our work would eventually lead to a change in the accepted standard, but I am sure that the adoption of this new table would probably meet significant resistance.”

Would that be thermal or cultural?

Back To Table of Contents