To resolve design issues, it may be necessary to question the root source of some fundamental accepted practices and assumptions.

From an early age, we all are taught some basic truisms. For example, we learn that the “normal” typical oral human-body temperature is 98.6⁰F (conveniently, that’s 37.0⁰C). Many decisions, such as concluding that someone has a fever and therefore should get some medical treatment, are based on this number (Figure 1).

We know that body temperature may differ among individuals and that exercise and circadian rhythms affect the reading, but that 98.6 number is the established marker. But what if it is not the right number at all? Think of all the implications this would have.

Why even think there’s a discrepancy? I recently came across the article “Decreasing human body temperature in the United States since the Industrial Revolution” in a reputable medical journal,  written by a team at Stanford University Medical School (don’t ask how I came across this article!). They looked at the history and source of the “98.6” data and raised some serious concerns about the number’s validity.

First, the data is from a large group of mostly men, and much (but not all) of it was taken in the 19^th century. Although you might assume the law of large numbers would average out sample errors, the circumstances under which the temperatures were taken and the inaccuracy of those thermometers are major issues (calibration was crude by our standards), so that benign averaging may not be the situation.

Second, the authors point out the data was collected “when life expectancy was 38 years, and untreated chronic infections such as tuberculosis, syphilis, and periodontitis afflicted large proportions of the population. These infectious diseases and other causes of chronic inflammation may well have influenced the ‘normal’ body temperature of that era.” In other words, nearly everyone back then lived continuously with low-grade infections that would have raised their temperature somewhat.

The authors concluded that the actual “typical” body temperature in our times is about a degree F lower. This difference can result in misleading health conclusions.

At first, I was shocked. It was like finding out that, whoops, it’s actually V = I × R^1.1 or E = mc^2.2. Then I realized that body temperature is not defined analytically by the laws of physics; instead, it is the result of accumulated “personal” data and subsequent analysis. As such, it can be a guideline but not an absolute and therefore should probably be applied with at least some caution.

Relation to engineering design

This caution applies to the inexactness of much design work as well. For example, many ground rules, rules of thumb, guidelines, and best practices are associated with attenuating the effects or generation of EMI/RFI.

These include where to terminate a shield, use of ground planes, placement of bypass capacitors, and establishing a single-point ground for analog and digital current-return paths, to cite a few. Still, there are many documented cases where these rules did not work or were counterproductive. Figuring out when to follow the rules and when to modify them is often one of the largest engineering challenges in taking a project that is 90% done and getting it over that finish line.

This feeling was reinforced by a recently published book, “Einstein’s Fridge: How the Difference Between Hot and Cold Explains the Universe” by Paul Sen (Figure 2). It’s a holistic look at the basic questions of what we mean by heat: how different experimenters, engineers, and researchers contributed to our present-day understanding of thermal fundamentals and misunderstandings and misconceptions from the earliest days of inquiry to the present.

One important theme that the book repeatedly shows is how our accepted ideas and concepts related to heat have evolved and the implications of these changes on further scientific progress. (Note: the first two words of the book’s title are very misleading, as is the cover artwork, but the book itself is good.)

Engineers and designers have to start somewhere and with something, of course, and doing so with accepted, long-standing assumptions is a good place to do so. At the same time, when things aren’t going as expected, it may be necessary to ask some tough questions such as “how do we know this is true?” and “Is this based on solid, verified data, or is it a case of ‘everyone knows’ assumptions that may have been somewhat incorrect, or perhaps things have changed, for some reason?”.

This is not a new concern. In Paul Dirac’s seminal book The Principles of Quantum Mechanics (1930), the brilliant physicist detailed this new, radically different, and highly controversial theory (many leading physicists were skeptical – with good reason). He outlined the contradictory evidence and conflicting theories, showed the inconsistencies and gaps between theory and data, and then ended the book with an admonition (in translation) “it seems that some essentially new physical ideas are here needed”, which is often re-phrased colloquially as “some new ideas are needed here.”

Sometimes, you have to step back and ask questions such as “where did we get those numbers?” or “how sure are we about this?” and even “what general rules are we following that may not be right?” to fully resolve a design issue. It’s prudent engineering and scientific practice to do that.

Related EE World Content

How can printed temperature sensors address a variety of applications?
The challenge of temperature measurement and sensor calibration, Part 1
The challenge of temperature measurement and sensor calibration, Part 2
The challenge of temperature measurement and sensor calibration, Part 3
Using optical fibers for temperature measurement, Part 1: Context
Using optical fibers for temperature measurement, Part 2: Principles
Using optical fibers for temperature measurement, Part 3: Variations
IC Simplifies Digital Temperature Measurements
A Micro-Thermometer To Record Tiny Temperature Changes
‘Smart Thermometer’ Improves Flu Forecasting
Bluetooth thermometers include data-logging functions

References

eLife Sciences Publications Ltd., “Decreasing human body temperature in the United States since the Industrial Revolution
Harvard Medical School, “Time to redefine normal body temperature?
Paul Sen, ““Einstein’s Fridge: How the Difference Between Hot and Cold Explains the Universe”