Perhaps we are all getting over-reliant on a number and trying to associate it with our sensation of thermal comfort which, contrary to the assertion in the original article is by no means simple. our bodies actually sense energy flow as opposed temperature and are affected by radiant heat, air flow, clothing (insulation), humidity etc., According to current research, external temperatures have more of a bearing on our sensation of thermal comfort than the idea of a fixed number providing a value of 'temperature'. So rather than being hoodwinked by the variable accuracies of commonly available temperature sensors there is an argument for a device without readout that we nudge up and down until we are comfortable, and removes the stress of trying to correlate a presumed numeric vslue with an expectation of how we feel!
All instruments "lie" to you. The good ones just tell you by how much--they have specifications for uncertainty. Temperature can be measured much more accurately than 10 degrees, using say a quartz crystal or platinum resistance thermometer you can get accuracy well below 1 degree (I don't remember exactly, but I'm guessing something less than 0.01 degree). Those cost thousands of dollars, however.
An even more complicated question--does the air temperature really tell you enough? Here's an old story/joke from my University that makes this point:
Students are sitting in Thermodynamics class in Winter Quarter. The air temperature is 72 Farenheit. The students are wearing long pants and long-sleeved shirts. The students are comfortable.
A number of students fail the class, and return to take it again in Summer Quarter. The air temperature is again 72 Farenheit. The same students are now wearing shorts and short-sleeved shirts, and are just as comfortable as in Winter Quarter. Why?
The answer is--heat transfer happens three different ways--convection, conduction, and radiation. The convection is the same (since the air temperature is the same). The conduction is roughly the same (since the objects in the room are the about same as the air temperature). However the radiation is different, since the walls are colder in winter than in summer.
The moral of this story is that even if you could measure the air temperature with zero uncertainty (and you can't), there are still other factors in how cold or warm you feel:).
Analog is always more precise than digital, have observed that while mesauring temperatures using normal analog thermometers, lab grades and digital versions..even the IR based sometimes give faulty readings..
I designed a product in 1980 which needed precision frost threat detection. Obviously nothing has changed since then! Same problem, same solutions on offer, same vendors even!
Is there a market for cheap but certified-precision (laser-trimmed?) thermometers/sensors?
BTW, those flexible printed thermometers (relying on melting wax) are very accurate but its hard to make use of the precision trip point even if there are many segments.
I used a Curie-point reed relay for my application in the end. Expensive but it meant no electronics was needed! They are probably still operating perfectly.
Finally, I hear they plan to include temperatures with the GPS signal starting next April ;-)
How about gas (petrol) gauges -- is it juts my imagination, or does the last 1/4 go down much faster than the first 1/4?
I come back to my point that, considering the advanced level of our technology, you would thing we would be better at measuring relatively basic things...
You should look at products made by Omega, they specialize in precise temperature measurement and control... and all their products are made in the USA. Their thermocouples are accurate within 1 deg C, RTDs are within a half degree.
Blog Doing Math in FPGAs Tom Burke 24 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...