Meteorological thermometers are usually enclosed in
white louvered enclosures (Stevenson screens—[link]) to
keep
them from being affected by radiative heat transfer. This
is not just heating by sunlight, but cooling by exposure
to the night sky—[link]. Stevenson screens generally work
okay,
but they're expensive and bulky.
I propose using two separate thermometers, equally
exposed to the radiative environment, where one has its
bulb
painted white (to absorb/emit little radiation) and the
other has its bulb painted black (to absorb/emit lots of
radiation). The black and white paints don't have to be
perfect; calibration can take care of any nonzero
emissivity/absorptivity of the 'white' paint and and
nonone emissivity/absorptivity of the 'black' paint. The
only
requirement is that the have very broad
emission/absorption spectra. Spectralon for white and
carbon black for
black would work well, I think.
By comparing the temperatures measured by each
thermometer and doing a bit of math, one can then
determine
both the radiative temperature and the thermal
temperature of the environment, even though both are
affected
by both (but in different proportions). (This is similar in
concept to wet-/dry-bulb thermometers used to measure
both humidity and temperature.)
To ensure equal exposure to the radiative environment
(i.e. not shielding each other), the thermometers could
be
rotated around each other continuously, or have their
absorbers/radiators intercalated like the colors on a
beachball. Other solutions are welcome too.
41/277 [2018-01-31]
[In addition to my count of how many ideas from my list
I've posted so far, I'm now going to include the date I put
the idea on the list, if I recorded that date.]