Jones, this issue has been discussed at great length, here and
elsewhere. My conclusions based on my own research and Bob Higgins'
excellent paper are that the emissivity of high-purity alumina at 6-12
um is reasonably constant at ~0.92 over the range of temperatures the
GlowStick operates, and that alumina is opaque to any IR at those
wavelengths originating from an internal heater coil.
Both the Optris camera and the Voltcraft IR2000-50D (used by myself and
me356) measure over 6-12 um and will yield accurate temperature data
from bare alumina if that emissivity is used. Calibration against
thermocouples has found this to be true within 1% or so. Using a coating
would require recalibration and testing, and would make comparison with
prior test data open to question.
The Williamson pyrometer was useful for independent measurement of
emissivity, and the accuracy I referred to applied to this use only. We
found it was unacceptably affected by near-visible IR to use as a
primary temperature sensor. Such instruments are meant mainly for
measurement of metals and other near-IR-opaque materials.
On 7/29/2015 1:38 PM, Jones Beene wrote:
AlanG,
Thanks for the complete explanation. I guess the most obvious follow-on is
that since the "Williamson two-frequency pyrometer, recommended to Bob
Greenyer by industry experts as the most accurate technique available" (the
one which has been used successfully in this older testing)- then why not
also include it now in the new test?
Is it not available? Assuming that the Williamson is not available where the
present test is to take place, can we assume that even so --- with the
longer wavelength IR testing, since the coating does have a smaller effect
(less than when a pyrometer is used) it will be applied in the spirit of
achieving the best possible result ?
Even if the coating is only a PR gesture, the numerous comments from experts
on the lack of a black coating would make one think that an uncoated tube
will be not acceptable.
