Physical Photometry

[Roger Breton]

I am going over my copy of CIE Publication 18.2 (1983) "The Basis of
Physical Photometry", Section 4.3 Irradiance and Illuminance. It is written
that Irradiance and Illuminance describe the flux incident on a surface at a
point.

Irradiance (symbol: Ee ; unit watt per square meter, W/m2) is a radiometric
quantity related to radiant flux by the equation Ee = dØe/dA.

Illuminance (symbol: Ev ; lux, lx; lumen per quatre meter, lm/m2) is a
photometric quantity related to luminous flux by the equation Ev = dØv/dA
and to the spectral distribution of irradiance Ee,lamba (lamba) by the
equation Ev = Km † Ee(lamba) * V(lambda) * d(lambda).

(I did my best to preserve the notation)

Can I take it from the definition of Illuminance that it is possible to
report Illuminance at a particular wavelength? Or is this kind of reporting
physically meaningless?

Why I ask is because I have an instrument which reports Lux as a function of
wavelength and I wonder to what extent is this an abuse of words or
confusion on the part of the manufacturer? The way this instrument makes
illuminance measurement is by pointing its head to a surface (of known
spectral reflectance) such that the instrument measures the light incident
on that surface, at every 5nm, from 380nm to 750nm.

I am trying to measure the direct sunlight entering my office at vaious
phases of the day. At 11:00am, (sun partially hidden by thin clouds) the
instrument says there is 68810 Lux falling on the surface I am measuring,
which I find reasonable. The software tells me that this light corresponds
tristimulus values of X=65955 Y=68810 Z=56441. (I think they're huge?) I see
that CAP Y and the reported Lux value are the same value (68810). I am just
not familiar with other makes of instruments designed to measure spectral
irradiance and illuminance to relate sensibly the measurements reported by
this instrument (SpectroCam by Avantes).

But I have a feeling that something is not kosher?

Any help is appreciated,

Roger Breton

 

[tlianza]

Roger Breton wrote:

Can I take it from the definition of Illuminance that it is possible to
report Illuminance at a particular wavelength? Or is this kind of reporting
physically meaningless?

/********************************************************/

This kind of reporting is physically misleading, as well as meaningless. It
doesn't reflect well on the people who wrote the software or instrument.
Illuminance is reported as a function of the integration of the V(lamda)
curve. Now if you have a single wavelength source, you might be justified
in reporting illuminance and record that it came from an illuminant of that
particular chromaticity, but wavelength really has no bearing in the
discussion of Illuminance.... at least after the integration...

--
Tom Lianza
Technical Director
Sequel Imaging Inc.- A GretagMacbeth Company
25 Nashua Rd.
Londonderry, NH 03053

 

[Roger Breton]

Roger Breton wrote:

Can I take it from the definition of Illuminance that it is possible to
report Illuminance at a particular wavelength? Or is this kind of reporting
physically meaningless?

/********************************************************/

This kind of reporting is physically misleading, as well as meaningless. It
doesn't reflect well on the people who wrote the software or instrument.
Illuminance is reported as a function of the integration of the V(lamda)
curve. Now if you have a single wavelength source, you might be justified
in reporting illuminance and record that it came from an illuminant of that
particular chromaticity, but wavelength really has no bearing in the
discussion of Illuminance.... at least after the integration...

That's what I think. But, I learn to hold on to my preconceived notions -- I
am far from knowing a great deal in color. There could be applications where
Lux at a certain wavelength is most revealing about a light source or
lighting system.

Roger Breton

 

[Danny Rich]

The definitions are correct but the symbology is a bit cryptic. The term
for light flux per unit area per nanometre is spectral irradiance. The
prefix spectral is used signify that the measurement is made a one or more
specific wavelengths. When the signals are averaged or summed over
wavelength the term is just irradiance. When the summation is weighted by
the V-lambda function the term becomes illuminance. Photometric quantities
are always integrals (summed) over wavelength. This is why illuminance at a
specific wavelength makes no sense.

Of course, the contrast to this is that a monochromatic light - such as a
laser beam - has a photometric response (illuminance) that is obtained by
integrating the spectral irradiance over all wavelengths where most of the
values will be zero.

Danny Rich