Re: On the effect of florescence
Re: On the effect of florescence
- Subject: Re: On the effect of florescence
- From: Robin Myers <email@hidden>
- Date: Tue, 1 Mar 2005 20:16:54 -0800
On 1 Mar 2005, at 17:26, Roger Breton wrote:
Using the UV Cut filter puts all the
paper measurements to the same reference point.
Robin Myers
Please excuse my ignorance but isn't there a limit to which a UV
filter can
'adequately' correct measurements in substrates with varying amount of
FWA?
The UV blocking filters I have examined are yellow in appearance with a
transmittance spectrum that has close to 0 transmittance in the UV
region, usually with a steep slope up to almost full transmittance
around 400 nm, give or take a bit. The almost zero transmittance in the
UV range has the effect of cutting all the UV light from the
spectrophotometer in the UV.
I tested the Spectrolino filters by taking some Hammermill Multipurpose
paper, a commonly available office paper with a FWA and measuring the
reflectance using each of three filters on the Spectrolino; Null, D65,
and UV Cut. The Spectrolino uses a tungsten halogen bulb for
illumination, which has a small amount of UV production, but more than
enough to excite the FWA. The expected result for the Null filter was a
peak in the blue spectral region. The D65 filter was expected to
produce a higher peak in the blue due to it's higher transmittance of
the violet-blue region than the rest of the spectrum. If the UV Cut
filter worked properly, it was expected to remove the FWA produced
peak, leaving the spectrum flat, or nearly so, in the blue region. The
paper's FWA produced a peak at 440 nm and here are the measured results
from the experiment:
Filter 440 nm reflectance
-------------------------------------
Null 1.028
D65 1.133
UV Cut 0.891
If anyone is interested, I can email them offline the graph of the
results.
You write that it 'puts all the paper measurements to the same
reference
point'. Fine. But wouldn't you rather say that it correct paper
measurements
by the same fixed amount? I fail to see what 'a fixed reference point'
would
be.
No, I would not say that it takes a fixed amount from the measurement.
Filters that have zero transmittance in a spectral region will
completely remove the light from the result in that region. Remember
that the illumination's spectral power distribution is multiplied by
the filter's transmittance distribution and the object's reflectance
distribution.
The same reference point to which I referred is that only the light
from the visible portion of the spectrum is involved in the print
measurement, the UV variable is removed.
Let's take the case of printing papers or inkjet papers or office
photocopying or laser papers. As you know, some of these papers are
very
close to neutral, many european litho paper, I'm told. Inexpensive
litho
papers (coated gloss #3) have a moderate amount of FWA, with b*
between -2
to -5. I've personally seen some inkjet papers with b* = -9 ! That is
quite
fluorescent, if you ask me. But, in this context of wildy differing
amount
of FWA in papers, I have difficulty to conceptualize that a fixed UV
cut
filter "will put all the paper measurements to the same reference
point'?
You don't mean to say that a UV cut filter will bring a paper with b*
= -9
or a paper with b* = -4 or -5 or -2 to the b* = 0 ? In my limited
understanding of physics and metrology, I would tend to think that the
UV
filter on the Spectroscan will 'substract' a constant amount of visible
light evergy over a fixed wavelength interval but that this fixed
substraction (please excuse my poor terminology) will not be the right
correction for all amounts of FWA in every paper that comes along.
Therefore, I would venture to say, there is some component of FWA that
will
not be corrected but the filter. And we'll still have problems with
gray
balance in profiling.
Actually, I expect a paper with an FWA, when measured with a UV Cut
filter to have a positive b* value. For example, here are the CIE
L*a*b* values for the Hammermill Multipurpose paper in the above
experiment:
Filter L* a* b*
-------------------------------------------------
Null 96.40 1.38 -3.54
D65 1.133 2.81 -6.78
UV Cut 0.891 -0.53 2.45
(all values measured and calculated with SpectraShop 2, CIE 1931
2-degree observer, D50 illuminant)
Why did I expect a positive b* value when the FWA's contribution was
removed by the UV Cut filter?
Answer: because most white papers have a lower deep-blue and violet
reflectance than in the rest of the spectrum. The human visual system
adapts to make the paper appear white, but the spectrophotometer reads
the lower blue-violet reflecting paper as slightly yellow. This is a
common issue with most of the white materials in common usage
(especially those made with titanium dioxide, the most common white
colorant).
Notice, too, the b* value for the D65 filter. It shows almost double
the blue result from the Null filter, thus showing that varying amounts
of UV excitation can cause large changes in the color of the paper.
Robin Myers
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden