[argyllcms] Re: More direct method of measuring FWA?
- From: Ernst Dinkla <info@xxxxxxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Fri, 23 Jul 2010 17:41:47 +0200
Graeme Gill schreef:
> Ben Goren wrote:
>> It just occurred to me. Most papers laced with FWA are coated on one
side only. The
>> back side of the paper is uncoated and has no brighteners.
>>
>> Graeme, might it be practical to manually incorporate a reading of
the back side of a
>> paper for the purposes of FWA measurement? I'm thinking something
like an instruction
>> at the beginning or end of reading a chart to flip the paper over
and read any clean
>> spot. Only done with an appropriate command-line switch, of course.
>
> The main weakness in using the FWA compensation is characterising the
illuminant. No
> affordable instrument is capable of measuring the Ultra Violet
wavelengths.
> In V1.2.0 I have created a new tool (illumread) that provides an
indirect way of
> doing this, by making use of the FWA algorithm. While it's a bit of a
fiddle,
> and may not give an accurate UV spectrum, it does more accurately
model the
> effect of the UV on the FWA in a given paper, and therefore improves the
> accuracy and usefulness of FWA compensation.
>
> G
My spectral measurements of front and back of inkjet papers will help in
selecting papers. That was the main goal. I don't think they will help
in creating better profiles. Meanwhile I learned more about the FBA
effects and their interaction with the paper substrate and other
whiteners. I think it will be difficult to apply a general rule on how
FBAs shift the spectral reflectance. Not only for the different
lightsources in use (including spectrometer lightsources).but also for
some papers.
Measurements on both sides of the paper are not that relevant for
profile creation. The paper base for example is generally more opaque
for ultraviolet light than for the longer wavelengths. A nice example is
the Ilford Galerie Smooth Photo Paper (IGSPP 11) that is loaded with FBA
in the paper base but much less or nothing at all in the inkjet coating.
The brightening effect is reduced if spectral reflection is measured at
the print side. There is some logic in creating a paper like that as the
FBAs are contained within the polyethylene RC layers and have more (gas)
fade protection, creating an optical brightener buffer in time but the
initial brightening effect is reduced. That in contrast with for example
Epson's EEM (naming shifted in time, even called archival once) that has
a high FBA content in the inkjet coating but nothing in the paper base.
What factors are considered in the creation of an FBA algorithm is
unknown to me but looking at spectral curves of FBA containing papers I
would think it can not be as simple as a negative b in Lab value. If the
absorption of UV is considered as well then another effect should be
taken care of. The excellent TiO2 whitening agent, not an FBA as we know
it, absorbs UVB light but that energy isn't emitted in the visual
spectrum but comes free as heat. It reflects more equal over the visual
spectrum, scatters/blocks more in UVA and is quite
opaque as used in paper coatings. TiO2 particle size has an influence on
its UV reflectance and absorption. A bad companion of FBAs as it is in
competition with them on UV light absorption and it blocks light in
general so blocks in and output of FBA deeper in the layer. A drop below
400 nm in the spectral measurements doesn't have to be FBA absorption
only but could indicate TiO2 as well and the last doesn't deliver the
brightness around 435 nm that we know of FBAs. Baryte is a better
companion for FBAs as it doesn't absorb UV light and is less opaque. An
FBA algorithm should not only consider the valley left of 410 nm (and
not all UV enabled spectrometers are specific there) but also for the
peak at 430-440 nm and how it stretches to longer wavelengths.
Brightness as measured in a band of 10 nm around 457 nm isn't enough
indication either.
--
Met vriendelijke groeten, Ernst
Dinkla Gallery Canvas Wrap Actions
| Dinkla Grafische Techniek |
| Tel.040-2920132 Mob.0649426098 |
| www.pigment-print.com |
| ( unvollendet ) |
Other related posts:
