Digital BW, The Print

Daniel Staver wrote:

> I had my UT7 inks go totally green on me once. 
> Definitely not normal.

It can happen, but the 3D toner is more greenish to start with because it is
designed to be used with the Y-position, Lab A toner also in the mix.  The
3D toner has more cyan relative to R800 Blue than does the UT7 cold toner.
(The R800 "Blue" is a very magenta looking "blue.")

> It was some sort of settling issue. Shaking the cartridges 
> got the color back to normal.

You'll notice that I'm using un-touched UC LM in my 7500 4K+ mix.

When I put drops of magenta -- MIS or Epson UC -- into a test tube of the
MIS base that appears to be the best I can get, the magenta will show signs
of settling.  The MIS magenta settles more than the Epson, but I think the
problem arises from the same basic factors.  Magenta is a relative weakling
when it comes to light fastness.  So, to make it more lightfast, I suspect
they grind it to a larger average particle size.  The larger the particles,
all else being equal, the slower the fading -- but also the faster the
settling.  To try to offset the settling, the ink suppliers appear to
increase the viscosity and specific gravity of the base used for that
particular pigment.  So, to see what the limits are of the Epson heads, one
thing I do is measure those parameters of the Epson magenta.  You'll recall
the MIS "GP" inkset.  It was made because the viscosity of the then new 7600
M was too high for the 1280 to pump correctly.

When we mix different color pigment inks and base, however, the viscosity,
etc. become more of an average, and not what is ideally needed for the
magenta.  This is the basis for my move toward non-blended inks for large
format -- if not all -- printers.  I'm no longer willing to deal with these
issues in large format printers. (And I expect the 3800 to become a high
volume, large format printer.)  

I can increase the viscosity and specific gravity of the base to help hold
the magenta in solution, but then the tone shift reverses itself.  I might
be able to balance 2 different pigment types, but it would take a lot of
work and be more of a compromise than staying with a one-pig-per-ink
approach.  When the pigment types go to three, I suspect it is virtually
impossible to get a base that balances them as well as can be done when only
one pigment type is in the mix.

Usually, the constant agitation of the carts that happens in desktop
printers smaller than the 4000 class is sufficient to take care of the
problem.  So, in the past, the problem was limited to the relatively low
volume large format printers, which are often in settings where they are
used regularly.  Additionally, something like the MIS Autoprint program
takes care of the problem.  But this was a real hassle for me, and the
problem will become very widespread as the 3800 volume picks up.

The R800 Blue is much better than the magenta, but it is not entirely free
from the above effects, nor is the R800 Red.  The carbon and cyan are the
pigments that stay in suspension the best in the MIS base.  This is probably
because those pigments are so lightfast they can be ground to very small
sizes.

At any rate, I agree the problem is real, but usually not a serious issue
for most desktop users.  With the advent of the 3800, however, I expect the
picture to change significantly.  I will not support a blended inkset on
that machine, and it does not appear to need one anyway.  Our B&W inksets
were, after all, made due to the deficiencies of the existing color
printers.  Those deficiencies are increasingly being taken care of by the
printer manufacturers.  The main problems I see with the k3 approach, for
example, is based more on the Epson software than its ink layout.  I don't
think more than the minimum color pigments should be used, and I don't think
B&W should require a different workflow than color.  But, for now, switching
out the yellow or using a rip is the best way around these "problems." 

Paul
www.PaulRoark.com