My main point is that I think this is a flawed approach. I know Sandy teaches it because it is easy for new people to use because there isn't a need to find the ink limits and partitioning points, but Using Curve_K as the only means of controlling the ink density removes several options for getting closer to a working profile.
I use the single gray value system for teaching purposes at workshops because,
1) it is very simple to explain, 2) students immediately understand how it works,
and 3) it works as well, or better, with the Epson K3 inks than any two or
three gray value partionned system I have ever worked with. By
better I mean, 1) a very small correction curve is necessary to linearize, and
2) tones are as smooth, or smoother, with minimal grainy look. It is also
very "robust" in that blockage of a few nozzles of the printer has
almost no impact on overall density range or banding since at all points on the
X-Y coordinate the density is formed by the same three or four UV blocking
inks.
The process is very simple. One merely sets the ink levels for the three or
four UV blocking inks in the ink set to establish necessary DMAX transmission
for the exposure scale of your process. Then your print a step wedge print,
read the input/output values, and place those in the K curve. Please have a
look at the Carbon-238 quad views (with curve and not curve) that I placed in
the Files section in the folder on QTR profiles for alternative printing, and
note especially the very gentle correction curve that is required. You can also linearize with the LINEARIZE
command, and/or the K Curve.
While other QTR tools, such as gray gama, gray highlight and gray shadow corrections can not be used with this type of single gray value system, they are essentially never needed since the original file is so close to being perfectly linear in terms of UV blocking density to begin with. And when the ink level of the three or four blocking inks is changed by an equal percentage, the actual UV blocking density will decrease or increase by the same amount at any point on the X-Y axis.
Sandy