Digital BW, The Print

I thought I had my workflow down until I ran into this strange problem.

I'm currently running an Epson WF1100 using Eboni ink and 18% gray (Paul Roark's K2/EZ method). It's been working pretty nicely for me. Recently I've been noticing that the deepest shades of black were blocking up which created a slight sheen. The prints will then have that solarized type of look. I thought, "no problem, I'll reduce the dmax and reprofile."

I went through my normal steps. I decided to use 87% (previously 89%) as my darkest black and proceeded to set my curve to 87% and applied to my rgb-raw-16 file. Then applied the same curve to my 21 step gray file and printed. The print looked good. I measured the scale using my spyder print spectro. I then dropped the raw file and the measurements onto the qtr ICC profile creator.

This is where it gets strange. When I tried printing using the profile, anything 100% black was just printed white (not printed?). What's also strange is when I proofed it using the new profile, it would show the same thing, 100% black was displaying as white.

I went through my workflow and tried different things. Created curves using r,g, and b curves separately, gray/rgb curve by itself. A combination of both. Applied curve as 21 step rgb and also 21 step as gray scale.

BTW when I say curve, I really mean a straight line from 0,0 input/output to 100,87 input/output.

Does anyone have any ideas why this would happen or what could cause this and how to fix the problem?

Thank you!!