QTR-Quadtone RIP

I'm having difficulty achieving the Dmax that I think the following ink/paper combination 
should be able to achieve:

7800, K3, Innova PSC.

When printing the ink separation page, the K values where as follows

65.00	17.93
70.00	17.89
75.00	17.41
80.00	17.15
85.00	17.20
90.00	17.59
95.00	17.36
100.00	17.53

Therefore, I have set the K ink limit to 75 and black boost to 80.  No matter what I try, I 
can't get anything close to 17.15 when printing the 51 step random target with the non-
linearized curve I have created.  I'm also getting reversals on the 100%K patch.  The 
following is the end of the data produced by QTR-Linearize-Data.app.  

90.00	1.589	18.26	0.82	2.00	-         L                   |  a    b                     +
92.00	1.596	18.07	0.81	2.13	-         L                   |  a    b                     +
94.00	1.599	18.01	0.79	2.17	-         L                   |  a    b                     +
96.00	1.603	17.89	0.84	2.13	-         L                   |  a    b                     +
98.00	1.605	17.83	0.76	2.40	-         L                   |  a     b                    +
100.00	1.594	18.13	0.71	2.50	-         L                   | a       b                   +

Even the 98.00 patch is at 17.83 - a far cry from 17.15 . . . or am I being too picky?  It 
seems that with the K ink limit of 75 and boost of 80, that the 100 patch generated with 
the curve should be the same as the 80 patch on the ink separation page, or am I missing 
something? 

Thanks for any insight you can provide.

Best.

Josh