QTR-Quadtone RIP

I have printed two negatives, which I then contact printed with gum. In the zzMisc folder under Files in this forum I have loaded a zip (Diginegs-gum.zip) which contains the two profiles, and four jpegs:
- Scan-140228-0001.jpg is the print generated from the negative printed with the profile r2880-gum-FaGF-WN-PM-v0.txt
- Scan-140228-0001_measure.jpg is the same as Scan-140228-0001.jpg but desaturated, with the white and black point set as per Ron Reeder instructions and blurred (Gaussian blur with an 8 px radius).
- Scan-140303-0001.jpg & Scan-140303-0001_measure.jpg are the same but the negative was generated with the profile r2880-gum-FaGF-WN-PM-v1.txt.
Now an explanations:
In a previous post I was asking for help with the conversion from Gray Gamma 1.8 to Gray Gamma 2.2. The first negative I printed was using a step tablet which had the incorrect gray densities due to the fact that I converted from GG1.8 to GG2.2; so when you look at the scan Scan-140228-0001, the densities value read:
(CV=correct value, IC=incorrect value after conversion, SR=scan read)
CV IC SR
0 0 0
5 4 0
10 8 0
15 13 0
20 17 0
30 25 0
40 34 0
50 43 5
60 53 10
70 63 87
80 73 95
90 85 100
100 100 100

For the second print I assigned the GG2.2 profile (not converted) and the gray densities remained the same. At the same time I made some modification to the profile and generated r2880-gum-FaGF-WN-PM-v1.txt, and the resulting negative was used to contact print Scan-140303-0001.jpg (Scan-140228-0001_measure.jpg was generated as above).
The resulting scan readings are:
(CV=correct value, SR=scan read)
CV SR
0 1
5 0
10 0
15 0
20 0
30 0
40 0
50 0
60 0
70 84
80 90
85 92
90 94
95 97
100 99
The changes I implemented from version 0 to version 1 are:
LIMIT_K - from 40 to 20
LIMIT_C - from 10 to 5
LIMIT_Y - from 10 to 5
LIMIT_LC - from 40 to 60
LIMIT_LLK - from 75 to 85
GRAY_VAL_2 - from 25 to 30
GRAY_VAL_3 - from 20 to 25
GRAY_GAMMA - from 1 to 0.7
I was wondering if you could give me some pointers on which value should I try to tweak before trying to linearize with a curve.

Thanks,
Luciano

To give a more comprehensive look I have also uploaded in the same zzMisc folder the file "Ink Pattern Page - R2880.jpg" which is the first test I contact printed to determine the UV behavior of my inks. As noted, I use Photo Black based on a mail exchange I had with Ron Reeder.