Ok, sorry, I forgot that you were having full dmax in 2 minutes - you've mentioned that in previous posts.
The fact is that you can never reach the same dmax under the negative, more so with self masking processes, and develop out pd still has some print out, therefore some degree of self-masking. Most people aim for 90% of the dmax to calibrate for, that way you keep the dark tones more in the straight line of the H&D curve. There's always some amount of tonal progression, the trick is to find the point where the marginal utility equals 0. This is more easy to observe with a step tablet test; my method is to significantly overexpose the test tablet, then scan the print, average each step and use the threshold tool to find the minimum step number where it merges with the previous one and decrease the exposure according the numbers.
Also, a too humid paper may compromise the negative, the silica coating is very very hygroscopic, will suck moisture and/or emulsion if it's not dry enough: try to not pre-humidify the paper or putting 1/2 mil impervious and transparent material (mylar for instance) between the paper and the negative.