Message
Re: using QTR-Linearize-Quad to create a new .quad file (solved) (to Jeff)
2015-10-05 by brian_downunda@...
I think your problem comes from using the 51 step patch, rather than the 21 step. I have tried it and found that it provides "tooooo much information", as the saying goes. That is, it wiggles around too much. This is the sort of reversal that can cause the quad file relinearisation droplet to fail. The 51 step patch may be useful for some things, but not this IMHO.
You're like to get more consistent results from the 21x4 patch. Although it's less precise, it's possible to have too much precision. I have had the droplet fail on the 21x4 once, but only on a curve that had quite bad curvature. The droplet does have its limitations.
On the question of averaging readings, I am less concerned about measuring the same patches multiple times than I am about reading multiple patches. That is, I'm less concerned about variability in the performance of the measurement device than I am about the consistency of the printer in laying down the ink, and also possible variation in the paper coating. So when I have used the 51 step patch, I've created a 51x3 patch and matching reference file. I sometimes check the variation between the three or four patch sets, and generally while not huge, it can be more that you'd expect, and demonstrates the advantage of averaging different patches. But this probably wouldn't solve your problem.