Ok - here is my first shot at capturing what I did.
Please let me know if you want more detail or if this is sufficient.
I have more data and images if it would be helpful - but I thought I would throw this up for now and see if anyone wants more.
Thanks again everyone!
Sean
file at Dead Link Removed
Sean, Clay, et al. -- It is great that you guys figured out a QTR profile for palladium that works on the 3800. I finally got around to reading Seans document and I have a couple of comments that you may or may not find useful. In the spirit of trying to make even better QTR profiles, here they are:
1) Clays approach is great if you already know where you are going (ie, already have a negative that works for palladium to compare to). It is not a general method for making a profile for a new process when you do not know what the finished curve should look like. And, in fact, using Clays process did not work the first time for Sean. It required a fair amount of empirical fudging to come up with Seans final curve. If you do not already know what the final result should look like, I think you have to go back to basics as described in my manual and go through that misery.
2) Sean talks about using an Xrite 810 to measure UV (ultraviolet) OD values and using that information to try and match the ideal curve. In fact, the Xrite 810 that I own only reads visible and color (RGB) values, not UV. And that probably explains why Sean found that the OD needed to print pure white was around 2.8 or so. When measured on an Xrite 316T in UV mode, you will find it takes an OD in the UV of about 3.2-3.3 to print pure white. This is somewhat important since the different Epson inks all have a different ratio of UV to visible absorbance. And UV light is what exposes the palladium, not the visible light.
3) The QTR profile derived by Sean/Clay is very nearly a black ink only profile. A little bit of yellow is used to kick up the dense end of the neg, but the other colors are set to very low ink limits and they do not absorb much UV. On my old Epson 4000 this would not have been good. With the 4000 using all the inks in similar amounts visibly led to a smoother negative. Whether or not it is important for the 3800 still has not been tested.
Anyway, after that carping, here is my confession. About a month ago I bought an Epson 3800 and, off and on, have spent the last month trying to make a useable QTR profile for it (for palladium). Until today I had failed miserably, despite heroic fiddling with ink curves. Then, today I gave up, turned off the LLK ink, and made a two part gray system profile pretty much like I used to do for the the Epson 4000. Worked on nearly the first try. In my opinion the LLK ink is more than useless for dig negs. It absorbs very little UV light and would only affect the darkest tones of the final print anyway. Trying to get enough density out of it to linearize the dark print tones was a fools game for me. Attached is the final QTR profile I came up with. It is for palladium with no contrast agent. Note that all of the inks contribute heavily to neg density at almost all levels of negative tone.
Cheers, Ron-san
