Prints are too dark


I made a few prints on Hahnemuhle paper and the middle gray tones are printing too dark. The prints do not look like palladium prints to me at all and they have more contrast than I expected. The final print does not look like the image that is on my computer screen.

I use two monitors that are connected. One is made by NEC and the other is my Macbook Pro. I calibrated them using the X-rite i1 software. I used D50 for the white point and 80 for brightness.

My exposure time in a UV printer is 5 minutes. I am developing the print for 1 and 1/2 minutes in PO heated for 3 and 1/2 minutes and the clearing is in three baths with EDTA, Citric acid, and Sodium sulfite…one Tbs of each in 1 liter of water. The time in each bath is 5 minutes. I used a water rinse between the developer and the first clearing bath.

I am using the default icc profile that came with the PiezoDN program. I thought about making my own ICC profile but the manual says it would add contrast and I do not need that.

Do I need to linearize again? My numbers in the L column would increase and then decrease and then increase. Should I try another paper?

Thanks for your help.





Hi Catherine,

All of the process details you listed look fine. My only comments on that would be that the length of time you heat the developer doesn’t matter, rather the actual temperature is what you should note, and the water rinse after the developer is not necessary though it probably doesn’t hurt. You didn’t mention what your coating mixture is. Is it just FO and Pd? Any Tween or Photo-Flo? And more importantly any Na2 or other contrast enhancer?

One question about your PrintTool/QTR setup comes to mind: Are you using the ICC with your linearized .quad or something else? In the first PrintTool setup page you have selected Application Managed (PrintTool Managed if you have upgraded), the relevant IJM-HahnemuhlePtRag ICC profile for your printer, and Perceptual intent. In the QuadToneRIP section of the next (Print) dialog you should have selected your custom linearized curve as Curve 1, along with the rest of the proper settings.

You are most likely seeing the normal effect of using an ICC profile. I think you misinterpreted what the manual says about them. Any ICC profile has the effect of deepening and compressing the dark tones and increasing contrast through the rest of the tonal scale. This means the IJM-made ones as well as any custom ones that you may make yourself.

I suspect that you would prefer the results from using your linearized .quad <span style=“text-decoration: underline;”>without</span> an ICC better. I do. I feel the ICC makes the print too dark and too contrasty for my taste. Others no doubt prefer the ICC look.

One thing that is lacking in the PiezoDN spreadsheet is a way to compare the various calibration stages in one graph. This is something that I find to be very useful so I made my own. Here is a sample plot showing 4 stages in my linearization of Platine. I think it is self explanatory but ask if you are unsure.

[attachment file=1660]



Keith, thanks so much for your helpful reply. My coating mixture is Fo and Pd…7 drops of each. I am not using any NA2 or Tween or Photo Flo though I noticed that you do use these in your printing. I figured I already had enough contrast. I am open to trying the Tween and the Photo Flo. The temperature of the PO is around 110 degrees

I am using my linearized quad with the ICC profile from IJM with Perceptual Intent. My Print Tool still says “Application Managed” I did not realize there was an upgrade. The Embedded Profile says Gray Gamma 2.2. What should be in the Format space? I do have the proper settings in the Print Dialog Box.

I did think that an ICC profile had to be used but now I know better.

Thanks for your graph but I do not understand what it is saying. If you do not mind explaining it I would like to know about it.



Most all Pt/Pd printers, myself included, have found that Hahnemuhle Platinum Rag performs much better with the addition of Tween 20 or PhotoFlo to the mix. We see the effect primarily in how smooth the tones are and, to a lesser extent, maximum density. If you are pleased with your results then no problem, but it might also be a case of “you don’t know what you’re missing”. :wink:

As for Na2 or other restrainers, I wrote about this at length in another thread. I do not use it to increase contrast, but rather at the minimum effective amount to eliminate fog. At this low amount it actually increase the tonal scale slightly over using none at all. If you are sure that you have no fog, then no worries. In the best of all possible worlds, that would always be the case. But most of us experience a slight amount of fog when using no restrainer at all, even if it is only visible when you have an image area that should be paper white directly adjacent to actual uncoated paper white. If you print with brush stroke edges showing you may never notice this even if it is occurring. To check for it, measure an uncoated spot on the edge of your test print and compare that to step 1. If they are the same then you are in good shape. If step 1 reads even slightly more density than uncoated paper then you have fog.

The PrintTool upgrade adds some features specifically for digital negatives. Unfortunately for PiezoDN users, they all involve inverting the image so we can’t use them since PiezoDN inverts via the curve itself. At least that’s how it looks to me.

Format Space shows your image size in pixels and the image mode, for example 5040x6840 Gray:16. This and embedded profile are informational only in that you can’t change them here. Thanks for mentioning the Embedded Profile! That is important and should always be Gray Gamma 2.2.

Try printing again using No Color Management and your linearized QTR curve, just like when you printed the linearization target. Which raises another question: When you printed the linearization target were you sure you used No Color Management? If you printed that with an ICC you will need to go back and redo it without.

In the graph, the wide gray line represents linear - this is what we are looking to achieve. Above the gray line is too light, below is too dark. There is some acceptable margin of error which is why I made it wider. The green line is the result from my first print with the Master quad. The upper highlights are good but the rest of it is too light, in particular the flat area from around step 9 through 50 or so is very compressed indicating little tonal separation in this range. The yellow line is the first linearization. It is close but still slightly on the light side. From the CGATS data derived from this print I then did 2 things. First, I made a second linearization, represented by the blue line; and second, I made an ICC profile, represented by the red line. The second linearization is almost perfect, and probably about as close to perfect as we can get with hand-coated processes. The ICC shows exactly what Walker said to expect: deepened shadows at the cost of some compression in that region where the curve flattens out approaching black from around step 110 through 129, and increased contrast through the rest of the scale.

In general when looking at these graphs, the steeper the line the higher the contrast. But increasing contrast in one region always decreases it somewhere else, and vice versa. Always. No exceptions.



Keith, I made a print yesterday without an ICC profile and it was much better. However some of the shadows were bl0cked up. I assume that I should be able to get the shadows where I want them without modifying the digital file. Is that correct? I will follow your suggestions above and see what results I get. If some shadows are still too dark, what can I do?

I know I used No Color Management when I printed the linearization target.

Thanks for explaining the graph. Now I understand.

Thanks for all the information you share. It has really helped.



Depends on how good your monitor calibration is and how good your linearization is. If either is off what you see on screen will not be what you get on paper.

The monitor settings that you posted above should be good. Did you verify your linearization by reading the target and pasting it into the Smoother? It should be very close to the target line through the entire tonal range.

One thing that I find very useful is to proof the file by making a digital print on paper. To me, the notion of a perfect screen to print match is a bit of a pipe-dream, especially when going through an intermediate step as we are doing here. I know Jon and Walker will disagree with me on this, and I agree that it is a worthy goal, but a transmitted light image on the screen will never look exactly like a print on paper.

If you must proof on screen, it is probably better to use a soft-proof setup with a custom ICC profile or the IJM profiles. In my experience this tends to over compensate, making the screen image too flat with blacks less black than on a print, but it might be worth a try even if you are not printing the negative with it. It does give a good representation of the image color you will get in a print.


In my time here in China (13 days or so now) I’ve finally gotten around to building PiezoDN proofing curves for platinum. These curves can be used to print an image with a 1.4 dMax using the PtPd default ICC to “hard proof” the final printed piece.

With upcoming pro ink PiezoDN curves you will even have the ability to tone the hard-proof to what color you need and also to print uncoated (using MK-HD ink with the printers that have 2 K channels) on the actual platinum paper you require.


This will be particularly useful for those who do not have tons of money to spend on proofing.





I will finalize and post an update when I’m back in the US.