Where did my blue go?

[GeraldB]

I'm puzzled. I'm pretty new to the printing game. Decided to try out 3 different Baryta papers.  I selected a picture that had a wide range of saturated colors. Printed it on 3 different papers each with their own vendor profile. All the colors came out very nicely except the blue which was desaturated big time. The same on all three papers. A nozzle check showed no blocked jets.

Here is what I did:

Windows 7 Pro 64 bit, LRM 3.4, CS5 12.03, Epson 7900.
Open in CS5 from LRM. Create duplicate, position side by side, turn on proofing on copy with correct profile with black point and paper simulation both relative and perceptual.
Found that there was almost no difference in the proof version except a little darker which I corrected by brightening a little.
Turned on gamut check, a few pixels out of gamut, hardly visible.
Save proof as tiff. Print from LRM. In the driver properties the mode is Custom (Off) No color adjustment.
All colors and luminosity good except Blue which is highly desaturated.

Any suggestions where I could be going wrong?

[Mark D Segal]

I'm puzzled. I'm pretty new to the printing game. Decided to try out 3 different Baryta papers.  I selected a picture that had a wide range of saturated colors. Printed it on 3 different papers each with their own vendor profile. All the colors came out very nicely except the blue which was desaturated big time. The same on all three papers. A nozzle check showed no blocked jets.

Here is what I did:

Windows 7 Pro 64 bit, LRM 3.4, CS5 12.03, Epson 7900.
Open in CS5 from LRM. Create duplicate, position side by side, turn on proofing on copy with correct profile with black point and paper simulation both relative and perceptual.
Found that there was almost no difference in the proof version except a little darker which I corrected by brightening a little.
Turned on gamut check, a few pixels out of gamut, hardly visible.
Save proof as tiff. Print from LRM. In the driver properties the mode is Custom (Off) No color adjustment.
All colors and luminosity good except Blue which is highly desaturated.

Any suggestions where I could be going wrong?

Presumably the blue is desaturated relative to what you see on your display. Is your display properly calibrated and profiled?

Can you print a MacBeth colour checker to see how your printed blues compare with those on that chart? Can you print some intense but in-gamut blue patches (create them in Photoshop) and measure the printed values to see whether your printer is reproducing them correctly?

These issues often need some analytics to learn where the problem is.

[GeraldB]

My monitor is a NEC 2690wuxi calibrated with the Spectraview II that came with it. Admittedly its been a few months since I last calibrated so I can try recalibrate and see if there is a difference. I'd be quite surprised if that is it.

How would I measure the printed value of a blue patch made in PS. Would I need a spectrophotometer for that?

[Mark D Segal]

I agree.

Yes.

[Alan Goldhammer]

Rather weird result.  Try printing out a test print allowing the printer to manage the colors to make sure it's not a bad profile (or in this case 3 profiles).  I've seen some manufacturers profiles do exactly what you have noticed (and you don't even need a spectro to measure anything, it's pretty evident just from a visual examination).  If the printer managing colors shows the same result it could be a corrupted driver.  I would delete the printer and reinstall making sure you have the most current driver.

[GeraldB]

Found the problem. Blocked blue nozzles. I ran a nozzle check and blue was mostly missing from the pattern. Ran a head clean, got a clean nozzle check pattern and then the print had its blues. I had run a nozzle check one week before and it was perfect. Rather disappointing Epson quality. I wonder if there is problem with the machine?

[Mark D Segal]

Nope - not likely a "problem" with the machine, other than the fact that these clogs are pretty much par for the course. If I don't use my 4900 for a few days in a row, I get clogs too - usually the Cyans. One cleaning of one or two pairs fixes it. Always do a nozzle check before printing, even if from the day before.

[howardm]

Mark,

Have you found any difference in nozzle check 'performance' due to using cheap bond paper for the check?

[Mark D Segal]

Nope - same information no matter what you print it on, but regardless of the paper it's best to look at all those little wiggly lines under a low-power magnifier just to make sure you see correctly what's going on. Yellow is the hardest of all, but it's hard whether you print that pattern on photo paper or bond paper.

[Garnick]

Nope - same information no matter what you print it on, but regardless of the paper it's best to look at all those little wiggly lines under a low-power magnifier just to make sure you see correctly what's going on. Yellow is the hardest of all, but it's hard whether you print that pattern on photo paper or bond paper.

I've always printed nozzle checks on bond paper, both lower and higher quality. I print on both ends and both sides. Then I trim the patterns from one end and use it again, so I get 6 nozzle checks per sheet.  Especially with the 9900 I always view the yellow with an 8-10x loupe and the others with a standard, illuminated magnifier. If in doubt I might print the pattern on either Luster or Gloss material, since the bond paper can sometimes appear to show slightly deflected patterns. I've always found that with Epson printers it's an absolute must to run nozzle checks at the beginning of each printing session. Because of the issues I've had with the 9900, including print head and pump/cap assembly replacements, nozzle checks and accompanying cleaning cycles are par for the course. Since the last service call it has been working well, except for the 720dpi issue, but I will not be delving into that scenario at this point. And I definitely agree with Mark that the Cyan is usually the first to show dropouts. I assume that when you referred to "blue" nozzles you meant Cyan. Since blue is a combination of Cyan and Magenta, you would of course see a very desaturated blue with almost no cyan ink firing. I would imagine that the rest of the print was looking rather warm also. I'm really surprised that you didn't see banding as well at some point. However, that can be rather subtle sometimes, depending on which ink is dropping out.

Gary  

[GeraldB]

Thanks for all the good advise. I thought nozzle checks were only something one does once in a while. Now I know different.
Garnick I did not notice the banding in the photo. The way I found the problem was to print some solid colour swatches and then I saw the banding and proceeded to do a head clean.