Solved: Dark prints and posterization


I recently converted my R3000 to use the K7 Special Edition inks. The matte prints on Photo Rag seem OK although a little darker than I expected. However, the prints on Epson Exhibition Fibre paper are terrible. They are way too dark, there’s no detail in the darks and I’m seeing posterization. I downloaded the glossy curves (from the links on the Technical Support page and received replacement curves from Jon Cone - but still the glossy output appears as if the inks are in the wrong positions or that the curves have been tampered with. To be clear, the curves have not been touched.

I’ve been through the linearization checker process and the charts do not look great at all.

I’m trying to upload a couple of at files to help demonstrate the issue but the Manage Attachment function won’t let me.


I uploaded some files to Dropbox Dropbox - Global - Simplify your life

Are you printing from Mac or Windows?
What operating system version?
What version QuadTone RIP are you using?
If Mac, are you printing from Photoshop or QTR Print Tool? If Print Tool, what version?
Please print the Ink Separation chart to check your ink shade placement, following the instructions here:

Please let me know, thanks~ Dana

Thanks for the links Brian. That second one didn’t help I’m afraid. I’ve applied GO (although there was no way it was going to suddenly fix the mess on the paper) and I’ve produced linearization curves. They are in my provided Dropbox link.

After reviewing your images, I see you’re using Windows.
Are you evaluating the glossy prints before or after printing the GO layer?


Window 8.0

I will print the ink separation chart now.

Thanks Dana.

OK, photos of both ink separation files in Dropbox. (I get errors when I try to upload them to this forum).

I posted that link to jeffg53’s thread because I’ve been working with him on his EEF issues and your problem sounded similar. But those linearisation plots look to me like either one or two channels are misbehaving fairly badly, or you have some inks in the wrong positions.

Can you explain what makes you say that? I’m looking at them but can’t interpret them enough to see anything obvious. What do you mean by channels?

I looked at these documents:

I’ve never seen anything like those, esp the EEF. The way that it drops precipitously and then partially recovers means that the upper highlights are way, way too dark, and then it recovers somewhat towards the mid-tones. What else could cause a pattern like that? Dana or Kelly will confirm or contradict my suspicions, but you could also print out the ink separation page in calibration mode and check yourself that the right shades are in the correct colour channels.

Thanks. Your linearizations are HORRIBLE (I’ve never seen one this bad), like as if you have a dark ink in the light positions…

I have positioned your Ink Separation printout over my known good one (yours on top/left, mine on bottom/right). The size and brightness is different, but just based on ink density, it looks like your ink shades are in the correct order, though I don’t know if they’re printing correct density without measuring (comparing each channel, and the density relationship compared with others, I think your ink densities look pretty good). Your yellow channel still has some strong color staining, which will require printing yellow flush images from QTR Calibration Mode, or doing more cleaning cycles.

Please keep me posted.
Warmly~ Dana

Hi Dana,

I know my linearizations look horrible but I still have no clue what to do about it.

I had real problems getting rid of the yellow. In the end, I ran a 2nd ink charge cycle and then printed a yellow head purge image from Photoshop. The print came out grey after the 2nd purge so I though it had all gone. I guess not.

You say to keep you posted but about what? What do I do at this point to improve the situation?


I have attached your QTR window, print tests and linearizations for easier viewing (Exhibition Fiber on top, Photo Rag on bottom):

What next steps do you recommend?

I just read thru the email thread between you and Jon, that he forwarded me to help figure out your problem. I noticed this attachment, which shows the color/density of your 21 step strip measurements. This visually shows the same thing as your linearizations- dark highlights then regular mid-tones to shadow range. I have again reviewed your ink separation, and edited it to make your paper white match mine for better comparison, and other than your yellow shade 7 don’t see anything really obvious that would cause such dark highlights…

How old is this R3000 printer? For some reason, yellow is always the worst for color staining. It’s worse with an older printer, as I think the yellow pigment particles collect in the nooks and crannies of the damper. When you flush or print, the surface yellow gets flushed, then more releases while the printer sits, which causes highlights to print yellow again. Without flushing the printer’s internal ink lines with PiezoFlush, the only sure way to eliminate color staining is to replace the ink selector unit (dampers), otherwise, printing yellow purge sheets or doing cleaning cycles to get rid of the yellow before each printing session until it no longer appears is the solution.

Please check your QTR version (the version listed at the top of the window is not necessarily the version #). To do this, open QTRgui, then select Help and choose About. The window that opens will list different versions for different parts of QTR, please let me know what version is listed for “Quad to Printer”.
Have you deleted and reinstalled your print driver and QuadTone RIP?

Another thing I would check is ink in the carts, removing the printer and printing process/settings from the equation. To do this, get some Q-tips/cotton swabs, and a piece of coated inkjet paper (a scrap piece from a bad print is fine). Shake your carts well, then line them up from shade 1 to 7 (black to yellow). Dip the end of a Q-tip so it’s wet with ink, then gently wipe it across the paper surface to make an ink smear. Mark the smear with the cartridge position/ink it came from, then use a new Q-tip to test the next ink, and continue until you have a sample smear of all inks. I have included an example below for you to see what I’m talking about.

Please let me know, thanks~ Dana

1 Like

It’s not an old R3000. Maybe 18 months. I had a serious head block problem just out of warranty and Epson wanted to charge me almost the full value of the printer to repair it. After doing some research, I bought some Magic Bullet and sorted it out myself. I then had a problem with the front loader that I wasn’t able to resolve so that’s when I decided to switch to a dedicated B&W printer (having purchased a 3880 in the meantime).

QuadToPrinter is QuadTone RIP is a recent install. I don’t see what deleting the print driver is going to do in this instance.

Good idea to check the ink in the carts. Presumably the cotton buds are dipped into the refill hole at the top of the carts? I’ll do this tomorrow and let you see the results.



Thanks for the additional information Phil.

Please let me know what you discover after doing the Q-tip/cotton bud ink smear test.

I have seen the Epson print driver cause very strange things to happen, so when faced with unusual problems like this, it’s good to start fresh with everything, including the drivers.

Please send me the exact 21 strip image that you printed to get the linearizations, as well as a jpeg of the waterfall image you posted print examples of for me to review. You can email them to me at

I’ll wait to hear from you, thanks~ Dana :slight_smile:

Hi Phil~

I’m just checking in to see what you discovered after doing the Q-tip ink smear test, if you have questions, or there’s anything else I can help you with.

Best regards~ Dana

Hi Dana,

Apologies for not getting back to you sooner. I didn’t get an opportunity to do the test until today.

Q-tip smears.jpeg is in the dropbox folder. I’m not sure what to make of it.

Not related to my particular issue I don’t think but Light Black has magenta in it. I don’t see how that is possible.