Grays are reddish

[renaldas22]

I am using colour-science.com test image for visual printer and monitor check. Few weeks ago I got satisfactory results. Today I repeated the same thing and I get totally different result. Colors are shifted, grays are reddish.
Now details.
Image dowloaded from:
http://www.colour-science.com/quali...es/Reference Print printer 2362x3543pixel.jpg
Opened in Photoshop, assigned Adobe RGB profile, saved as tiff.
Printed with Flexi8 on XC 540, on orajet 3164g calendered vinyl. At that time I had no appropriate profile so I used the one witch looked right (for Glosy calendered vinyl). I made notes of settings I used. Image looked OK than. Nozzle check was OK.
Today I decided to use right profile for orajet 3164g vinyl. So I downloaded it from oracal.com and installed as instructed in pdf manual. Printing the same image from Flexi8 with only difference - this new profile. I get reddish image.
Than I printed bunch of the same image using different profiles - all reddish.
Than what confused me finally was that I get reddish image even if I repeated all settings as I printed before (weeks ago).
Nozzle check is OK. Does temperature settings can affect colors? But I am sure I did not changed temperatures. I am totally clueless.

Any ideas?

 

[Rooster]

Temperature and humidity can and will affect print quality and color reproduction. It doesn't necessarily have to be the heat settings on your machine either. Ambient temp and especially humidity need to be controlled for accurate results.

 

[eye4clr]

Who makes a color test file with no embedded profile?

A few years ago, I was having BIG color shifts in paneled print jobs. I was going crazy relinearizing and even reprofiling for every other tile. Not much was really keeping things tight. Then I started to track temperature and humidity changes by writing down the values when each tile was printed. We found that our shop has a big temp and humidity shift in the early afternoon when the sun hits our west wall and changes everything.

I was getting delta E shifts of 10-12 within a 20 minute time frame. The rest of the day things stayed very stable.

So yes, temp especially can shift color. Not sure how much humidity plays a role with solvent printing, but temp sure does.

 

[sjm]

I would trust the image as the grays are not affected even though there is no profile attached.

You can determine that for yourself by not colour managing the image and then by assigning the Adobe RGB profile.

I would look to other variables that may have changed, for example environmental conditions and also never rule out that from batch to batch vinyls will vary in consistency effecting results.

 

[eye4clr]

sjm! That was a completely lucid, on topic post that actually contributed something.

Please allow me to be the first to congratulate and thank you.

 

[sjm]

eye4clr, I am flattered. Trust me I too give credit when credit is due. Cheers!

 

[renaldas22]

After two days random test prints I found a solution (I hope).
Things become right when I changed linearization values. Instead of 85,85,70,85 I used 70,50,50,70. At the same time I changed temperature settings from 40,40,50 to 45,45,50. Grey look more natural now and whole image looks better. Thanks for suggestions.

 

[Neil]

Instead of 85,85,70,85 I used 70,50,50,70.

Me no like.

Why did you take the axe to all the colours?
You could just pull the magenta back a tad, leaving the others alone.

 

[sjm]

That's the beauty Neil, when you use a stock profile, there's no right or wrong. What you think works for you doesn't work for some else. i.e. no two printers are the same nor operational envirnoments.

Ergo, that's why a $2,000-$3,000 investment in a profiling packaging makes sense. Though I regress you can play with CMYK values all you want until you hit the sweet spot.

 

[renaldas22]

Me no like.

Why did you take the axe to all the colours?
You could just pull the magenta back a tad, leaving the others alone.

Because the image is better now. Even better than the one which I considered OK weeks ago. And its lighter.

Sad that I had to find the solution in such a dumb way (randomly changing settings and printing the same image). I doubt the company will ever consider buying any profiling package. They were and are happy with XC540 performance.

I just recently begun to work in sign business. I had printing experience before. What I figured out - these guys at our sign shop are not very concern about colors as long as the image looks acceptable. They used to match custom colors by randomly printing series of color swatches to get closest one. They used the same output profile for every media

And I found the Lm and Lc cartridges empty. They said these are sitting in the printer for ever. And XC540 didnt tell these are empty. I figured it when begun to troubleshot why nozzle check shows no half Lm and half Lc nozzles. There was no ink in cartridges!!! Thats why.

Changed the cartridges and Lm was cured. But with Lc was long story. Had to suck some ink with syringe, which temporally helped. Two weeks later had to change cap top and even damper. Anybody needs advice how to do this - just ask. Its not so hard and not expensive. Finally I got all nozzles up and firing.

XC540 was not cuting sheet off. Fixed this issue by limit initialization.

Lots of stuff "how to" I found online. But still there is a lot to figure out, to learn. But I am getting there.

Does anybody knows the meaning of "dot type" whitch is 1 to 5 in "drivers settings" under "Advanced" tab when printing to XC540 from flexi8.

:signs101:

 

[Neil]

Because the image is better now. Even better than the one which I considered OK weeks ago. And its lighter.

It sounds like you needed to make a global ink restriction first - or whatever they call it in your package. Max ink, density, ink limit etc...
To reduce the total amount of ink laid down to suit the media better.

Then, once you've done that (which won't affect the linearization curves), you could reduce just the magenta only - to compensate for the reddishness if needed.

Probably similar results in the end but you won't be changing the original linearization curve data that the profile was based on.

 

[Rooster]

That's a big adjustment. Just out of curiosity, how much time and materials did it take to finally arrive at these settings?

 

[renaldas22]

That's a big adjustment. Just out of curiosity, how much time and materials did it take to finally arrive at these settings?

:Big Laugh

spare time and left over space.

I know that it was dumb way . But it worked. Now I know the issue is right profile. Will work on it (in spare time).
I noticed Flexi Production manager said this new orajet profile I imported is not set jet. It has no linearization curves and etc. Can I adjust profile just visually. We dont have any profiling devices.

I believe with some help :signs101: I'll be able to get this printer print right.