dE Tolerance for Pantone solids

[Prepper]

Hi all,

We publish a library of books with a different PMS background color for each year. We have 14 PMS colors picked for this and have printed them now for 11 years. When we started out we didn't have a spectro to measure them with and over the years colors have drifted thru new ink companies, papers, etc. and just plain on press variations which seem to be worse lately. We now have a spectro and can measure the Lab values of these inks and have a "good" reference point to use going forward.

My question is, I know the press printing tolerance for cmyk solids for Gracol is dE of 5, what would you suggest for press tolerance of Pantone solids?

Most all our colors are darker shades. Some of our main PMS #s are 5605, 5255, 229, 140.

Any suggestions on what would be an acceptable dE range for these on press?

Thanks

 

[gordo]

How are the Spot colors specified? How are they proofed? What is your "good" reference.
(Measuring deltaE is not tolerancing)

Best gordo

 

[Point918]

We have dozens of spot colors that we try to hold to dE 3 and usually achieve this value. One very particular customer requested dE of 1. We objected stating dE 5 was more of an industry standard so settled on dE 3.
We verify this with a visual check of an ink drawdowns, that are stored in a dark climate controlled area when not on press, and of course with standardized stored spectro readings that are recorded and reported (if customer requests) throughout the run.
When we changed ink suppliers we required the new vendor to match ALL existing drawdowns, spectro values and visually, as jobs came up for re runs.
With these packages on retail shelves next to one another and not necessarily from the same run customer tolerances are understandably more stringent.
Sheetfed offset packaging shop here.

 

[Prepper]

Well Gordo, they aren't spec'd or proofed exactly. Long version of long story...to start with, we designed and print these in-house for ourselves...The colors were picked from swatch books originally, and the design ended up being that we lay down the solid dark pms, then it's overprinted with a very light colored version of that pms, 100% coverage, just to give a hint of a shade in a duotone photo on the back cover and to put a bit of tone down where the dark pms is a vignette and fades to 50% at the top of the book covers. I know, I know, one of the most difficult designs we could have come up with, but its what we do. Result is a 6-color job, 4-color images, 2 pms inks, reversed type out of all 6 colors, aqueous gloss coating over all. To make it even more complicated, the dark pms background color for a particular year, when print our foreign translations, gets printed on a tab on the spine of those books to denote the volume year that one is from. Sorry but you asked, well maybe not for all of this. The quality is good and it is a nice looking book. We have done this now, like I said for 11 years, about 55 languages totaling around 7 million books a year.

We had no proofing system to speak of up until the first of this year when we purchased an Epson 7890 with EFI ColorProof rip. Now we can produce verified proofs but our dark pms background isn't exactly a perfect match to the swatch in the Pantone books and is overprinted by a very light version of itself. I guess I could take the Lab readings from press sheets and use those numbers for proofing?

I'm in prepress, I have an i1 spectro that I can measure colors with to give us a reference to establish a baseline going forward. The press dept. has never measured the inks or printed sheets to check that what they're printing lines up with what's been done before, they tend to just run everything to the same densities they did before and whatever comes out is what it is, this has worked pretty well amazingly but variations in the dark pms seem to be getting larger, too light to too dark. It of course shows up in the bindery when they cut, fold and collate and see it all and causes friction then between the departments and I would like to help them and at least have something that I can go by when asked why this variation is happening. Some of it is normal variation and some isn't, I would like to be able to take up for them if it's okay and help them figure out what is going on if it isn't close enough.

So...my question, what would you suggest as a press tolerance for variation on solid PMS colors?

Cover sample attached

Thanks

 

[Prepper]

Yes Point918, sounds like exactly what I'm wanting to do here, establish a baseline and keep it as constant as we can. These books end up in our libraries and offices around the world, from 10s to thousands together on shelves and we want to keep them as close as we can but still know the limits of the process and understand what is and isn't acceptable, which is hard when you're aiming for perfection, you have to realize the printing process falls way short of that sometimes, even when you do everything right.

Thanks

 

[Kaoticor]

Remember as well that certain colors you will notice a difference more than others as far as Delta error goes. I'll let the others comment on the particular details and the proofing aspect but here is what we have adopted in our shop:

Delta 3 to 5 - Not acceptable on spot colors, possibly acceptable on solid patches for 4 color work. Depends on the artwork.
Delta 2.5 - Acceptable for spot colors. Depending on the color we might stretch it to a 3, but never over 3.

Sometimes, when it comes down to it, we still have to make a judgement call, despite what dE is. But for us personally, we just dont seem to like the results once you start getting
over a dE of 3. That has seemed to be the best "baseline" for us personally. Mind you, I'm not necessarily recommending anything here, just letting you know what we have done here. We are faced with the same challenge: that our books end up in libraries next to each other (along with any color variation!)

 

[chevalier]

<3dE spot color accuracy is considered "acceptable" at many (most?) North American packaging facilities. Note that metallics are basically irrelevant when measured this way.

Remember as well that certain colors you will notice a difference more than others as far as Delta error goes.

I'd add that in many cases certain colors can have comparatively higher dE values between reference and sample before they are "visually" different.

Sometimes, when it comes down to it, we still have to make a judgement call, despite what dE is

Unfortunately this is very true for even those of us with in-plant ink shops and the latest color control and measuring technology.

 

[Stephen Marsh]

What dE method is going to be used?

Is the spectro uv cut or unfiltered?


Stephen Marsh

 

[Prepper]

1976 most likely and its not uv cut, although I am hoping to upgrade to i1 pro2 this month which is supposed to do both I think?

Why do you ask? One method or uv type better than another for measuring spot colors?

 

[gordo]

@Prepper,

Before you can establish a press tolerance for variation on solid PMS colors - especially deltaE - you need to establish an effective workflow for dealing with spot colors and understand the points of failure.

Based on your workflow description, very briefly, I'll lay out what many offset packaging people do as they have basically the same problem as you (spot color specification and tolerancing).

There are two issues you need to deal with:
1- specification of color hue/establishing the color target
2- defining an acceptable variation

There are two areas in which this is done prepress and pressroom. They don't necessarily deal with the issues the same way.

Here's the process.
Prepress:
1- Prepress identifies the hue of the target PMS color using CIEL*a*b* coordinate values. This should be derived by measuring ink draw downs on your preferred substrate of the spot color mixed according to Pantone's formula. Since you aqueous coat you will need to coat the draw downs since aqueous coating may shift the ink hue.
2- You need to specify and standardize the base inks that are used in the spot color mix.
3- You measure the hue of the draw down and that is your CIEL*a*b* color target.
4- If you have the ability to run higher and lower densities with your ink drawdowns the do so. You need to establish the high/low solid ink density range that delivers acceptable color. Alternatively you can use process ink SID tolerances ( +/- 0.10 )
5- This material is then documented, archived and used to provide color targets for the pressroom (i.e. your own spot color swatches for the pressroom).

Pressroom:
The press room does NOT use CIEL*a*b* values nor does it use deltaE for production. This is because with a spot color the press cannot shift hue - it can only shift density.
1- The pressroom confirms the solid ink density value by running the spot color to the the ink drawdown density. If hi/low SID values were provided then that is the range within which the press is to be held.
2- The pressroom may also create hi/low SID range samples during makeready for future use.

If that is done then the press operator can make a visual comparison between the drawdown and the presswork to determine the match. If there is no visual match then it is likely that the spot color ink has been contaminated in the ink trough due to an ineffective washup or that the ink was incorrectly mixed in the first place. A spectro can be used to determine the cause of the mismatch and the ink trough can either be emptied and cleaned again or the ink batch remixed.

best, gordo

 

[Stephen Marsh]

1976 most likely and its not uv cut, although I am hoping to upgrade to i1 pro2 this month which is supposed to do both I think?

Why do you ask? One method or uv type better than another for measuring spot colors?

The reason that I ask is that dE really needs to be qualified. What dE is being discussed? dE76=5.399 vs. dECIE2000=3.713 for the same colour comparison. dE76 is not "weighted" for the colour being measured, for example for neutral grays, one needs a low dE76 value (<1.5*), while for primary colours one can get away with a higher value as acceptable tolerance (<5*) [*as one finds for say Fogra 39 colour bar confirmation/validation].

Just curious. I ask about the spectro as I see a lot of UV i1 pro units out there, and when folk wish to measure against say Fogra39 CMYK they don't like to hear that they should have an uncut device!


Stephen Marsh

 

[Prepper]

In working on this and actually doing it for the first PMS color job we ran, they printed me a range of densities from 30 low to 30 high in 5 point increments, I measured them all, checked each for dE against the reference color and found that to stay within a dE of 3 works out to be a range of 10+\-. Our target density has been 178 for this color so 168-188 will be our new production tolerance.
Also thought while doing this, why not do that for the CMYK inks also? We use the G7 method and previously just ran solids to a certain density where the Lab color was good and then used our normal tolerance of +\- 10 on press. Why not just run a density range for these colors and establish a production tolerance that will maintain a dE of <5?

Any thoughts?
Thanks

 

[gordo]

In working on this and actually doing it for the first PMS color job we ran, they printed me a range of densities from 30 low to 30 high in 5 point increments, I measured them all, checked each for dE against the reference color and found that to stay within a dE of 3 works out to be a range of 10+\-. Our target density has been 178 for this color so 168-188 will be our new production tolerance.
Also thought while doing this, why not do that for the CMYK inks also? We use the G7 method and previously just ran solids to a certain density where the Lab color was good and then used our normal tolerance of +\- 10 on press. Why not just run a density range for these colors and establish a production tolerance that will maintain a dE of <5?

Any thoughts?
Thanks

Because you'd be doing it the good old-fashioned way rather than the new-fangled fancy-pants you need a certified consultant way.

LOL

Best, gordo

 

[meddington]

Why not just run a density range for these colors and establish a production tolerance that will maintain a dE of <5?

You might find a production tolerance range to maintain a delta Eab of 5 (alone) would give you a very wide density window for process work...often too wide to maintain other metrics (secondaries, dot gain, graybalance, etc). I personally feel a tolerance of +/- 0.10 is fairly wide, but may be ok for solid spots.

 

[Prepper]

I personally feel a tolerance of +/- 0.10 is fairly wide, but may be ok for solid spots.

Ok, me too on the +\-.10. I was just saying that in reference to G7 and Lab colors of CMYK being less than dE of 5, isn't that what it says? If you're trying to establish a density baseline and then a production tolerance, you can find that density that gets you closest to ideal Lab color and then run a range to find what tolerance would keep you <5 dE, while keeping an eye on the secondaries also. At press that is the main measurement they are watching is density. What would you recommend as a production density tolerance for G7 process since we have no spectro at our press, just a regular scanning densitometer?

On tolerances, we print for ourselves, basically an inplant, so trying to establish some tolerances that would be a save\toss for sheets coming off the press. Currently we have no set tolerances for that, if during the runs the solids vary up or down 15-25 points it still just all gets saved and sent to bindery. Do most places set a hard tolerance or is it all saved and up to the customer to see if he finds a problem in the job or not? That's kind of where we are now and being our own customer there's really no one to say I'm not paying for this, so there's no accountability, other than there should be some pride in what you're producing.

Maybe posing the question this way, and I know it will probably be a vastly different answer depending on whether you are in prepress or are a pressman, how much variation would you consider acceptable, or in other words, you'll pay for as a print buyer without asking for a credit or reprinting?

Personally I don't want to be trying to just stay within "acceptable", I'd like our tolerances to be better than that, but realizing there are rather large inherent variables to the printing process, it seems we could still find a line somewhere that we could determine is our save\toss for sheets.

 

[meddington]

Ok, me too on the +\-.10. I was just saying that in reference to G7 and Lab colors of CMYK being less than dE of 5, isn't that what it says?

Yes, but keep in mind there is also a requirement (per ISO12647-2 upon which G7 is based) for a delta H of less than 2.5, to minimize the difference in hue, which isn't necessarily accounted for in a +/- 0.10 density tolerance. Moreover though, delta E is a poor metric for a press operator to run toward. G7 had previously shied away from defining production tolerances as its really a process agnostic methodology, and one set of tolerances does not fit all processes. I prefer to determine tolerances based on what the process is capable of maintaining...sort of a “6 sigma” approach based on a press test or series of press tests of an individual process where the goal is to determine the tolerances themselves based on the capability of the process. There are consultants and software that really facilitate this. There’s no sense in designing tolerances that are either too loose or too tight.