Best way to specify color to print vendors?

[chad.trent]

A little background first...

I've been working in the printing industry for more than 20 years - mostly in prepress in the offset sheet-fed sector. For the last year I've been at a plant that does corrugated boxes/packaging/displays/etc. We do flexo printing in house, but order out our litho sheets and top sheets. We work with multiple vendors due to size limitations, costs, etc. It's pretty common for us to order part of a job from one vendor and part from another. Obviously this is opening itself up to problems when we have to guarantee that the printing we get from vendor A matches the printing we get from vendor B.

A lot of our work is spot color or CMYK+spot. Spot colors have been our biggest hurdle. So, our procedure as of now is to specify a PMS color and then also provide LAB values for that color and tell the vendors we expect all printed material to be within 2 delta E of the lab values we provide. We also spec the paper we want our jobs printed on.

So far that has worked, but we are hearing now that LAB values are calculated differently by different devices, and that if the spectro we use is not the exact same as the printer uses, there is no guarantee we are going to be getting the same color.

We have been researching the best way to specify color for over a year, and we can't seem to find good information on the best way to do it. We have talked to three or four "experts" and received three or four completely different theories as to how to do it, and each person told us why their method was correct and the other methods were flawed.

So, I'm trying to find out what is the best way to guarantee that a certain PMS color from one vendor is going to be the same from another vendor and what is the best way to communicate that to our vendors so that we can get consistent color.

 

[Alith7]

leading with this - I am NOT an expert.

that said, being prepress manager for a mid-size sheet fed shop that is frequently at the receiving end of those orders here is my 2 cents.

the first and most obvious answer is of course to use the same printer. I know that that is not always possible, so here are a few suggestions.
1. provide a color swatch for matching to both printers. Having a specific visual that YOU know matches should help. Or, depending on turn time, provide a press sheet from one printer to the other. edit add: (I know they seem to be outdated, but Pantone's chip books were awesome for this! just REALLY expensive. But this situation is what they were designed for)

2. LAB can be inconsistent, as can density and most other "digital" methods of matching. it all depends on calibration, the equipment, and many other variables. Not to mention different presses lay down ink differently and what one press needs to run for density may not be the same as a different press needs to achieve the same color. That being said, LAB will get you closer than most other methods.

3. Make sure supplies are the same. Ink, paper brand & type. Different ink vendors, or different paper can make a HUGE difference. Most printers should honor your request for a specific paper type, but it may be more expensive from one printer depending on their negotiated prices from their supplier. As an option, supply the paper yourself. it will save you on mark-up for the stock, and then you KNOW the papers will be consistent.

that's all I have. Mainly, try to limit the number of printers you are out sourcing to, and if possible, provide a color match sample. To my knowledge, there is no fool proof "system" out there to guarantee that things will match.

 

[gordo]

We have been researching the best way to specify color for over a year, and we can't seem to find good information on the best way to do it. We have talked to three or four "experts" and received three or four completely different theories as to how to do it, and each person told us why their method was correct and the other methods were flawed.

So, I'm trying to find out what is the best way to guarantee that a certain PMS color from one vendor is going to be the same from another vendor and what is the best way to communicate that to our vendors so that we can get consistent color.

The traditional way (and IMHO the best) is to supply a draw down - i.e. a physical sample of the ink on the substrate. Anything else is an abstraction and problematic. Tolerance for deviation is provided by high and low solid density samples - because solid ink density - not L*a*b* values - is the only control that the press operator has. So, three sample patches; the target color, acceptable maximum lightness and maximum darkness from that target.
If you can specify the ink supplier (and formula) to use then you eliminate ink spectral differences.

gordo

 

[Point918]

They way we deal with this is to get an ink drawdown on the stock the job is running on with the coating or varnish on top. This is approved by the customer. This becomes the visual standard that the press must match.

PMS books & Lab values are nice tools but can and do result in a visual mismatch. We see it all the time. You are not selling Lab values or PMS swatches. Substrate and coatings can have a huge affect on color. Of course 5k lighting in proper environments is essential when making these judgements. Remember color is: Object, Observer & Illuminate. Change any one and the color changes. Good luck with the Observer..... ;-)

 

[Alith7]

...You are not selling Lab values or PMS swatches. ...

Ha!! forever ago when I ran press the catch phrase around the shop was we don't sell trim marks... sorry, made me laugh.

 

[chad.trent]

The problem we have is that if we get two different drawdowns from two different printers, they can be off from each other. Say drawdown A is darker than drawdown B. Now when the press is running, even if they stay within 2 delta E of the drawdown, if printer A is on the darker side of the drawdown and printer B is on the lighter side, we have a possible difference of 4 delta E in the color, which won't work. We have had this exact situation happen in the past running a job in Warm Gray 11, and the two orders were nowhere near the same color and had to be rejected.

 

[Alith7]

The problem we have is that if we get two different drawdowns from two different printers, they can be off from each other. Say drawdown A is darker than drawdown B. Now when the press is running, even if they stay within 2 delta E of the drawdown, if printer A is on the darker side of the drawdown and printer B is on the lighter side, we have a possible difference of 4 delta E in the color, which won't work. We have had this exact situation happen in the past running a job in Warm Gray 11, and the two orders were nowhere near the same color and had to be rejected.

I believe the suggestion was to provide the printers with draw downs that YOU have made, not them.

However, in that situation, I think the best bet would be to get a press sheet from one printer and provide to the other printer for matching.

 

[chad.trent]

We don't have any way to make drawdowns. We are a 100% flexo plant, and we don't have any offset ink here.

I'd love to get a press sheet from one printer to send to another, but we generally have turn around times of 1-2 days on printing jobs, so the time just isn't there. Once we get the sheets printed, they then come back to us and are laminated onto board, then die cut, assembled, or whatever, so while we may have 2 weeks for a job, the printing can only take a couple days (including shipping) or it throws off the schedule for all the other operations involved in the job. Luckily our printers know this and are willing to work with us on it.

By the way, I realize I'm doing the "yeah, but..." thing with every suggestion, and that's not in an attempt to be argumentative. We've been trying to nail down a process for over a year, and have been through all the options we have come up with. So far, specifying LAB values, paper, and illuminant conditions seems to be working okay, but we have only been testing the process on a few jobs. We don't want to go too far down the rabbit hole with this method if it isn't going to work in the long run.

I know that some of the big companies have set color standards that they send to their print vendors. People talk about them all the time. We just haven't been able to see any of these standards to base our standards on, so we're trying to play catch up without much go to on. We'd love to get info from our print vendors on this, but they have been reluctant to supply any suggestions.

It would be nice if we had a local printer that could do this so we could run over there and work with them in person, but the closest one is three hours away, and the largest press they have is 28x40, so less than half of our jobs go to them.

 

[Point918]

Are the various printers you are using have a common ink vendor by any chance?
Would they be willing to work with a common ink vendor?

I realize many shops mix their own in house.

I guess what I am getting at is put the problem on the vendors
That's what our customers do to us!

If not you need to determine if a little more turn time is worth an exact color match.
Color matching can be a nebulous area.

 

[Alith7]

Ouch!
No problem with the "yeah, but"s...I know how it goes.

I did notice you threw one variable there that can change things. Laminating. Do you mean laminate as in the process of applying a C1S litho to a corrugated or actual plastic lamination? Because different pigment bases react differently to having laminate applied. we run into that here. Especially if all we have for a reprint is a laminated sample. Orange colors seem to be the worst for color shifting when laminating.

Have you considered going old school and getting Pantone Chip books?
Pantone - SOLID CHIPS Coated & Uncoated
if you have a few set printers you use, maybe getting a chip book and tearing out full strips and sending a set to each printer? not ideal, but then at least you know each printer has EXACTLY the same dye lot for a PMS book, and you won't have to wait for overnighting or anything since they will already have the full set.

and you probably won't get a good response from your printers. Most just see that you are sending work elsewhere and get bully. They look at it as why should they help you to send work somewhere else. It's not helpful, but its a common thought process.

From what I have gathered, the bigger companies use systems like Fuji's ColorPath Sync to make sure cross-plant printing matches. however, that requires a level of cooperation that would be hard to achieve through different companies. And, each shop has to have all the calibration equipment checked and settings adjusted to accomodate for the slightly different readings.

Here is a thought though... Get a set of spectrodensitometers from maybe Techkon, and have them calibrated so that they are all reading the same from the manufacturer, and loan them to the printers. then, if you specifiy LAB values and stock, you know that they should be reading the same.

Other than that... I'm out of ideas.

 

[Stephen Marsh]

Ha!! forever ago when I ran press the catch phrase around the shop was we don't sell trim marks... sorry, made me laugh.

There is a huge untapped market for offcuts, just waiting for a vendor. Just make sure that the trim marks are in register.


Stephen Marsh

 

[gordo]

I believe the suggestion was to provide the printers with draw downs that YOU have made, not them.

However, in that situation, I think the best bet would be to get a press sheet from one printer and provide to the other printer for matching.

You get one draw down from one printer and the other matches their ink to that. You provide info re ink brand and formula.

Gordo

 

[rich apollo]

We don't have any way to make drawdowns. We are a 100% flexo plant, and we don't have any offset ink here.

Get yourself a Little Joe press, or work with any local ink vendor. You can bribe a pressmen to do anything with beer, donuts, or cheeseburgers. For an ink guy, try scotch.

To your original post - I think that spec-ing the colors in L*a*b* is THE way to go. L*a*b* is objective and device-independent. PANTONE colors are all L*a*b* based, now. Large packaging groups are using online repositories of color data - Maxwell, Pantone Live, MatchMyColor - to communicate this information around the globe.

Different instruments don't necessarily "calculate Lab" differently, at least that's not how I would phrase it. You will run, absolutely, run into questions of inter-instrument agreement (or disagreement). Two instruments measuring the same color patch will probably come up with slightly different readings. Is it enough to say the system doesn't work? I don't think so. The instrument manufacturers work very hard to minimize the variability within devices, and AMONG devices. Folks CAN purchase traceable standards tiles to check their devices and/or send their spectros in every two years for re-certification (as is recommended). It costs a little money, but how much do you lose on a reprint?

Send a drawdown (OF YOUR OWN MAKING - I hate chip books) and now you have a common starting point. You don't care if your vendor's spectro says that gray is blue, or up is down, or that Uncle Bob is your Aunt Fannie; as long as it ALWAYS says that gray is blue, or up is down, or that Uncle Bob is Aunt Fannie.

I say specify your colors in L*a*b*, specify your tolerance(s) in ∆E(2000) and hammer out a demonstrable system of accountability with your client(s) and vendor(s).

 

[chad.trent]

Ouch!
No problem with the "yeah, but"s...I know how it goes.

I did notice you threw one variable there that can change things. Laminating. Do you mean laminate as in the process of applying a C1S litho to a corrugated or actual plastic lamination? Because different pigment bases react differently to having laminate applied. we run into that here. Especially if all we have for a reprint is a laminated sample. Orange colors seem to be the worst for color shifting when laminating.

Have you considered going old school and getting Pantone Chip books?
Pantone - SOLID CHIPS Coated & Uncoated
if you have a few set printers you use, maybe getting a chip book and tearing out full strips and sending a set to each printer? not ideal, but then at least you know each printer has EXACTLY the same dye lot for a PMS book, and you won't have to wait for overnighting or anything since they will already have the full set.

and you probably won't get a good response from your printers. Most just see that you are sending work elsewhere and get bully. They look at it as why should they help you to send work somewhere else. It's not helpful, but its a common thought process.

From what I have gathered, the bigger companies use systems like Fuji's ColorPath Sync to make sure cross-plant printing matches. however, that requires a level of cooperation that would be hard to achieve through different companies. And, each shop has to have all the calibration equipment checked and settings adjusted to accomodate for the slightly different readings.

Here is a thought though... Get a set of spectrodensitometers from maybe Techkon, and have them calibrated so that they are all reading the same from the manufacturer, and loan them to the printers. then, if you specifiy LAB values and stock, you know that they should be reading the same.

Other than that... I'm out of ideas.

I mean laminating C1S to corrugated (sometimes top sheet to single face). Our spectro is an Xrite Exact. All the printers we use use some sort of Xrite machine, but I don't know which one right off.

The PMS chip book is a good idea, until we run out of chips and have to buy a new one. Or it sits here for a year and the colors change.

Get yourself a Little Joe press, or work with any local ink vendor. You can bribe a pressmen to do anything with beer, donuts, or cheeseburgers. For an ink guy, try scotch.

To your original post - I think that spec-ing the colors in L*a*b* is THE way to go. L*a*b* is objective and device-independent. PANTONE colors are all L*a*b* based, now. Large packaging groups are using online repositories of color data - Maxwell, Pantone Live, MatchMyColor - to communicate this information around the globe.

Different instruments don't necessarily "calculate Lab" differently, at least that's not how I would phrase it. You will run, absolutely, run into questions of inter-instrument agreement (or disagreement). Two instruments measuring the same color patch will probably come up with slightly different readings. Is it enough to say the system doesn't work? I don't think so. The instrument manufacturers work very hard to minimize the variability within devices, and AMONG devices. Folks CAN purchase traceable standards tiles to check their devices and/or send their spectros in every two years for re-certification (as is recommended). It costs a little money, but how much do you lose on a reprint?

Send a drawdown (OF YOUR OWN MAKING - I hate chip books) and now you have a common starting point. You don't care if your vendor's spectro says that gray is blue, or up is down, or that Uncle Bob is your Aunt Fannie; as long as it ALWAYS says that gray is blue, or up is down, or that Uncle Bob is Aunt Fannie.

I say specify your colors in L*a*b*, specify your tolerance(s) in ∆E(2000) and hammer out a demonstrable system of accountability with your client(s) and vendor(s).

Well, we were going that way, but our flexo ink supplier (who is one of the big names in flexo ink apparently) is telling us that LAB colors are not the same from device to device. He's telling us that the different devices use different math to translate the reflectance curves into LAB values, and that different devices will return different LAB numbers for the same color. He couldn't tell us how different they would be, and he suggests we supply our print vendors with a .cxf file from our spectro that contains reflectance curves. I feel like that is over complicating things, but if that's how it needs to be done then that's what we'll do.

Unfortunately making drawdowns ourselves is not a possibility right now.

Thanks to everyone who made suggestions so far. Some good information to think about...

 

[Alith7]

good luck!

 

[chad.trent]

Yeah. I have a feeling I'm going to need it...

 

[gordo]

Snip

Well, we were going that way, but our flexo ink supplier (who is one of the big names in flexo ink apparently) is telling us that LAB colors are not the same from device to device. He's telling us that the different devices use different math to translate the reflectance curves into LAB values, and that different devices will return different LAB numbers for the same color. He couldn't tell us how different they would be, and he suggests we supply our print vendors with a .cxf file from our spectro that contains reflectance curves. I feel like that is over complicating things, but if that's how it needs to be done then that's what we'll do.

Here's more info related to what te ink tech was telling you:

The Print Guide: Top reasons why color instruments don't agree

Best, gordo

 

[Colt]

Any small or medium print shops out there making their own ink drawdowns? If so, what equipment do you use?

 

[Point918]

Our in house ink guy uses a Little Joe.

In all respect to Rich Apollo's comment "Is it enough to say the system doesn't work?"
I'll say it - IT DOES NOT WORK.

Our ink tech, pressroom manager & manufacturing manager all with over 25 years experience concur. LAB standards work some of the time. But we need it to work all of the time. (We are one facility of a large packaging company.)

Specific colors, pastels and blues are particularly bad. I am of course talking infinitesimal differences in colors to the layman but, not to our print buyers

LAB may work fine for your customers after all THEY should be defining your level of quality. Anything more is a waste of money.

 

[maas]

Not sure why you undertake this task alone, as every supplier in the chain mentioned stands to benefit by not having to deal with a disenfranchised customer get them involved, and include your ink supplier in the discussion, you may find they can supply a limited batch of ink specific for this client and process to mitigate the different lab values of the substrates, all that you need to do then is run to density, print is all about putting an ink film on paper, a press does not create colour.