Process Work

[dkelley]

Can anyone explain the process for doing process work
Such as color matching the printed sample. What do you match it to if there is no previous run (new job).

 

[mattbeals]

Approved contract proof for press guidance. Or, an in house standard of one variety or another. A lot of shops will run to standard densities and call it good.

 

[gordo]

Can anyone explain the process for doing process work
Such as color matching the printed sample. What do you match it to if there is no previous run (new job).

The presswork is aligned with (matched to) whatever proof has been agreed to by the printshop and customer. If there is no proof supplied (yikes) then the presswork is run to an industry specified print characteristic (e.g. ISO 12647-2) or the shop standard for solid ink density and you get what you get or whatever the press operator thinks is correct.

best, gordo

 

[jotterpinky]

Does the specification specify what the densities should be, e.g. a 50% halftone creates a 65% dot on the sheet? or is the standard density equal, e.g. a 50% dot in the artwork equates to a 50% dot on the finished sheet?

 

[gordo]

Does the specification specify what the densities should be, e.g. a 50% halftone creates a 65% dot on the sheet? or is the standard density equal, e.g. a 50% dot in the artwork equates to a 50% dot on the finished sheet?

There is standard ink density for the solids and there is dot gain/TVI.

The historical TVI values for a #1 paper are K 22%, C 20%, M 20% and Y 18%

Presswork is not normally linear - i.e. a 50% request in the art, or a 50% dot in the artwork, should not result in a 50% dot on the finished sheet.

best, gordo

 

[David Milisock]

In a shop that has their ducks in a row. The plate setter/press combination, is linerized and calibrated to the shop paper establishing ink densities based on the ink manufacturers recomendations and the paper. Usually an instruction set for coated and uncoated paper for each press is created.

Then the pressman will run to pre-established ink densities. It's really just that simple.

Some people want to calibrate the plate setter/press process to a standard such SWOP or Gracol using ICC profiles, others use a postscript color management, depending on the caliber of the technician either of these process can be dubious. Unfortunately in some cases even manufacturers tcechnicians can be suspect.

With an ICC system the color is converted to the ICC standard used for the press with an ICC profile controlled system and can allow the use of NON-CMYK color models in the RIP process with REASONABLE conversions from non-CMYK models to CMYK.

With a postscript color managed process the same thing happens and while the CMYK gamut of the press may be expanded slightly ONLY CMYK source color data can be processed properly

Color management is in the hands of the file creator as the press system once calibrated simply requires maintenance.

 

[gordo]

In a shop that has their ducks in a row. The plate setter/press combination, is linerized and calibrated to the shop paper establishing ink densities based on the ink manufacturers recomendations and the paper.

IMHO, in 99% of the cases there is no need to "linearize" the plates before applying a tone reproduction curve (it's a redundant step). Generally the SIDs come first since they are a print industry specification and are integral to the mechanics of the process. The ink vendor will then help the printer select the ink series that performs best at that spec.

best, gordo

 

[David Milisock]

IMHO, in 99% of the cases there is no need to "linearize" the plates before applying a tone reproduction curve (it's a redundant step). Generally the SIDs come first since they are a print industry specification and are integral to the mechanics of the process. The ink vendor will then help the printer select the ink series that performs best at that spec.

best, gordo

Maybe you and I are on the same page but reading the book in a different lanquage. What I mean by linearize the plate is running the plate setter base calibration process. This assures that the plate setter is responding to the input signal properly by verifying the plate output against a known target.

 

[TerryWyse]

What I mean by linearize the plate is running the plate setter base calibration process. This assures that the plate setter is responding to the input signal properly by verifying the plate output against a known target.

How would one know that the "plate setter is responding to the input signal properly" and what is "known target"? Seems rather ambiguous...can you be more specific?

Terry

 

[David Milisock]

How would one know that the "plate setter is responding to the input signal properly" and what is "known target"? Seems rather ambiguous...can you be more specific?

Terry

The last Heidelberg device I worked with had a target file you imaged on the plate and read it with a densitometer. How else would you know if the plate setter is operating properly?

 

[Lukas Engqvist]

There are many CTP plates that do not have a pigment that is measurable with densitometer (In the way that film was), usually the ones that are not dependant on a chemical developer. Some systems have a digital check wedge to verify laser and focus.

 

[David Milisock]

There are many CTP plates that do not have a pigment that is measurable with densitometer (In the way that film was), usually the ones that are not dependant on a chemical developer. Some systems have a digital check wedge to verify laser and focus.

I have not done any of those however one has to be able to be sure the plate setter is functioning with nominal parameters before one calibrates for color. So I would assume the device has a manual and a procedure for just such purposes. Of course in todays climate I'm sure there are technicians that just fial tocover thatbase before they waste their clients money.

 

[TheProcessIStheproduct]

Don't want to highjack thread, but this is a similar question. We are moving from a house ICC profile for proofing to an industry standard and would like to hear what people like best; SWOP, G7, GRACOL, OTHER? My main criteria is pleasing process work, easy to get to on press, and repeatability. We will be using EPSON 7900 with closed loop spectro and ORIS color tuner. This is first time we are printing to a standard, I have for years been a proponent of matching proofer to press with a custom ICC profile, but now with the new epsons and the spectro I hope to keep the proofs very constant and match press to it...

 

[TerryWyse]

The last Heidelberg device I worked with had a target file you imaged on the plate and read it with a densitometer. How else would you know if the plate setter is operating properly?

It just seemed to me that was a lot of words to simply say "check to see if the plate is properly calibrated". And, of course, a densitometer is completely the wrong instrument to use on a plate if you're wanting to accurately read dot percentages. You need a high resolution video/camera type device (CCDot, iCPlate, Techkon Spectro-Plate, etc.) to get an accurate reading....in my opinion at least.

To check if a platesetter is "operating properly" would usually take a technician that's intimately familiar with the device....most of the time a correctly exposed plate in terms of laser energy, maximum run length, etc., is decidedly non-linear. Once the platesetter is set for correct exposure, you usually need to come back with a plate linearization curve to give you the final "correct" dot percentages, assuming that a linear plate is what you want to start with.

Terry

 

[gordo]

And, of course, a densitometer is completely the wrong instrument to use on a plate if you're wanting to accurately read dot percentages. You need a high resolution video/camera type device (CCDot, iCPlate, Techkon Spectro-Plate, etc.) to get an accurate reading....in my opinion at least.

I don't know about now but in the early 2000s Heidelberg (the company David Milisock mentioned) was happy to use densitometers to measure plates. If there is enough contrast between coating and base aluminum they work quite well. (Quality In Print: Using a densitometer to measure plates)

To check if a platesetter is "operating properly" would usually take a technician that's intimately familiar with the device....most of the time a correctly exposed plate in terms of laser energy, maximum run length, etc., is decidedly non-linear.

Absolutely

Once the platesetter is set for correct exposure, you usually need to come back with a plate linearization curve to give you the final "correct" dot percentages, assuming that a linear plate is what you want to start with.

Nooooooo! You come back with a plate curve that gives you the desired tone values on press. Linearity is irrelevant.

best, gordo

 

[Alith7]

There are many CTP plates that do not have a pigment that is measurable with densitometer (In the way that film was), usually the ones that are not dependant on a chemical developer. Some systems have a digital check wedge to verify laser and focus.

That's the EXACT reason I'm hesitant to switch to the Kodak plates! how can you make sure your system is in line with specs if you can't read the dot until it's inked up and on press?!?!?!? they haven't been able to give me a good answer. Just simply "that you do".

we run a Xitron Sierra rip (upgraded from a Xenith Extreme) and this is the process we use.

1. Linearize your plate burn with the pure 0-100 test strip sent from the rip with NO calibration. check to make sure that when you have NO curve applied, the laser is imaging the correct dot. No, you don't usually want to run linear, especially given that most of our color standards are from the neg days when you would have a certain percentage of dot gain just from the neg to the plate. If things aren't reading correctly, we apply a Plate curve that applies to ALL plate of that thickness.

2. Once we've made sure that the laser is imaging correctly 25% reads 25%, 50% reads 50%, etc. with a PLATE READER, then we run the test job at linear on the press to measure the presses's added dot gain and see where things are coming out from a specific press, on a specific type of paper (coated, uncoated, cover, text, etc.) and create as the rip calls it a "Characterization" for that specific press/paper that will apply a color specific curve for each separation.

3. the last thing we do once everything is in line on presses and papers across the board is build profiles for the Epsons so that we can have a reasonable approximation of a color proof. as I tell my customers, they are only about 95% color accurate.

No, I don't think you should run linear, but you DO need to make sure that your CTP is imaging correctly, and that's the best way to do it to set up a bench line for checking. unless you're one of those lucky people with the knowledge to calibrate the CTP laser in the machine and such.

 

[Alith7]

To check if a platesetter is "operating properly" would usually take a technician that's intimately familiar with the device....most of the time a correctly exposed plate in terms of laser energy, maximum run length, etc., is decidedly non-linear.

no, they're not, and the laser DOES fluctuate with a dust spec, time wearing down the laser strength, etc. which is why, if you set the base line for just the plates to linear then calibrate per press/paper, you have a mark that you can check for and adjust without having to recalibrate EVERY profile in your system every time the laser fluctuates.

If your color curves are applied based on the CTP imaging your data linear, ie: you want to push the 50% to have a dot gain at 57%, apply the curve BEFORE you send to burn so that the CTP gets data to burn a 57% dot. now, if your CTP is burning Linear and the laser power drops from age where you're now getting a 53% dot, you can easily adjust the plate curve ONLY to bring it back up to the correct dot and all your profiles are still valid and stable.

If you just apply plate curves "on the fly" per profile, if your laser drops, you now have to re-profile EVERYTHING!

at least, that's my 2 cents worth of input.

 

[meddington]

... the laser DOES fluctuate with a dust spec, time wearing down the laser strength, etc. which is why, if you set the base line for just the plates to linear then calibrate per press/paper, you have a mark that you can check for and adjust without having to recalibrate EVERY profile in your system every time the laser fluctuates.

I understand your point and why that works for you, but I wouldn't consider fluctuation of this nature to be a characteristic of a consistent/well maintained imaging device. During the last 6 years since introducing CTP at my old company, we've had to have the laser adjusted a few times as part of maintenance, but only have have to change curves when a new plate type was introduced, and we forgo plate linearization.

Moreover, on occasions when the laser power was in question (or a chemistry issue for that matter), it resulted in banding first. In almost no occurrences did we have an instance with a significant/measurable drop in plate dot percentage that didn't first show a banding anomaly. in these instances, plates would measure fine, but visible banding would be present. I'm still a proponent of measuring plates of course, but in 6 years of doing so at that company, visual analysis caught more issues than a measurement device (which I can't name one occurrence).

 

[gordo]

@ Alith7

This has been discussed quite a bit before, but, there is no need to linearize the plate before applying a curve. It is redundant.

The tech will calibrate the CtP/plate combo for best exposure/processing..
The result is likely a non linear plate.
The non-linear plate is run on press to determine the press tone response.
From the press tone response a curve is created (based on 2 things - the requested tone in file and tone response on press)
The compensation curve is applied to when plates are imaged.

For QC, the plate is measured with curve applied.
E.g. Let's say that @ the 50% the curve should result in a 57% on the plate.
Then every time the 50% patch is measured it should be 57%.
If it changes then determine why - e.g. forgot to apply the curve, chemistry drifting, dirty laser etc.
Then decide if a change in the curve is required to bring the 50% back to 57% or whether some other correction needs to be applied.

(posted just as Meddington posted - he is correct)
best, gordon p

 

[TerryWyse]

I don't know about now but in the early 2000s Heidelberg (the company David Milisock mentioned) was happy to use densitometers to measure plates. If there is enough contrast between coating and base aluminum they work quite well. (Quality In Print: Using a densitometer to measure plates)

But that's only because that's all they had....therefore it had to work
I've been there as well....I believe a plate can be measured *consistently* with a densitometer but not *accurately*.....but that's just me. But with a densitometer, you had to deal with kludges like using the correct Yule-Neilsen Factor for a particular brand of plate....that's because a densitometer can never be a dot reader....it's a density/tone calculator that thinks it can measure dots.

Nooooooo! You come back with a plate curve that gives you the desired tone values on press. Linearity is irrelevant.

I see your "Nooooooo!" and raise you....something.

I will SORT OF agree with you (given the respect I have for you, it would be foolish to completely disagree with you!).....this was sort of a long-standing debate in the G7 world early on...to linearize prior to determining tone curves or not. My "argument" for the case of applying a plate linearization curve prior to press curves is simply one of practicality.....when the G7 Curve software reports that my 50% dot should be a 46% dot, I want to be able to measure/verify that DIRECTLY on the plate and not have to go through mental gymnastics compensating for the non-linearity that the curve calculation software knows nothing about. Of course, if I could some how TELL IT what my "raw" plate curve actually looked like, giving it the ability to compensate for this and give me the actual on-plate value, I'd be OK with that....but for now the software I'm using assumes 50%=50% from the outset.

...and no disrespect intended.

Regards,
Terry