Press Tending and CTP Curve compensation

[AngusPady]

I am tasked with establishing a common colour appearance for 5 presses. As each press is added we start with a G7 press run. This is establishes the presses starting point. We have CTP curves for each press. I'm very familiar with the G7 process and NPDC's. I am fortunate that we only have 2 medias, this is an offset packaging environment. What I am looking for is a process to monitor and update the curves over time. I don't have data collection at the press as all the presses are density only closed loop and I only have a limited amount of real estate for colour bars. Usually I only get a 25, 50, 75, 100. One thing that I have learned is I can't use one or two press runs to change the curves. It takes a bunch to see a trend. There's no way I can put the P2P back on press.

This industry needs a SIMPLE piece of software that can take the 25, 50, 75 CMYK individually and a set of CMY's to determine grey balance. From these measurements one can average and smooth the data and recommend curves adjustments for TVI or NPDC. I don't want a cloud solution, I want to local. It needs to independent of all the big players. It should show trending and TVI spread for each of the colours from the data received. Does this exist?

Why do we spend so much time calibrating the most stable piece of equipment, the inkjet but not press calibration, it is so important and is so often neglected. Just my 3 cents.

 

[Deleted member 16349]

Why do we spend so much time calibrating the most stable piece of equipment, the inkjet but not press calibration, it is so important and is so often neglected. Just my 3 cents.

Why? Because the industry has not wanted to make the offset process consistent. There is still variation within a run and variation of printing the same screen combination within the same repeat. One can not calibrate a process that is not consistent and reasonably repeatable.

The other problem is with the method of calibration. If the goal is to get matching colours in a press or between different presses, G7 and TVI is not capable of doing that. It may come close but that is just mainly luck. Mathematically it can not predict the colours.

 

[mastegman]

I am not sure that industry doesn't want consistent press performance but there are just too many variables that are hard to control to achieve some sort of ideal state. That's why we have tolerances in the standards and specifications for different substrates. Essentially, it's much easier to control a device that squirts ink onto a substrate with extreme precision at the speed of an inkjet than it is using an offset process to spit out thousands of impressions in an hour.

 

[gordo]

[SNIP]What I am looking for is a process to monitor and update the curves over time. I don't have data collection at the press as all the presses are density only closed loop and I only have a limited amount of real estate for colour bars. Usually I only get a 25, 50, 75, 100. One thing that I have learned is I can't use one or two press runs to change the curves. It takes a bunch to see a trend. There's no way I can put the P2P back on press.

This industry needs a SIMPLE piece of software that can take the 25, 50, 75 CMYK individually and a set of CMY's to determine grey balance. From these measurements one can average and smooth the data and recommend curves adjustments for TVI or NPDC. I don't want a cloud solution, I want to local. It needs to independent of all the big players. It should show trending and TVI spread for each of the colours from the data received. Does this exist?

It exists but is still in alpha and beta testing - i.e. not ready to ship yet, but it will be coming to market.

Is that a service/solution that other Forum members would be interested in (assuming a reasonable implementation cost)?

 

[AngusPady]

Why? Because the industry has not wanted to make the offset process consistent. There is still variation within a run and variation of printing the same screen combination within the same repeat. One can not calibrate a process that is not consistent and reasonably repeatable.

The other problem is with the method of calibration. If the goal is to get matching colours in a press or between different presses, G7 and TVI is not capable of doing that. It may come close but that is just mainly luck. Mathematically it can not predict the colours.

Yes there is variation and yes it is an inconsistent device but if you do a long term trend analysis, say 30-40 measurements over a few months you will see how the press moves. I've been doing this in excel and then plotting the values. I have attached a 90 trend for the dot gain for one press. You can see that the Magenta was in a good place in around 22% but moved up 26 to 28%. This required a CTP adjustment. Once done the colours fell into line again.

Gord: I'm not sure the Beta you are talking about will do what I want. In that case they do all the analysis and then tell me what I need. I want to control the process. Assuming we are talking about the same company.

I just want a stupid simple interface that will do all the math for me. As I said, I'm doing this already in excel and it works but its time consuming and prone to human error.

Ok don't laugh, I uploaded a VERY primitive concept. Now someone with interface smarts pls build this.

 

[gordo]

Gord: I'm not sure the Beta you are talking about will do what I want. In that case they do all the analysis and then tell me what I need. I want to control the process. Assuming we are talking about the same company.

I just want a stupid simple interface that will do all the math for me. As I said, I'm doing this already in excel and it works but its time consuming and prone to human error.

Ok don't laugh, I uploaded a VERY primitive concept. Now someone with interface smarts pls build this.

Your attachment seems not to want to download/be viewed.

Yes we're talking about the same company. There will likely be more than one implementation/solution as the project progresses. I have to be fuzzy about this for obvious reasons.

 

[AngusPady]

No Title

I messed up uploading last time.

 

[Deleted member 16349]

It exists but is still in alpha and beta testing - i.e. not ready to ship yet, but it will be coming to market.

Is that a service/solution that other Forum members would be interested in (assuming a reasonable implementation cost)?

I guess I know which company you are referring to also. This is of course a development/marketing project and as such, the outcome and its costs are not estimable.

There is a double problem here. One is if the technology can be developed that will work properly and the other is if there is going to be any demand for it. Both are really difficult to determine.

The dangers in doing innovation is that there can be a tendency to underestimate the negative issues and over estimate the potential positive issues. There can be a mindset that does not want to hear negative views because they do not support the desired goals. It is a trap that innovators get into all the time but it is emotional. To reduce risk of all kinds, one has to be very aggressively realistic in looking for factors that will prevent the desired goal from being achieved. It is much better to know as early has possible if some of the assumptions one started with are not valid and the project needs reevaluation or even stopping.

Nothing wastes more money and time than a little bit of ignorance in a critical area. Thinking one is say 80% of the way in a project when one is not aware that the final 20% is not possible really means you are at 0%.

Well thank heavens that in the printing industry, really good marketing will cover up much of the inadequacies of poor innovation. If the innovation does not meet the desired goals, then one can market it like hell and make users think it is their fault that things don't work.

I have great empathy for those who work to innovate technology and knowledge. That does not mean I always agree with them but the innovation process is difficult and dangerous. It is a serious business even though it can be fun and exciting if one is on the right track. It is torture if you find out later that much was wasted for nothing. Resources can run out long before solutions and valid knowledge are obtained.

 

[gordo]

I guess I know which company you are referring to also. This is of course a development/marketing project and as such, the outcome and its costs are not estimable.

There is a double problem here. One is if the technology can be developed that will work properly and the other is if there is going to be any demand for it. Both are really difficult to determine.

The dangers in doing innovation is that there can be a tendency to underestimate the negative issues and over estimate the potential positive issues. There can be a mindset that does not want to hear negative views because they do not support the desired goals. It is a trap that innovators get into all the time but it is emotional. To reduce risk of all kinds, one has to be very aggressively realistic in looking for factors that will prevent the desired goal from being achieved. It is much better to know as early has possible if some of the assumptions one started with are not valid and the project needs reevaluation or even stopping.

Nothing wastes more money and time than a little bit of ignorance in a critical area. Thinking one is say 80% of the way in a project when one is not aware that the final 20% is not possible really means you are at 0%.

Well thank heavens that in the printing industry, really good marketing will cover up much of the inadequacies of poor innovation. If the innovation does not meet the desired goals, then one can market it like hell and make users think it is their fault that things don't work.

I have great empathy for those who work to innovate technology and knowledge. That does not mean I always agree with them but the innovation process is difficult and dangerous. It is a serious business even though it can be fun and exciting if one is on the right track. It is torture if you find out later that much was wasted for nothing. Resources can run out long before solutions and valid knowledge are obtained.

Very true.

 

[rich apollo]

Angus, would pressSIGN work for what you're wanting to do? You can monitor performance and do curve adjustments with a colorbar that is simpler than the P2P. It seems like it would do you're asking, but I may be missing something.

 

[gordo]

Angus, would pressSIGN work for what you're wanting to do? You can monitor performance and do curve adjustments with a colorbar that is simpler than the P2P. It seems like it would do you're asking, but I may be missing something.

Could you post an image of the colorbar it uses or email to me? Does one use that colorbar rather than a standard colorbar? Does it work with standard press side scanners?

 

[Cornishpastythighs]

Just built a preeSIGN colour bar for our older Heidelberg press. Has the usual suspects, 25-50-75%, Trap, Neutral gray, Solids. Also input a very basic Heidelberg colour bar containing 70% and solids for our hundreds of jobs that contain that colour bar. You can select from many different colour bar configurations. Works great with Intellitrax

 

[AngusPady]

I really like PressSign, I will revisit it on Monday. I have uploaded the three bars that are used. We use the different bars based on how much space is on the sheet. I can't use the press colour bar at the moment. The press is density only and I can't access that data. All data collection is done post press run.

In an ideal scenario I would be able to use the data from any one of the three bars. Currently I have to do the measurements in ColorPort, then I average the results in excel & then I can plot them all in a trending graph.

 

[Deleted member 16349]

I really like PressSign, I will revisit it on Monday. I have uploaded the three bars that are used. We use the different bars based on how much space is on the sheet. I can't use the press colour bar at the moment. The press is density only and I can't access that data. All data collection is done post press run.

In an ideal scenario I would be able to use the data from any one of the three bars. Currently I have to do the measurements in ColorPort, then I average the results in excel & then I can plot them all in a trending graph.

Angus, you say the press is run with closed loop density control. Just wondering, are the runs always aimed at the same density targets or do they change them sometimes?

 

[AngusPady]

Erik,

Generally they run to standard densities but will change the densities from time to time to hit a specific colour.. I don't measure any runs that are more than 4 DE from our target densities.

 

[Deleted member 16349]

Erik,

Generally they run to standard densities but will change the densities from time to time to hit a specific colour.. I don't measure any runs that are more than 4 DE from our target densities.

Are you saying that the tolerance is +/- 4 DE?

What would that be in terms of density variation that you would accept measurements?

 

[rich apollo]

gordo,
pressSIGN will use custom colorbars. The minimum requirements for a colorbar are a solid and a paper patch. You can expand from there. It has an optional interface that is exactly like a press console interface. You can build sheet-wide colorbars and use the feedback for press operation. It will work directly with some press-side scanners and will also ingest data from others that it does not drive directly (Heidelberg). It works with devices like the i1, SpectroDens, SpectroJet, eXact, Intellitrax, et cetera.

Here's a typical colorbar.
View attachment pressSIGN_CMYK_4CX_Colourbar_ISO_12647-2_Coated_White_Backing_v5_labels_below.pdf

 

[rich apollo]

Erik, the density variation equivalent of 4 ∆E (by the way, there is no +/- with ∆E) would depend on the ∆E calculation being used and the color involved. I don't know a way to answer your question satisfactorily.

 

[Deleted member 16349]

Erik, the density variation equivalent of 4 ∆E (by the way, there is no +/- with ∆E) would depend on the ∆E calculation being used and the color involved. I don't know a way to answer your question satisfactorily.

Rich, yes, very good point. +/- delta E is not a correct way to express the range.

In the simplest form of deltaE, one can think of it as a circle around the target. 4 deltaE would be the radius in Angus's case. The whole range would be represented by the diameter and for this case, that would be 8 deltaE. So one could have two points that would be acceptable and they could be at most 8 deltaE apart. Not really +/- but potentially the whole range apart.

Again you are right that one can not answer the question relating density to deltaE values. There is no absolute relationship.

In my question to Angus, I was more interested in the range of densities that were in tolerance for his specific runs. I was just trying to get a rough idea.

If tolerances for density are fairly large, one would expect at times to have printed colours that have larger ranges of deltaE since dot gain changes with changes in ink film thickness.

 

[gordo]

Erik, the density variation equivalent of 4 ∆E (by the way, there is no +/- with ∆E) would depend on the ∆E calculation being used and the color involved. I don't know a way to answer your question satisfactorily.

My soeculation is that Angus measures the Lab values and if the SIDs result in a color that is more than deltaE 4 from his target then he rejects those sheets from his evaluation. I think the GRACoL tolerance is 5 deltaE.