Old School vs New School

Ask or answer questions related to the flexo prepress environment.

Moderator: Frank Burgos

Old School vs New School

Postby Red_Right_Arm » Mon Nov 14, 2016 1:05 am

I run a prepress department and have been getting a lot of backlash for inaccurate plates lately. It seems to me that there is a disconnect between three upper managers and myself as to what is the correct way to create accurate plates for press. The three upper managers come from 35 years of experience printing in flexo. They come from a time before computers, RIPs, or digital ∆E measurements. All they know of flexo printing is that you have absolutely no idea how a job will run until you get the plates on press and begin seeing what is coming out at the end. From there you evaluate the results and decide if you want to change your anilox, impression, ink density, or line screen on the plates. But until you start printing that live job everything is an unknown and it is the pressmen's job to monkey with the press until it starts to look correct. And so I will describe two scenarios below. I'd love to know A) If I'm wrong, and if so where do I go astray. B) If I am correct, how have people overcome this in other companies.

Scenario 1 (From the three upper managers)
- Digital proof is approved.
- A "color accurate proof" is printed. This proof runs through one set of color management that is the same for all substrates on all presses.
- The digital file is sent to the RIP.
- The RIP has one generic dot gain curve that is meant to be good enough for most substrates on most jobs.
- The plates are made using this curve.
- The plates go on press and begins producing labels to be evaluated.
- Based on the labels coming off the press, the pressman will change anilox rolls, ink densities, impression settings, and even different line screens on the plates.
- Once the various adjustments are made the pressman evaluates if they are making the labels look better or worse compared to the proof.
- After some combination works, the labels are measured to be within 2.0 ∆E of the proof or the Pantone book, depending on the colors.
- Prepress's job is to instinctively tweak the dot gain curve in order to anticipate the issues they are going to have on press.
- At times Prepress ought to know to make a variety of plates at a variety of line screens with a variety of dot gain curve adjustments so that the pressmen have options to try out as they are setting up the job.
- Color management can't possibly work for what we do because there is no way to know ahead of time which anilox, density, line screen, or impression will work to produce this job correctly.

Scenario 2 (The one I keep saying we should be using)
- Fingerprint all the presses on our most popular substrate, which should get us 80% of the way there.
- Create different generic profiles for each press based on these fingerprints.
- Decide which anilox to use first. Adjusting it is a last resort, not a first move.
- Keep consistent ink densities. Adjusting it in a controlled way is a last resort, not a first move.
- Decide which line screen to use on the plates first. Adjusting it is a last resort, not a first move.
- The pressman can play with the impression to see if it makes things better or worse, but that's about it.
- Allot time into the workflow for measurements and adjustments and a new set of plates.
- After taking measurements from the produced labels, the dot gain curve is duplicated and that's the one that gets tweaked for this specific job. The tweaks are based on the dot gain measurements coming off the press for this configuration.
- The adjusted plates are made and should run more accurately on press this time.
- If not, changing the anilox, ink density, or plate line screen still shouldn't be done.
- Measurements are taken, the plate is adjusted farther, and a new set of plates made.
- Repeat as many times as needed until new plates resolve the issue.
- If new plates aren't helping, then start messing with the anilox, densities, impression, or line screen.
- It means that every new job will need a new set of adjusted plates at first. But until we decide to do extensive fingerprinting for each variable on each press this is as good as it'll get.
Posts: 1
Joined: Fri Mar 04, 2016 10:54 pm

Re: Old School vs New School

Postby Fuzz » Tue Nov 15, 2016 12:31 pm

Need to finger print to give prepress AND press any kind of shot at getting it right. Sounds like all of you are making this more difficult than it needs to be.
Posts: 354
Joined: Mon Jan 19, 2009 4:37 pm

Re: Old School vs New School

Postby wdustin1 » Thu Nov 17, 2016 4:01 pm

The way we do it is we put the best pressman on the typical press and set the curve for all the presses based upon that one operator. I'm not sure how other places operate but i cannot schedule the same jobs for the same presses and be able to keep track of which plates and which person set up that fingerprint because our pressmen move from press to press a lot. I also can't see the benefit of making adjustments to the curve for every single job and having a new set a plates made based upon how the job looked. We did 3 separate finger print trails and set the curves for every job based on that and we haven't had much issue with any process jobs, we did one on a film and one on a paper, the operators do have to change anilox from time to time but not very often. I do agree with you on trying to control the processes and and there are some things you can do the keep it controlled, we measure dot gains and density on all process jobs and keep them the same, we also make sure that all the operators set up the jobs the same way. For example, everybody has to follow C,M,Y,K, in that order, but i also think that flexo press operating is an art more than a science and you do have to depend a lot on the operators instincts and knowledge.

I don't understand why you have different line screens for different jobs, we use one line screen for everything, we don't even present changing the line screen as an option. Have you found that changing them helps in some way? I'm curious as to why you do that.
Posts: 63
Joined: Wed Oct 02, 2013 12:54 pm

Re: Old School vs New School

Postby richflexo » Wed Nov 23, 2016 1:44 pm

I feel your pain! Scenario 1 is all to familiar with me. What kind of colour management tools and software are they basing your generic curve on and how long ago was it done? What processes have changed since then and are they keeping up with the maintenance of the cylinder, anilox, bearings/bushings etc. Scenario 1 works great if you have a great track record and keep your machines in pristine condition but its not like that. Furthermore, how often are you doing a QI on your plate making equipment? When was the last time you did a stain test on your plates or a focus test? Are your exposure bulbs at the end of their cycle? How are your curing lamps? Solvent nice and clean? All this places a huge part no matter what scenario you are up against.

In a perfect world, your scenario is the way to go but it sounds like you face greater challenges with your production team for not planning properly. As the prepress manager perhaps you could, if you don't already put a SOP before the plates go down. You will come off as a huge prick at first but unless all the proper information isn't given to you in the beginning, how are you supposed to send down plates and artwork that preform properly? Set up new protocols and if you're the one in charge and taking all the flak, its your job to make them follow your procedures. THEN if they fail, you have some explaining to do. But you sound like you know what you're doing and if you do it properly you will come up on top and your jobs will look great and next thing you know your pressman will be buying you rounds at the pub after work!

Good luck!

PM me if you need any help with anything.
Posts: 87
Joined: Fri Jun 01, 2012 3:01 pm

Re: Old School vs New School

Postby guy de cafmeyer » Thu Nov 24, 2016 2:30 am

I am from a repro house , here in South Africa ;
This is how we do it here :
Using the best printer , we made a digital dot proof , of the fingerprint plates .
We had the printers' ink supplier match the solid shade and densities of our Epsom proof printer , at the printers' preferred viscosity .
Supplied uncompensated plates in "generic" screen ruling , with pics and step wedges .
From print result , we then re-supply fingerprint to (now) match proof .
From this print , we then adjusted our Epsom profile to suit result .
Now our proofs are matched within 95% on the first pull .
This is a very basic summary , but is essentially how we do it . . .
guy de cafmeyer
Posts: 72
Joined: Tue Apr 14, 2015 3:13 am

Re: Old School vs New School

Postby Acmegv » Fri Jun 02, 2017 10:43 pm

You are definitely on the right track.
If it were me this would be my approach and it is affective.
I would take a step back and look at what anilox standards you would prefer to use for profiling. Recommend they all be the same LPI and BCM.
Then I would run an optimization of each press using same anilox LPI and BCM single color (black). Evaluate samples and see if presses are printing the same. If not adjust anilox tooling to get them to match.
Build your curve at this point.
Once they all match run your fingerprint on each machine. Measure and evaluate fingerprints they should be very similar to each other.
Now this is the part nobody wants to do because we want to see results but it is important.
Go back into press with the same setup and run the fingerprints again.
Measure and evaluate
Compare all runs. Were you repeatable?
If you have a repeatable process now take your fingerprints and average them together and create your ICC profile.

Your thinking this guy is crazy we can't do this for every substrate. Well you don't have to. Do this on a substrate that fit majority of your work and use profiling tools to adjust the profile for your other substrates. Then run a single fingerprint on a press measure and evaluate back to the profile. They should match within reason.

Good luck!!
Posts: 6
Joined: Wed May 31, 2017 9:55 pm

Return to Prepress

Who is online

Users browsing this forum: No registered users and 2 guests