Crazy unpredictable gain

[longlimb]

Here is the situation. We have switched to a new plate (from Sword to Electra) and I keep getting the strangest behaviors of dot gain on press. (but this was happening with the Sword as well)

I was doing some readings on press yesterday and noticed my dot gain for cyan was around 24% running a really light SID and my magenta was right where it should be at the SID we shoot for inhouse. We run staccato 20M and my plate curve is set for about 18%-19% gain on CMY

Today I was doing some readings on press again and now my cyan is right where is should be as far as dot gain and my magenta with pushed SID is at about 10% gain???

Since this is a completely different plate line I now think it's not plate related. I have been staying on top of checking my plates since I run a plate control strip on all plates and read them daily. They are not moving.

The pressmen say it's the same set of blankets as yesterday only we have fresh ink this morning.

We did a test job about a month ago when we switched to this plate and it was dead on once I built my plate curve. Hit my target gains perfectly and matched the Matchprint perfectly!

What else can cause this flux beside paper? I know paper and weather can affect this, but by this much??

the two jobs I was comparing was on the same paper only different weights?

The plant manager says there is nothing wrong with the press. But we are having the same exact issues with two completely different plate lines?? Come on...

 

[sveibl]

I have seen this happen & found it to be ink consistantcy from ink company.
Our Pressroom super agreeded & created a curve for ink. Noticed pressman bragging on "fresh Ink"

 

[longlimb]

That would make sense. I asked the Plant manager about the ink a while back and he swears it's the best ink on the market and nothing is wrong with it.

Is anyone else using Pertech Inks and having this problem?

 

[dazenmonkey]

Gain

Gain

Hi again longlimb. I assume you have created a footprint of the press and developed a plate curve based on that footprint. If so and also assuming that the dots you think you are putting on the plate are actually the dots you put on the plate, then one must assume that your house is in order and there is some problem (either mechanical or chemical) with the offending unit.
Here is where diplomacy might come in handy.
Try a large 50% screen on each unit of the press. You will see every imperfection in every unit of the press.
Measure the plate before you mount it and then the sheet after printing.
Compare this to your footprinting data and you will find the problem.
All inks seem to have some inconsistency in manufacturing but I have worked in shops for years that use Pertech with good results.
Do your homework though and make sure that your process is rock solid before proceeding.
Good luck. Hope things turn out well.
Peace.

 

[dimitri]

Check the packing height and the check roller stripes to make sure.
It's kind of tough these to change overnight however... but everything is possilbe.

 

[longlimb]

Thanks Dazen. Yeah I have verified all the mentioned. When we switched to this plate I was lucky enough to have a Kodak Color Services guy here with me and we went over everything. The dots looked good. The plate read where it should be (and continues to).

I measured the 50% screen on what I am getting now and checked it against my footprint sheet I have saved. I've found they are different, just don't know why. If we get some time I'll do the 50% screen on the whole sheet and see what that looks like.

Could it be a fountain solution incompatibility? Can anyone using the Electra XD plate tell me what solutions they are using on press? We tested everything we have and can't find any problems.

Could the gum be blinding the plate? But it wouldn't just affect one plate out of a set of 4 would it??

I'm totally scratching my head here...

I have had them check the stripes but they looked good. Packing may be something to look into...

 

[dazenmonkey]

Longlimb,
Talked with our pressroom foreman. He suggested the following.
Ink, PH, or a packing issue on the cyan unit.

Is the tac of the cyan ink you are using now the same as when you did your test?
20 mic does require special tac.
Also the PH could be a factor. Are all the units are feeding off the same batch of fountain solution or are they separate? The Ph can drastically affect the quality of 20 mic.
In addition we have seen here that if the ink gets to loose (or old) we get unusually high gains.
Also consider the packing in that unit.

Ok. I am out.
Peace.

 

[longlimb]

hi dazen-

I'm not sure how to test tac (pardon my ignorance). However, this problem appears to affect all the colors, but sporadically. The cyan was reading a really high dot gain one day with a low SID and then the next day the cyan was behaving normally and the magenta was getting a really low dot gain with a really high SID. Same lay down order.

I think you may have hit the nail on the head though about the PH. I have asked the plant manager over the years about PH and he always says PH doesn't make any difference. I could never argue though because I didn't know any better. I'm going to get something to measure the PH because they don't maintain any kind of standard when it comes to it.

Where do I get a PH meter? Or I should say, how do you measure PH?

 

[dimitri]

Stochastic plates are more sensitive and could have more variation at different locations of the sheet. This could be caused due to the exposure of the plate. It is also possible to have different amounts of deviation from the target in the midtones, 1/4tones, and 3/4tones. This would indicate a problem in the platesetter. Moving from 20 to 25 could provide more stability, at the cost of coarser dots.
Both of these comments are broad assumptions; not better to use other than as a rough rule of thumb. But in any case, since you said that you measured the plates and you are hitting the targets you're after, then how could the plates be an issue?

I like more the fountain solution reasoning or tack mentioned earlier.
One question: are you a UV printer?

 

[longlimb]

We aren't UV.

Plate readers are rather rock solid, correct? I'm using an Xrite ICPlate2 reader. About a year old.

I just have a hard time accepting it could be a plate problem because wouldn't all the plates have the issue and not just one at a time... from time to time?...

 

[dimitri]

I agree. You said though that you are incosistent on a random basis. I quote: "However, this problem appears to affect all the colors, but sporadically. The cyan was reading a really high dot gain one day with a low SID and then the next day the cyan was behaving normally and the magenta was getting a really low dot gain with a really high SID. "
We were having issues like that with stochastic 20 microns (bound to the platesetter head we were using), and we switched to 25, plus added an 'S' shaped linearization curve on the platesetter and we are now fine.
I think that that measuring device is switchable between reading FM or AM screens. This could mess you up. Make sure you measure the patch from the same angle everytime, as you place the instrument down. I've seen slight deviations based on that.

You need to measure both pH and conductivity. Usually, if conductivity is off target, pH would be off too. pH isn't very sensitive to variation and has a rather flat response, so it's more difficult to show that something is wrong. You can find devices that measure both online rather easily. Just make sure you get one that adjusts the readings based on the temperature automatically.

 

[longlimb]

Dimitri,

How did you nail down that it was a platesetter issue? If you measured each plate would they show the variance? I only spot read them. I do measure at the same angle and have it set for FM though.

What do you mean by S shaped curve? I have a plate curve that I created from an uncalibrated run. Is a S curve something different?

I appreciate all the help guys!

 

[dimitri]

If you measure the plates daily and they do not move, then it's not the plates. Hope I'm not confusing you; the fountain solution ideas, or other press issues, seem more reasonable since you monitor your plates.

Anyhow, I was sending a linear black (or magenta, doesn't matter really) plate twice per day for a couple of days, before and after we changed the chemicals on the exposure unit. This were supposed to read 50% at the 50% dot area, 60% at the 60% dot area and so on. I saw that when we were supposed to get 25% we were getting 20%, and when we were supposed to get 75%, we were getting 80% (arbitrary numbers). These deviations were bigger for 20 microns than 25 microns. I was also measuring different areas across the plate. Then, I made a linearization plate to make the 25% read 25% and so on, that had the shape of an 'S', since the plate was gaining in the shadows and losing at the highlights, and then applied any plate curve from a new fingerprint on top of it. I was reading 10% increments.

 

[Deleted member 16349]

Longlimb,

I have been following the posts with interest. I could not find anywhere a comment on the type of press you have this probem on. Maybe I missed it. What press type is it? Thanks.

 

[longlimb]

It's a Mitsi 28" 6 color with AQ. Not UV in our shop.

Dimitri-

I always thought you didn't need to run a linear curve and then a print curve. The run on press off the uncalibrated plate could just be adjusted from that point and you can skip the linearization of the plate.

What kind of imagesetter are you using?

 

[LauraMinter]

I think the problem with not running a linear curve and then the plate curve is "what basis do you start from?" I agree that it is double work, but basically if you have optimised your platesetter, so there is no more than 2% deviation across your plate, and your internal 50% is reading 46% for example, and you then apply your dot gain curve that is great. But what happens if on your next batch of plates your 46% is now 44%? It is going to affect your dot gain curves on the press. Surely it is easier to just relinearize the platesetter and always start working off the same base, ie 50% is 50% on plate than having to redo the dot gain curves?

 

[Deleted member 16349]

It's a Mitsi 28" 6 color with AQ. Not UV in our shop.

I am assuming sheetfed.

 

[longlimb]

Ahh. I see your logic now. That makes sense. That way I'm only re-doing the plate curve and don't have to go back to press to adjust a print curve.

 

[cocfvc]

Sounds like a fountain solution issue, this should be monitored and there should be a standard for your pressroom to follow. You might want to try taking a plate that is performing poorly (cyan unit) and switch it with one that is performing correctly (magenta unit), don't change inks just plate from one unit to the next and re-reading the gains. This will confirm if the plate is at fault or it is something else in the system. Remember there are far more variables to maintain and control in the pressroom than in prepress.