Plate Curves

[Farabomb]

No two presses are the same. That 4c breakdown in that book is just a ballpark. It's impossible to create a book that represents all presses.

[Joe]

I'm referring to the "Pantone Solid to Process Guide" or the "Pantone Process Guide". These books have the Spot color printed next to process equivalent. So you can see the difference in how it would print with a PMS or with the 4/C build.

I'm not talking about using just a solid PMS ink.

Yes they are telling you approximately how much dot gain they had to account for to get those CMYK values to match the solid Pantone. But it is still  up to you to know how much dot gain is on your press for the particular paper you are printing on. Uncoated newsprint has more dot gain than a coated stock. Paper matters. In a perfect world we would all be using the same presses, paper, ink, blankets, and 8 billion other variables. But this isn't a perfect world.

[David]

But this isn't a perfect world.

say it ain't so Joe!!!

I coulda swore it was perfect. I think I still have the email.

[Joe]

But this isn't a perfect world.

say it ain't so Joe!!!

I coulda swore it was perfect. I think I still have the email.

Maybe Texas is?

[David]

it's as close as it gets....
you want some?

[Farabomb]

Let's bring this back as I'm trying to get me head around it. I'm going to have to take it in steps or I'll get lost.

1: My old time pressman said you get more press gain as it goes through units. Say a 4c job black would gain more (K in first unit) than if it was just a single color black job. Fact or fiction?

2: In terms of prinergy the plate curve linearizes the plate. Then you need to create a curve that pulls more at upper dot values to compensate for gain. If it gains 10% @ a 5% dot that's not much of an increase. The same 10% gain @ 80% dot is much more of an increase.

[Joe]

Not sure if I agree with number 1. Each color has gain so if there are 4 colors it might appear to be more gain but each color is probably still only gaining so much.

On 2...you have a problem if you are getting 10% gain at both ends of the scale. On newsprint we get approximately 20% at a 50% patch and probably 25% at the 75% patch. Anything over about 80% goes solid on newsprint anyway. But at the 5% patch we only get about 2-3% gain. If you get 10% gain at a 5% there is a problem.

[Farabomb]

I've managed to convince the boss now that we are going to the AGFA plate to attempt to get things correct. I now have plate curves to linearize the plates. Now I have to figure out how to create a curve to make this plate print like the old plate.

I don't have a dot reader that can read paper. Do I need this or can it be done visually? Can I just whack the BS curve I've been using on the old plate and have at it? I don't think that's possible but after today I'm not sure what to think.

[Farabomb]

I was just using 10% as a number. I have no way to measure the gain on the paper. The boss is under the impression you get X amount of gain across the board. It can vary more or less at different dot percentages?

[Joe]

Well it depends on how you look at it. If you are talking about a 10% gain of a 5% patch it is only a 1/2 percent gain so a 5% would become a 5.5%. But if you mean a literal 10% gain where a 5% become a 15% that would be a huge gain. When I'm talking dot gain I am talking about literal gain where a 5% may become a 7%. That is 2% gain to me but as a percentage of 5% that is about a 40% gain.

[Joe]

I've managed to convince the boss now that we are going to the AGFA plate to attempt to get things correct. I now have plate curves to linearize the plates. Now I have to figure out how to create a curve to make this plate print like the old plate.

I don't have a dot reader that can read paper. Do I need this or can it be done visually? Can I just whack the BS curve I've been using on the old plate and have at it? I don't think that's possible but after today I'm not sure what to think.

You can do it without a dot reader but it is more of a guessing game. Form my Harmony dot gain curve I only create about 5 points. 5% 25% 50% 75% 98%. I take the 5% down 1%, the 25% down 4%, the 50% down 8%, the 75%down 10% and keep the 98% at 98%. This is for newsprint. For a good coated stock I would cut it back less.

[Farabomb]

Ok, I think I have a handle on it. We're going to run a test form like we have in the past and I'll have to do my best to eyeball it to get it where we need it.

[Farabomb]

Well that failed miserably.

I have a linear plate curve for both my plates now. Those aren't coming close to our proof and look like shit on press. Not a big surprise as we never had a plate curve in the past. The highlights are gone, not even close and in fact for some reason I had 3 colors that read in the file 8/5/6 cmy, linear plate was 3/4/5, old plate was 8/5/6. No wonder the pressmen were having issues.

I need to get the new TS plates to look the same as the old LH-PJ plate. I was assured by the salesman this would be a drop in plate replacement. We all know how that goes.

I attempt to use our old curve that was on the LH-PJ. I figured since I made test plates before using the same curves and exposures as the LH, with the only difference being having to invert the image (LH is write to background, TS is write to image) it would work. It worked on the test plates. No dice. So very heavy in the mids.

Now here is my question. The LH was blowing out the background to create the image, the TS is writing to the background to create. Should the curves be opposite? It makes sense in my head but I've also been sick since saturday.

Hopefully the AGFA tech will be here tomorrow to help sort out this. When I said I was ok with changing plates, fingerprinting the press was to be included. I said that every time changing was mentioned. Now I have to attempt to fix it on the fly in order to get jobs out.

Just like when we kicked out Kodak and by some miracle I made a 5% bump work close enough on the fuji plate. Now I have to make this work. All the boss remembers is we fingerprinted the press... 7 years, 2 ink suppliers, 4 house stocks, a different location FFS and new fountain solutions ago. Is anyone in printing interested in doing things right or is good enough, good enough?

[Joe]

Ummmm...if the file read 8/5/6 and your "linear" plate read 3/4/5 then you don't have a linear plate.

[Joe]

Now here is my question. The LH was blowing out the background to create the image, the TS is writing to the background to create. Should the curves be opposite? It makes sense in my head but I've also been sick since saturday.

Different plates. The old curves mean nothing with the new plates.

I think you need to go back to work on your linear plates. Whatever the patch reads in the original should be what the plate reads. If they are not...you are not linear. Once you get it linear then you calibrate to make the new plate read the same as the old.