Registration or doopid pressmen?

[plategirl]

Hello, we had a job today, a tabloid, that had issues and one pressman is on my back about it being the pdf's fault and got the production manager on for the ride. Registration was off. I mean, not so that it could be adjusted. The black on only the one side seemed stretched, is how I'd describe it, at the bottom of the one page. The other side of the plate came in OK. We use a trendsetter news and thermal plates, if that matters, and no serious registration issues on another tab we ran just 3 nights ago.

They had me reimport a new pdf and said it was all better, but this was after they retorqued the blanket etc.

So anyway, I guess I just want an answer, not pass the buck. IS there any chance of this issue being due to a case of a "magically stretching pdf"? CAN my trendsetter be to blame? I'm all for having it recalibrated if necessary, but my supervisor isn't, and the buck stops with her.

Sometimes when we run standard jobs, a similiar thing happens...the odd page will be in, but the even page on th eleft will be "off", though ususally by a very small amount, almost invisible to even the trained eye.

Blanket issue, CTP issue, or pdf issue? It can't be the plate puncher becasue other wise they could just slide the plate over to fix it.

If it helps, we are a newspaper. Most jobs are on 23 inch paper, and this job was 30". One of the guys, a pressmen but not the doopid one, thought maybe the blanket was compressed in the middle and this was why the image was warping at the ends and not in the middle. Our usual press supervisor was off tonight, so we ended up being 5 hours late on this job.

 

[rbailleu]

blanket issue. and he moved the plate to get one corner on. the tail is spreading due to water and pressure.

 

[plategirl]

Thank you. I hugely suspected that but it's good to have confirmation, especially if this issue is brought up by the production manager (who has neither prepress nor press experience). "Magical jumping pdf", indeed!

 

[JaimeZ]

It is not the PDF, like rabailleu confirmed it's the blanket and if it were not the blanket then it would be the platesetter or bad plates. PDFs have their own problem but misregistration is not one of them

 

[Deleted member 16349]

The black on only the one side seemed stretched, is how I'd describe it, at the bottom of the one page. The other side of the plate came in OK.

They had me reimport a new pdf and said it was all better, but this was after they retorqued the blanket etc.

If it helps, we are a newspaper. Most jobs are on 23 inch paper, and this job was 30". One of the guys, a pressmen but not the doopid one, thought maybe the blanket was compressed in the middle and this was why the image was warping at the ends and not in the middle. Our usual press supervisor was off tonight, so we ended up being 5 hours late on this job.

Plategirl,

Even if the plates are perfect, the print can have out of register problems for both sheetfed and web presses. There are some differences between what happens in a sheetfed press and a web press but since you are printing a newspaper, I assume it is a web.

With a web, it is also related to whether the gap in the blanket is small and always has good contact with the web or if the gap is larger and the contact with the web is lost for a short time.

For a web press there are two related issues. One is (print) repeat length and the other is print length.

The repeat length is the distance from a point on the print to the same point on the print in the next repeat. This is measures on a table with the paper having no tension in it. This measurement can be a problem especially is the paper is light and has absorbed water, which would expand it, but there are ways around this.

The print length is the distance from one point in the print to another point in the print, within the same repeat (within the gap to gap points). This is also measured in the machine direction on a table with no tension in the paper.

Here is the important thing. Normally the ratio of print length to the repeat length should be constant (on average) in a web press that has a very small gap. This is independent of the tension levels that the web is printed.

Think of it this way. If you have an image that is 50% of the circumference of the plate cylinder, that will always print 50% of the repeat length (on average) if the blanket gap is small and when the blanket is always in contact with the web. This is independent of very many factors.

If I understand your problem, you are printing longer on one side of the web than on the other side. Say left side compared with right side.

This condition violates the theoretical consistency of the ratio of print length to repeat length. So what can cause this.

Certainly blanket movement could and having a loose blanket might be the cause.

On a web press, if one changes the impression squeeze, this will result in a change in the pulling performance of the blanket on the web. You normally run 24" web and now you ran a 30" web and one pressman thought that the center might have been compressed a bit. This would tend to mean that the outside edges of the 30" web are running with a higher squeeze than the center and therefore would be pulled along differently. Also it was not clear if the web is always centered or if the wider web is running off center.

From my experience, more squeeze makes the blanket cylinder have a larger effective diameter and therefore pulls more.

Having an uneven pulling condition across the web is not sufficient to give you the problem you have but if the press also has a gap that is large and therefore the web is not controlled in the gap area, an uneven pulling condition could result in the out of register problem you see. What would happen is that on each repeat, the one side prints a different length due to the change in pulling performance but as a larger gap arrives at the nip, the paper can relieve the tension and the cycle starts again.

If the gap is small but the blanket is not tight enough and can move on the blanket cylinder, a similar cycle could happen but in this case the blanket moves due to the uneven pulling force and recovers after the gap passed the nip.

It is an interesting problem and probably the re-torquing of the blankets did the trick.

This is of interest to me since I have done some work on this subject. I presented a technical paper in 1997 at the TAGA conference on the related subject of print repeat and web tensions in offset presses. This was from work I did in about 1986. A complicated subject but it was interesting at the time.

 

[plategirl]

Erik, thank you for that very informative comment. Wow! Some of it went over my head but I did get the jist of it. When the head pressman comes in I might just show it to him, and the others too.

JamieZ, if I can borrow just one more minute of your time, can you expound on what conditions in which the Trendsetter or plates, might be responsible for a warped image, besides the obvious ones that I can think of, such as a dented drum, or plates, and what it might look like? Our plates are on the cheap end (!!) and anything I can do to get us better quality plates would be a plus. I don't think the imagesetter was the problem this time around, but I want to learn everything I can in case something another situation comes up, so I can at least defend myself with the cold, hard facts. God, I love Print Planet!

 

[maxon]

... can you expound on what conditions in which the Trendsetter or plates, might be responsible for a warped image, besides the obvious ones that I can think of, such as a dented drum, or plates, and what it might look like?...

Plategirl, is it internal punching or offline ? Could be inconsistent loading due to worn out grippers, or simply incorrectly cut plates. I can't think of any other geometry issues caused by the TS itself being an external drum machine.
We use a flatbed ECRM Mako CTP (violet), the pressmen have complained once for plates being randomly out of register, that was last year. We were convinced it's not the ctp to blame as it's a three point detection completely and totally foolproof just can't go wrong. We didn't have a video punch system available to measure the consistency of image positioning x-y, or a long electronic caliper/ruler so I had to devise a way to check the image placement. I took a thin sharp screwdriver and a steel ruler and extended all corner register marks on all four plates down to the edge of the aluminium. Then stacking all four plates I could see from the sides that all scratches aligned beutifully, so the plates were ok. Turned out in the end the offline punch machine was out of order having one register pin wobbly.

 

[nemo]

A Trendsetter can never "warp" a plate, period.
Image may be crooked, shifted, scaled, out of registration, with banding, etc. but not warped.

 

[maxon]

A Trendsetter can never "warp" a plate, period.
Image may be crooked, shifted, scaled, out of registration, with banding, etc. but not warped.

Absolutely so, since it's an external drum.

 

[oxburger]

OMG. A stretched digital file? Who could even come up with this concept? I thought I heard everything, but WOW. The ONLY way it would be the case is if the pdf was created by using CMYK components from different pdfs and compositing them into one. Either that or something happened when plating. It is possible that the platesetter had a glitch, but otherwise.... weird.

 

[plategirl]

OMG. A stretched digital file? Who could even come up with this concept? I thought I heard everything, but WOW. The ONLY way it would be the case is if the pdf was created by using CMYK components from different pdfs and compositing them into one. Either that or something happened when plating. It is possible that the platesetter had a glitch, but otherwise.... weird.

What's weirder? What's weirder? They just gave this doofis my job, based on part on this incident. Oh, and the press room just lost us 3 commercial jobs due to quality so poor the customer copy looked like start-up waste. Yeesh. Fries anyone?

 

[otherthoughts]

It's been many years since I've run a press, maybe 18 years now. However even back then, our sheet-fed presses had split tail-clamps to address "sheet fan-out". To explain what sheet fan-out is, it's the progressive growth of the sheet's width at the tail end as it travels from the first down unit to the last down unit.

To illustrate, imagine the first down color you are concerned with is the black printer. With each successive "Squeeze" the sheet fans-out like bread dough squeezed by a rolling pin. This 1st down black ink image printed on the sheet is also fanned-out more and more as the subsequent impressions occur. In other words, for a sheet-fed KCMY ink sequence, the first down black ink would seem to grow progressively wider near the tail end with respect to the downstream units, typically with the last down unit seeming to print the most narrow.

The split tail-clamps I mentioned earlier, allowed me to force the tail end of the 1st down
plate inwards slightly and stretch the downstream plates outward slightly as needed to counteract any fan-out.

A web press may have a method of achieving a similar task via cylinder skewing or some other method(maybe not). One thing I suspect is common to both web and sheet-fed presses is that the greater the "squeeze" applied to the sheet, the greater the fan-out. Another factor that i suspect is common to both is that, the thinner and larger the stock, the greater the fan-out.

The best way I can think of to determine if it is indeed fan-out you are dealing with, is to compare the first down ink's plate against the resulting printed sheet. Trim the sheet/cutoff as needed near the tail end and compare it carefully with the 1st down inks plate(you should probably remove the plate from the press in order to determine anything or convince anyone). If a sheet/cutoff's 1st down imprint on the sheet/cutoff has grown in width near it's end when compared against the plate that printed it in the first place, how else would you explain it?

If the very same sheet/cutoff compares accurately to the last down inks printing plate, it is further proof that fan-out occurred between the intervening units. For further proof, carefully trim the 1st down plate as needed near the tail and lay it upon the last down unit's plate and see if they match widths with one another, if they do, why don't they match on the press sheet/cutoff? Finally introduce the sheet/cutoff as well to settle things and see what matches what.

As many have indicated before me, the press-room is the much more likely suspect for this problem. I hope that the tests I suggested will reveal the source of the problem. The tests I suggested will likely require enough skill in execution so that one does not warp the plates or the sheet/cutoff. A scope with a 25x magnification or better would likely be needed as well. I should also mention that the tests I suggested should be considered as evidence only, they are only as good as the person's skill performing them. Regardless, no one in their right mind would consider the results of these tests absolute proof.

A dubious test is better than no test at all, the way I see it.

 

[Oscar Ramirez]

If you want to eliminate the CTP causing the registration issues you can create a job the has cross hairs in 4 corners of the image area and place a few more in the middle, top and bottom of the plate. Image the plate, depending if its positive or negative,take a look the image and burn the same plate 3 more times, then process. Use a loop to check the reg marks-show the pressman.

 

[Alois Senefelder]

Salient points

Salient points

Hi my friend "Otherthoughts" some more PDFs for your enlightenment - also for the rest of the Readers, regarding "Mis-register" on a Web-offset Press,


Regards, Alois

 

[Alois Senefelder]

1 more PDF !!!

1 more PDF !!!

Hello everyone,

Obviously making sure the Plate has been "Imaged" correctly and the "Control Strips" are correct.


Regards, Alois

 

[otherthoughts]

Hi my friend "Otherthoughts" some more PDFs for your enlightenment - also for the rest of the Readers, regarding "Mis-register" on a Web-offset Press,


Regards, Alois

Thanks Alois.

I never ran a web press myself or worked for a company that had one. But you knew that

I do have one question remaining after reading your attachments regarding RPL on web presses.

On sheet-fed presses a change in print length can be accomplished by migrating small amounts of packing from the plate cylinder to the blanket cylinder or vice versa as required while still maintaining the proper squeeze between the two cylinders.

The documents you attached indicated that RPW is easier to control than RPL on web presses. So what I am wondering is if a packing change like I mentioned above for sheet-fed presses would have any effect on a web press? It's probably an ignorant question, but I am like that sometimes.

Thanks again
otherthoughts

 

[klh]

I have to share this. . . a press trainer was on premises for one year. Once the press was up and running for about 5 months the trainer (not our employee) sent the pressman in to tell me that "my magenta plates were not hitting". I asked him what he meant and he said that the register marks on the magenta plates did not match up with the other three - that the trainer said I needed to fix my magenta mark! He thought that each line in the register mark was individually drawn. I showed him how they are made and he still was insistent that it was either me or the platesetter. I went out, pulled the set of plates, got a ruler and measured everything - which was all ok - while the trainer went upstairs to the owner and told him that I could not make a magenta plate that could hit.

I got called into the main office, explained that everything from my department checked out, but that I would call in a tech to confirm it. (Because you know, 20+ years of experience, being a female in this business, I still don't know anything.) Anyway, I called one of my favorite, trusted techs - he told me it had to be the blanket on the press. . . I relayed this to the pressman, who told the trainer, who said, "bullsh#t". The next morning, before the trainer arrived the pressman changed out the blanket, hung a set of plates and the problem was gone. . . I told the trainer that I must have mixed up the unexposed plates in the drawer and used two cyans in a row, and that I would never do that again!

He was so smug, then the pressman told him he changed the blanket. . . stoopid pressroom trainers!

 

[Deleted member 16349]

Thanks Alois.

I do have one question remaining after reading your attachments regarding RPL on web presses.

On sheet-fed presses a change in print length can be accomplished by migrating small amounts of packing from the plate cylinder to the blanket cylinder or vice versa as required while still maintaining the proper squeeze between the two cylinders.

The documents you attached indicated that RPW is easier to control than RPL on web presses. So what I am wondering is if a packing change like I mentioned above for sheet-fed presses would have any effect on a web press? It's probably an ignorant question, but I am like that sometimes.

Thanks again
otherthoughts

It's a good question because the answer is not so obvious. The conditions on a web press are different than on a sheetfed press and that results in slightly different outcomes.

On a sheetfed press if you increase the packing under the plate, the print length should decrease. If you increase the packing under the blanket, the print length should increase. When you decrease the packing under the plate and increase it under the blanket on a sheetfed press, you get the affect of both actions.

A web press will be different. Changing the packing under the plate will change the print length but changing the packing under the blanket will not affect the print length relative to the repeat length.

The above description is based on having no unusual slipping conditions in the press.

To get a better idea of this problem, think of the print length being exactly 50% of the circumference of the plate cylinder surface (surface of the plate on the plate cylinder). If you added packing, then the print length on the plate which is a fixed length, becomes less than 50% of the circumference of the plate cylinder due to the increase in diameter resulting from the added packing.

The print length, which is now less than 50% of the plate cylinder, prints onto the blanket. Since the blanket cylinder is geared to the plate cylinder, it then must print the print length onto the blanket at less than 50% of the blanket circumference. It has no choice. Even if the blanket is over packed or underpacked, the print length percentage must be what ever it was for the plate.

The blanket in turn prints the same print length percentage (<50%) onto the paper. In this case the print repeat represents the reference for the print length. Changing the blanket diameter with changes in packing will not affect the print length percentage but will affect the web tension.

In web presses, the first unit determines the print repeat and all the following units must print the same repeat. Web tensions can go up or down in the press after the first unit but the print repeat will be the same on average. On top of this basic condition one can add the changes in print repeat due to changed in moisture in the paper. As was shown in the PDF, tensions can vary due to changes in moisture.

When one is Insetting, where the web coming to the press was printed before, a control system is required to match the preprinted web to the print in the press. This is an unstable condition and MUST have a control system to maintain any accuracy. With these kinds of control systems for Insetting, the infeed unit of the press changes speed slightly to maintain the position of a control mark which is preprinted on each repeat. The change in speed of the infeed, changes the web tension going into the first unit and that changes the print repeat of the press to match the print repeat of the preprinted material.

With a web press there is a bit more dynamics involved related to print length and repeat lengths than with a sheetfed press.

 

[otherthoughts]

If I am following you correctly Eric, then plate packing does affect the print length, at least with respect to the impression transferred from the plate to the blanket. That much seems to be the same for both a sheet-fed press with bearer to bearer contact as it is for a web press with it’s plate to blanket cylinders geared together.

However, from that point forward, the repeat timing coordination between the inking units serves to register one unit to the next unit regardless of the blanket packing and takes precedence over blanket packing as well with regard to print length. I am still with you so far I believe.

The area I find troubling is that if the printing units synchronize their repeat length with minute speed variations between the units, wouldn't any speed changes primarily be used to synchronize the basic timing between units and secondarily be used to alter the print length?

In my mind, synchronizing all the units to some reference point on the cutoff makes great sense. However, lets say that retarding the Magenta unit slightly, just for example, in order to increase it's print length, somehow bothers my sensibilities.

Maybe it's just me and my sheet-fed way of thinking, but wouldn't setting the plate and blanket packings be something that a pressman would define firstly in order to minimize web tensions between units job to job, day to day? And secondarily introduce any speed differences between the units to handle print length as needed?

But then what the hell do I know about web presses? and be it noted that I can't see out of one ear or hear out of one eye to boot! I suspect that what the real problem here is, that it's hard to teach an old sheet-fed pressman new Web tricks! Now we are getting to the bottom of things.

Elementary my Dear Watson, it's just a simple case of deductive reasoning.

Thank you for your patience, one and all.
otherthoughts

 

[Deleted member 16349]

The area I find troubling is that if the printing units synchronize their repeat length with minute speed variations between the units, wouldn't any speed changes primarily be used to synchronize the basic timing between units and secondarily be used to alter the print length?

In my mind, synchronizing all the units to some reference point on the cutoff makes great sense. However, lets say that retarding the Magenta unit slightly, just for example, in order to increase it's print length, somehow bothers my sensibilities.

Maybe it's just me and my sheet-fed way of thinking, but wouldn't setting the plate and blanket packings be something that a pressman would define firstly in order to minimize web tensions between units job to job, day to day? And secondarily introduce any speed differences between the units to handle print length as needed?

otherthoughts

It is a bit difficult to picture what happens and providing the mathematical description will tend to make things more confusing. Fortunately there are simpler ways to look at the problem and I will try to answer your questions in that way.

First let me clarify some terms or comments so we are discussing the same thing.

There is a difference in the rotational speed of a print unit, the effective surface speed of the blanket and the speed of the web.

Basically in all presses, each print unit is synchronized to have the exact same rotational speed. This is done via gears or a drive shaft or servo motor drives, etc. This has to be true because if one unit was rotating faster than the others, its print would advance while the other didn't. So for register accuracy the units must have the exact same rotational speed and be positioned angularly so the print is in register.

The web going through the units does not have to go the same speed. The units do not have to have the blankets have the same effective diameter. The effective diameter of the blanket cylinder is its diameter, which can be measured with a blanket gauge relative to the bearer plus the minor affect of squeeze and web tension difference before and after the blanket nip. The blanket diameter can be changed by adding or subtracting packing. From the tests I have done, increasing the squeeze increased the effective diameter up to a point, and then it can drop off and make the effective diameter decrease. The difference in web tension before and after the blanket nip can increase the effective diameter if the tension difference is positive, ie higher tension after the nip or decrease the effective diameter if the tension difference is negative where the tension after the nip is lower than before it.

The amount the squeeze and the tension difference have on changing the effective diameter is very much related to the blanket type. Some blankets will show a larger effect than others to these variables and that is one reason why it is not a good idea to mix blanket types in a press.

The effective diameter is the mathematical values that is used to calculate the effective surface speed of the blanket with respect to the web. It is used to model how the blanket actually performs with respect to how fast it is pulling the web into the nip. The effective diameter times the rotational speed is then the effective speed of the blanket surface acting on the web.

An important thing to understand is that even though the operator sets the press up as well as they can, due to minor differences in blanket properties, blanket diameter, squeeze, etc. all print units will have slightly different effective diameters. This is normal.

So the picture is: We have a web press running a web that enters the first print unit at some infeed web tension value.

If we for the moment forget about the affect of water absorption in the web we can say:

At a nice steady state condition, the web will increase or decrease in web tension as it goes through the press depending upon the effective diameters of each unit. If one unit has a larger than average effective diameter, it will pull the web faster and the tension will increase before that unit and decrease after that unit. When the web tension between units increases there is also an increase in web speed.

So here is this press running the web at a nice steady state condition and the web tension levels between each unit can be different but are steady.

This sounds very complicated because there are changes in web tension and changes in web speed but the simplicity of this is that they all balance out. This will be true even if the moisture of the paper changes, which will change its elasticity and even its final length.

We can now look at this from the very simple perspective.

You have a press that prints 1000 meters of paper. All print units turn the exact same number of revolutions. Therefore, all units MUST print the same print repeat. They have no choice.

Therefore, your question about retarding the Magenta print unit will have no affect on print repeat or print length. Print length is a ratio or percentage of the print repeat length and it is determined by its percentage of the plate circumference. I define print length as a distance between any two points in the machine direction, in an image which is within the same print repeat but it is relative to the repeat length. Print repeat, I define as the distance measured at zero tension, say on a table, from one point in the image to the same point in the image in the next repeat, in the machine direction.

On a web press, if one wants to change the print repeat and therefore change directly measured print length, this can be done in a number of ways. Change the blanket type. Change the packing to change the blanket diameter or change the infeed tension. Changing the infeed tension is the easiest to do. Increasing the infeed tension will result in a shorter print repeat and therefore a shorter print length. Lowering the web tension will increase the print repeat and therefore the print length.

The conditions you set up the first unit to will result in a print repeat that all the other units must follow. Changing the infeed tension into the first unit will change the print repeat for all units but when this is done, the register will have to be adjusted because the images position in a the press changes slightly and the following units will need to follow that movement. That is what register controls do so well.

I hope this help you understand what I was trying to say.