Thursday, August 25, 2011

A Printer's Lullaby

Dwane Hollands, owner of Hollands Print Solutions (http://www.hollands.com.au/) is a sometimes music composer.So, I challenged him with the idea of creating a piece of music using the sound of a printing press to create the underlying musical theme. He chose the rhythm of a Heidelberg press (hear that sound by clicking HERE).

Dwane names his composition "Printer's Lullaby" (although I think it's more of a lament). He describes it in these words: "After a hard days work, those machines need to have a good rest at night. Excited from the day and all the adventures they've been through, they take a while to settle in. After they've drunk their warm milk and cookies and brushed their teeth, it's time for the Printer's Lullaby to get them snuggly off to sleep. Night, Night..."

Click on play arrow to view Dwane's 2 minute composition that I've enhanced with some of my photos taken at a few of the printshops that I've visited over the years.

You can listen to more of Dwane's compositions by clicking HERE

Tuesday, August 16, 2011

Misregistration on press

You may have the most accurate CtP, the best and newest offset press, however, you are still covering absorbent paper with ink and water and squeezing under pressure in the press. The result is that there will always be a chance of misregistration in the presswork.

Here are a few of the most commonly encountered types of misregistration:

Back sheet flare (a.k.a. flare-out or coning)
Back sheet flare usually happens when a lightweight paper is printed with heavy ink coverage. As the sheet is pulled through the press the tail corners of the sheet fan out slightly causing misregister in the corners.

Back stretch
Back stretch can occur if blanket and packing on one or more units may have been compressed during the pressrun, causing change in the print length.

Waggle
Waggle can occur if there's a problem with the on press register system and/or side guides.

Web growth
One type of web growth occurs on web press as the paper streams from ink unit to ink unit. The paper edge gives up moisture at a different rate than the middle of the web. This changes the size of the paper and hence causes misregistration.

Fit
Misregistration occurs when one or more colors do not fit with the others in a set. It is typically caused by the film, or plate in a CtP device, changing size due to environmental temperature changes. As the temperature increases the film/plates expand so the image on plates imaged at different times may be different.

Despite the fact that some degree of misregistration is normal in the printing process, and that it impacts the integrity of the final presswork - as far as I can determine - it's surprising there are no official specifications for what is an acceptable tolerance for misregistration.

In multicolor printing, when all the layers of inks are in perfect register, one is not aware of the individual ink layers, only the image created by their combination.

However, if one or more of the individual ink layers begins to move out of register, the image begins to appear softer, with lower detail definition. Color and tone may subtly shift. And if the ink layers continue to move further out of register, color fringes begin to appear at the edges of detail, and finally the color image breaks up.

There is no absolute point at which the loss of register between the ink layers causes the color or image integrity to become unacceptable. And while misregistration can be expressed in thousandths of an inch or millimeters or in angular degrees, its effect on color register will change depending on several factors. For example, misregistration is most visible in reverse type within a four-color process area. The sharp edges of the type and the high color contrast between the inks and the bare paper exaggerate the visibility of any color fringing caused by misregistration. But color images with little sharp detail or where a black border hides the edges can tolerate a greater degree of misregistration before the reproduction is seen as unacceptable.

Experienced press operators have an intuitive understanding of these factors which has led - in the absence of industry specifications - to a pragmatic approach to the problem of misregistration. The historic, trade practice has been that misregistration is deemed acceptable up to 1/2 row of dots. That is enough to shift a rosette from clear-centered to dot-centered. So, for example, at 133-150 lpi that's about 0.0033ths of an inch (0.0083 cm at 60L/cm).

In register - clear-centered rosettes with a uniform edge.

Out of register by a distance of one half a row of dots - clear-centered become dot-centered rosettes with, in this case, one color layer hanging outside the other three creating a color fringe.

How the press operator measures misregister
Of course, press operators cannot directly measure the amount of misregistration. Instead, when they look closely at the press work through a loupe they employ a simple visual trick:
The press operator visualizes the width of a row of halftone dots divided into quarter distances. The solid black lines show the width of one row of halftone dots. The long-dashed black line goes through the center of the dots. The short-dashed lines divide the width of the dot row into quarters. For a 150 lpi halftone, the spacing between adjacent dotted lines is 0.0017" (0.04 mm).

Although trade practice has been to hold misregistration to within one half row of dots - which seems to be a very tight tolerance - there is still a negative impact on the integrity of the press work.

As the dot structure goes from clear-centered to dot-centered as it drifts out of register, the rosette becomes twice as visible effectively halving the screen ruling. So, a dot-centered 150 lpi rosette structure is as visible as a coarse 75 lpi screen. This is often seen like a rash on important skin colors where small amounts of misregistration make the rosette appear then disappear. And as the rosette drifts from clear centered to dot centered it’s like intermittently turning a light on and off and back again - so the color goes intermittently darker and lighter through the run.

Lastly, misregistration changes the overprint ratios of wet and dry trap - causing a color shift. So while the image is going darker and lighter it is also shifting color from bluish to redish and back again. For example, this can be particularly troublesome in, for example, car brochures where there are large expansive areas of neutral grey made up of 4/c process.

A guide to print misregistration - symptoms and causes.

Symptom
If, in a pull of inspection sheets, only one sheet is misregistered out of tolerance.
Cause
Likely no action is needed as the misregister may be just a chance occurrence.

Symptom
If, in a pull of several consecutive inspection sheets, one or more sheets is misregistered out of tolerance.
Cause
Check the press register system and check the paper for wavy edges.

Symptom
Registration is good on several pulls of inspection sheets, but shows a drift toward the limits for misregistration in one or more colors.
Cause
Check the press gripper edge and side-guide register settings.

Symptom
While the gripper edge registration is good, there is a consistent sidewise misregister along trailing edge. The paper may occasionally show wrinkles.
Cause
Check for excessive impression/blanket squeeze which may be stretching the paper. Also check the paper for wavy edges.

Symptom
Misregistration varies from sheet-to-sheet, mostly in the around the press cylinder direction and is worse along the trailing edge of the sheet.
Cause
This usually happens due to mechanical stretch in lightweight papers. It can also happen with loose blankets or it the ink tack is too high.

Symptom
Misregistration appears to be random and occurs in different areas on different sheets within the run.
Cause
This sometimes occurs when printing heavily embossed papers. The impression pressure may need to be reduced.

Symptom
Specific colors are consistently longer or shorter from gripper to trailing edge.
Cause
The blanket and packing on those units may have compressed during the pressrun causing change in print length.

Symptom
Consistent misregistration at the gripper edge or side-guide edge.
Cause
Check the press register system.

Symptom
Random misregister at the gripper edge or side-guide edge.
Cause
Check the paper edge as the trim may be bowed convex or concave or the paper may have wavy edges.
Alternatively the press speed may be too fast for register system to function properly.

Symptom
Misregistration occurs lengthwise along the gripper edge.
Cause
The paper may be grain-short, its moisture content too low, or the pressroom humidity is too high.
Check the side-guide and sheet-forwarding mechanisms.
The plates may have been imaged in different temperature conditions and may need to be reimaged.

Symptom
Registration is good except for a consistent misregister in one area of sheet.
Cause
If the area of misregistration is on off-guide side trailing edge, check paper for wavy edges.
Check the plates as they may have a built-in misfit caused during imaging in which case new plates will need to be made. There may be a large solid on an earlier-down color near the area that's misregistered, causing paper stretch in that area.

Sunday, August 7, 2011

The Wayback View - Print Ephemera - Printer's Invoices

Old printer's invoices, if you can find them, can provide a fascinating glimpse into how business was conducted "back in the day."

Click on the images to enlarge

Printers often diversify their operations to attract new business. However this 1924 invoice from Buxton and Skinner shows a diversification that is a bit extreme. I wonder what "radiation," "meter rental," and "condensation" actually refer to?
In 1916 you could get 1,000 envelopes for $2.50 (and apparently pay no tax).
This Amsterdam New York invoice from 1944 includes an interesting note to the effect that income tax was withheld on wages. I wonder why they noted this on the invoice?
Perhaps this invoice from 1897 inspired a famous Vulcan with it's motto: "Print and Prosper."
A 1944 job estimate from G. Claridge & Co. of Bombay (Mumbai) appears to be a form letter with the text in red being preprinted while the text in black being customized for each project.
The total of this 1918 invoice for presswork came to $18.25. Translated into 2011 dollars that $18.25 is equivalent to $291.97.
Advertising rates in The Journal published by John W Eedy in 1897 were very reasonable and its presswork "Neatly executed. [with] Prices Moderate."
Selma Printers in 1870 also prided themselves on "Printing Neatly Executed."
M.J. Sullivan differentiated himself from his competition by being a "Practical Printer" (I assume that he considered other printers as impractical). However, it appears that the order for 100 name cards placed August 24, 1922, was a bit slow in being fulfilled since the invoice shows a billing date of December 1, 1923 - over one year later.
Companies are always trying to reduce their printing spend. Here, Rand McNally has saved a bit of money by using an invoice from 1890 and with a simple crossing out of the "8" and adding a "0" has transformed it into a 1900 invoice. Another interesting thing about this invoice is the purchaser's title: a certain George M. Beadle, Mining Expert, "etc." I suppose that Mr. Beadle had other expertise that the accountant at Rand McNally just couldn't remember.