What happens to the dot gain curve when solid ink density changes?

[LarryT]

Hi,

A book that we typeset for a client came back from the printer with washed out and dull looking images. Since there are no control patches in the f&gs for this job, which happens to be black only, I indirectly measured the tiny slivers of gray in a pie chart to get some idea of what happened to the tones in the our pdf (see attached graphic).

The solid black patch came in at around 30L* (1.2D) instead of around 23L* (1.4D) for our coated sheet – light, and no surprise here. The tone reproduction curve, however, showed TVI peaking at around 20% instead of 50% to 60% that we had expected.

Is this kind of leftward-leaning dot gain curve typical of not putting enough ink on paper (sorry, I don’t know the proper printing term)? Or are the readings more likely due to something else altogether?

Also, can anyone point me to a book, magazine article, or some other source that shows a family of dot gain curves as a function of solid ink density?

Thanks for any help with this.

 

[rath307]

Tone curves do not usually affect solids however if the ink density is not
correct that will affect the solids

 

[gordo]

The solid black patch came in at around 30L* (1.2D) instead of around 23L* (1.4D) for our coated sheet – light, and no surprise here. The tone reproduction curve, however, showed TVI peaking at around 20% instead of 50% to 60% that we had expected.

Is this kind of leftward-leaning dot gain curve typical of not putting enough ink on paper (sorry, I don’t know the proper printing term)? Or are the readings more likely due to something else altogether?

Also, can anyone point me to a book, magazine article, or some other source that shows a family of dot gain curves as a function of solid ink density?.

As the solid ink density changes so changes the shape of the dot gain curve. The shape it takes is dependent on several factors including the halftone screening that is used. The peak is rarely exactly at the 50% point. Usually it's around 40%-45% so it leans to the left. Looking at published dot gain charts of what happens when SIDs change won't help you much. That being said, System Brunner has published such curve comparisons (ISO contours) in their literature (haven't checked if they're on the web).

BTW in the image you posted was of a scan of the printed page. Note the show through. I'm assuming that you used a densitometer or spectrophotometer (not the Kodak chart included in your scan) to measure the dot gains? If you measured the graphic with a white background your measurements may have been skewed because of that show through. Measure against a black background.

More details here: Quality In Print: Eliminating show-through in scans and/or photocopies

best, gordo

 

[Lukas Engqvist]

I was going to suggest something in your press is not right, maybe too much dampening which is giving you both low SID and high dot gain in low%. Since it is thin paper there may even be stretching when fibres are wet… you should examine the print for slurring the dots… that would give appearance of dot gain, especially in low%.

Does your pressman have any ideas? You really need control strip to do any quality control, and that would also include checks for slurring.

 

[LarryT]

Gordo: Thanks for the useful lesson. The Kodak chart was my link to the real-world. The small slivers of gray were too small to get at with my i1Pro, so I did the “measuring in Photoshop” and used the spectro on the physical Kodak chart. I overlooked the fact that show thru will make my readings lighter.

Lukas: The early word from the printer is that the images were all in K-only deviceCMYK format in the pdf, but their front end expected deviceGray for black only jobs. Seeing a cmyk source, the software did a cmyk to cmyk conversion. Hmm... (Update: This might not be it, so I'm passing your suggestion along to the printer. Thanks very much!--Larry)