Black And White Digital Printing Part 3b

Introduction

After nearly two years of faithful service the black channel in my R200 clogged permanently after sitting unused for several weeks. No amount of coaxing and trickery has been able to unclog it. This printer has been my only BO print producer since I swapped my 2200 for a 2400 in August, 2005. So my BO print production came to an end.

Buying a replacement is difficult because the current R260 model doesn't have a Black Only option, and the few 200/220 models on Ebay are going for premium prices. Being reluctant to spend the money, I began to wonder if Paul Roark's QTR driven 3-channel BO method (3MK) for the 1800/800 could be adapted to the R200. The 3MK method is based on the 1.5 pl droplet size of those printers. Even though the R200 has 3 pl dots, in just thinking about the principle of how it works it seemed that using a 2nd MK channel to minimize the graininess and dither banding ought to work to some degree and produce a better BO print than the traditional single channel approach (and allow me to continue using this printer).

As I am not inclined to tinker with RIPs and curves very much, it remained a purely mental exercise until one evening recently when my good friend and BW Forum member Steve Karafyllakis informed me that the LC, LM and Y channels of the 200/220 produce finer droplets than the K, C and M channels, and perhaps if these finer dot channels were used the multi-channel BO approach might be effective. Intrigued by the idea and with his help we fiddled with carts and chips and set up the Y and LC channels with Eboni ink (I had only two Eboni carts on hand at the time).

For both channels I used a curve I had created back in 2005 in a BO attempt with the 2400. The time consuming work to emulate a BO ramp had already been done, so this seemed like a good starting point, and indeed it turned out to work very well.

To make a long story short, the experiment succeeded beyond my wildest expectations. The resulting prints are nearly as smooth as Paul's 3MK samples, and are enormously better than traditional single channel BO prints.

As for using two channels instead of three, my feeling at this point is that since this printer just has 3 pl dots, there would be little improvement by adding a 3rd channel, and the added complexity wouldn't be worth it.

The following sections contain a detailed outline for setting up the process. It is assumed the reader has basic knowledge of using QTR, refilling carts and swapping chips.

Setting Up The printer

It is a good idea to determine the best two of the three small-dot channels, LC, LM and Y. Some channels can have more banding than others, and it can vary among printers. It might also be a good idea to do a head alignment before doing this.

In QTR, choose the Tools/Options/Calibration Mode menu item and print out the calibration chart. This prints a ramp for each channel. Inspection with a loupe of the light end of each ramp will show that these three channels do have finer dots than the others. It is possible to see some fine banding in these ramps, and this is what we want to avoid as much as possible. Choose the highest quality channel for the primary ink. For my R200 this was Y, which had no banding at all. The next best channel (slight banding) was LC, so this was chosen for the secondary channel.

Next, prepare Eboni carts with appropriate chips for the two channels. Install the carts and do the usual nozzle checks and cleaning cycles as needed.

The Primary And Secondary Curves in QTR

The following instructions are for creating the two curves. The primary curve will have to be fine tuned with a Photoshop .acv curve which can be downloaded from my web site here. The zipped file is named
CJ-2KBO-R200-Ordered.acv. It should be unzipped and placed in a relevant folder somewhere. You will be navigating to it later from within QTR (the theory for all this is explained in the next section below).

1) In QTR open the Curve Creation window and select Files/New to create a new curve definition file. For Printing Model select "QuadR200"

2) In the Ink Setup tab, set the Default Ink Limit to 100 and the Black Boost to 100.

3) Locate the channel that will be used for the primary curve (Y for my R200) and select the "Gray Ink" item from the dropdown picklist. Set Density to 99 and Limit to 48.

4) Then locate the channel that will be used for the secondary curve (LC for my R200) and select the "Toner" item from the dropdown picklist. Set Density to 99 and Limit to 12.

All other channels should be "Not Used". The result should look like this (your actual channel selections may be different)

5) In the Gray Curve tab, for Highlight enter 2.5, for Shadow enter 1.8 and for Gamma enter 2.1. Leave the Overlap box empty.

6) Then click the Browse button and navigate to the .acv file and select it.

The result should look like this (the file name will contain the drive letter and folder as well)

7) In the Toner Curve tab (secondary curve), enter the same three values as above: Highlight=2.5, Shadow=1.8 and Gamma=2.1. Do not add the acv curve here. Just leave it blank.

8) At this point the curves definition is complete. Click the Show Curve button, say Yes to save, and give it a descriptive name that will mean something later. I named mine CJ-BO-200-LCY, which means my custom curve for BO on the R200 using the LC and Y channels. This name will now show up in the QTR curves list. Here's what it should look like in the curves window

How These Curves Work

The primary curve is doing most of the work. The combined ink limit of both curves is 60 (Y=48, LC=12). 60 is what is needed to get full dmax from the Eboni ink on EEM and PhotoRag. The Y curve by itself, with an ink limit of 60, will produce a single channel BO print. By dividing the work among two channels, the two different dither patterns combine to eliminate nearly all dither banding. They also create a smoother overall tone because the smaller dots can be placed closer together, in a more even distribution, at any given point on the ramp while maintaining the same black/white ratio that produces the tone. The combination results in a much better BO print.

As for the difference in ink limits, my first experiments (before adding the acv curve) had each curve carrying half of the ink load, and the resulting print was coarser than a traditional BO print. My theory is that the curves were essentially the same and therefore were both attempting to place dots in nearly the same places, basically doubling up on the same pattern and making it worse. There were some really ugly squiggle patterns in the dither. The further I separated the curves the smoother it got. After a lot of experimentation the current split of 48/12 produced optimum results. Adding the acv curve to the primary curve further changed it's shape, and the resulting print was yet a bit smoother.

The acv curve was necessary to control the contrast (I was matching the 2KBO print to a control print made earlier with traditional BO printing - the goal was to produce a print with equal contrast and density). The Shadow, Highlight and Gamma curve settings do not allow such fine tuning of the curve. By keeping the secondary curve at a low level, it's influence was benign enough that an acv curve was not needed there.

It might be advantageous to do further experiments with other variations in ink limit ratios (and using three channels) for smoothness. Different papers may benefit from different total ink limits for best dmax, so a separate file for each paper may be worthwhile.

Note: I highly recommend not to change the Shadow/Highlight/Gamma settings when experimenting with the curves. These create the curve shape that this technique requires, and are the foundation of the whole thing. Even slight changes make a radical difference in the print. Any contrast changes should be done with the acv curve in Photoshop. Any changes in the ink limits will require a corresponding adjustment in the acv curve.

It's also important to note here that these curves were developed using the "Ordered" dither algorithm and the "Better" (slower) Speed choice, both selected in the main QTR window.

Making Prints With QTR

All that remains now is to make prints. In the Main QTR window your curve file should be selected from the Curve 1 picklist, with Resolution at 2880 and Speed at Better (Faster can be used for proofs to save time, as the contrast and density are the same).

Conclusion

I am extremely pleased to find that this 2kBO technique works so well on the inexpensive R200/220 printers. It makes pure carbon printing, with it's significant archival advantages, a far more viable option than traditional BO printing. However, some of BO's simplicity and convenience is lost, due to the need for a RIP. There is never a free lunch.

This configuration works well, but should be considered a starting place. More experimentation could lead to even better results.