Making a B&W ink set for 6-color Epson printers

[Lucas28]

For Canon there is a much easier way to obtain B&W photos. Grey dye ink is available for Canon printers like the MG6250.
I've filled the color cartridge of an iP90 with this grey ink from Sun Chemical. All three colors got the same grey ink.
The photos appeared to be very neutral, and no density correction was needed. The blacks looked deep enough, although the iP90 is a CMY printer. The blacks will be deeper with the use of extra photo black.

 

[martin0reg]

For Canon there is a much easier way to obtain B&W photos. Grey dye ink is available for Canon printers like the MG6250.
I've filled the color cartridge of an iP90 with this grey ink from Sun Chemical. All three colors got the same grey ink.
The photos appeared to be very neutral, and no density correction was needed. The blacks looked deep enough, although the iP90 is a CMY printer. The blacks will be deeper with the use of extra photo black.

Are you sure this result is consistent and repeatable with other photos? 3x the same gray in all three C-M-Y positions...a good quality b&w print would be a matter of luck I think..
But anyway I ordered a bottle of grey canon ink.


My first try with lucasz's approach for canon printers is not satisfying to me yet...

martin0reg,

3 x Yes

Turning off ICC (or other CMS) allow for direct mapping of "computer" (additive) colors (red, green, blue) to "printing" (subtractive) colors (C, M, Y).

For our purpose it is essential to know direct mapping of CMY colors:

• 100% Magenta is made from rgb(100%,0%,100%)
• 100% Cyan is made from rgb(0%,100%,100%)
  
• 100% Yellow is made from rgb(100%,100%,0%)
• Background is made form rgb(100%,100%,100%)

In unmodified ink system it allow for printing with sole ink, eg. I want to test Magenta for starving and I draw entire sheet of paper with rgb(100%,0%,100%) (usual name: "Fuchsia"). And after disabling ICC/CMS in printer driver, entire page will be printed with Magenta ink (and only with Magenta).

Halftones are made as follow:

• N% Magenta is made from rgb(100%,100% - N%,100%)

Eg: If I want 33% Magenta I have to draw it with rgb(100%,67%,100%).

Some examples maybe, grayscale image and it chromatic counterpart intended to be printed with BK ink from M ink tank slot:
View attachment 1551 View attachment 1550

And transformation made with IrfanView "Shift+G" command
(please note R and B values set to "255" instead of "0"):
View attachment 1552


I left rest of carts in their places. Nozzle print looks like this:

Replacing C with BK will act same, but black point have to be shifted to rgb(0%,100%,100%) (usual name: "Aqua"). In most photo printers C and M are optimized to deliver superior halftones.
Replacing both C and M is possible (black point have to be shifted then to rgb(0%,0%,100%) ), but overall image will be to dark. This is because each of CMY colorants blocks 1/3 of spectrum. If BK is used instead any of CMY colorants, entire spectrum is blocked. If both C and M are set with BK, entire spectrum is blocked twice, leading to darkening midtones.
MG5150 is dye CMYK with "UCR like" CMY-to-BK replacement. Replacement is engaged when every of CMY colorants reaches ~60%. Chromatic colors are not affected with "UCR like" replacement.

This method may be considered as poor man's B&W printing, but it also allow to use partially damaged printheads (some non-critical electronics failures or particular color clogged).

Ł.

...Missing deep blacks and poor contrast.
But I will tweak my settings of conversion ...

 

[The Hat]

@Lucas28 overall you have got a great blend of greys but but no cigar yet, I think the photo is just to washed looking, you need to add a bit of black into one or more of your carts, so keep trying.

Smashing looking motor dough..

 

[Lucas28]

The photo looks better than the scan, but I agree that the photo black ink is missing.
Good B&W prints can be achieved with the grey dye ink, as I experienced with the Canon MG6250 which has grey in the ink set.

But the results with the K3 pigment ink is great! Reason why my latest experiments are with that ink.

Martin: About the OCP K3 prints. The three photos are printed above each other on one A4 sheet. The upper photo is printed with light cyan in the yellow cart, with no other corrections. I too think that one has the best tone.

 

[martin0reg]

@Lucasz
While my first try with your approach for canon printers (black ink in magenta cart and converting the image to magenta) is not satisfying to me yet, do you have any additional tips and hints?

In short my prints are lacking deep blacks and show poor contrast in general, at the same time they are too dark.

Is your instruction to convert the image (posting #20) all I have to do?

What do you think about replacing yellow with canon gray ink too...and what other conversion of the image would be needed - if there is any conversion possible to mach this?
I am referring here to what PeterBJ asks in the other thread:

...
I wonder if also replacing the cyan and yellow cartridges with dye black will remove the colour cast? ...

 

[Łukasz]

@martin0reg

Poor D-max may be hard to overcome without putting another ink into set.
Maybe changing setting of "Manual Color Adjustment" dialog called "Brightness" to "Dark" would allow for better D-max.

Poor contrast (darken midtones) may be due to dots overgrow, similar situation is when PGBK is substituted with dye ink (D-max is poor and details in shadows are lost).
PeterBJ has proposed simple correction to overcome this issue:

I had to reduce intensity by 20 units in the printer settings

So, looks like intensity is gamma like correction, and if properly set it is a chance to recover lost contrast.

I'm going to put into action iP4850/iP4950 with two dye inks: CLI-GY and CLI-BK on cyan and magenta positions, and it seems to be not easy.
iP4950 is in working order, but 1 pl print from Cyan is still quite subpar.

Ł.

 

[Łukasz]

Hi,

I'm going to put into action iP4850/iP4950 with two dye inks: CLI-GY and CLI-BK on cyan and magenta positions, and it seems to be not easy.
iP4950 is in working order, but 1 pl print from Cyan is still quite subpar.

I got fairly good results in two-colorant method (GY on Magenta position and BK on Yellow position - both Cyan and dye Black nozzles found to be unusable on this specifically damaged printhead).

Some specific transfer curves was used.

Now I'm thinking how to put results in most usable way (especially how to easy create those curves, cause this seems to be most problematic).

Ł.

 

[Łukasz]

Let's go

There are numerous problems to discuss.

First is general condition of printhead in iP4950 used for test.

As mentioned before, small droplets of Cyan was partially clogged, and this was not resolved.
What is more, CLI-BK (aka photo black) is affected by distortion (it will be discussed later).

Another thing is construction of testing procedure.

Testing procedure is organized into stages. Each stage was intended to test certain ink tank combination, including using dummy ink tanks (filled with cleaning solution) and masking some pins on printhead connector board.

Performed tests (and their short names):

1. Regular setup: each ink tank on its natural position (“s01”).
2. Regular setup but both PGBK and CLI-BK disabled (“s02”).
3. Regular setup, all disabled except for CLI-BK (“s03”).
4. Two colorant setup: GY on CLI-M, and BK on CLI-Y (“s04”).

Unfortunately because of printhead condition, two most interesting stages wasn't performed:

• Regular setup with all inks disabled except for CLI-C
• Two colorant setup: GY on CLI-C and BK on CLI-M

In each stage there was a batch of prints (and their short names):

• nozzle check on plain paper (“nc pp”)
• nozzle check on lightweight super glossy photo paper (“nc sg”)
• test image on glossy RC photo paper* (“timg”)
  
  
• test grayscale on glossy RC photo paper* (“tgs”)
  
  
• custom image on glossy RC photo paper* (“cimg”)
  

*Some word is needed about glossy RC photo paper.
It is 3-rd party paper, A4 size, with heavy curl (possibly cut from web). Curl was on long sides leading to hampering printhead (very remarkable sound).
It was cut to A5, curl is now on leading and trailing edge. To allow safe printhead movement, quite big margin was added to testing images.

Last but not least, some limitations in forum engine was found.

• only 10 attachments allowed per post
• forum engine is interfering into images content making fine details useless; to prevent that, images are provided as zip files

Next topic will be Stage 1. Regular setup: each ink tank on its natural position (“s01”).

Ł.

 

[martin0reg]

I appreciate your efforts .. but it seems to be a very special approach .. only for users who are willing and able to manipulate the printhead..

For me the special ink set from mikling for canon is more practical:
http://www.printerknowledge.com/threads/b-w-on-canon-printers.4962/
- 3 matching shades of gray (probably HP compatibles) for the color carts C-M-Y, high quality B&W prints without fuss (except the conversion of the color printhead to B&W only).

For the moment I am more interested in converting an epson A4 (because there is a R285 in front of me which has to be unclogged...) - like pharmacist originally has suggested in his first postings.

As soon as my epson is ready I will try pharmacists approach of diluting black/grey ink for all color carts. Possibly with dye and pigment as the epson takes it all..

We are talking here in this thread about epson and canon ... two different threads may have been better to follow ..
But anyway there is much information here regarding B&W conversion of a color printer.
Which is a important subject for all who appreciate real B&W without any color shift.
Remembering dark room silver gelatine prints...?

 

[Paul W.]

@pharmacist:

This thread may have run its course... but I'm new here and am slowly making my way through all 60-some pages of threads! I'm learning lots of good tips here... got my start in the Yahoo! groups, special thanks to Paul Roark. I use his Eb6 inkset, mix my own with the same percentages you list in your opening post. My black is Eboni. However, my base is not gloss optimizer but what he calls Cb6 (recommended for the 1400 printer). It consists of 59% distilled water, 29% glycerol, 11% PhotoFlo and 1% Edwal LFN. I'm curious about using GO as a base... does your mix work on glossy paper? I'm using matt paper myself.

Hope I can catch up with you folks!

TIA,

Paul