FNO - Fading in black/cyan/"photo" dye based inks

[canonfodder]

Grandad35,

I am about ready for you to try my color swatches! Some will be ready for mailing on Tuesday.

They are on strips of printer paper. I am excited about the prospect of seeing the difference in inks AND in printer papers, because the color swatches may show differences in both. In addition to tests on your spectrophotometer, the swatches are suitable for visual comparison by a good set of eyes. They can also be scanned, but whose scanner do you trust? Not mine.

I have considered 6 different ways to produce color swatches. None of the 6 included painting or spreading the ink by hand. I have done trials on 3 of the 6 methods. The last of the 6 is working. I have written a paper on the methods.

I wrote up the details on all 6 methods considered, but that is too much. After finding some success with number 6, the other 5 don't really matter to anyone but my records.

I am condensing the written material into a simple procedure for making the color swatches. That will be ready in a few days.

 

[Grandad35]

canonfodder,

When you say "printer paper", I assume that you are talking about "Photo Paper", not generic bond paper used for text based documents. While we would get readings on bond paper, I am not sure how the colors would relate to printing photos.

The first test should include a validation of the repeatability of your printing technique. I would like to include 5 "identical" paper samples, each with the same ink swatches. If there is ANY operator influence on how they are printed, two additional samples should be supplied that are printed by two other people to test this effect.

PM me and I will send you my address.

 

[canonfodder]

Yes, that is definitely Photo Paper. I have been testing with Kirkland Professional Glossy and a small amount of Canon Photo Paper Plus Glossy. I am out of the Canon. I have some HP and some Epson paper, but I think they may be older processes and not represent current technology.

I have been making some swatches at what I think are full intensity or saturation, and also some at 50% of that level. It would seem to me that better measurement results might be obtained at something less than full saturation. Of course when a level is picked, all inks would be treated alike, not adjusted individually. The desire to compare inks or monitor changes in inks will only be served by treating all to the exact same process.

At about what intensity should swatches be made for your spectrophotometer? I guess you could tell better by trying some that I have made at the two levels. Maybe I will include another level between the two. We will call it 75%. The method is readily adjustable.

My present swatches are about 1/2 inch by 3/4 inch. I know that is smaller than you suggested, but the consistency of color across the swatch is rather good, so a more limited averaging should be acceptable. We can talk more about this as things develop.

 

[Grandad35]

canonfodder,

The Kirkland paper is a good choice, since it is widely available and many people on this forum use it as their standard - it is my standard paper for "snapshots" that aren't destined to be framed and displayed for a long time.

Selecting the ink laydown rate is a difficult problem, in that it is very easy to overdo it. For example, if 100% just gives full saturation, 125% won't show much difference. I would start by printing this image (http://www.nifty-stuff.com/img/files/Fade_Test_Chart.jpg) and using its saturated colors to get an idea of what colors your printer can achieve. Since the purpose of this work is to develop a test that will allow us to compare the colors of various ink sets, it is not necessary to achieve maximum saturation on your test patches. For this reason, I would try various ink laydown levels to find where the saturation is slightly below the maximum levels achieved on the test print. As a one-time exercise, it would also be interesting to run test sets at 50%, 75%, 100%, 125% and 150% of this level to see what happens as the ink laydown is changed.

I can use the 1/2" x 3/4" samples, but I would like to see several samples of each color printed close to each other to get an idea of the repeatability (at least until we develop confidence in the printing/coating process).

I look forward to reading about your coating process.

 

[canonfodder]

Grandad35

The color swatches are in the mail. By this point in my trials I believe that my "100%" intensity is too high, and I may have to set a new 100% about where the present 50% is. I have used Kirkland glossy photo paper and MIS CSI-8 ink for my tests.

I am still working on the write-up of the method used in making the swatches, but I will describe its basis.

The method uses a tightly controlled procedure to dye a strip of glossy photo paper using the dye type inkjet ink as the dye. An even intensity across the swatch results.

The equipment required:
A test tube or other similar container, a 1mL syringe or similar precise pipette, a timer or clock with sweep second hand, newspaper, and facial tissue. 1/2 mL or less of each ink is used for the make up of a sample, and a sample can make multiple swatches at multiple intensities.

In addition to making color swatches for analysis by your spectrophotometer, the swatches make good visual samples of the ink's colors. Mixtures trying for red, blue, and green can be done. Swatches can also be scanned if your scanner is good enough, or scanned to see how bad your scanner is, like mine.

 

[canonfodder]

Grandad35,
When we are ready to post the swatch prep details, should a new thread be started? We have gotten pretty far off the subject of "Fading in black/cyan etc. "

 

[canonfodder]

Grandad35 and other members,

Slowly some questions have built up in my mind about this making of color swatches. I am not at all sure of the answers. I know that I should have asked these questions before starting my search, but I operate in a disorganized mode at times.

A. What properties of the ink are we trying to test? I would suggest some possibilities:

1. Color "purity". How near is the color to the intended color. "Is the cyan at the correct wavelength?" etc.
2. Color "intensity". The strength of the ink color.
3. How the ink "works" with a photo paper.
4. How well the inks of a set mix for producing RBG.

B. What is the testing to enable?

1. Measure the consistency of a supplier's ink.
2. Make comparisons of different supplier's inks.
3. Quantify or qualify how the different inks "work" with a photo paper.

I started out with an open mind and dreamed up several approaches to making color swatches, but now I have more ideas that branch off somewhat, and I realize that a clarification of the goal or goals is needed.

 

[Grandad35]

Canonfodder,

My original goal is stated in post #3 in this thread, specifically:
"The goal of testing these color swatches was to start a catalog of CIELab colors for known inks so that we could (hopefully) identify an unknown ink by matching its colors to a known ink. It might also be possible to individually select each ink color from the supplier with the closest match to the OEM inks for those who don't want to deal with profiles and were willing to run a "mixed supplier" inkset. At the very least, we would have a scientific basis for deciding which inks caused which color shift."

1. None of the inks that I have seen gives exactly the desired color, probably because there are no dyes/pigments available that perfectly match pure cyan or magenta or yellow. The inks are compromises that try to get close to the desired color, but never really get there.
2. The inks don't "mix" on the paper - they produce distinct color dots that blend together because they are so small (http://www.nifty-stuff.com/forum/viewtopic.php?id=392). How these dots are printed is why the "dither" and "diffusion" options in the printer driver can affect the appearance of printed colors. For more details, also see (http://www.nifty-stuff.com/forum/viewtopic.php?id=455).
3. At best, this test can only measure the color of an inkset relative to other inksets, not how it will actually print.
4. As stated above, for those who don't profile there is the hope that they will be able to select inks from different suppliers that most closely match the OEM inks.

 

[canonfodder]

Grandad35,

Thanks for your answer.

O.K. I do think we are on the right track with the swatches I made.

I don't think that any comparison of photo papers would be meaningful with the method I am using. It might be interesting, but there would be no way of quantifying on any scale. We'll leave paper variations out of this and just concentrate on Kirkland photo paper.

 

[Grandad35]

canonfodder,

I received and tested your samples, but my PMs to you are apparently bouncing due to a "persistent delivery failure" - a message that I sent 32 hours previously just bounced. Do you have an alternate email address that I can send messages to? Is there a chance that your spam bucket is so full that you don't have any more space in your mailbox?

The consistency of your test samples is much better than anything that I have tried in the past, with a typical "Delta-E" (dE) between readings of 0.5. For reference, it is usually stated that a dE of 1.0 is the smallest color change that a human can perceive, and that getting a dE of 2 or less in printing is on the edge of what is possible.