• Epson 3880
  • 3rd party refillable cartridges
  • Colormunki Photo

Just thinking creatively here.

Is it possible to replace an ink color in an Epson 3880 with a completely different color in order to increase color gamut in the ranges that I typically use in my prints?

For example, I never print B&W and usually print landscapes, so is it possible to replace the LK or LLK and another ink with green and orange inks? (Of course new profiles are needed—I have the Colormunki Photo.) I am concerned that there is something in the firmware of the printer or the driver that makes assumptions as to the general color loaded in each slot.

If it's possible, how do I make the best choice for which inks to remove?


Due to responses, I wanted to be clarify that I have no interest to re-engineer an ink system that is "universally" better than what Epson has done with their 9 ink system (the 3880 has 9 inks). Rather, I would like to "tune" the combination of inks to the specific color gamuts I typically print. When plotting gamut as a 2D or 3D shape on a graph, that means removing available gamut on some sides and adding it to others (in theory). No matter how good the OEM ink combination is, they can't make a "one-combination-fits-all-prints" solution. I think it would be cool if the printer firmware allowed for this.

  • 1
    \$\begingroup\$ I'm pretty sure what you are proposing is well beyond the limits of even the most flexible calibration methods. You'd have to rewrite the entire color management section of the firmware for the printer in question. \$\endgroup\$
    – Michael C
    Commented May 1, 2018 at 2:45
  • \$\begingroup\$ You're probably right. I was hoping to discover someone who attempted something like this and check their result! I think I'd need to know more about what exactly is in an ICC profile. If the "which shade to which cartridge and ink volume" conversion happens in the firmware, it sounds like there's no getting around it. \$\endgroup\$
    – mach
    Commented May 2, 2018 at 5:08
  • \$\begingroup\$ Increasing the gamut is relatively easy; but, it may not help with the correct rendition of the image. The gamut is the overall area of possible hues, not where a specific hue will appear in the image. For that, you'll have to re-write the software for the colour breaks. \$\endgroup\$
    – Stan
    Commented May 2, 2018 at 13:40
  • \$\begingroup\$ Printer ink isn’t just a dye in water. Fluid properties are critical. The odds of you formulating a working ink are small. \$\endgroup\$
    – Eric S
    Commented May 2, 2018 at 18:27

5 Answers 5


The Epson 3880 uses a set of 9 cartridges. It's still a CMYK sort of system, but more like CCMMYKKKK. With four different varieties of black, it's understandable that you might be thinking that replacing "light light black" with "dark cyan" might give deeper blue evening skies, or adding "dark magenta" might allow for prettier roses. And yes, perhaps adding "true green" or "vivid orange" might let you shift the printer's gamut in other ways.

But there's obviously more to it than just swapping in a new tank. First of all, you'd have to find a tank with the color you want, and it's not like Epson probably makes non-standard colors. Next, you'd have to tell the printer about the new color, and it'd need to know how to use it. I don't have any particular experience with this printer, but the ones I've used expect a specific color in each position -- there's no provision for changing colors.

So I think you've got a great idea, but I don't think it's feasible without some serious firmware hacking.

  • \$\begingroup\$ You're probably right about the firmware needing changed. I wanted to mention Epson does make green and orange inks in their more expensive 11 ink printers, so 3rd parties like InkOwl and such make green and orange inks in bulk available to use with refillable cartridges. \$\endgroup\$
    – mach
    Commented May 2, 2018 at 5:04
  • \$\begingroup\$ Unable to edit comment for correction - I can't find green and orange on InkOwl's site, but I know I've seen them available somewhere. \$\endgroup\$
    – mach
    Commented May 2, 2018 at 5:16
  • \$\begingroup\$ I wonder if the reason for multiple colors has more to do with avoiding over-saturating the page with ink than improved color reproduction. \$\endgroup\$
    – xiota
    Commented Jul 9, 2018 at 21:00

Get the chromaticity coordinates of the ink and plot them on a CIE Chromaticity diagram.
It's the colourful one shaped like a lung.

The coordinates are given as an "X" and a "Y" which you can locate along the x-axis and the y-axis of normal graph paper. If you want something fancy, you can download a full colour one with the X and Y axis labeled with a scale.

For a 4-C process, you'll have one pair for each ink. Plot the points. Connect them and you have the gamut enclosed inside the boundary.

Now, you can start your research. Find another ink with a point outside the ones you have and that ink will extend the gamut of the ink set according to the plot. The white-point will be at x= 0.301, y= 0.301. If the point you want to experiment with falls in line with the standard printer ink AND the white-point, you have yourself a genuine useable ink-set with a greater gamut than the original factory version.

How can this happen? Some inks are made from expensive components by non-standard procedures not commercially viable although superior in one or more ways. There are aqueous and solvent inks. There are concessions that limit quality or quantity of products. Just because a corporation is large, it is driven by small development teams of two to five individuals that do heavy lifting intellectually.


Modern color imaging is based on the 19th century trichromatic color theory of color vision proposed by Young. The first color picture was made in 1861 by James Clark Maxwell using three pieces of black & white film exposed sequentially thru red, green and blue filters. Black & white slides were prepared and projected on a screen, one projector equipped with a red filter, another green and the third with blue. When the three images were superimposed, the world saw the first color photograph. Gabriel Lippmann demonstrated color photographs, no filters, using an interference process. Dr. Land of Polaroid fame demonstrated a two color method using white and red light. The Lippmann and Land methods are curiosities that were never commercially practical. The Maxwell method prevails today.

TV and TV projectors and computer screens use an additive color method which entails fracturing a vista into tiny points of red, green, and blue and displaying them for viewing. This method controls the intensity of each of the millions of red, green and blue pixels (picture elements) that comprise the image. While this works for projected and screen displays, it fails when the image is presented as ink, dye or pigment on paper.

For images on paper we resort to dots of color using the subtractive primaries. These are the complements of the additive primaries. Complement in this usage is “opposite”. The complement of red is cyan, a red light blocker. The complement of green is magenta, a green light blocker. The complement of blue is yellow, a blue light blocker. The CMY system uses the subtractive primaries to control how much of the three primary colors will reflect from a surface.

Why cyan, magenta, and yellow? Each consists of two colors. Cyan is blue + green. Magenta is red + blue. Yellow is red + green. Each acts as a filter that limits how much of its complement (opposite) will be reflected. In other words, cyan is a red blocker, magenta is a green blocker and yellow is a blue blocker.

These subtractive primaries work OK, but they have a deficiency. They should when overlapped make black. Sorry to report that they can only make dark gray. This is because the cyan dye/pigments and the magenta dye/pigments are slightly off hue. Nobody yet has found the right colors. Yellow however is very good. Because the three when overlapped fail to make black, we add black when we can, and this adds contrast. The added black is called by the printing industry a “Kicker”. Thus the term CMYK.

It’s OK to play around, changing the colors of the inks used in your inkjet. Just remember, some of the best minds in the business have been working on this for more than a few hundred years. Not to discourage you. Who knows what will come out of experimentation! I think, you will need to change the software and the hardware to get improved results. Also, high end printing (lithography and printing press) often add additional colors to the CMYK as enhancements.

  • \$\begingroup\$ That's all good info, but I think it misses the point of the OP's question. If adding more ink colors (like light cyan or light light black) improves the printer's real-world color rendering capability, can other colors be used to shift the gamut in a different direction? \$\endgroup\$
    – Caleb
    Commented May 1, 2018 at 20:22
  • 1
    \$\begingroup\$ @ Caleb - I misses nothing -- For years engineers have been trying to improve on the CMYK method. OK to experiment - what are the are the chances he will improve? Slim to none I think! \$\endgroup\$ Commented May 1, 2018 at 20:52
  • \$\begingroup\$ It is a lot of background info, but partially answers the question of which inks to replace, if attempted. Removing the darkest black ink is out of the question, for example. I'm a little confused how Epson's "light" and "light light" colors add to the gamut if they are overshadowed by their darker brothers. \$\endgroup\$
    – mach
    Commented May 2, 2018 at 5:00
  • 1
    \$\begingroup\$ @ mach -- Its a good question --- It’s a more complex science then you think. The color you see is a combination of the color of the ink, the size of dots of ink, how the dots overlap, the angle of the laydown for each color, the degree of white space between the dots etc. Because it is difficult to control the dot size because of capillary action of the paper pulp, the intensity can be controlled by using light and dark ink/pigment. Look up lithography and half-tone as these subjects are the randfather of the concept of the inkjet. \$\endgroup\$ Commented May 2, 2018 at 6:20
  • \$\begingroup\$ Interaction of ink and paper is referred to as dot "gain." Dot gain is a "single-merit" factor in printing quality. It is the factor that determines the necessary dpi for a given paper stock for optimal "acutance." Newsprint (high d.g.) is printed at low (65) dpi. Baryta coated stock (low d.g.) can be printed at high (300) dpi without "muddy" shadows. \$\endgroup\$
    – Stan
    Commented May 2, 2018 at 13:15

Just thinking creatively here.

Is it possible to replace an ink color in an Epson 3880 with a completely different color in order to increase color gamut in the ranges that I typically use in my prints?

Possible, no. Instead of "a completely different color" you'd want to use "a somewhat similar color". Darker or lighter, and run the paper through twice; also write the ICC profile and use Colormunki.

I wouldn't expect to beat engineering teams with millions of dollars behind them. You are counting on discovering something that they don't know about printing and color management.

You probably can get good results if you stick with spot colors (regular colors and one or more customised shades). The more cartridges in your printer the finer the control over shading, and the greater difficulty.

Some work experimenting in this area has been done with edible inks (food coloring).

Kopykake has a .PDF offering this advice:

"The Epson Eco Tank line of printers does not use troublesome ink tanks with microchips. Instead, these printers have ink reservoirs that are easily filled and refilled or topped off with Kopykake’s Edible Ink Refill bottles.".

You'll have an easier time mixing food coloring instead of inkjet ink and can eat your mistakes.

Some articles on reengineering the color management system and developing your own inks and color profiles (along with tweaking the printer to work this way with Windows software) can be found by searching for "edible printing cake" or similar search terms.

It is so popular that it's even possible to buy Canon or Epson printers from Walmart especially for this purpose, at a reasonable price. Professional cake printers can cost thousands.

Once you get the hang of mixing safe inks, creating new colors and profiles, and producing desirable results, then you can switch to mixing inkjet ink.


Yes... if:

  • You can revere engineer the internal printer's drivers and color interpretations of the matrix sending the signal to the tank.

  • You can design a color separation mode.

  • You can fabricate a chemically correct pigment that is "more green or orange" that have the correct particle size, and solvents so the ink is compatible with the printer, besides toxic and stability issues.

  • You have a workflow that is able to calibrate the color output, prepare the color profiles, color modes conversion matrixes...

No, because:

  • Nobody makes those inks. If they were they would be selling them everywhere.

  • Mixing current inks (cyan and yellow and magenta and yellow) has no sense because they are the exact same inks.

  • There is no current hexachrome system. There was one by Pantone some years ago, which had a special color separation but it is closed since 2008. https://en.wikipedia.org/wiki/Hexachrome

If it were feasible these days, some brand would be selling them and Pantone would not have closed Hexachrome.

But you could try to spend some millions on research and probably patent issues. Then probably yes.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.