"Carbon on Canvas," 32 x 40 inches
(Carbon toned with the minimum and best cool pigments to make a neutral print.)
For my "carbon on canvas" project, click here.
First, decide whether you need to print glossy paper, including the "baryta" type papers. If you are OK with just matte paper printing, then the "Eboni" MK carbon based inksets are the best and least expensive. If you mix your own dilution base, the cost can be about 1% of the cost of ink in small Epson carts at full retail price. The Eboni based, matte carbon inks also clog the least. I think the Eboni Variable Tone inkset is the best there is -- as long as matte paper is your target. Moreover, for display under glass, acrylic or mylar (that is, where the image is protected), matte paper is the best and easiest to deal with. See http://www.paulroark.com/BW-Info/3880-Eboni-Variable-Tone.pdf for a typical setup.
Thanks to Alain Oguse (firstname.lastname@example.org), the 3880 Carbon Variable Tone PDF is now available in French. See http://www.paulroark.com/BW-Info/3880-Eboni-Variable-Tone-01-2017-fr.pdf.
Below I discuss the glossy version of this inkset approach. I have it loaded in an Epson 9800 for the purpose of printing on large canvasses for a September 2017 show at Gallery Los Olivos (Los Olivos, California, just north of Santa Barbara -- think the movie "Sideways"). However, I must say that I actually am preferring the matte canvas under acrylic. After the show and any orders for the glossy canvas, I may well pull the inkset and revert to what I consider the absolute best, which is the matte, "Eboni" MK based variable tone inkset mixed with the generic "c6b" base.
The various dilutions of Eboni are sold pre-mixed by MIS Associates, Inc (aka InkSupply.com) as part of its Eboni-6 inkset.
In general, bulk inputs for the inksets I formula can be found at https://www.inksupply.com/roarkslab.cfm. I am not connected to MIS, nor do I receive any royalties. Everything I do with respect to inks is open source, and there are alternative sources and inputs. That said, I am happy to have MIS make these well tested inputs available to us in reasonable quantities and at good prices. I often buy in larger quantities from the wholesale supplier, STS Inks, but these inks are not sold from its web page. MIS/www.Inksupply.com is usually the appropriate retail source for U.S. purchasers.
Second, if you want to print on glossy paper, as well as matte paper, then the Glossy Variable Tone approach using MIS glossy carbons is what I use and recommend. See, for example, http://www.paulroark.com/BW-Info/Glossy-Carbon-Variable-Tone.pdf for a typical setup. (For a French language version of this PDF, click here.)
The glossy carbon inksets cost more to mix because you cannot use the generic matte base that I've formulated. (Regarding that base, I recommend you use version c6b. For Eboni as well as Canon and HP pigments, it has worked perfectly for me for years.)
Glossy carbons will also clog a bit more because they have binders in them to stick the pigments to the glossy paper. The inks are still among the best in this regard, but the more binders (glue) the more clogging, in general.
The glossy carbon inksets I use are mixed from MIS ("UltraTone") PK and "amber" base. These are sold from MIS's https://www.inksupply.com/roarkslab.cfm page.
Universal neutral toner: For both the matte and glossy carbons we are most fortunate that the same toner can be used. MIS sells a version of this premixed. I use Canon Lucia Blue and Cyan pigments with generic base c6b for my printing. The same profiles work with both. I expect the MIS premixed toner to be as stable as most third party B&W inks. The Canon based toner with carbon, particularlly on the best matte paper, makes what I believe to be the best neutral B&W -- more stable than the old silver prints. But 100% carbon is, of course, even more stable.
A single light blue toner is what I've found is the best compromise between flexibility and ease of use and profiling. I have found that very few can master the profiling of a two-color (cyan and magenta) ink system.
The best printer to start with is probably the Epson 1430 (1500 outside the USA). I have had 3 of this series, and they have been the most reliable printers I've ever used. Most Epson printers are compatible with the inks I use and recommend. HP and Canon printers are not.
More detail on carbon printing and the approaches is found below.
In the past, we B&W photographers prided ourselves in the archival nature of our lightly selenium toned silver prints. With the advent of digital tools, the inkjet print was, for good reasons in the early days, considered second rate. Most today still are if one is interested in fine art black and white. While today's best color pigments are very good and better than the "wet darkroom" color processes, those pigments are not up to the silver print standard. Developing the digital and inkjet processes to achieve "silver print" class image stability and quality has been a significant effort of mine and some other like-minded people.
Today, the best carbon pigment prints we are making from dedicated B&W printing systems have exceeded the silver print
in a number of ways, including image stability.
Just to get your attention, here is a lightfastness comparison of an inkjet 100% carbon pigment print versus a selenium-toned
silver print (fiber base, not RC). The testing was done by Aardenburg-Imaging.com.
The midtone, Lab L = 50 test patch delta-e values
(total change in density and color -- lower is better) after 100 Mlux-hours of light exposure
(about 50 years of display) were as follows:
100% carbon pigment print: Delta-e = 0.1
Selenium toned silver print: Delta-e = 1.2
That is, this (MIS "Eboni") carbon print was, by this measure, 12 times more stable than the silver print. The point is really to stress that with the right materials, inkjet prints can be significantly more stable than the classic silver prints. But, the materials and processes are critical when it comes to making the best black and white prints. The higher the carbon pigment content, the better the likely stability. Of course the desired print tone may necessitate some added color, but for stability, that color needs to be held to a minimum and the quality of the color pigments used must be very high.
(See a screen shot of these test reports for 100 Mlux-hours light exposure here. For the full test reports, see Aardenburg Imaging and Archives. )
The visual image quality of, particularly, B&W prints from inkjet printers designed mostly for color printing, and composed mostly of bright color inks, has also been a problem. A good, appropriately toned, B&W image is best and most reliably obtained by having a B&W printing system that is dedicated to B&W, having carbon inks in most if not all of its channels. I use an approach that has carbon in all but one of the positions (8 carbon inks) and one light blue pigment toner position that offsets the slight carbon warmth when, for example, I want snow or clouds to be totally neutral white.
In short, a predominantly carbon image is best for B&W, and a printing system made just for B&W printing can make a print that is objectively better than the old silver print technology. The need to serve the mainstream color printing market has made the OEM systems very good at color printing, but for B&W the higher percentage of carbon used by specialized carbon inksets can provide greater quality as well as value.
For my top prints I use "Eboni-6" carbon pigment inks (developed by me on an open source basis and sold by MIS Associates). These are made on cotton-based fine art papers.
That said, I also have developed a B&W dye printing process for shorter term, un-glazed display, and a new project will explore carbon on canvas for very large displays.
These pages document the printing approaches I have developed and used since the transition from darkroom to digital.
High carbon content images on cotton-based papers are for the long haul, while B&W dyes for shorter term display and canvas for very large sizes can certainly impress in the short run.
The best B&W printing approach I have found for fine art is to use an "Eboni" MK carbon based inkset with a single light blue toner. I call these by various names, such as "Eboni Variable Tone." In 2015 I set up an Epson 1400, a 7800, a 3880, and, finally, a 9800 with these types of inksets. All of the PDFs relating to these projects, linked above, are similar. There are two main variations. With the 1400 and 3880, the toner is in the Yellow position so that control with the Epson driver is very easy. With the 7800 and 9800, the toner is in the LLK position because I use the Y position for a second MK, which is needed to achieve the best dmax with Arches watercolor paper. I assume people using this wide format professional printers will be comfortable with QuadToneRip.
All of the above inksets are variations of "Eboni-6," which is sold by MIS Associates or can be mixed (diluted from the MK) by the user. The light blue toner is a blend of Canon Lucia Blue (55%) and Cyan (45%) pigment, then this is diluted with the generic base, which is sold pre-mixed by MIS. The toner itself is not sold by MIS and needs to be mixed by the user. The 1400 and 7800 setups diluted the Canon pigment with 90% base. The more recent ones dilute the Canon mix with 75% base, which is plenty light and puts less water and chemical on the paper.
For most who do not want to do any mixing, Eboni-6 alone works very well, and MIS has a number of older B&w inksets that I've designed. However, these Variable tone Eboni inksets are, I believe, the best ones yet.
In reading my various PDF, it's important to note that, as new information is developed about the materials we use for printing, my views have changed and may well change again in the future. As such, the dates of my various PDFs as well as my stated preferences here need to be taken into consideration. I try to base the opinions expressed here on the latest information available, including the findings of Aardenburg-Imaging.com, Wilhelm Research, the Digital Print Preservation project (dp3project.org), my own fade testing, as well as many other sources. In that respect, those who follow the forums I frequent will note that my initial top blue toner did very well on a fade test but failed in a long term ink stability test. So, my current top toner is a Canon-only pigment blend that proved itself to be a very stable suspension, and in fade testing showed itself to be essentially equal to a selenium toned silver print. So, while 100% carbon is the best, carbon toned to neutral is still very good, and if I find a better approach, I will let people know and follow that better approach. So far, nothing has come close to the stability of carbon as the basis for modern B&W printing.
See http://www.paulroark.com/BW-Info/Eboni-6.pdf for general information about the Eboni-6 100% carbon inkset. Eboni-6 and its variations will run in a number of Epson printers from the WorkForce quads to 8-ink K3 printers.
Beware that some companies call their inks "carbon" pigments even though they may be blends of carbon plus weak color pigments. There is no such thing as a neutral 100% carbon pigment ink. Carbon pigments print with a warm tone by their nature. What is unique about Eboni carbon is that it is capable of making more neutral matte prints than the other popular carbon ink approaches.
For those who expect their prints to be considered worthy successors to the silver (or noble metal) prints, the issue of longevity is very relevant. For those who don't care about long term image stability, the ease and economy of carbon pigment printing may still be enough to make the approaches I recommend worth considering.
I generally mix my own carbon inks using the formulas specified at http://www.paulroark.com/BW-Info/Ink-Mixing.pdf. I call these DIY mixes as "Carbon-6" to distinquish them from the MIS trademarked "Eboni-6" inks. They use the same profiles.
One of the most critical problems of the third party blended carbon inks is print tone shifting, often to greenish, as one color (magenta) fades more quickly than the other (cyan). See, for example, this comparison of a 100% carbon print versus a typical third party blended carbon plus color inkset after 140 Mlux-hours of light exposure. In theory, we'd like to have a single, very high quality blue pigment that directly offsets the carbon warmth. Then, as it faded faster than the carbon, the print tone would simply warm in a straight line to the natural carbon warmth, which with Eboni is not bad to start with. Unfortunately, that ideal blue pigment does not exist. So, we have had to rely on alternative approaches, but using cheap cyan and magenta inks is not one of the successful ones.
HP's approach with its Z3200 PK, LK and LLK was to very carefully match the fade rates of the colors it uses in those blended inks. Fade tests of its Vivera inksets have shown impressive resistance to color shifting. Thus in several recent inksets, I have used the HP cool gray inks as the "toners." See, for example, http://www.paulroark.com/BW-Info/7800-EbHP-2013.pdf for the description of an Eboni plus HP toning option. However, the toning range when using the HP gray inks is minimal, and HP must start with warmer carbon (smaller particles due to the need for glossy compatibility and the thermal head technology) than I can use for matte printing only using Epson piezoelectric heads.
In my current variable-tone inkset approach I focus on finding the best color inks that come closest to the ideal blue color and thus minimize the "hue angle" between the two. This search has caused me to select Canon Lucia EX blue and, now (toner v. 2) the Canon Lucia Cyan (actually somewhat bluish itself) pigments for the toner. The hue angle between them is only 50 degrees, as opposed to the typical hue angle of 120 degrees that separate the C, M and Y inks in the typical color inkset. See http://www.paulroark.com/BW-Info/LAB-color-wheel-2.jpg for a visual description of the strategy behind this approach. With this smaller hue angle difference, less color is needed and any differences in fade rates will have much less impact on the image.
The reality is that most inkjet papers will probably fail before the carbon image fades. As such, having a small amount of the highest quality color inks may have little impact on the print's expected life. 100% carbon is the ideal, but it is not needed for the vast majority of prints that most of us make. So, being able to print a truly neutral tone print has significant real world value. On the other hand, 100% carbon on Arches *watercolor* paper should outlast all of the inkjet prints, as noted below.
Acid free and buffered paper is needed for any print that is expected to have a long life. Lignin in paper made from wood is typically the source of acids in cheaper paper. Even if there is sufficient buffering in the paper to absorb the acids formed by the breakdown of the lignin, the lignin itself will yellow the paper.
Cotton-based paper is often considered the best, in part, because cotton is virtually lignin free. However, without buffering in the paper, air-borne acids will still attack the cotton fibers. This is what I've seen happen to my well-processed wet darkroom prints that were stored in a manner that allowed air-borne acids to reach them.
OBA's (Optical Brightening Agents) are dyes added to papers that convert UV into visible light and make the papers appear brighter. They are widely used even in cheap "plain" papers made for copying and text printing. However, not only will the OBA dyes fade, causing the paper to warm, but they also appear to be implicated in reactions that will cause staining of dark stored prints. This issue is under-reported in the materials and testing most photographers rely on. The best explanation I've seen is that of Mark H. McCormick-Goodhart, Director, Aardenburg Imaging & Archives, in his August 29, 2015, comment on the Luminous Landscape forum, near the bottom of the page.
The bottom line for OBAs is to avoid them for prints desinged to last a long time.
Good, deep blacks are a characteristic I, like most B&W photographers, expect in a top quality B&W print. Here, I have found and verified with 1 degree light meter readings that the best matte papers usually have a deeper black than do the glossy papers when both types are displayed indoors, on the walls, under glass/acrylic in normal office and home environments. Glossy papers can have a great dmax in ideal and very bright lighting (including spectrophotometers). However, in typical indoor display environments, and when protected under standard picture glass or acrylic, the reflections from the glossy surfaces and glazing usually give the matte papers an edge. It's all about the reflections.
On the other hand, if the print is going to be viewed with no glass or acrylic over it, the "glossy" papers can have a substantial visual advantage. B&W dyes on Red River's glossy "metallic" paper is my formula for the prints with the most "pop" and "three dimensional" feeling. Dyes go into the paper coating rather than lying on top. The latest Epson dyes are very good, and with a proper acrylic spray the coating and dyes can be significantly protected. Here, the particular characteristics of the paper surface can make a major difference. Some papers seem to have an anti-reflection coating built in. Some have hard surfaces that resist fingerprints and abrasion significantly better than others. These may not do well in torture testing, but for medium term home or office display, they can be very good indeed. Pigments and an inappropriate protective spray on top of the highly-engineered surface of a modern glossy paper can seriously affect the potential of a paper-ink combination. As usual, which approach is best varies with the circumstances.
The amount of color needed to neutralize carbon warmth tends to be least with matte papers and most with glossy papers. The natural, un-coated papers tend to be more neutral by nature. With MIS Eboni-based inksets, there are a few coated matte papers that are also quite neutral. Glossy papers tend to print sepia tone with just carbon. As such, they require heavy toning with color pigments.
Physical deterioration of the paper is a factor that is not well tested. While it is likely that an image composed of 100% carbon would outlive the paper base, finding objective data on the relative expected longevity of different paper types is difficult. The coatings and laminations on inkjet papers appear to be the most likely to cause problems. Differential expansion and contraction caused by humidity and temperature fluctuations tend to crack and separate laminated paper substrates over time. These factors are not being tested in popular long term storage testing protocols, where humidity and temperature are kept constant and not cycled. Historical evidence is probably our best guide. See http://www.dp3project.org/preservation#crackinginkjetlight for the Digital Print Preservation Project's look at cracking of inkjet coatings. See also http://www.luminous-landscape.com/forum/index.php?topic=87926.0 for a comparison of how different types of inkjet paper did after 6 months of sun exposure through a window. Both of the tests, above, involve treatment of the paper that no one would subject a valued photograph to, but they probably also indicate what will happen eventually to these coated papers.
In the past, the photographic "RC" papers had the worst reputation. The extent of that reputation that was due to specific photo paper characteristics as opposed to the nature of that type of laminated substrate is unclear.
In my own old photo digital restoration work, I find what I call "micro-cracking" of the emulsion rather typical of many of the oldest photos. This surface deterioration is often the limiting factor in pulling information off the old prints. Some conservators have opined that all coated or laminated substrates will eventually crack and/or delaminate.
Arches un-coated watercolor papers have an outstanding reputation in the painting world, where they have been used for hundreds of years by artists. And they can make excellent carbon pigment B&W prints. In general Arches is not quite as smooth as coated inkjet papers, but more than adequate for large display prints. Note that the backside of the paper is much less likely than the front side to have fibers sticking out of the surface. While there are other high quality watercolor papers, Arches achieves the highest dmax (deepest blacks) of any un-coated paper I've tested. In fact, Eboni v. 1.1 achieves a dmax that is better than the vast majority of inkjet papers with OEM MK. With no coatings or laminations to crack or flake off, these un-coated watercolor papers are likely to age much better than coated inkjet papers. While I suggest this paper as a benchmark, and I use it, uncoated watercolor papers require special workflows and inks to make good images. They are not what I recommend for most people. There are many very good inkjet papers. The best matte papers look excellent and will last a very long time with proper handling and display. I note and profile many of these in the various PDFs I have posted on printer and ink workflows.
With respect to the unique visual quality of the dyes, it is probably the fact that pigments lie on top of the glossy surface, whereas dyes take the image into the coating and off the high gloss surface that makes the difference. The pigments on top of the paper result in a number of artifacts and relatively veiled highlights. The unique look of the metallic paper with dye results in an image quality that is, in some important respects, beyond what I've seen before.
Note that advances in the molecular structure of dyes have made the best dyes reasonably lightfast. Epson's advanced dyes (perhaps based on Fujifilm R&D) have been rated by Wilhelm Research at close to 100 years of display under glass. Epson's latest "dry lab" professional printer uses the "UltraChrome D6" inkset, which uses dyes and is marketed as producing "archival" prints, although their literature also indicates a projected display life of "over 80 years," which some might think is a bit shy of what the vague term "archival" usually implies. Other, more detailed and critical tests suggest that, particularly when a protective spray is used (and I do for all display dye prints), a reasonable, what I would call medium term display life is likely with advanced dyes, and photo album life may, in fact, be good enough to justify Epson's claimed "generations."
A priority of mine has been to find ways to improve the quality of prints made with advanced dyes. In this regard, in 2014 I changed my (Epson-Noritsu) dye inkset approach using an Epson 4000. Ironically, I switched from a dedicated B&W approach to a typical OEM "K2" ink arrangement. Unlike with pigments, the color Claria/Noritsu dyes are more lightfast than the black dye. This change should result in a significant increase in the likely display life of the dye-based prints I make. See this current approach, which should work for any Epson K2 and K3 printers, at www.PaulRoark.com/BW-Info/4000-Noritsu-2K.pdf.
I have been using the Epson 4000 and the dyes with a 17" roll of Red River Polar Pearl Metallic paper to make most of my display prints during the last year. After spraying them with Lascaux Fixativ, I dry mount the prints on foam core and use a simply black frame for display -- no glazing and no mat board.
The main visual weakness of this medium is that it can display color shifts when viewed under different lighting conditions. This was mostly a problem with old-style fluorescent lights. Older lights with Color Rendering Indices (CRI's) of less than 80 may result in a green/cyan tone shift that is significant. I and the gallery I most often show in have switched to LED lights. For me, that has essentially solved this problem.
I have made B&W profiles for the Epson 1400 and its OEM Claria inks, using QTR, and have outlined my recommended approaches in a PDF at http://www.paulroark.com/BW-Info/1400-Claria-BW.pdf. I currently use a 1400 with Claria and Noritsu inks, in the standard (color) arrangement both as my color printer (using the Epson driver in its usual manner) and B&W printer (where QTR is really needed) for cards, brochures and other high glossy output.
Another alternative is a dedicated B&W approach with the 1400/1430. This does requier the mixng of an "LK" inkd. With this approach the Epson driver becomes a viable way to print good B&W with the dyes. Dedicated B&W inksets like this are much easier to profile, though, ironically, the black dyes are not as lightfast as the color dyes. See http://www.paulroark.com/BW-Info/1400-Claria-Noritsu-2K2LK.pdf.
"Carbon on Cotton" for best image stability and low cost;
MIS 100% carbon pigment inksets;
-- Eboni-6 for smoothness;
-- Eboni-6 details, including for the 1400;
Ink Mixing, including for Carbon-6, HP, and dyes; for former darkroom workers and others so inclined;
High Sierra Workshops at the historic Golden Trout Camp;
Solvang, CA, USA