Monday, October 18, 2010

Ink Diary, 18 October

After a month of anticipation, I finally boiled my first batch of linseed oil this past Saturday. On the way upstate I was particularly alert to morbid possibilities. Tales of disaster abound in the manuals I have been reading and they have steadily gained influence over my thoughts. Whole sections of London decimated by flames due to a careless ink maker. Certain pots which, when agitated, produce cascading streams of burning oil. Pre-Industrial towns mandating ink works to operate beyond the city walls because of their incumbent danger and stench. The recipes I have read that involve burning the oil put the process in an ominous context: If the oil has not already ignited on its own. Similarly, recipes that do not encourage burning the oil caution against allowing it to auto ignite, which will surely happen unless the ink maker remains vigilant. The basic recipe that I have decided to follow is a three stage process—boiling, igniting while boiling, and boiling some more—and I approached my first attempt with more than a little angst. During the weeks prior to our trip upstate I had regular visions of molten jets of burning oil erupting from my Dutch oven; decimating, at best, a couple of vegetable beds or, at worst, part of my face. To try to contain the imminent disaster I decided to cheat a little. Rather than starting with regular linseed oil I would try boiling a few cups of already thickened stand oil just to see what would happen. I bought myself a handy Mr Max butane burner figuring I could control it better than I could an open fire; and set up a workstation on a metal covered table that Annie had outfitted in the gravel drive. After a fitful night, I set to work. The process was more like running a deep fryer at a county fair than scrambling to contain Vesuvius. In barely 37 minutes the varnish appeared to be ready—a wooden stick charred immediately when inserted into the oil; and, when cooled, the varnish produced strands over an inch long when stretched between my forefinger and thumb—and I was left with a cast iron pot of thick, sticky goop full of discolored bits of burnt oil. I took the pot off the stove and placed it on the gravel driveway where, to my great frustration, it refused to catch fire. Within an hour a hardened skin similar to thick Saran wrap formed over my thickened batch of oil.

A couple of weeks ago, the printer Arthur Larson turned me onto C H Bloy's A History of Printing Ink, Balls, and Rollers 1440-1850. In it, Bloy has assembled sixty-nine ink recipes ranging from a Chinese block-printing ink from AD 251 to an account of William Morris's ink made by Jänecke of Hannover. The recipes are a fascinating distillation of the literature of a craft in that they show the near infinite distinctions possible with so few ingredients. Oil (linseed or nut), lampblack, turpentine, amber, onions, and crusts of bread. Add to these diligence, time, and fire, and you end up with sixty-nine competing arguments for how to proceed. There are few consistent instructions but there are some general trends: the British say Burn the Oil! and the French say Don't Burn the Oil! Onions or crusts of bread are helpful to either 1) absorb the oil's greasiness or 2) allow the ink to dry well, though some Encyclopædists question their effectiveness toward either end. One recipe quoted in de Champour and Malpeyre's Nouveau Manuel Complet de la Fabrication des Encres [Paris: 1856] is very specific: add 2 kilos bread to each 8 litres of oil in small slices, 3 or 4 at a time. One ingredient that most recipes agree upon, and that is pointedly absent from Baskerville's recipe, is turpentine [see below]. I am a little suspicious of Bloy, though. He has dramatically cut and re-written the Baskerville recipe that appears in Hansard's Typographia. This would be fine except that he does not admit to altering the recipe. What else has he left out of other recipes?

The three most common methods for determining that your varnish is ready are 1) it produces long threads when stretched between forefinger and thumb; 2) the crusts of bread you have inserted into the oil have charred; or 3) a wooden stick chars when inserted into the oil. The time stipulated to produce a good varnish for summer printing is consistently 5 hours. In recipes that call for burning, you boil for 2 hours, burn, and then boil for 3 more. Some recipes call for burning 5 minutes, others for 30. Some for burning while boiling, others for burning off the heat. They all emphasize that the process takes time and attention. After preparing Annie for what would be a long day boiling oil, imagine her surprise when after 45 minutes I came in and said, "I think it's done." For myself, I was deeply suspicious. I had assumed that using the thickened stand oil would reduce the time, but 37 minutes? Further, what was now in my pot didn't look like the beginnings of a fine ink. For one, the clinging, darkened skin seemed an inauspicious omen. Perhaps I could lift it off like fat from a chilled broth and beneath find a smooth, thick varnish? Repeatedly throughout the day I rubbed my fingers along the surface, trying to convince myself that this would work.

Most ink recipes of any length make special note of how important it is to not overheat the oil and to stir it consistently with an iron ladle. I threw mine on a burner turned to high and stood back, awaiting disaster. There is simply no way that I have produced a superior varnish. Next time, I will work at lower heat and stir with the same attention I would give a risotto. In the meantime, I spent an hour in my shop seeing what I could make with my first batch of mealy varnish. It is very sticky, so much so that it resists mixing or handling of any kind. After adding some amber varnish, which is 50% turpentine, it loosened up considerably and I was able to get it to accept the lampblack, though it is not fully integrated. The ink is still too tacky to make a proper draw down but I am hopeful that adding turpentine will bring it to a better working viscosity. I will try this later in the week.

Wednesday, September 29, 2010

Ink Diary, 29 September

Ink 002
Mixed 29 September 2010
First test 29 September 2010

30ml Ink 001 + 1ml Grove's 16th century amber varnish*

I received my 80ml bottle of "16th century amber varnish" from Grove's last week. Although I know the ink I mix with it today will be unusable, I am curious to see the effect of the amber on the ink by doing side by side draw downs with Ink 001 and Ink 002. Grove recommends one drop for a spatula knife full of oil paint. I have know idea what that would translate to in ink. I am starting with 1:30 amber to ink because a 1/4 teaspoon is 1ml and two tablespoons is 30ml (two tablespoons will yield enough ink for my experiments).

Working characteristics:
On working the ink with the brayer it had the same sloppy, splattering consistency of Ink 001. However, on clean up there was noticeably more resistance to the California Special Wash I use. I imagine I'll need to switch to Kerosene for wash up.

The draw downs are dry after only a few hours (with the exception of thicker swells).

The amber varnish has a good deal of turpentine in it. I guess you come to love what you know: although I can practically bathe in denatured alcohol I have always found the smell of turpentine revolting. I suppose I'll have to come to know turpentine, and love it.

* I had considered naming this ink "Ink 001a" because it is simply Ink 001 with an additive. I opted for the name "Ink 002" because all of my ink experiments will be variations on only two or three additives. I would prefer that my final ink be named "Ink 025" or "Ink 026" rather than "Ink 001y" or "Ink 001z."

Wednesday, September 22, 2010

Ink Diary, 22 September

Ink 001
Mixed 21 September 2010
First test 22 September 2010

Total of 1/2 cup of ink made with 1 rounded tablespoon of lamp black and 450 dPas linseed stand oil. So far, Hebert's Encyclopædia is the only mention I've found of the proportion of pigment to oil. his recipe calls for one part pigment to seven parts oil by weight. I mistakenly mixed 1:7 by volume which will result in a highly diluted pigmentation.

Made one draw down at 9:30 am on 21 September. Within 24 hours the ink was dry to the touch with the exception of a few thicker swells.

Before printing:
Many of the 19th century articles on ink manufacture mention the practice of adding prussian blue pigment to black ink to compensate for poor quality black pigment. This practice continues today. Nearly every commercial black ink I have used has a high blue content, making a true matte black impossible to achieve. What I am striving for with my ink is to make a black that reflects as little light as possible, turning each letter into a miniscule black hole around which the white paper shimmers. My first draw down of Ink 001 is a promising step toward my goal. The ink has a lovely matte finish and no noticeable trace of blue.

Working characteristics:
As I had suspected the ink is too fluid—it even splatters a bit when worked with a brayer. I hate to think what would happen if I put it on my press. The consistency and tack is closer to an intaglio than a relief ink. The linseed oil simply isn't thick enough. On Brian Donnell's website I found my first specific numbers for boiling linseed oil. According to Donnell, the linseed oil needs to be brought to 720-750°F before it makes an acceptable ink. That is a considerably higher temperature than stand oil is boiled to. There is no point in continuing until I boil some oil upstate.

Tuesday, September 21, 2010

Ink Diary, 21 September

21 Spetember, Progress on linseed oil

I first read an abridged version of T. C. Hansard's recipe for Baskerville's ink in F. E. Pardoe's, John Baskerville of Birmingham: Letter Founder and Printer [London: 1975]. The general vagueness of the recipe either assumes a prior experience of ink-making on the reader's part or attempts to conceal a lack of specific knowledge on the part of Hansard (it was probably a little of both). Baskerville's ink recipe was a closely guarded secret and the information that Hansard relied upon was gathered either from Baskerville's foreman Robert Martin or, more likely, Thomas Martin, the elder printer's nephew. What first confounded me about the recipe was the method of incorporating amber into the ink but, after further reading, my focus has shifted to the preparation of the linseed oil, the base medium of the ink. Here is the complete recipe as printed by Hansard:

He took of the finest and oldest linseed oil three gallons, this was put into a vessel capable of holding four times the quantity, and boiled with a long-continued fire till it acquired a certain thickness or tenacity, according to the quality of the work it was intended to print, and which was judged of by putting small quantities upon a stone to cool, and then taking it up between the finger and thumb; on opening which, if it drew into a thread an inch long or more, it was considered sufficiently boiled. This mode of boiling can only be acquired by long practice, and requires particular skill and care in the person who superintends the operation, as, for want of this, the most serious consequences may occur, and have very frequently occurred.* The oil thus prepared was suffered to cool, and then a small quantity of black or amber rosin was dissolved in it, after which it was allowed some months to subside; it was then mixed with the fine black, before named, to a proper thickness, and ground for use.

*Here Hansard mentions that linseed oil should not be boiled in the printing office because of the ease by which it catches fire.

When I took my first exploratory trip to Kremer Pigments last week, I showed the clerk Hansard's recipe and inquired after fine, old linseed oil that I could boil. In its stead I was shown Kremer's linseed stand oil, which is a pre-boiled linseed oil that they produce in two thicknesses. I ordered a couple of liters of the thicker oil. Today I mixed some of it with finely ground lamp black which is collected in the same manner as it was in the 18th century. Over the previous couple of days I have played around with little bits of the oil on my fingers and, though it is quite thick and tacky, I have a gut feeling that it is still not thick enough. With that said, the only way I can test it is to make some ink and put it through the press. Intrigued in the meantime to learn more about the preparation of the linseed oil, I consulted Hansard's full text on printing inks in, Typographia: an historical sketch of the origin and progress of the art of printing; with practical directions for conducting every department in an office: with a description of stereotype and lithography [London: 1825]. In it, Hansard quotes from the recipe for Black Ink in Rees's Encyclopædia:

The oil is boiled in an iron pot, capable of holding at least half as much more, because it swells very much; when it boils it is kept stirring with an iron ladle; and if it does not itself take flame, it is kindled with a piece of lighted paper, or burning wood, in order to increase its consistence and tenacity, and to diminish its greasiness. The oil is suffered to burn for half an hour or more and the flame being then extinguished by covering the vessel close, the boiling is afterwards continued, with a gentle heat, till the oil appears of a proper consistence; in which state it is called varnish....

The burning of the oil is intriguing. In Luke Hebert's The Engineer's and Mechanics Encyclopædia [London: 1846], Rees's method of linseed oil boiling is repeated, but Hebert asserts that the burning of the oil is necessary for the ink to dry properly. The fact that linseed oil dries on its own, without the aid of additives, puts his claim in doubt. Rees's assertion that the oil burning lessens the inks greasiness—a consistent goal in every ink recipe I've read—and thickens the oil to the proper consistency makes more sense. I have no doubt that Kremer's stand oil is as grease-free as I am likely to find but the consistency worries me. The ink I made today (which is more properly oil paint) is too fluid to print small type well. ¶These issues of ink consistency will simply need to be resolved through experimentation. Despite my knowledge of Baskerville's ink when printed, I know nothing about how it looked or felt in his iron pot. Further, considerations of ink viscosity change dramatically when using a power driven ink cylinder, as I do, rather than a hand-held baren. Time, or more precisely Speed, is the common disconnect I am encountering. I want to print with my inks tomorrow; yet every published description of Baskerville's ink emphasizes the long periods of time he allotted for its maturation. Not only did he allow his oil to "subside" for months after boiling but he allowed his mixed ink three years to settle before grinding it for use. Baskerville's refined and delicate typefaces only printed well because of his mechanical improvements in press, paper, and ink. One of the reasons his competitors (as if he had any) were so critical of his type's fineness was undoubtedly because they knew they could never print it as well he could: Baskerville's type required Baskerville's ink. I suspect that one of the reasons he let his ink gestate for so long was so that it would thicken to the proper viscosity for printing his type.

I will publish notes on the first ink mixture after I test it tomorrow.

Monday, September 20, 2010

Ink Diary Week 1, Entry 2

20 September, Progress on amber

Since my initial visit to Kremer Pigments, I have felt increasingly uncomfortable with the idea that, on the advice of two of their employees, I was going to put turpentine and amber in a pot and cook it on my kitchen stove. In my twenties I spent a few years working in art supply stores and, if I may be considered a representative cross-section, there is about a 90% chance that the advice I received at Kremer was wrong. So last night I spent a few hours reading and have had some of my concerns addressed. It turns out that the clerks I consulted at Kremer were misreading Kurt Wehlte's entry, Amber, a Hard Resin, in The Materials and Techniques of Painting:

Some amber is exceedingly hard, softening at approximately 250°C and melting between 290 and 300°C. This fossil resin cannot be dissolved directly, even in hot drying oils, but must be fused or "run" first.... Attempts to dissolve amber directly in turpentine succeeded on a small scale in the laboratory, thanks to some ingenious tricks, but these were without practical value.

The phrase that stuck out as most troubling (besides the extremely high temperatures and "without practical value") was "must be fused or 'run' first." There is no mention of melting in turpentine at heat or of the conditions required for fusing the amber. Additionally, I have no ingenious laboratory tricks up my sleeve. After a few internet searches I found a couple of helpful references on violin making and woodworking blogs. This entry posted in May 2007 by someone calling themselves Fridolin is particularly insightful:

I have recently started making my own amber varnish. I have found the descriptions given in Geary Baese's book on the old italian varnishes to work for me. The difficult part is to fuse the amber, which requires A LOT of heat. In my experience 300°C is not enough. I use special lab equipment (glassware and electric heat source). You also get a lot of toxic and smelly fumes (It can not be done inside). Once you have fused the amber, it easily dissolves in linseed oil (at a much lower temperature) and it gives a very thick varnish that dries within 24h in my UV cabinet.

In the 18th century, when amber varnish was more widely used, it must have been obvious that the amber needed to be fused before adding it to the oil. In fact, Fridolin's choice of words, "Once you have fused the amber, it easily dissolves in linseed oil," matches Hansard's Baskerville recipe exactly. Clearly, my little jam jar of linseed oil and amber will never transform into Baskerville's magic ink of its own accord. Quickly, I began searching for Mr. Geary Baese's magnum opus on old Italian varnishes. Google books has scanned one copy but does not have the rights to make it available. They list Baese as the publisher and that the copy they scanned is at the University of Michigan. Searches on bookfinder, abebooks, and Columbia University's online catalogue yielded no copies. Searching for information on Baese I found this foreboding entry from 1995 on the website of Alf Studios Concert Violins:

The pure resin varnishes he uses cost more than $10 an ounce from Geary L. Baese, 610 W. Mountain Ave., Fort Collins, CO 80521, but don't bother contacting Geary unless you are a committed violin maker.

Not only am I not a committed violin maker but.... Google searches yielded no further information on Baese. ¶My next search for amber fusing techniques brought William Theodore Brannt's, Varnishes, lacquers, printing inks, and sealing-waxes: their raw materials and their manufacture, (1893) to my attention. After reading this passage on page 45, I began to have some serious second thoughts:

In fusing it [amber] suffers decomposition. It gives off water, succinic acid, marsh gas, and a mixture of liquid hydrocarbons (known as oil of amber), while a more or less dark-colored substance, the so-called amber-colophony or fused amber, remains behind. This amber-colophony is the substance which is especially prepared for the manufacture of lacquers by roasting amber.... The oil of amber, which in fusing amber may be obtained as a by-product, forms in a refined state a pale-brown fluid of a strong, disagreeable odor. The crude oil is dark brown, and possesses a very repugnant odor.

After reading this enticing description, I began searching for where to buy already fused amber rather than searching for how to fuse amber. The first hit was a 10ml (2/3 tablespoon) vial available from Dick Blick for the discounted price of $133 and change. Successive searches found a number of small workshops making amber linseed varnish at a more reasonable price. I have decided to leave that stage of the process to other craftsmen.

Sunday, September 19, 2010

Ink Diary Week 1

19 September 2010

A few weeks ago, after reading too much news about various corporate trespasses, I decided that my small autumnal rebellion against corporate malfeasance would be to manufacture my own printing inks. Looking, of course, to John Baskerville for guidance I dredged up T C Hansard's recipe for Baskerville's ink from F. E. Pardoe's biography. Neither Baskerville nor his widow Sarah would divulge the secrets of his lustrous black ink but various sources were culled together into a recipe by Hansard in 1825, fifty years after Baskerville's death. On the surface, the recipe sounds fairly straightforward—a bit of lamp black collected from furnace burners mixed in with thickened linseed oil—but for the one predictable exception: "The oil thus prepared was suffered to cool, and had then a small quantity of black or amber rosin dissolved in it, after which it was allowed some months to subside..." [my emphasis] On a close reading of the recipe, the problems begin to multiply. First of all, what does a small quantity of rosin mean? Is it one part per ten or one part per hundred? The issue is one of elasticity: too much rosin will result in an inflexible ink which could potentially crack when dried. The only way to discover the proper proportion is through experimentation. ¶Second, although I can obtain all of the ingredients Hansard mentions from Kremer Pigments, the staff at Kremer assures me that amber has to be dissolved in turpentine (which is highly flammable and noxious) at a temperature of over 500 degrees fahrenheit. The resulting mixture is a shellac used predominantly in oil painting and violin making. Considering Baskerville's years in the Japanware business it makes sense that he would be well versed in the use of shellacs. But Hansard never mentions shellac. The rosin is not added to the linseed oil as a liquid but as a solid, it is dissolved in the oil. Which brings up the issue of time and heat. Does, "after which it was allowed some months to subside," mean that it took some months for the rosin to dissolve in tepid oil? Was the oil re-heated to aid the dissolution of the rosin and then allowed some months to subside? Against the better judgement of Kremer's clerks, I am opting to stick with Hansard for the moment. I have added 15ml of solid amber rosin to 150ml of linseed stand oil at room temperature. If the amber has not dissolved at all in a week I will try the same proportions over heat.

Friday, August 20, 2010

The Color of the Tiber

A couple of weeks ago I finished printing Swan & Hoop 2: A Roman Inscription on Santi Giovanni e Paolo al Celio. Like Mediæval in Padua before it, A Roman Inscription features a map that has been paired down to the essential landmarks that are discussed in the book. Unlike the map in Mediæval in Padua, though, the Rome map is entirely black with one red structure denoting Santi Giovanni e Paolo al Celio and a curving bluish line where the Tiber River bisects the city. All of this seemed straight forward enough until I began mixing the ink. The trouble with urban Italian waterways is that the water in them is green (or brown when it rains). This presents a host of problems when printing maps of Italian cities. I have never considered myself a realist but to print the Tiber in a bright aqua would be visually confusing to anyone who has walked along the river's banks. Despite its status as an ever flowing source of myth the Tiber is simply not blue. And therein lies the rub. To print a river in muddy green ink, however accurate, would throw off the visual balance of the map, disrupting the legibility of the image and causing the viewer to momentarily miss the point. So the goal was to make a color that is at once blue and ambiguous, evocative without being sentimental or misleading. My solution to this was to mix a blue that is actually a blue interpretation of brown, an ink composed of blue, green, yellow, red, black, titanium white, and transparent extender.