Ink for a Vintage Cryptographic Machine

Hi. I hope that you folks do not mind this off-topic request. I’m trying to learn about inks for an application that involves printing with metal type, and I am hoping that letterpress operators might be able to point me in the right direction. I’ve had a hard time finding details about ink formulation for my unusual application so far.

One of my hobbies involves collecting, restoring and operating vintage cryptographic hardware (i.e., code or cipher machines). Some of these machines have printing mechanisms which print encoded and decoded text on paper tape, such as my World War II era M-209B. I have pictures of this machine on my web page:

http://www.nf6x.net/2009/02/converter-m-209-b/

After 70 years or so, the original ink rollers are naturally dry. Many collectors rejuvenate their old rollers with a bit of WD-40, diesel fuel, water-based stamp pad ink from the local office supply store, etc. I’m trying a different approach: Making brand new ink rollers.

The printing mechanisms in these machines use metal type wheels which pick up ink from an inked felt roller, and then stamp a letter onto paper tape. I’m hoping that this is similar enough to letterpress printing that letterpress printers might be able to educate me a bit. I’ll try to attach a picture showing the print mechanism. The inked felt roller is in front, followed by a wheel with metal type, with a strip of paper tape behind it. A print hammer behind the paper tape presses the tape against the inked metal type wheel.

I believe that I can figure out how to fabricate new felt rollers using materials from my favorite industrial supplier. The next task will be for me to identify the right kind of ink, and then buy or formulate it as needed. I suspect that the ink should be oil-based, so that the roller doesn’t dry out too quickly. All of the stamp pad inks that I’ve found at office supply stores seem to be water-based, and they dry out in a matter of days when I apply them to a felt roller. They also are too thin to transfer to the metal type properly, and the printed results are not very good. I’m looking for something that can remain exposed to air for an extended time (months to years) and still print well.

I tried buying some oil-based block printing inks from an art supplier. They turned out to be a thick paste that smells like it has a linseed oil base to me. I don’t know if that’s the right kind of oil for this application; wouldn’t it tend to cure in the roller just like paint? I also bought some “burnt plate oil” which is supposed to be the right thing for thinning this type of ink. It also smells like linseed oil to me, and it’s quite viscous. I don’t know if or how block print ink differs from letterpress inks.

I’ve experimented with thinning this ink with various kinds of oils in order to soak it into a felt roller. I haven’t found the right mix yet, assuming I’m even on the right track.

Recently, I’ve gotten the idea of trying to thin the thick oil base temporarily by heating it, and then vacuum-impregnating it into the felt roller. I haven’t tried it yet, but I’m gathering the equipment to experiment with this.

Do any of you folks have any insights into what sort of ink might be appropriate for this application?

I’m also interested in figuring out where to buy and/or how to formulate oil-based ink that’s suitable for re-inking ribbons for old typewriters, computer printers, teletype machines, etc. This might be a different kind of ink than I need for the cipher machines, since an inked ribbon is pressed against paper by metal type instead of transferred to the surface of metal type and then pressed against the paper.

Thank you all in advance for any insights that you might provide, and I hope that this off-topic request is not unwelcome.

image: M-209-B Print Mechanism

M-209-B Print Mechanism

Log in to reply   15 replies so far

I have no direct experience with this, but I expect Bates Numbering Machine Ink would work just fine in this instance. The ink for Bates hand numbering machines is thicker than standard stamp ink, but still much thinner than printing ink. It’s also meant to sit for a long time in a pad in the numbering machine. It’s at least worth looking into.

Michael Hurley
Titivilus Press
Memphis, TN

Thank you, Michael! I’ll order a bottle to experiment with. I found it at Amazon. I think that one of the hurdles that I’ve been having trouble with is simply figuring out what to even search for.

Does this forum have a mechanism to email me when a reply to my topic is posted?

That’s a relatively easy fix….. Michael is right. The correct ink to use would be Bates Numbering Machine ink, since it doesn’t actually “dry”. Instead, the vehicle absorbs into the paper…. so it can stay on the felt roller for a long time without gumming up.

I wouldn’t recommend using any sort of thinned printing ink since they are all meant to dry… and as anyone here knows, you can’t leave that sort of material on your press indefinately.

The M209 series were indeed quite remarkable machines. It’s good to see folks taking an interest in keeping them alive.

Cool. I should have a bottle of black Bates ink to experiment with by this weekend.

I’m not positive, but based on the color of the old dried up rollers that came with my machine, I think that the M-209 ink was originally purple. So, I’ll probably still need to experiment with ink formulation, but now I know that letterpress printing ink is probably the wrong stuff to start with. Do letterpress inks “dry” because their base is an oil that polymerizes in air like linseed oil? Would I instead need to look for a different kind of ink that uses an oil base that’s stable in air?

I wonder if the Bates ink might also be suitable for re-inking typewriter/printer ribbons that are no longer available? Plain reel-to-reel ribbons are still not hard to find, and in some cases, I think it’s possible to rebuild an oddball cartridge with the guts from a more common cartridge. But then there are odd cases like the re-inking cloth ribbon cartridge for my DEC LA-12 Correspondent terminal. It has a relatively short ribbon loop that gets re-inked by an ink-soaked foam rubber roller. The old roller is hard as a rock now, either because of dried ink or because the rubber has decomposed. I plan to try to clean it and re-ink it to see if I can resurrect the terminal. I believe that the original ink was oil-based because of the oily film that the roller left on my fingers when I handled it.

I’m happy to see that somebody here has heard of the M-209 before! I’m also interested in other Hagelin-designed machines like the CX-52 and CD-57.

I agree with the comments you have gotten so far.

Regarding the linseed oil which you have mentioned, that is a “drying oil,” meaning that it will fairly rapidly react with the oxygen in the air and form a hard skin. That is not, of course, what you want. If you were going to experiment with an oil, I would go with a petroleum oil like a heavy mineral oil, which I think you can get in the drug store if you ask the pharmacist for it.

If you wanted to experiment with making a water miscible ink, I would try propylene glycol as a base. Propylene glycol dries roughly 33 times slower than water, meaning that you would have to evaporate around 33 cups of water, one after the other, in the time it would take to evaporate one cup of propylene glycol. You may also be able to get propylene glycol at the drug store since it is used in pharmaceuticals.

http://en.wikipedia.org/wiki/Propylene_glycol

Propylene glycol, or derivatives of it, have also been used in various kinds of inks, and I would not be too surprised if the stamp pad type of inks use it as well.

Do you have any suggestions about where I might be able to find suitable pigments or dyes (particularly black and purple) that I could try mixing with various oil bases? I’ve experimented with using various oils to thin the printing ink that I already have (which probably has a drying oil base), but I haven’t figured out where to get raw pigments to formulate my own inks yet.

I’ve also seen at least one reference to castor oil being used as an oil base in the past, so that’s something else I can experiment with if drug stores still carry it.

Since you only need a small amount, printing ink companies are probably not going to be interested in helping you unless you can find a sympathetic person in one of them. That being said, I would probably try a large art supply store (or other artist supplier) which might have pigments and dyes for people who are interested in making their own oil or water based artist paints.

I think there is a good chance that the purple is a dye. I think that even some printing inks partially use purple dyes, although we usually stay away from dyes because they are less light fast, less bleed resistant, etc. One difference betwen dyes and pigments is that dyes dissolve completely in the ink, leaving no actual particles. Pigments are very small particles which are suspended in the ink. For black, printing inks generally use carbon black pigment (like lamp black, which for instance, forms above oil lamp flames if they are turned up too high). This is good for ink because the particle size is very small, which is necessary in order to make good pigmented printing ink.

Just a thought: there is a chance you might be able to use ink jet printer black ink. This ink, I’m sure, dries very fast, but if you mixed it with mineral oil, or propylene glycol (whichever it would dissolve in, I don’t know), you might be able to slow the drying rate way down that way. It might be worth a try, although I think some ink jet ink, might be toxic, so try not to breathe it, or get it on your skin, etc.

Please keep us posted, this is interesting!

I’m glad that you’re finding this interesting, too! I was a bit worried about clomping into a random forum I found while googling and then asking off-topic questions. I’ve tried discussing this on a cryptographic collector list I’m on, and the responses were nearly hostile at my audacity to try making reproduction ink rollers rather than the accepted method of taking some random modern calculator ink roller from Staples, cutting it with a razor blade, and cramming it in there. :)

It makes sense that the purple would have been a dye. I have an impression in my mind that purple inks have been quite common in printing mechanisms in the past, and I’d bet a dollar that it’s because there was some specific purple dye that was inexpensive to manufacture. I’ve made a few attempts in the past to figure out what sort of dyes and pigments would have been common in the 1930s-1940s, but I haven’t figured out how to find that information or people who know about it yet.

Hmm, I ought to search around to find if I can locate a source of carbon black, such as from a chemical supplier. Maybe I can just fabricate my own typewriter ribbon ink from scratch? Could it be as simple as mixing lamp black with something like soy oil or castor oil (a couple of ink bases I’ve seen mentioned in my searches so far)?

Pigments are standardized across the industry. It is the vehicle (oil, water, acrylic, etc) that changes for the particular use. The pigments that you get with printing ink are the same available for oil paint and watercolors, or for that matter house or auto paints. The fineness of the ground pigment, and the way it is to be applied determines what additives are used by the manufacturer. You are looking for an ink that has the property of staying liquid on the roller, while drying on a sheet, which seems to me to be very similar to rubber stamp inks. It is possible to buy ink dispensers to refresh ink stamp pads, although I don’t know if a purple color is available. I would start looking at that particular branch of the industry to find what you need.

Paul

Carbon black from a chemical supplier is probably going to be in like a 50 lb bag minimum. Plus, you probably don’t want to get the dry powder form because it is so fine that it can be carried by air currents like dust and could make your whole room black. I once was at a talk by an ink supplier who had a small jar of colored pigment. He said something like, you can open the jar if you want, but if you are going to, tell me first so I can get out of the room.

I would try an art supply store and try to get black pigment that is already in some kind of a liquid or paste base, ready to be incorporated in an ink or paint formula.

Regarding castor or soy oils, I think they may also be drying oils, and since they are vegetable oils, they also may get rancid after a while. I would try a mineral oil, as I said, but that’s just a hunch.

I thought maybe the purple dye might have been an aniline dye, because liquid aniline dyes were used with flexography (another type of printing) many years ago. But I wasn’t able to confirm it. There is a lot of info on purple dyes and pigments on the web in various places, like here, if you scroll down in it:

http://en.wikipedia.org/wiki/Purple

or here (for Easter eggs, but who knows, it might work for you)….

http://www.pysanky.info/Chemical_Dyes/History.html

The latter site even lists some suppliers (scroll down to DYES):

http://www.pysanky.info/Links/Supplies.html

I like the anecdote about leaving the room before the jar is opened! I’ve ground up charcoal before, and the stuff just won’t hold still in an open container. :)

Based on the Wikipedia article on Purple, manganese violet pigment and Mauveine (aniline purple) dye look like good candidates for what they might have used in purple inks during WW2. Purple adding machine inks and so forth that I’ve encountered during my own lifetime might be made from newer dyes/pigments, though. I’d think that the most likely reason for many machine printer inks to have been purple would be availability of some particular purple dye/pigment that was cheaper to manufacture than good old carbon black, but that’s just speculation on my part. Or maybe aniline dye had some other properties that made it preferable to inks with suspended solid pigments, and maybe Mauveine was cheaper to make than other dyes?

You could very well be right…….I’ll be interested to hear what you come up with.

When we make inks for letterpress, we not only use pigment and solvent, but we use a vehicle (or you could call it a varnish), whose function it is to surround the pigment and protect it and bind it to the sheet. With a pigment ink, and without any vehicle, the pigment will likely sit on top of the sheet and then rub off easily after the solvent soaks into the sheet. With a dye, since it is dissolved in the vehicle and/or the solvent, I would think it would be more likely to soak into the sheet and less likely to rub off.

Just food for thought…..

Interesting discussion to see here — in addition to letterpress, my interests extend into the world of teletype machines, which are quite closely related to encryption machines. As an aside, there is a direct link between teletype technology and letterpress printing:

http://www.circuitousroot.com/artifice/letters/press/compline/literature...

Back on topic, a while back, I got to wondering what typewriter ink was made out, and I came across this:

http://www.sarahnorris.net/Papers%20&%20Research/Typewriter%20Inks%20Ann...

While you’re not looking to re-ink ribbions, I imagine that many of the properties are the same.

Thank you for the typewriter ink reference! I am also interested in re-inking ribbons to help keep my old machines up and running.

My interests include crypto machines, teletype gear, vintage computers and related hardware, military radios, military trucks, etc. There are lots of overlaps between those, as well as connections to other things.

My interest in crypto machines also led me to interest in generating keying materials, which led to interest in simple bookmaking, which led to acquiring 450 pounds of military surplus cast iron in the form of a 1952 Chandler & Price paper guillotine… which is yet another peripheral connection to printing. Who knows? Maybe I’ll even catch the letterpress printing bug someday!

Oh, wow! After reading the typewriter ribbon ink bibliography, I think I need to try to obtain copies of most of those references. It looks like a gold mine of the kind of information about inks that I’ve been searching for. And now I’m also curious about thin-layer chromatography…