so, for the past billion years or so, i’ve been working on a typeface. Flapjack.

Flapjack started out digital (it was called Bismarck at that point). then i wanted to use it for letterpress printing, so i cut it in linoleum (that’s when it was redubbed Flapjack by a colleague). i like it so much that i thought that there should be a woodtype version and that’s where the trouble began.

i acquired a small pantograph machine which i thought/hoped would do the job. and then came the unbelievably difficult task of acquiring end-grain maple that had been surfaced to the required .918″ for printing. this just about broke our bulldog pluck, but we at the press figured that we could make the stuff ourselves, so we set out to gain valuable woodworking skills to assist us in this vital task.

along the way we learned a whole bunch of stuff which we will be detailing later on, but the most persistent question from those who would not understand what we needed to do was this : why don’t you just cut the type on a CNC machine? my answer to this was usually a two-word vulgarity followed by : because that would be too easy.

i think i knew i had to cut it in the way i knew i had to cut it in was because i just knew that i had to do it that way. doing it the long, painful and difficult way would satisfy some deep-seated need to do it that way and it would (i thought/think) achieve the look and feel that i knew i needed.

to that end, i’m almost ready to bring this woodtype experiment to a beginning and (hopefully) and end. but not before my wife intervened with the nerdbot3000.

the nerdbot is a 3d printer. she had purchased on for her school that turned out to be quite a lemon and didn’t want to bother her with the idea for generating type this way. but when she got the nerdbot3000 v. 2.0, and it was working much better than the first one, i decided to plunge in.

first off, i took the vector drawing of the uppercase G of Flapjack and manipulated it in a software program called Autodesk123D. to save build time on the nerdbot, i wanted to create a piece of type that could be mounted on a piece of .75″ plywood that i had used for the linoleum version. i extruded the raised part of the letter to 10 mm, left the base at 5 mm, exported the results and fed it into the nerdbot.

to my surprise things worked out pretty well. below you can see the actual printing of the honeycomb that formed the middle part of the Flapjack letter sandwich.



the printing too about 45 minutes to complete. the letter itself was attached to a printed substrate which allows for a smoother build and better adhesion of the material to itself.


after the detaching the letter from the substrate, this is what i had :


i held the letter up to the light so you could see the structure of the build a little bit better. this clearly shows the honeycomb pattern that the printer uses to build in strength while cutting down on build time and material usage. i was a little worried that the letter would simply crumble when it was proofed on the press.


i took it out to the print shop to ink it up and proof it. here are the first results :


the observant observer will of course see that i made an amateur mistake by forgetting to flip the letter in reverse at the beginning of the process. but i will tell that observer to suck eggs and remind them that i was excited to test the build and it wasn’t about whether the letter printed correctly the first time. i would be lying.

you can see that the printed image is rougher than a badger’s arse, so i did some benchtop sanding with 220 grit sandpaper to smoothing things out a bit and see how the edges would print. below is my next result.


i’m not really convinced that the nerdbot is the way to go for printing type. or at least to get a smooth, flat surface that i am used to with woodtype. the way that the printhead lays down the materials is just not conductive to traditional letterforms. i have seen one example on the old interwebs (actually, a friend sent it to me in a taunting manner because i had previously dismissed 3d printed type as being too rough for letterpress after experiments with my wife’s first printer) that factored the way the printer printed into the design of the letters themselves — they created ‘wireframes’ for the letters which meant a smooth surface wasn’t necessary.

when i looked at the final product the traditionalist in me was frankly appalled, but i understood what they were trying to get at. i’m not sure if they used an over-the-counter sort of 3d printer that i did (the article only mentions a ‘polyjet’ printer) and the photos are maddeningly lacking the detail that i need to determine if it’s the same basic build structure as mine is. but i think it’s safe to say that it’s similar.

i was quite surprised that the letter held up the way it did to a number of proofings. it seemed sturdy enough to withstand a short press run, but i would have to mess around with that a bit more.

the big factor for me was the time involved. 45-minutes for one letter means that printing all 26 letters would take approximately 20 hours. that’s without any punctuation or numerals.

it almost makes using the pantograph seem easy.