Did you know that you can cut letter press relief or intaglio plates on the NYCR laser cutter? The laser cut acrylic holds a decent edge and is far less expensive than copper plate. Here is a short guide to how to make engraved invitations using the intaglio process:



First the plates need to be cut. Using inkscape with a thin font like Zapfino, layout the text and vector art. Briar Press is a great source for EPS that can be converted to PDF to import into inkscape. For intaglio process the etched or cut pieces will be black; relief process is the opposite.

I’ve had the best results with the text as solid black and raster cut at 80% power, 20% speed, while the artwork with fine lines were vector cut at 5% power, 10% speed. The cuts do not need to be very deep (unless you want to emboss without any ink), so the low speed is to ensure that the laser cutter’s steppers make smooth lines. I use my own GPL’ed epilog driver to give me better control over the cutting versus etching, but this should work fine with Corel as well. Here is a short video of the laser cutter as it vector cuts the mirror image text.

Which brings us to the first rule of making printing plates: Be sure to flip the entire page. Otherwise your prints will be backwards!

A second rule of making plates is to include registration marks on the outside border. The paper will be larger than the final size since it is very difficult to align while placing it on the plate. These registration marks can be used once the paper is dry to ensure that all the pieces have consistent centering.

And my third suggestions is that the piece of acrylic plate should be much larger than the oversized paper. It is easier to handle if it has extra space and produces better results if the paper is able to lay flat without being bent across the edge.

Scrub the plate after it is cut to remove any acrylic residue and dry it well. Then apply a small dollop of etching ink, such as Speedball Printmaster, to the plate. You don’t need very much, but there isn’t any harm in over applying. With practice you can figure out how much is necessary so that you don’t waste ink.

Using tarlatan cloth (or rough paper towels), wipe the ink into the etchings with circular motion until the entire plate is covered. A soft spatula can be used to help spread the ink, but metal ones can scratch the acrylic and leave unsightly marks on the resulting prints.

Using a clean piece of tarlatan or paper towel, wipe the plate clean with the same sort of circular motion. Occasionally refold the cloth to get a clean corner and keep wiping until it is spotless. The ink in the grooves won’t be picked up by the cloth, so keep scrubbing!

Using the adjustment knobs above the rollers, set the pressure setting on the press. This will require adjustments based on the thickness of the plate, the paper and the amount of padding. Be sure that both sides of the press have the same value and expect to do many experiments during the setup until the best height is found. Since there is no ⌘Z command, I like to make notes on the test pieces so that I have a way to revert to a previous setting.

Put the plate on the press with the ink side up and a slightly damp piece of oversized paper on the etched area. To dampen the paper I’ve found that the best results come from placing it between two wet paper towels for a few minutes while I ink the plate. Just like the pressure adjustment, expect to make a few dozen test prints before figuring out the right level of dampness. To prevent the paper from sticking to the sacrificial newsprint, I like to use a layer of aluminum foil above the paper.

Smoothly cover the plate and paper with the newsprint and felt, then crank the press’ wheel. Intaglio needs lots of pressure, so it will be difficult to push the plate through the roller. If the gear train skips, adjust the roller height to be 0.5 mm higher and try again.

Only run the paper and plate through once. If it is backed through the roller a second time there is a chance that the paper will skip and a second impression will be made.

Carefully lift the felt, newsprint and foil to reveal the printed paper. If all goes well, it will have transfered everything from the plate with no missed spots or water bleed.

But, for your first few plates it probably won’t be perfect. The one on the left was too dry and the paper was not pushed far enough into the etched grooves to pickup the ink. The one on the right was too wet and caused the water soluble ink to bleed. The one in the middle was just right. Re-ink the plate and give it another try.

If you realize that you’ve made a mistake in the artwork or text, rinse off all the ink, turn the plate over and use the laser to etch the other side. 3mm acrylic is strong enough that the etchings on the reverse side will not be transfered through, so you can reuse the piece.

Some inks aren’t right for etching. This Caligo Carbon Black formed “threads” when the paper was removed from the plate — the engraved ink should be smooth and shiny as it dries. Some inks don’t contain drying agents, so they will still rub off several weeks later. Also, for ease of clean up, you should only use water soluble inks.

Stay tuned for another blog post on relief printing, coming sometime soon.

Some friends at a local university reached out to us recently and offered to let us rescue a robot from a junkyard fate. Not being in the business of turning down free robots, we quickly agreed. Three of us showed up on Monday in a Ford Escape. We left in a U-Haul.

This beast of a machine weighs in at 550lbs. It’s a Gilson Cyberlab C400 Automated Plate Preparation Workstation. We’re not exactly sure what that is, but we do know that it has a huge robotic gantry meant to move at high speed with accurate positioning. And it has neuroprobes. What are we going to do with it? Maybe it will be the next BarBot. Or a 3D Printer. Or maybe some sort of exercise machine. We’re not sure yet. Check the vids.

Adam and I upgraded Hexascroller to control 5 m of Adafruit RGB LED strip through a spare serial port connected to a Teensy 2.0 that drives the strip via SPI. Now when a new message is displayed, the accent lights switch to a bright flashing mode to attract attention, then they will return to soothing, slow color changing mode.

Click the “Read more” to see additional photos of the installation and setup.

We’ve had a few knitting machines rattling around the space over the years, but when fellow gadget lover Josh dropped off the lace carriage for our Toyota K747 knitting machine, I figured it was time to take it past basic stockinette stitch and explore the machine’s punch card mechanism.

Like most decades-old machines, a layer of dried oily gunk coated many of the moving parts. There are also a few broken/missing pieces. I’m in the process of cleaning / fixing the machine, and trying my best to document it as I go along. I got a hold of the K747 Service Manual, and started diving in. It’s oiled up now, and the next steps are to repair the broken needle selector and write up a program to generate punch cards for the laser.

 The picture above shows the offending broken lever. These 12 little blue levers tell the machine which needles to push out for the patterning, but lever #3 is broken so it never engages. Getting in to replace it is a bit daunting, in fact just getting to that lever in the first place was a bit of an event, I’m going to attempt to fix it with Sugru first. If that doesn’t work, then I’ll be makerbotting a replacement and praying I can get it all back together after dismantling it.

Since the machine is loud and takes up space I generally only work on it during the day when there’s more room, but if there’s sufficient interest I’d be happy to demo it at an upcoming craft night.

You can see more shots of the inside of the machine over on my blog: Knitting Machine Teardown Part 1 and Part 2.

I’m Kellbot! I’m a member of NYCR, and I love gaming.

My friend Katherine and I have been working like crazy on a game project for the last few months, and we’re thrilled with what we’ve accomplished so far. Now, we’re trying to bring it to an open beta while making it as fantastic as possible. We’ve started a Kickstarter campaign to get us there.

City of Epic is an RPG based around real-world exercise. Rather than get all TL;DR, I’ll let this “high quality” video explain it:

If you like some combination of gaming, fitness, and awesomeness, please consider donating!

Inspired by Ranjit’s instrument-a-day (Make coverage), I’m writing a new wrist watch face every day for my progamable inPulse watch (more background). The full sources are posted online for others to build on.

Day 1 was a fixed point 3D rendering engine with a rotating Utah Teapot, day 2 was a rolling odometer or aviation altitude ticker display and day 3 is a 24-hour analog clock. This last one still needs some work.

Some friends of mine have a house in the mountains in Steamboat Springs, Colorado.  In the winter they dig a luge track in the (many feet of) snow in their yard and have timed sled races to see who can make it down the fastest. They call it the Mario Cup. I went out there last week with my buddy Mario (a member at Artisan’s Asylum and the namesake of the race) and we made some improvements to the system. Details below.

As you can see, the luge is several hundred feet long and on the side of a mountain, so it’s tough for a person with a stopwatch to know when to start and stop the time.  Here I am testing the track before we had finished it; at the end, I’m a tiny little dot in the snow.

Mario and his girlfriend also helped calibrate the track.

Recently, the competition has become fierce and the accuracy of a stopwatch-wielding human isn’t enough to conclusively pick a winner.  So we automated it.

We designed a system so a “race official” would start the clock by hitting a button on a controller, which would sound a loud countdown of beeps and then start the clock.   The luger would take off from the start line down the track.  Arduinos hooked to sensors would detect when the luger passed the start, midpoint, and finish lines.  The arduinos were also connected to Xbee radios to communicate the time at which the racer passed by.

It took a while to sort out which sensor would work best.  We tinkered around with tilt sensors, photointerrupters, Hall effect sensors, and proximity sensors.  We ended up with plain old photoresistors wrapped in plastic wrap, sealed with tape, and buried in snow under an upside-down disposable plastic food container.  That way, the luger would slide right over the sensor, block the bright white light coming off the snow, and trigger the arduino to note the time and send it back to the controller.  Here we are setting up the sensors and tuning them in fast forward.

And here I am, soldering up voltage dividers in the snow. Potentiometers are difficult to find in Steamboat, but the snow keeps beer cold while you resolder new resistors, so I’m calling it a wash.

Finally, we got the sensors tuned just so.

The timing data was sent by radio back to the controller, which showed the time on an LCD display.

Now truth and justice will prevail in the Steamboat Backyard Luge Association races.  More information can be found here: http://mariocup.tumblr.com/

Next steps for the project include adding in two sensors at the start, midpoint, and finish so we can know the luger’s speed at each point, and adding in some fancy blinding-bright luxeon 1-watt LEDs to give the luger a proper starting signal like a christmas tree at a drag race.  Any excuse to go back will do.

The inPulse watch is a great platform to hack on. It has an ARM7, 32 KB of flash and 8 KB of RAM, Bluetooth, a buzzer, an OLED screen and a button. Not much by today’s standards, but plenty to play around with. The programming environment is very much like a microcontroller; no multitasking, no dynamic memory, and very constrained memory/cpu. That is, of course, what makes it so much fun.

My multi-app framework attempts to make it possible to write multiple apps that can share the limited RAM by a technique similar to FORTRAN COMMON blocks. The main app.c file has the mainloop that calls the loop() method of the current app. This main loop also does a static allocation that fills the entire BSS; each app that is linked in rewrites it with its own local state while it is running. When the user holds the button for two seconds, the main app selects the next linked in app and call its init() method to allow it to perform its setup.

Since some of the apps need trig functions and there is no floating point support, I’ve written a fixed-point sin() and cos() function that are within +/- 0.0075 of the real sin function. Close enough for most government work… The function and lookup table consume 157 bytes of flash and no precious RAM. I haven’t benchmarked the performance, although most of the apps already have time-delays to only redraw the screen at 30 fps.

There are still several things that need to be done to make the watch usable as a watch. The most important is power management. Even in “powersave mode”, the battery runs dry in about three hours. Clearly it is not saving much by turning off the display.

The other two major requirements are a local build environment and a more accurate simulator. The cloud-compile environment does make it easy to get up and running, but there are many things that a local build would make easier. For instance, the Bluetooth stack is always linked in even if it is not used, leaving less RAM and flash to the user. Or figuring out why calling snprintf() breaks everything. Hopefully they’ll release the local SDK soon.

Until then, the developers hang out on the inPulse forum and have been super helpful with debugging some of the getting-started difficulties.

A few weeks ago, NYC Resistor had quite a showing at the first Worlds Maker Faire in Queens.  We showed off a number of projects, including Ranjit’s MIDI Player Piano, Mr. Stabby, Raphael’s Twitchies, and Chris’s Cray-on-a-FPGA.  One of the new projects shown was the NYC Resistor Atari 2600 demo, thrown together the night before by me, Ben Combee.

The hardware you see there has a few homebrew components.  The console is an Atari 2600 Jr, the smaller version that was on sale in 1985 through 1990.  I’ve installed an AV modification from The Longhorn Engineer to get composite, S-Video, and stereo audio outputs.  The monitor is an older 21″ Dell unit that has composite inputs.  The demo was running off a Harmony Cartridge, a very cool homebrew development board done by people at the AtariAge website.  It lets you load a bunch of ROM images on a SD card and select which one to run at boot time.  For the Faire, I used a special autorun mode where it would always immediately start with the demo instead of showing the menu.

The app wasn’t written directly in 6502 assembly.  Instead, I used a great development tool called Batari Basic. It’s a BASIC language wrapper around the 2600′s hardware with prewritten display kernels. While you can’t do everything with it, it’s a great way to get an idea up and running on the system.

If you want to download the code or the binary to run in your 2600 emulator, it’s part of the NYC Resistor github depot along with many of our other projects.

In need of a better storage/organization scheme for my Lego collection, I made these meta legos: boxes that look and function like legos which hold my legos. The full details are available on Kellbot! but, long story short, I used OpenSCAD to create plans for boxes in any brick dimensions I like.

If you haven’t used OpenSCAD, it’s super handy for generating .dxfs for laser cutting, and is great for designing parts to be MakerBot’d as well. I’m teaching an Intro to OpenSCAD class next weekend, it’s only $25 and no prior programming experience is required!

Plans and more details can be found at [Kellbot!]