The White/Superba 1602 I purchased sadly did not include a full-bed cast on comb, which makes casting on things wider than swatches somewhat tricky. There are a few floating around eBay, but they all seem to be $50 and/or in France, and shipping a three foot long steel comb from France to the US isn’t particularly cheap. Also, I am impatient. I’m told that a knitting machine store in Toronto also has some, but again, impatient.

So I set out to make my own. The comb consists of small metal teeth which go in between each stitch, and a rod which is threaded through holes in the top of the teeth to trap the yarn in place. I made a 12″ long test comb to see if it would work before committing to a whole yardstick worth of bobby pins and hot glue.

I purchased 4 packages of bobby pins and some wooden rulers at the local dollar store. I then arranged the bobby pins at 5mm intervals along the ruler, since the Superba machines are 5mm machines.

I used a second ruler (they were 3 for $1!) to make sure the pins were all at an even height and that they retained their 5mm spacing. Then I ran a line of hot glue along both sides to hold things in place.

To use the comb to cast on ribbing, you run the carriage/yarn across the needles to establish a zig-zag of yarn across the bed. Then you push the cast on comb up between each stitch, and run a rod through the top of the comb. This traps the yarn between the rod and the ruler below, letting you appropriately tension the yarn for the first row of stitches.

As you knit, the comb pulls the fabric down between the two beds. A weight is attached to each end to provide tension. In this photo, I’ve lowered the front bed to make it easier to see what the comb is doing. You can just barely see the small copper rod at the bottom of the fabric.

I used it to knit a gauge swatch for a sweater I plan on knitting soon. It’s a mini-cable rib pattern. The cables are worked by hand, which is pretty tedious though I’m slowly getting faster at it.

My dollar store comb is by no means as good as the real thing, but it works pretty well for $2 worth of materials. It has a few key issues which make it a little tricker to use than a normal cast on comb. First, the end of the bobby pins are just a little bit too wide, I have to set the bed spacing all the way up to 6 in order to get them to fit through. Second, the bobby pins get in the way of the holes used to hang the weights. And my DIY comb is nowhere near as durable as the steel one. But it works well enough for now.

I don’t have a knitting machine problem, I could quit right now, I swear.

While working with my Toyota 747 I decided to try and find a ribber for it, which led me to the Ravelry Knitting Machine Sales group. I did not find a Toyota ribber, but I did find someone in Brooklyn who was selling a White/Superba 1602. He was selling it at a reduced price because the electronic selector box was not functioning.

The Superba knitting machines, which were also sold under the brand names White, Singer, and Phildar, are really interesting machines. Instead of a main bed and an optional removable ribbing bed like most Japanese machines, the Superbas have two permanently fixed identical beds. This makes it much easier to get consistent, even ribbing.

Mechanically, the machine is in good shape. Stockinette stitch, ribbing, and jacquard patterning (done by manually moving the needles into place) all work well.

The selector box, which can be seen in Patrick’s photo stream, works by reading stitch patterns off mylar sheets. A photoresistor detects either a light or dark square and sends a signal to the machine which moves the needles accordingly.
In addition to maintaining a comprehensive site on Superba machines, Patrick also was kind enough to supply me with the users manual, service manual, exploded part diagrams, and logic flow charts for the electronics.

As soon as I opened up the machine and took out the circuit boards, it was clear things weren’t working properly:

I superglued the board itself back together, and then used lumps of solder to repair the broken traces. Jumper wires would have been better, but admittedly I was too lazy to get up and find a spool of wire.

Once the traces were repaired, Phooky helped me test the output voltages. Since it takes in 110V mains power, I was nervous to start poking at it by myself. But we only made the electricity arc between the multimeter probes once. Have I mentioned that mains power is kind of terrifying?

Anyway, of the four pins that connect to the card edge, two are tied together to ground, and the other two provide 24v for the motor drive (which feeds the mylar sheets) as 12v for the COP420 microchip.

Speaking of the COP420, my first instinct was to try to get a firmware dump off the chip and try to reverse engineer the firmware. Not that I have any experience doing that, but luckily Trammell does. Unfortunately, he found out that the COP420 is a mask-programmed device, meaning that the program is put into ROM when the chip is created, in contrast to something like the Atmega chips used in Arduinos. If you’re lucky, the “test mode” on the chip was initially enabled, making it possible (if somewhat of a pain) to read out the firmware.

The more I think about it, the more it makes sense just to redesign the selector box from the ground up. Most of the bulk and power of the box is related to the scanning and advancing of the mylar cards, which are adorably archaic but not exactly convenient. A USB interface would be vastly preferable, and would cut down on about 2/3 of the circuit. I need to do some investigating to see if the whole thing could be USB powered, eliminating the need for a separate power cord and bulky transformer.

Yeah, ok, I have a knitting machine problem.

After purchasing a fresh sponge bar for the Toyota K747 I started to put it through its paces. It knits stockinette beautifully, so I started working through the tuck stitches.

At first I couldn’t get the I and II buttons to stay in place at all. After some disassembly and a lot of oil, we got things moving again and it works great. But I quickly noticed that the 3rd needle selector wasn’t working. So I started to disassemble the machine to figure out why.

With the top cover off you can see some of the neat mechanisms in there. The zigzag gear (right) physically moves the needle selector up or down the needle bed. This means if you have a simple shifting pattern like a zigzag or checkerboard, you don’t need a punch card at all. You can just move the zigzag back and forth with each row.

Each of the blue needle selector levers rotates down to catch a small lever on the bottom of the machine. When these levers are caught, they cause the corresponding shafts (anyone know what these are really called?) to raise, which then push the needle butts forward. The small red lever releases the needle selectors.

Getting into the machine was a little tricky, you have to remove the card and zigzag knobs to access all the screws. Thankfully there’s a useful diagram in the K747 service manual. With the knobs removed, you can see the discoloration of the plastic over the years.

I had to fully remove the needle bed from the machine casing to get at the needle selector levers. Unfortunately by that time my hands were absolutely covered in grease so I couldn’t take many pictures. But once I got it out, I could flip it over and see the problem part:

The two little tangs on the bottom of the 3rd needle selector have somehow broken off, though I’m really not sure how. Unfortunately disassembly/repair of that part of the machine isn’t covered in the service manual. I see two options going forward: 1) I can carefully disassemble the needle selector mechanism, 3D print a replacement lever, and pray to God that I can get it all back together again, or 2) Fashion some replacement tangs with Sugru and hope they stay in place.

Since option 2 doesn’t require completely disassembling the machine, I’m going to try that first. There are a lot of little springs and I’m not sure I have the dexterity to get them all back in place again. I’d also like to better document the machine teardown, so sometime in the new year I’ll recruit an assistant and take apart / photograph the whole thing.

In the meantime the machine works great, and I can come up with plenty of patterns that avoid the 3rd needle selector.

A few years ago I was given a Toyota K747 knitting machine, a gorgeous blue knitting machine that features a punchcard patterning system. It is by all counts a fascinating machine, and I was excited to receive a lace carriage for it today.

It’s been a while since I’ve devoted time to the knitting machine – they’re a bear to set up and take down so without somewhere to leave it set up all the time it can be daunting to use.

I noticed that one of the screws on the side wasn’t biting into anything, and decided to take the machine apart to figure out what was missing. I took out 6 screws on each end, and pulled off the end caps.

With the caps off I could stare into the dark abyss of needles and gears. I discovered it’s really challenging to take photos down a long, dark tube when using a point and shoot camera, so forgive the image quality of the following photos.

There are two “compartments” inside the machine, the front area which is all metal and has slots at the top for the needle tabs (called the needle butt) to stick out of, and the back which is mostly plastic where the needle butts rest. This is also where the patterning shafts (for lack of a better term) are located.

Here’s the front compartment, the front of the machine is to the left. If you have trouble seeing the needle tabs, click the image for an annotated version.

Here’s the back. You can see there’s some greasy dusty areas. I’m not sure how to get in there to clean it. The needle shanks are pointing down to the “shafts”, which if you look closely you can see there are 12. The Toyota K747′s repeat is 12 stitches wide.

The missing part in question is a small rectangular nut which slides into the end cap. I’m really hoping this is something I can source rather than having to mill/tap one myself. Any suggestions for shortcuts are appreciated.

Aside from the missing nut, the machine is in great shape, and the nut isn’t super crucial to the machine’s operation. My next step is to find a scale version of one of the punch cards so I can cut some out on the laser cutter. I found a few places that claim to sell punch cards for it but none of them had them in stock.

This is part two of our home theater PC adventure. If you’ve just arrived you may want to start with part 1, Hello Xbox 360 Wireless Gaming Receiver.

Hooray, Windows can finally see the controller. We fire up the control panel tool for gaming devices and confirm that all buttons and axes work. We’re just about there! XBMC even helpfully includes keymappings for the xbox controller by default! We should be pretty much plug-and-play from here! I open up XBMC and…. nothing. The controller does nothing. I smash all the buttons, and still nothing.

The first step in debugging is to open up the log file and see what’s going on. The log file starts fresh every time you boot XBMC, and if you’re lucky you’ll see a line like this shoved in there:

19:00:17 T:2904 M:876916736 NOTICE: Enabled Joystick: Controller (XBOX 360 For Windows)

This means that XBMC can see the controller. It also tells us that it thinks the controller is named “Controller (XBOX 360 For Windows)” and depending on your OS and a few other seemingly random factors, it may be named something different.

This name is critically important, because it’s how XBMC knows which keymapping profile to pick up. When I went into the XBMC system/keymappings folder and looked at the existing 360 controller profiles, none of them was an exact name match. So I copied one of them and pasted it into the user keymappings folder.

The next step is to go into our copy and replace with or whatever your system thinks it’s called. That node pops up a bunch so you’ll probably want to use find+replace rather than manually copying and pasting all the time.

I’m not sure exactly what happened, at some point while I was messing around with all of this I managed to un-pair the controller from the PC without realizing it. After much cursing and whining, “why doesn’t it woooooork,” I realized what had happened and rebooted things.
Once I had renamed the joystick and it was actually communicating with the PC, I fired up XBMC and something magical happened: both the A and B buttons were functioning as Select and Back, respectively. Hooray! And the right analog stick was working as a volume control (though the axis was inverted). Not much else seemed to do anything, but IT WAS WORKING. Hooray!

The next task was getting the D-pad to work for navigating menus. Let me take this time to say that like most people who grew up with Nintendo controllers, the d-pad on the Xbox 360 controller is a source of scorn and hatred. But I wasn’t quite ready to tackle the analog stick, so the d-pad would have to do.

At this point I turned on debugging in XBMC and then proceeded to methodically press every button on the controller (and swivel each stick axis) exactly once. This worked great for the buttons but none of the axis data showed up in the debugger at all. Great. The keymap xml file I copied incorrectly identified the d-pad as “buttons” when it is in fact a “hat” according to Windows, so once I replaced the “button” nodes with “hat” nodes I was able to map the directions on the d-pad to Up/Down/Left/Right commands.

I should mention that I spent a lot of time googling about my problem, and mostly found forum threads where one person said “my xbox controller doesn’t work” and another person said “use xpadder,” or someone says “the d-pad doesn’t seem to work” and again the reply is “use xpadder.” After reading that response in about 20 threads, I was really starting to think the whole system was just too bugged to be viable. But in reality, people on the xbmc forums just aren’t willing to get their hands dirty.

Only in the process of writing up this post did I find someone who had actually taken the time to map out which buttons, hats, and axes where which. I wish I had found that post last night, it would have saved me about two hours.

I’m still deciding what the analog sticks should do, and trying to figure out how to get the controller to turn off when I’m done, but we got things to the point where it was good enough to navigate around to Dexter, and that was enough for one night. And I will say, navigating with the Xbox controller feels nice, much nicer than breaking out the clunky keyboard.

We recently set up Xbox Media Center (XBMC) on the computer that’s attached to the TV. XBMC actually has nothing to do with xbox – it was a homebrew media center application for the original xbox that has since been ported to Windows/Linux/Mac/iOS.

We’re pretty close to what we feel is the “perfect” setup, but getting here was a headache and a half so I’m putting up a series of posts documenting our journey.

Part 1:  Hello Xbox 360 Wireless Gaming Receiver
Part 2: Getting XBMC to work with our Xbox 360 Wireless Controller

Hello Xbox 360 Wireless Gaming Receiver

We’re currently using a wireless keyboard/mouse combo with our media PC, which is clunky and sort of a pain to use. I started looking into media center remotes. We also have a collection of wireless 360 controllers, and only very rarely use all four at once. And so my mission began.

Microsoft did at one point make a wireless 360 controller receiver for the PC, a little white dongle that plugs in to a USB port. Unfortunately, they discontinued it some time ago. You can still find the receiver bundled with a controller some places, but I had a hard time finding an OEM standalone receiver.

Fortunately, the Chinese knock-offs on eBay are still readily available. I got one for $13 including shipping, it’s a little white wireless dongle with a USB cable. Unsurprisingly, it did not include a driver disk. Equally unsurprisingly, Windows had no idea what it was when we plugged it in.

We downloaded the official Microsoft drivers for the official dongle. You can find them here: http://www.microsoft.com/hardware/en-us/downloads , click under the gaming heading.

Our media PC is running Windows XP. Everyone else seemed to have a perfectly fine time installing the drivers by hand. I guess we’re just unlucky, but the damned thing just wouldn’t install.

Turns out, we had the same problem this guy had: the dongle wasn’t identifying itself correctly, and thus Windows didn’t recognize that it belonged to the driver.

The fix involves editing the driver’s .ini file, which sounds a lot scarier than it is. Here are the steps I took:

Step 1: Find your errant dongle in the hardware manager.

1. Go to your Control Panel
2. Click on Hardware and Sound
3. Click on Device Manager under Devices and Printers
4. Scroll down the list to Other devices
5. Double-click the Unknown device. If you have multiple unknown devices, you may need to use the process of elimination to figure out which one it is.
   

Step 2: Find out your device’s PID

1. Click on the Details tab
2. Look for something like this USB\VID_045E&PID_0291&REV_0107
3. The bit you’re looking for is PID_0291, or whatever number is in the place of 0291
4. If the number you see is 0719, you have a different problem than I did. Sorry.

Step 3: Edit the driver .ini file to reflect the actual PID

1. Figure out where the driver installed itself (probably somewhere within c:/program files) and open that folder
2. Open up Xusb21.inf in Notepad or your favorite text editor
3. Find and replace any instances of ‘Pid_0719′ with ‘Pid_0291′ or whatever your device’s number is
4. Save the file

Now if you’re lucky, you can run through the update driver dialog, point it at the folder containing xusb21.ini, and Windows will recognize the device as a Wireless Xbox 360 Receiver. Hooray! You are now halfway to tv-controller bliss.

On to the next part, getting the stupid thing to work with XBMC.

I’m a member of NYC Resistor, and while I think we’re a pretty awesome group of people, we’re not without our bikeshed arguments. These kinds of arguments, in any group, can range from mild time-killers to arguments so intense people end up feeling personally hurt.

One of the biggest problems I’ve noticed is when people start shooting other people ideas down. It starts a negative feedback loop, and everything goes downhill from there. But lets face it, some ideas really are terrible, and maybe we think it’s worth a bruise to the person’s ego to save us all from a huge mistake. How do you skirt a bad idea without killing a friendship?

Personally, I’m going to work on shifting away from saying “no” to people. Which is not the same as saying yes. I’m banning “no” from my discussion vocabulary, and replacing it with the phrases “what if, instead/additionally…” and “I’d be OK with that if…, ” followed by an explanation of what would make me more comfortable with the project.

For [an absurd] example, let’s say someone in the group wants to buy an iguana and keep it at the space. I think this is a terrible idea, the space is for people, not animals. Instead of saying, “no, an iguana at the space is a disaster waiting to happen, it will totally die,” I could say “before diving into herpetology, let’s get a few plants for the space and see how that goes.” Or maybe, “I’m worried it will be forgotten and die, what could we do to make sure Iggy is cared for?” This gives people a way out of the stalemate and on to continue the discussion.

It takes less than a second to say no. Yet we spend too much time reciting our many and varied reasons to say no, rather than listening and considering what we could do to find a compromise.

So just like giving up “you should,” I’m giving up “no.”  I invite you to call me out on it when I backslide, and hope you’ll consider ditching “no” too.

This week I makerbotted for the first time! I know, I know… I should have done this much much sooner. Everything Tiny and Makerbot were actually founded in the same room, the old NYC Resistor location, and it's been really exciting to see things take off for them. But until recently, partly due to the success of Everything Tiny, I never really had time to sit down and get personally aquanited with the wide world of 3D printing.

Part of my probelm was that I always had ideas which were large and complicated, and I never finished the designs. So in an effort to actually produce something, I set myself to a very simple first project: a plastic organizer insert for a mint tin.

I'm using it to organize my miniature pompom collection. Because I have one of those.

Overall the printing process was pretty easy. Adam, Matt, and Pax were nice enough to help me get over the few problems I ran into. Matt, who is the developer in charge of ReplicatorG (Makerbot's printing software), gave helpful instructions like "now click the picture of the potato with an arrow coming out of it." He then asked if I knew any UI designers looking for work.

Here are the issues I ran into on my maiden print:

The model didn't adhere to the bed.  Because my design is pretty thin, it didn't stick very well to the bed of the makerbot, even with the heated build platform. This was easy enough to fix, we turned on the raft (a layer of plastic which goes down evenly before you start the real print, and is removable later).

The plastic was coming out goopy. The main issue with my first two attempts was that the plastic was coming out much too thick, and kind of lumpy/grainy. This caused two problems: first, it looked terrible, with little lumpy bits everywhere. But more importantly, the lumpy goopy bits would build up and then harden. Then when the extruder head came by again, it would hit these plastic lumps and move the model, throwing off the registration.

There were two suggested fixes here: first was to increase the speed, so that the build platform would move faster (allowing for less material to build up). This was met with a certain degree of success, but Adam suggested that the plastic we were using, which had been sitting out at Resistor for quite some time, had absorbed too much water from the air to really be useful. The moral of the story here is to store your plastics in airtight containers with some desiccant. In the end, we switched plastics.

My machine wouldn't talk to the Makerbot. This ended up being an issue with my machine, a somewhat unhappy eeePC. When I switched to my mac, all was happy again. Also, the eee is way too underpowered to really be generating gcode for prints. I was able to cold-boot my mac, install RepG, install the drivers, and generate the print's gcode from scratch in the time it took my eee to get halfway done generating the same gcode.

The build platform wouldn't heat. This one took us the longest to debug. One of the connectors was visibly damaged (names were named but I won't reprint them here), but it worked just often enough to make us think something else was the problem. After some thorough testing with the  multimeter, and some careful coaxing of the connections, power was restored to the build platform.

Here's the source model, which I've uploaded to Thingiverse. Note that this is meant to fit mint tins that I purchase wholesale, and as such may not fit the Altoids tins. I created the object in Google SketchUp and then exported it to STL for use with RepG. GoogleSketchUp is OK for doing things quickly, but there are a lot of things I wanted to do with the design that I couldn't convince SketchUp to let me do. Particularly, I wanted nice rounded edges on the top of my model.

I'm happy with the final print, and really stoked about Makerbotting more models in the future!

It's hard to imagine that 15 years have passed since Hackers came out. It seems like only yesterday we were grimacing through the bad acting and worse "technology" references that Hackers brought to life.

Whether you love or hate the movie, everyone can agree that the 15th Anniversary Party is the perfect excuse to dress up in ridiculous looking retro clothing and drink with a bunch of other wannabe cyberpunks. In other words, it will be a typical night in Williamsburg.

More information on the party Kickstarter Page

Every now and then I come across something that I can’t believe someone has taken the time to write.

In this case, it’s a DOS emulator for the Nintendo DS, the DSx86.

As if there weren’t enough options for extending the DS (homebrew, NES emulation), one man has taken it upon himself to write a DOS emulator for the DS. It will run most games that run on a 286/386, with some caveats. Not all the opcodes have been written in yet, and sound blaster support requires a little bit of finesse. But it’s under active development, which is exciting.

You load a game the same way you did back in the day… from the command line. Since the DS doesn’t have a keyboard, DSx86 includes one for you on the touch screen. It’s hilariously adorable.

Once the game is loaded you can either continue to use the “keyboard” or swap screens so that your game is shown in the touchscreen and you can use the stylus as a mouse. Holding down left/right on the d-pad to click took a bit of getting used to. There’s also a tap-to-click mode, but I found it difficult to use for gaming. It was a good illustration of how programs not designed for a touchscreen can be infuriating to use on one. Because most DOS games run at a slightly larger resolution than the DS, you have the option of either panning or resizing the screen. I found panning to be the most useful, and all it really cut off was the title/menu bar at the top.

Seeing the old Sierra logo along with the MIDI-tastic intro music on the DS was ridiculous and awesome. I’m just about finished my 20th or so lifetime play through of The Island of Dr. Brain. If you’re dying to take some of your old DOS games with you (you did save them all, right?) , check the compatibility list and give DSx86 a try.