[Adam] was tired of plopping the same components over and over into his Arduino-based designs. He spent part of his weekend laying out a small board that would host everything he needed and could be built as a single component for all future projects. Above you can see the project he calls SMDuino, an Arduino clone that can be used as a surface mount part.
The contacts on four sides of the board break out the pins. They’ve been designed with 0.1″ pitch which means they will work with standard pin headers. But since they’re plated through from top to bottom they are easy to solder to surface mount pads as well. The project is open source, so you can order your own boards (he used DorkBot PDX) or email him if you want to get in on a pre-order. That is for unpopulated PCBs only. But there’s few components used here so it’s pretty inexpensive to throw together. You’ll need four caps, four resistors, a crystal, an LED, the ATmega*8 of your choice (an ATmega328 is used here), and a low dropout regulator. Of course it is possible to go without the crystal oscillator.
Does this remind anyone else of the Basic Stamp 2?
[Joris Laurenssen] has been hanging onto this robotic arm for about twenty years. His most recent project uses some familiar tools to add voice control for each of the arm’s joints.
The arm has its own controller which connects via a DB-25 port. [Joris’] first task was to figure out what type of commands are being sent through the connection. He did some testing to establish the levels of the signals, then hooked up his Arduino and had it read out the values coming through the standard parallel connection. This let him quickly establish the simple ASCII character syntax used to command movement from the device. There’s only eight command sets, and it didn’t take much work to whip up a sketch that can now drive the device.
The second portion of the project is to use voice commands to push these parallel signals to the arm. Instead of reinventing the wheel he decided to use the speech recognition feature of his Android phone. He used Scripting Layer for Android (SL4A) and a Python script to interpret commands, push them to his computer via Telnet, and finally drive the arm. We’ve embedded the video demo after the break. He gives the commands in Dutch but he overlaid comments in English so you can tell what’s going on.
Continue reading “Adding speech control to an old robotic arm”
Talk about reducing the costs of a build, this tricopter uses cardboard as a frame and has one less motor than its quadcopter relatives. There are almost no details other than those shared in the video after the break so we’re just going to guess based on what we see (feel free to share your own insight in the comments).
The smooth curves of this integrated landing pad makes us thing the frame was cut either with a CNC device or a utility-knife wielding ninja. Two of the three motor supports look just like what is shown above, but the third has a hinged mounting bracket attached to a servo motor. This way the propeller can be tilted around an axis running parallel to the support arm. We’d bet this feature is mainly for adjusting the yaw of the aircraft.
The video comments mention that this can hover when the throttle is at 45%, showing that there’s a lot lift available when needed. That is until you really weigh it down by adding plastic cages around the propellers. It’s kind of neat to see the thing ‘sticking’ to the ceiling at the end of that clip by driving the throttle wide open and using the cages as top-sided landing gear.
Continue reading “Cardboard framed tricopter”
[Mike Field] got his hands on this Syma S107 helicopter with the intention of hacking it. After playing around with it for a while he set out to build his own infrared controller for the toy. It seems there is some protocol information about it published in various forum posts, but he decided it would be more fun to figure it out for himself.
He started off trying to capture the IR signals using Adafruit’s tutorial which has come in handy on a number of other projects. He could get his television remote to register, but not the toy’s controller. This didn’t stop fun, instead he tore open the controller and grabbed a logic sniffer to see what’s being pushed to the IR LEDs. The signals are a bit curious. It seems two different packets are sent with each command which [Mike] thinks is for use with two different models of the toy. In addition to that the frames are not synchronized. But a bit of 10 MHz sampling helped him to figure everything out, and he believes he’s got a more accurate version of the protocol than had previously been discovered. To prove it he developed an FPGA-based controller using VHDL which he shows off in the clip after the break.
Continue reading “Decoding, then cloning an IR helicopter toy’s control signals”