The Etch-a-Sketch was a popular toy, but a polarizing one. You were either one of those kids that had the knack, or one of the kids that didn’t. [Micah] was pretty firmly in the latter group, so decided to roboticize the Etch-a-Sketch so a computer could draw for him instead.
The build uses a pair of stepper motors attached to the Etch-a-Sketch’s knobs via 3D-printed adapters. It took [Micah] a few revisions to get the right design and the right motors for the job, but it all came together. A Raspberry Pi is charged with driving the motors to draw the desired picture.
Beyond the mechanics, [Micah] also does a great job of explaining the challenges around drawing and the drive software. Namely, the Etch-a-Sketch has a major limitation in that there’s no way to move the stylus without drawing a line. He accounts for this in his code for converting and drawing images.
The robot draws slowly but surely. The final result is incredibly impressive, and far exceeds what most of us could achieve on by hand. We’ve seen some similar builds in the past, too. Video after the break.
Hackaday.io user [mabe42] works during the week away from their home city and rents a small apartment locally to make this life practical. However, the heating system, a night-storage system, is not so practical. They needed a way to remotely control the unit so that the place was habitable after a long winter commute; lacking internet connectivity, they devised a sensible solution to create an SMS-controlled remote heating controller.
The controller runs atop an old Raspberry Pi B inside a 3D-printed case. Seeing such an old board given a real job to do is nice. Connectivity is via a USB UMTS stick which handles the SMS over the cellular network. The controller knob for the heater thermostat (not shown) is attached via a toothed belt to a pully and a 28BYJ-48 5V geared stepper motor. Temperature measurement is via the ubiquitous DS1820 module, which hooks straight up to the GPIO on the Pi and works out of the box with many one-wire drivers.
The software is built on top of Gammu, which handles the interface to the UMTS device. Daily and historical temperature ranges are sent via SMS so [mabe42] can decide how to configure the heating before their arrival. The rest of the software stack is in Python, as per this (German-language) GitHub project.
When using a manual machine tool such as a lathe or milling machine, there can be a lot of pressure to read the position and feed the axes at the correct rate. That’s why modern machines typically have some form of digital read-out (DRO). [Stefano Bertelli] has created a simple Raspberry Pi based DRO with an additional twist, that of a linked motor drive output.
Realtime encoder position reading and motor control are best done with a dedicated microcontroller, ideally with a proper RTOS.
The axes that need to be monitored should be mechanically attached to a position sensor like a linear encoder or a rotary type. Using a linear sensor with a linear axis instead of a rotary encoder on the downstream dial is better. For the readout unit, [Stefano] used a WaveShare 7-inch touchscreen module with a Raspberry Pi 3 for the UI of the readout unit. The Pi has a custom-designed HAT, that performs power conditioning and provides a robust RS485 interface. Connected via that RS485 link is another custom PCB based on an STM32F411 with a few supporting power supplies and interfacing components. The job of this board is to interface to the position encoders, reading positioning pulses using interrupts. There is an additional stepper motor drive courtesy of a ULN2003 Darlington driver to allow the control of a single motorised axis. An additional motor driver module is required, which should be no surprise since driving a milling machine axis will require a fairly beefy motor. This GitHub repo contains the FreeRTOS-based firmware for this board. This motor drive has the ability to be connected to a measuring axis in a programmable way, enabling one axis to be adjusted to follow or jump in controlled steps with another. This feature can significantly simplify certain types of machining operations, as [Stefano] elaborates in the video.
Lastly, the Raspberry Pi runs a simple Python application with Kivy for the GUI. As [Stefano] explains in the video below, this makes debugging and modification quite simple.
Parol 6 is a 3D-printed six-axis robot arm created by [Petar Crnjak] as a combination of the principles from a few previous projects. Aside from a pneumatic gripper, each axis is driven by a stepper motor, with at least a few of these axes being driven through a metal planetary gearbox for extra precision and torque.
For the last ten to fifteen years, optical drives have been fading out of existence. There’s little reason to have them around anymore unless you are serious about archiving data or unconvinced that streaming platforms will always be around. While there are some niche uses for them still, we’re seeing more and more get repurposed for parts and other projects like this tabletop laser engraver.
The build starts with a couple optical drives, both of which are dismantled. One of the shells is saved to use as a base for the engraver, and two support structures are made out of particle board and acrylic to hold the laser and the Y axis mechanism. Both axes are made from the carriages of the disassembled hard drives, with the X axis set into the base to move the work piece. A high-output laser module is fitted to the Y axis with a heat sink, and an Arduino and a pair of A4988 motor controllers are added to the mix to turn incoming G-code into two-dimensional movement.
[legolor] brings us a great, cheap rotary axis to add to your small 3 axis CNC mills. How are you going to generate G-Code for this 4th axis? That’s the great part, and the hack, that [legolor] really just swapped the Y axis for the rotation. To finish the workflow and keep things cheap accessible to all there’s a great trick to “unwrap” your 3D model so your CAM software of choice thinks it’s still using a linear Y axis and keeps your existing workflow largely intact. While this requires an extra step in Blender to do the unwrapping, we love the way this hack changes as little of the rest of your process as possible. The Blender script might be useful for many other purposes too.
The results speak for themselves too! We thought the 3D printed parts were suspect in a CNC setup, but for the small scale of game pieces and milling wood, the setup is stable enough to produce a surprisingly accurate and detailed finish. If you want to try the same approach with something larger or a tougher material, [legolor] has a suggestion of a tailstock setup that’s still under $100 USD. Continue reading “This $12 CNC Rotary Axis Will Make Your Head Spin”→
Magnetic loop antennas are great if you are limited on space since they are just a potentially small loop of wire. The problem is, they are sharply tuned. You normally have an adjustment capacitor to tune the antenna to different frequencies. [TekMakerUK] built one with a motor and an Arduino that he can tune from an Android phone. You can see more about the project in the video below.
If you want to transmit, the capacitor is often the weak part of the system. Luckily, some old gear yielded a capacitor with multiple sections and enough plate distance to handle the 5W desired. Of course, motor driving a capacitor isn’t a new idea, but this setup is nice since it uses a stepper motor and a rotary encoder.
