Battling weeds can be expensive, labor intensive and use large amounts of chemicals. To help make this easier [NathanBuilds] has developed  V2 of his open-source drone weed spraying system, complete with automated battery swaps, herbicide refills, and an AI vision system for weed identification.

The drone has a 3D printed frame, doubling as a chemical reservoir. V1 used a off-the-shelf frame, with separate tank. Surprisingly, it doesn’t look like [Nathan] had issues with leaks between the layer lines. For autonomous missions, it uses ArduPilot running on a PixHawk, coupled with RTK GPS for cm-level accuracy and a LiDAR altimeter. [Nathan] demonstrated the system in a field where he is trying to eradicate invasive blackberry bushes while minimizing the effect on the native prairie grass. He uses a custom image classification model running on a Raspberry Pi Zero, which only switches on the sprayers when it sees blackberry bushes in the frame. The Raspberry Pi Global Shutter camera is used to get blur-free images.

At just 305×305 mm (1×1 ft), the drone has limited herbicide capacity, and we expect the flights to be fairly short. For the automated pit stops, the drone lands on a 6×8 ft pad, where a motorized capture system pulls the drone into the reload bay. Here a linear actuator pushes a new battery into the side of the drone while pushing the spend battery one out the other side. The battery unit is a normal LiPo battery in 3D-printed frame. The terminal are connected to copper wire and tape contacts on the outside the battery unit, which connect to matching contacts in the drone and charging receptacles. This means the battery can easily short if it touches a metal surface, but a minor redesign could solve this quickly. There are revolving receptacles on either side of the reload bay, which immediately start charging the battery when ejected from the drone.

Developing a fully integrated system like this is no small task, and it shows a lot of potential. It might look a little rough around the edges, but [Nathan] has released all the design files and detailed video tutorials for all the subsystems, so it’s ready for refinement.

A button that stopped working has probably led to more than a few smashed remotes over the years. Fortunately [pescado99] has shared a beautifully simple cure for dead or dying remote buttons: graphite dry lubricant.

Most remotes operate by pushing a conductive carbon coating on the back of the button onto a pair of contacts on the PCB. Unfortunately, that conductive coating can wear off, leaving you with a dead or dying button. The video after the break [pescado99] demonstrates how to use a cotton swab to apply powdered graphite to the rear of the buttons to make them conductive again. A soft pencil can also be used, but the graphite works better.

This beautifully simple hack is too good not to share and could save many remotes from landfills. If you’re more interested in upgrading remote, you can build your own universal remote or replace it with a web browser.

Gyroscopes are perfect to damper short impulses of external forces but will eventually succumb if a constant force, like gravity, is applied. Once the axis of rotation of the mass aligns with the axis of the external torque, it goes into the gimbal lock and loses the ability to compensate for the roll on that axis. [Hyperspace Pirate] tackled this challenge on a gyroscopically stabilized RC bike by shifting a weight around to help keep the bike upright.

[Hyperspace Pirate] had previously stabilized a little monorail train with a pair of control moment gyroscopes. They work by actively adjusting the tilt of gyroscopes with a servo to apply a stabilizing torque. On this bike, he decided to use the gyro as a passive roll damper, allowing it to rotate freely on the pitch axis. The bike will still fall over but at a much slower rate, and it buys time for a mass on the end of the servo-actuated arm to shift to the side. This provides a corrective torque and prevents gimbal lock.

[Hyperspace Pirate] does an excellent job of explaining the math and control theory behind the system. He implemented a PD-controller (PID without the integral) on an Arduino, which receives the roll angle (proportional) from the accelerometer on an MPU6050 MEMS sensor and the roll rate (Derivative) from a potentiometer that measures the gyro’s tilt angle. He could have just used the gyroscope output from the MPU6050, but we applaud him for using the actual gyro as a sensor.

Like [Hyperspace Pirate]’s other projects, aesthetics were not a consideration. Instead, he wants to experiment with the idea and learn a few things in the process, which we can support.

Although we think of air-to-air radar as a relatively modern invention, it first made its appearance in WWII. Some late war fighters featured the AN/APS-13 Tail Warning Radar to alert the pilot when an enemy fighter was on his tail. In [WWII US Bombers]’ fascinating video we get a deep dive into this fascinating piece of tech that likely saved many allied pilots’ lives.

Fitted to aircraft like the P-51 Mustang and P-47 Thunderbolt, the AN/APS-13 warns the pilot with a light or bell if the aircraft comes within 800 yards from his rear. The system consisted of a 3-element Yagi antenna on the vertical stabilizer, a 410 Mhz transceiver in the fuselage, and a simple control panel with a warning light and bell in the cockpit.

In a dogfight, this allows the pilot to focus on what’s in front of him, as well as helping him determine if he has gotten rid of a pursuer. Since it could not identify the source of the reflection, it would also trigger on friendly aircraft, jettisoned wing tanks, passing flak, and the ground. This last part ended up being useful for safely descending through low-altitude clouds.

This little side effect turned out to have very significant consequences. The nuclear bombs used on Hiroshima and Nagasaki each carried four radar altimeters derived from the AN/APS-13 system.

Applying solder paste to a new custom PCB is always a little nerve-racking. One slip of the hand, and you have a smeared mess to clean up. To make this task a little easier, [Max Scheffler] built the Stencil Fix Portable, a compact self-contained vacuum table to hold your stencil firmly in place and pop it off cleanly every time.

The Stencil Fix V1 used a shop vac for suction, just like another stencil holder we’ve seen. The vacuum can take up precious space, makes the jig a little tricky to move, and bumping the hose can lead to the dreaded smear and colorful language. To get around this [Max] added a brushless drone motor with a 3D printed impeller, with a LiPo battery for power. The speed controller gets its PWM signal from a little RP2040 dev board connected to a potentiometer. [Max] could have used a servo tester, but he found the motor could be a little too responsive and would move the entire unit due to inertia from the impeller. The RP2040 allowed him to add a low pass filter to eliminate the issue. The adjustable speed also means the suction force can be reduced a little for easy alignment of the stencil before locking it down completely.

We love seeing tool projects like these that make future projects a little easier. Fortunately, [Max] made the designs available so you can build your own.

Steam turbines have helped drive a large chunk of our technological development over the last century or so, and they’ll always make for interesting DIY. [Hyperspace Pirate] built a small turbine and boiler in his garage, turning fire into flowing electrons, and learning a bunch in the process.

[Hyperspace Pirate] based the turbine design on 3D printed Pelton-style turbines he had previously experimented with, but milled it from brass using a CNC router. A couple of holes had to be drilled in the side of the rotor to balance it. The shaft drives a brushless DC motor to convert the energy from the expanding steam into electricity.

To avoid the long heat times required for a conventional boiler, [Hyperspace Pirate] decided to use a flash boiler. This involves heating up high-pressure water in a thin coil of copper tube, causing the water to boil as it flows down the tube. To produce the high-pressure water feed the propane tank for the burner was also hooked up to the water tank to pressurize it, removing the need for a separate pump or compressed air source. This setup allows the turbine to start producing power within twelve seconds of lighting the burner — significantly faster than a conventional boiler.

Throughout the entire video [Hyperspace Pirate] shows his calculation for the design and tests, making for a very informative demonstration. By hooking up a variable load and Arduino to the rectified output of the motor, he was able to measure the output power and efficiency. It came out to less than 1% efficiency for turning propane into electricity, not accounting for the heat loss of the boiler. The wide gaps between the turbine and housing, as well as the lack of a converging/diverging nozzle on the input of the turbine are likely big contributing factors to the low efficiency.

Like many of his other projects, the goal was the challenge of the project, not practicality or efficiency. From a gyro-stabilized monorail, to copper ingots from algaecide and and a DIY cryocooler, he has sure done some interesting ones.

Venus hasn’t received nearly the same attention from space programs as Mars, largely due to its exceedingly hostile environment. Most electronics wouldn’t survive the 462 °C heat, never mind the intense atmospheric pressure and sulfuric acid clouds. With this in mind, NASA has been experimenting with the concept of a completely mechanical rover. The [Beardy Penguin] and a team of fellow students from the University of Southampton decided to try their hand at the concept—video after the break.

The project was divided into four subsystems: obstacle detection, mechanical computer, locomotion (tracks), and the drivetrain. The obstacle detection system consists of three (left, center, right) triple-rollers in front of the rover, which trigger inputs on the mechanical computer when it encounters an obstacle over a certain size. The inputs indicate the position of each roller (up/down) and the combination of inputs determines the appropriate maneuver to clear the obstacle. [Beardy Penguin] used Simulink to design the logic circuit, consisting of AND, OR, and NOT gates. The resulting 5-layer mechanical computer quickly ran into the limits of tolerances and friction, and the team eventually had trouble getting their design to work with the available input forces.

Due to the high-pressure atmosphere, an on-board wind turbine has long been proposed as a viable power source for a Venus rover. It wasn’t part of this project, so it was replaced with a comparable 40 W electric motor. The output from a logic circuit goes through a timing mechanism and into a planetary gearbox system. It changes output rotation direction by driving the planet gear carrier with the sun gear or locking it in a stationary position.

As with many undergraduate engineering projects, the physical results were mixed, but the educational value was immense. They got individual subsystems working, but not the fully integrated prototype. Even so, they received several awards for their project and even came third in an international Simulink challenge. It also allowed another team to continue their work and refine the subsystems.

[bitluni] seems rather fond of soldering lots of LEDs, and fortunately for us the result is always interesting eye candy. The latest iteration of this venture features 8 mm WS2812D-F8 addressable LEDs, offering a significant simplification in electronics and the potential for much brighter displays.

The previous version used off-the-shelf 8×8 LED panels but had to be multiplexed, limiting brightness, and required a more complex driver circuit. To control the panel, [bitluni] used the ATtiny running the MegaTinyCore Arduino core. Off-the-shelf four-pin magnetic connectors allow the panels to snap together. They work well but are comically difficult to solder since they keep grabbing the soldering iron. [bitluni] also created a simple battery module and 3D printed neat enclosures for everything.

Having faced the arduous task of fixing individual LEDs on massive LED walls in the past, [bitluni] experimented with staggered holes that allow through-hole LEDs to be plugged in without soldering. Unfortunately, with long leads protruding from the back of the PCB, shorting became an immediate issue. While he ultimately resorted to soldering them for reliability, we’re intrigued by the potential of refining this pluggable design.

The final product snapped together satisfyingly, and [bitluni] programmed a simple animation scheme that automatically updates as panels are added or removed. What would you use these for? Let us know in the comments below.

Lenz’s Law is one of those physics tricks that look like magic if you don’t understand what’s happening. [Seth Robinson] was inspired by the way eddy currents cause a cylindrical neodymium magnet to levitate inside a rotating copper tube, so he cast a spherical copper cage to levitate a magnetic sphere.

Metal casting is an art form that might seem simple at first, but is very easy to screw up. Fortunately [Seth] has significant experience in the field, especially lost-PLA metal casting. While the act of casting is quick, the vast majority of the work is in the preparation process. Video after the break.

[Seth] started by designing and 3D printing a truncated icosahedron (basically a low-poly sphere) in two interlocking halves and adding large sprues to each halve. Over a week, the PLA forms were repeatedly coated in layers of ceramic slurry and silica sand, creating a thick shell around them. The ceramic forms were then heated to melt and pour out the PLA and fired at 870°C/1600°F to achieve full hardness.

With the molds prepared, the molten copper is poured into them and allowed to cool. To avoid damaging the soft copper parts when breaking away the mold, [Seth] uses a sandblaster to cut it away sections. The quality of the cast parts is so good that 3D-printed layer lines are visible in the copper, but hours of cleanup and polishing are still required to turn them into shiny parts. Even without the physics trick, it’s a work of art. A 3d printed plug with a brass shaft was added on each side, allowing the assembly to spin on a 3D-printed stand.

[Seth] placed a 2″ N52 neodymium spherical magnet inside, and when spun at the right speed, the magnet levitated without touching the sides. Unfortunately, this effect doesn’t come across super clearly on video, but we have no doubt it would make for a fascinating display piece and conversation starter.

Using and abusing eddy currents makes for some very interesting projects, including hoverboards and magnetic torque transfer on a bicycle.

A regular comment we see on electric aircraft is to “just add solar panels to the wings.” [James] from Project Air has been working on just such a solar plane, and as he shows in the video after the break, it is not a trivial challenge.

A solar RC plane has several difficult engineering challenges masquerading as one. First, you need a solid, efficient airframe with enough surface area for solar panels. Then, you need a reliable, lightweight, and efficient solar charging system and, finally, a well-tuned autopilot to compensate for a human pilot’s limited endurance and attention span.

In part one of this project, a fault in the electrical system caused a catastrophe so James started by benching all the electricals. He discovered the MPPT controller had a battery cutoff feature that he was unaware of, which likely caused the crash. His solution was to connect the solar panels to the input of a 16.7 V voltage regulator—just under the fully charged voltage of a 4S LiPo battery— and wire the ESC, control electronics, and battery in parallel to the output. This should keep the battery charged as long as the motor doesn’t consume too much power.

After rebuilding the airframe and flight testing without the solar system, [James] found the foam wing spars were not up to the task, so he added aluminum L-sections for stiffness. The solar panels and charging system were next, followed by more bench tests. On the test flight, it turned out the aircraft was now underpowered and struggled to gain altitude thanks to the added weight of the solar system. With sluggish control responses,[James] eventually lost sight of it behind some trees, which led to a flat spin and unplanned landing.

Fortunately, the aircraft didn’t sustain any damage, but [James] plans to redesign it anyway to reduce the weight and make it work with the existing power system.

We’ve seen several solar planes from [rctestflight] and meticulously engineered versions from [Bearospace Industrues]. If long flight times is primarily what you are after, you can always ditch the panels and  use a big battery for 10+ hour flights.

Sending teams of tiny drones to explore areas and structures is a staple in sci-fi and research, but the weight and size of sensors and the required processing power have long been a limiting factor. In the video below, a research team from [ETH Zurich] breaks through these limits, demonstrating indoor mapping with a swarm of tiny drones without dependence on any external systems.

The drone is the modular Crazyflie platform, which uses stackable PCBs (decks) to expand capabilities. The team added a Flow deck for altitude control and motion tracking, and a Loco positioning deck with a UWB module determining relative distances between drones. On top of this, the team added two custom decks. The first mounts four VL53L5CX 8×8 pixel TOF sensors for omnidirectional LIDAR scanning. The final deck does handles all the required processing with a GAP9 System-on-Chip, which features 10 RISC-V cores running on just 200 mW of power.

Of course the special sauce of this project lies in the software. The team developed a lightweight collaborative Simultaneous Localization And Mapping (SLAM) algorithm which can be distributed across all the drones in the swarm. It combines LIDAR scan data and the estimated position of the drone during the scan, and then overlays the data for the scans for each location across different drones, compensating for errors in the odometry data. The team also implemented inter-drone collision avoidance, packet collision avoidance and optimizing drones’ paths. The code is supposed to be available on GitHub, but the link was broken at the time of writing.

The Crazyflie platform has been around for more than a decade now, and we’ve seen it used in several research projects, especially related to autonomous navigation.

Commodity bearings are a a boon for makers who to want something to rotate smoothly, but what if you don’t have one in a pinch? [Cliff] of might have the answer for you, in the form of 3D printed bearings with filament rollers.

With the exception of the raw filament rollers, the inner and outer race, roller cage and cap are all printed. It would also be possible to design some of the components right into a rotating assembly. [Cliff] makes it clear this experiment isn’t about replacing metal bearings — far from it. Instead, it’s an inquiry into how self-sufficient one can be with a FDM 3D printer. That didn’t stop him from torture testing the design to its limits as wheel bearings on an off-road go-cart. The first version wasn’t well supported against axial loads, and ripped apart during some more enthusiastic maneuvers.

[Cliff] improved it with a updated inner race and some 3D printed washers, which held up to 30 minutes of riding with only minimal signs of wear. He also made a slightly more practical 10 mm OD version that fits over an M3 bolt, and all the design files are downloadable for free. Cutting the many pieces of filament to length quickly turned into a chore, so a simple cutting jig is also included.

Let us know in the comments below where you think these would be practical. We’ve covered some other 3D printed bearing that use printed races, as well as a slew bearing that’s completely printed.

Hydrofoils have been around for several decades, but watching a craft slice through the water with almost no wake never get old. In the videos after the break, [rctestflight] showcases his ambitious project: transforming a standup paddleboard into a rideable hydrofoil with active stabilization.

Unlike conventional electric hydrofoil boards that depend on rider skill for balance, [rctestflight] aims to create a self-stabilizing system. He began by designing a small-scale model, complete with servo-controlled ailerons and elevators, dual motors for differential thrust, and a dRehmFlight flight controller. A pair of sonar sensors help the flight controller maintain constant height above the water. The wings are completely 3D printed, with integrated hinges for flight control surfaces slots for wiring and control components. It’s better suited for 3D printing than RC aircraft since it’s significantly less sensitive to weight, allowing for more structural reinforcement. The small scale tests were very successful and allowed [rctestflight] to determine that he didn’t need the vertical stabilizer and rudder.

The full-sized version features a scaled up wing, larger servos and motors attached to an 11-foot standup paddleboard — minus its rear end — mounted on commercially available e-foil booms. A foam battery box stores a hefty LiFePO4 battery, while the electronics from the smaller version are repurposed here. Despite only catching glimpses of this larger setup in action at the end of the video, it promises an excitingly smooth lake ride we would certainly like to experience.

We’ve seen several 3D printed hydrofoils around here, but this promised to be the largest successful attempt. Don’t fail us [Daniel].

Marble machines are a fun and challenging reason to do engineering for the sake of engineering. [Engineezy] adds some color to the theme, building a machine to create 16×16 marble images automatically. (Video embedded below.)

The core problem was devising ways to sort, lift, place, and dump marbles in their correct positions without losing their marbles—figuratively and literally. Starting with color detection, [Engineezy] used an RGB color sensor and Euclidian math to determine each marble’s color. After trying several different mechanical sorting mechanisms, he settled on a solenoid and servo-actuated dump tube to drop the marble into the appropriate hopper.

After sorting, he faced challenges with designing a mechanism to transport marbles from the bottom hoppers to the top of the machine. While paddle wheels seemed promising at first, they tended to jam—a problem solved by innovating with Archimedes screws that move marbles up smoothly without clogs. The marbles are pushed into clear tubes on either side of the machine, providing a clear view of their parade to the top.

Perhaps most ingenious is his use of constant-force springs as a flexible funnel to guide the marbles to a moving slider that drops them into the correct column of the display. When a picture is complete, sliding doors open on the bottom of the columns, dumping the marbles into a chain lift which feeds them into the sorting section. Each of the mechanisms has a mirrored version of the other side, so the left and right halves of the display operate independently.

The final product is slow, satisfying and noisy kinetic testament to [Engineezy]’s perseverance through countless iterations and hiccups.

Marble machines can range from minimalist to ultra-complex musical monstrosities, but never fail to tickle our engineering minds.

Multiple motors or servos are the norm for drones to achieve controllable flight, but a team from MARS LAB HKU was able to a 360° lidar scanning drone with full control on just a single motor and no additional actuators. Video after the break.

The key to controllable flight is the swashplateless propeller design that we’ve seen few times, but it always required a second propeller to counteract self-rotation. In this case the team was able to make that self-rotation work for them to achieve 360° scanning with a single fixed LIDAR sensor. Self-rotation still need to be slowed was successfully done with four stationary vanes. The single rotor also means better efficiency compared to a multi-rotor with similar propeller disk area.

The LIDAR comprises a full 50% of the drones weight and provides a conical FOV out to a range of 450m. All processing happens onboard the drone, with point cloud data being processed by a LIDAR-inertial odometry framework. This allows the drone to track and plan it’s flight path while also building a 3D map of an unknown environment. This means it would be extremely useful for indoor or undergrounds environments where GPS or other positioning systems are not available.

All the design files and code for the drone is up on GitHub, and most of the electronic components are off-the-shelf. This means you can build your own, and the expensive lidar sensor is not required to get it flying. This seems like a great platform for further experimentation, and getting usable video from a normal camera would be an interesting challenge.

Drones that charge right on the power lines they inspect is a promising concept, but comes with plenty of challenges. The Drone Infrastructure Inspection and Interaction (Diii) Group of the University of South Denmark is tackling these challenges head-on.

The gripper for these drones may seem fairly straightforward, but it needs to inductively charge, grip, and detach reliably while remaining simple and lightweight. To attach to a power line, the drone pushes against it, triggering a cord to pull the gripper closed. This gripper is held closed electromagnetically using energy harvested from the power line or the drone’s battery if the line is off. Ingeniously, this means that if there’s an electronics failure, the gripper will automatically release, avoiding situations where linemen would need to rescue a stuck drone.Accurately mapping power lines in 3D space for autonomous operation presents another hurdle. The team successfully tested mmWave radar for this purpose, which proves to be a lightweight and cost-efficient alternative to solutions like LiDAR.

We briefly covered this project earlier this year when details were limited. Energy harvesting from power lines isn’t new; we’ve seen similar concepts applied in government-sanctioned spy cameras and border patrol drones. Drones are not only used for inspecting power lines but also for more adventurous tasks like clearing debris off them with fire.

In an educational project with ethically questionable applications, [ChromaLock] has converted the ubiquitous TI-84 calculator into the ultimate cheating device.

The foundation of this hack lies in the TI-84’s link protocol, which has been a mainstay in calculator mods for years. [ChromaLock] uses this interface to connect to a tiny WiFi-enabled XIAO ESP32-C3 module hidden in the calculator. It’s mounted on a custom PCB with a simple MOSFET-based level shifting circuit, and slots neatly into a space on the calculator rear cover. The connecting wires are soldered directly to the pads of the 2.5 mm jack, and to the battery connections for power.

But what does this mod do? It connects your calculator to the internet and gives you a launcher with several applets. These allow you to view images badly pixelated images on the TI-84’s screen, text-chat with an accomplice, install more apps or notes, or hit up ChatGPT for some potentially hallucinated answers. Inputting long sections of text on the calculator’s keypad is a time-consuming process, so [ChromaLock] teased a camera integration, which will probably make use of newer LLMs image input capabilities. The ESP32 doesn’t handle all the heavy lifting, and needs to connect to an external server for more complex interfaces.

To prevent pre-installed programs from being used for cheating on TI-84s, examiners will often wipe the memory or put it into test mode. This mod can circumvent both. Pre-installed programs are not required on the calculator to interface with the hardware module, and installing the launcher is done by sending two variables containing a password and download command to the ESP32 module. The response from the module will also automatically break the calculator out of test mode.

We cannot help but admire [ChromaLock]’s ingenuity and polished implementation, and hopefully our readers are more interested in technical details than academic self-sabotage. For those who need even more capability in their calculator, we’d suggest checking out the NumWorks.