A crystal radio is a common enough science fair project, but the problem is, there isn’t much on anymore. The answer is, of course, obvious: build your own AM transmitter, too. AM modulation isn’t that hard to do and [Science Buddies] has plans for how to build one with a canned oscillator and an audio transformer.
We don’t imagine the quality of this would be so good, but for a kid’s science project it might be worth a shot. Maybe something like “What kind of materials block radio waves?” would be a good project statement.
[Ben Finio] designed this project as a way to get kids interested in learning about science and engineering. Is it bad that we just want to build one of our own? It’s a light following bristlebot which in itself is quite simple to build and understand. We think the platform has a lot of potential for leading to other things, like learning about microcontrollers and wireless modules to give it wireless control.
Right now it’s basically two bristlebots combined into one package. The screen capture seen above makes it hard to pick out the two toothbrush heads on either side of a battery pack. The chassis of the build is a blue mini-breadboard. The circuit that makes it follow light is the definition of simple. [Ben] uses two MOSFETs to control two vibration motors mounted on the rear corners of the chassis. The gate of each MOSFET is driven by a voltage divider which includes a photoresistor. When light on one is brighter than the other it causes the bot to turn towards to the brighter sensor. When viewing the project log above make sure to click on the tabs to see all of the available info.
This directional control seems quite good. We’ve also seen other versions which shift the weight of the bot to change direction.
Riley, an 8 lb pug, has more beauty than brains, and a palate as unrefined as crude oil. While we hate criticizing others’ interests and tastes, his penchant for eating cat poop needed to stop. After a thorough exploration of a variety of options, including cat food additives that make its excrement taste worse (HOW? WHY? Clearly taste wasn’t the issue!), automatic litter boxes that stow the secretions, and pet doors that authenticate access to the room with the litter box, [Science Buddies] eventually settled on a solution that was amenable to all members of the family.
The trick was in creating a door mechanism with a blacklist of sorts rather than a whitelist. As the cat didn’t like to push the door open itself, the solution needed to have the pet door open by default. A magnet on Riley’s collar would trip a sensor attached to an Arduino that would control servos to swing the door shut immediately if he attempted to access the defecated delights. Of course safety was a consideration with the door swinging in Riley’s face.
Did you know that water can drip UP instead of down? It’s true! Okay, okay- it’s a bit of an optical illusion, but one that’s mesmerizing no less, and it’s one that is especially awe-inspiring for kids. As [Science Buddies] explains in the video below the break, it’s also achievable for anyone with some basic supplies.
On first glance, the “water dripping upward” illusion looks like it must be extremely complicated with precisely timed drops, and perfectly triggered strobing lights and the like- right? Well, not so much. [Science Buddies] demonstrates a highly simplified experiment using only an aquarium pump, a basic frame, a smart phone with a strobing app, and naturally, water. The experiment is presented in a simple manner that would allow a young person to replicate it without too much adult intervention.
The video goes into such concepts as frequency, duty cycle (pulse width modulation), and other basic engineering principles. The experiment can be completed for just a few dollars for the pump and tubing, and the rest can be improvised. What a great way to get a young one started on their way to engineering!
Multirotors, or drones as they’re popularly called, are so ubiquitous as to have become a $10 toy. They’re no less fun to fly for it though, and learning how they work is no less fascinating. It’s something [Science Buddies] has addressed in a series of videos examining them from first principles. They may be aimed at youngsters, but they’re still an entertaining enough watch for those of advancing years.
Instead of starting with a multirotor control board, the video takes four little DC motors and two popsicle sticks to make a rudimentary drone frame. Then with the help of dowels and springs it tethers the craft as the control mechanisms are explained bit by bit, from simple on-off motor control through proportional control to adding an Arduino and following through to how a multirotor stays in flight. It’s instructional and fun to watch, and maybe even for some of us, a chance to learn something.
When you think of high-throughput ptychographic cytometry (wait, you do think about high throughput ptychographic cytometry, right?) does it bring to mind something you can hack together from an old Blu-ray player, an Arduino, and, er, some blood? Apparently so for [Shaowei Jiang] and some of his buddies in this ACS Sensors Article.
For those of you who haven’t had a paper accepted by the American Chemical Society, we should probably clarify things a bit. Ptychography is a computational method of microscopic imaging, and cytometry has to do with measuring the characteristics of cells. Obviously.
This is definitely what science looks like.
Anyway, if you shoot a laser through a sample, it diffracts. If you then move the sample slightly, the diffraction pattern shifts. If you capture the diffraction pattern in each position with a CCD sensor, you can reconstruct the shape of the sample using breathtaking amounts of math.
One hitch – the CCD sensor needs a bunch of tiny lenses, and by tiny we mean six to eight microns. Red blood cells are just that size, and they’re lens shaped. So the researcher puts a drop of their own blood on the surface of the CCD and covers it with a bit of polyvinyl film, leaving a bit of CCD bloodless for reference. There’s an absolutely wild video of it in action here.
In late June of 2021, GitHub launched a ‘technical preview’ of what they termed GitHub Copilot, described as an ‘AI pair programmer which helps you write better code’. Quite predictably, responses to this announcement varied from glee at the glorious arrival of our code-generating AI overlords, to dismay and predictions of doom and gloom as before long companies would be firing software developers en-masse.
As is usually the case with such controversial topics, neither of these extremes are even remotely close to the truth. In fact, the OpenAI Codex machine learning model which underlies GitHub’s Copilot is derived from OpenAI’s GPT-3 natural language model, and features many of the same stumbles and gaffes which GTP-3 has. So if Codex and with it Copilot isn’t everything it’s cracked up to be, what is the big deal, and why show it at all?
