Extracting Data From Smart Scale Gives Rube Goldberg A Run For His Money

October 31, 2021 by Donald Papp 20 Comments

[Kevin Norman] got himself a smart body scale with the intention of logging data for his own analysis, but discovered that extracting data from the device was anything but easy. It turns out that the only way to access data from his scale is by viewing it in a mobile app. Screen-scraping is a time-honored method of pulling data from uncooperative systems, so [Kevin] committed to regularly taking a full-height screenshot from the app and using optical character recognition (OCR) to get the numbers, but making that work was a surprisingly long process full of dead ends.

First of all, while OCR can be reliable, it needs the right conditions. One thing that ended up being a big problem was the way the app appends units (kg, %) after the numbers. Not only are they tucked in very close, but they’re about half the height of the numbers themselves. It turns out that mixing and matching character height, in addition to snugging them up against one another, is something tailor-made to give OCR reliability problems.

The solution for this particular issue came from an unexpected angle. [Kevin] was using an open-source OCR program called Tesseract, and joined an IRC community #tesseract to ask for advice after exhausting his own options. The bemused members of the online community informed [Kevin] that they had nothing to do with OCR; #tesseract was actually a community for an open-source 3D FPS shooter of the same name. But as luck would have it, one of the members actually had OCR experience and suggested the winning approach: pre-process the image with OpenCV, using cv2.findContours() to detect and create a bounding box around each element. If an element is taller than a decimal point but shorter than everything else, throw it out. With that done, there were still a few more tweaks required, but the finish line was finally in sight.

Now [Kevin] can use the scale in the morning, take a screenshot, and in less than half a minute the results are imported into a database and visualizations generated. The resulting workflow might look like something Rube Goldberg would approve of, but it works!

Raspberry Pi Reads What It Sees, Delights Children

October 31, 2021 by Donald Papp 13 Comments

[Geyes30]’s Raspberry Pi project does one thing: it finds arbitrary text in the camera’s view and reads it out loud. Does it do so flawlessly? Not really. Was it at least effortless to put together? Also no, but it does wonderfully illustrate the process of gluing together different bits of functionality to make something new. Also, [geyes30]’s kids find it fascinating, and that’s a win all on its own.

The device is made from a Raspberry Pi and camera and works by sending a still image from the camera to an optical character recognition (OCR) program, which converts any visible text in the image to its ASCII representation. The recognized text is then piped to the espeak engine and spoken aloud. Getting all the tools to play nicely took a bit of work, but [geyes30] documented everything so well that even a novice should be able to get the project up and running in an afternoon.

Sometimes a function like text-to-speech is an end result in and of itself. This was also true of another similar project: Magic Mirror, whose purpose was to tirelessly indulge children’s curiosity about language.

Seeing other projects come to life and learning about new tools is a great way to get new ideas, and documenting them helps cross-pollinate among creative types. Did something inspire you recently, or have you documented your own project? We want to hear about it and so do others, so let us know via the tips line!

Continue reading “Raspberry Pi Reads What It Sees, Delights Children” →

A Builders Guide For The Perfect Solid-State Tesla Coil

October 31, 2021 by Dave Rowntree 14 Comments

[Zach Armstrong] presents for your viewing pleasure a simple guide to building a solid-state Tesla coil. The design is based around a self-resonant setup using the UCC2742x gate driver IC, which is used in a transformer-coupled full-wave configuration for delivering maximum power from the line input. The self-resonant bit is implemented by using a small antenna nearby the coil to pick up the EM field, and by suitably clamping and squaring it up, it is fed back into the gate driver to close the feedback loop. Such a setup within reason allows the circuit to oscillate with a wide range of Tesla coil designs, and track any small changes, minimizing the need for fiddly manual tuning that is the usual path you follow building these things.

Since the primary is driven with IGBTs, bigger is better. If the coil is too small, the resonant frequency would surpass the recommended 400 kHz, which could damage the IGBTs since they can’t switch much faster with the relatively large currents needed. An important part of designing Tesla coil driver circuits is matching the primary coil to the driver. You could do worse than checkout JavaTC to help with the calculations, as this is an area of the design where mistakes often result in destructive failure. The secondary coil design is simpler, where a little experimentation is needed to get the appropriate degree of coil coupling. Too much coupling is unhelpful, as you’ll just get breakdown between the two sides. Too little coupling and efficiency is compromised. This is why you often see a Tesla coil with a sizeable gap between the primary and secondary coils. There is a science to this magic!

A 555 timer wired to produce adjustable pulses feeds into the driver enable to allow easily changing the discharge properties. This enables it to produce discharges that look a bit like a Van De Graaff discharge at one extreme, and produce some lovely plasma ‘fire’ at the other.

We’ve covered Tesla coils from many angles over the years, recently this plasma tweeter made sweet sounds, and somehow we missed an insanely dangerous Tesla build by [StyroPyro] just checkout that rotary spark gap – from a distance.

Continue reading “A Builders Guide For The Perfect Solid-State Tesla Coil” →

Dante’s Inferno Arcade Reveals Your True Fate

October 31, 2021 by Lewin Day 6 Comments

Many of us are vaguely familiar with the levels of hell described in Dante’s epic poem from the 14th century, even if we’ve never visited ourselves. It’s natural to wonder in which circle of hell one might end up, but that’s a question that [scubabear’s] arcade build seeks to answer.

The stand-up cabinet was built for The Magic Castle, Hollywood’s exclusive private club for the magic set. The design is loosely inspired by old-fashioned love testers, the sort of which you might have seen in that Simpsons episode. The club has traditionally issued members with unique RFID tags in keychains, which can be used to trigger special objects in the facility. In this case, when a member scans their keychain and places their hand on a handprint, the machine starts up.

The hand is detected by an Adafruit touch sensor board, and the machine begins determining the fate of the member while playing a short musical interlude. Once calculated by the Raspberry Pi within, the user’s name is read out based on the RFID data, and their destiny is read aloud. They’re then given a receipt stating their destination in hell, along with a quote from Dante’s Inferno.

It’s a fun game and one that we’d love to try out if we find ourselves visiting The Magic Castle one spooky day. It’s made all the better by the sheer quality of the cabinet and the associated illustrations from [Jeremy Owen.]

For those keen to do something similar, [scubabear] hasn’t skimped on the details of the construction nor the electronics that make it all work. We’ve seen other great builds too, from the carefully crafted to the glowiest you’ve ever seen. Video after the break.

Continue reading “Dante’s Inferno Arcade Reveals Your True Fate” →

A Fascinating Plot Twist As Researchers Recreate Classic “Primordial Soup” Experiment

October 31, 2021 by Dan Maloney 46 Comments

Science is built on reproducibility; if someone else can replicate your results, chances are pretty good that you’re looking at the truth. And there’s no statute of limitations on reproducibility; even experiments from 70 years ago are fair game for a fresh look. A great example is this recent reboot of the 1952 Miller-Urey “primordial soup” experiment which ended up with some fascinating results.

At the heart of the Miller-Urey experiment was a classic chicken-and-the-egg paradox: complex organic molecules like amino acids and nucleic acids are the necessary building blocks of life, but how did they arise on Earth before there was life? To answer that, Stanley Miller, who in 1952 was a graduate student of Harold Urey, devised an experiment to see if complex molecules could be formed from simpler substances under conditions assumed to have been present early in the planet’s life. Miller assembled a complicated glass apparatus, filled it with water vapor and gasses such as ammonia, hydrogen, and methane, and zapped it with an electric arc to simulate lightning. He found that a rich broth of amino acids accumulated in the reaction vessel; when analyzed, the sludge was found to contain five of the 20 amino acids.

The Miller-Urey experiment has been repeated over and over again with similar results, but a recent reboot took a different tack and looked at how the laboratory apparatus itself may have influenced the results. Joaquin Criado-Reyes and colleagues found that when run in a Teflon flask, the experiment produced far fewer organic compounds. Interestingly, adding chips of borosilicate glass to the Teflon reaction chamber restored the richness of the resulting broth, suggesting that the silicates in the glassware may have played a catalytic role in creating the organic soup. They also hypothesize that the highly alkaline reaction conditions could create microscopic pits in the walls of the glassware, which would serve as reaction centers to speed up the formation of organics.

This is a great example of a finding that seems to knock a hole in a theory but actually ends up supporting it. On the face of it, one could argue that Miller and Urey were wrong since they only produced organics thanks to contamination from their glassware. And it appears to be true that silicates are necessary for the abiotic generation of organic molecules. But if there was one thing that the early Earth was rich in, it was silicates, in the form of clay, silt, sand, rocks, and dust. So this experiment lends support to the abiotic origin of organic molecules on Earth, and perhaps on other rocky worlds as well.

[Featured image credit: Roger Ressmeyer/CORBIS, via Science History Institute]

A Hackable Keyboard That Even Has Screens

October 30, 2021 by Lewin Day 4 Comments

There are a huge number of available keyboards out in the world these days, catering to all of the plainest and the most advanced desires. However, if you want something that’s just right, sometimes it pays to build your own. [Zach] did just that.

One of the key features of [Zach]’s build is that it diverges away from the Cherry MX switch form factor. The design uses low-profile switches instead, which help with keeping the keyboard low enough to avoid it causing wrist problems. The keyboard also uses IO expanders to hook up all the key switches, helping to reduce the incidence of ghost keys. The board can also be split in half, allowing it to be repurposed as a smaller macropad when desired.

It’s all wrapped up in a cool 3D printed case, and there are even three OLED displays on the right-hand side. They’re soldered to the PCB on special cutouts that allow the displays to flex and trigger tactile switches, acting as giant pressable buttons.

[Zach] does a great job explaining all the nifty engineering decisions he made to cram maximum functionality into the design. We’ve seen some other great DIY ergonomic designs too. Video after the break.

Continue reading “A Hackable Keyboard That Even Has Screens” →

Handwriting Robots Are Sending Snail Mail

October 30, 2021 by Matthew Carlson 40 Comments

As a kid, you might remember taking a whole fistful of markers or crayons, gently lining them all up for maximum contact, mashing them into the paper, and marveling at the colorful multitude of lines. It seemed like an easy way to write many times more things with less effort. While not quite the same idea but in a similar vein, [Aaron Francis] shared an experience of creating handwriting robots to write thousands of letters.

Why did [Aaron] need to write thousands of letters? Direct mailing, of course! If you were sending someone a letter, if it looked handwritten they’re much more likely to open it. What better way to make it look handwritten than to use a pen rather than a printer? They started off with Axidraw, a simple plotter made by EMSL. Old laptops controlled a few plotters and they started to make progress. As with most things, scale became tricky. Adding more plotters just means more paper to replace and machines to restart. An automated system of replacing paper is fiendishly difficult so they went for a batching system. A sheet of plywood that can hold dozens of sheets of paper became the basis of a new mega-plotter. 3D printers and laser cutters helped make adapters and homing teeth. A Raspberry Pi replaced the old laptops and they scaled up to a few machines.

All in all, a pretty impressive build. If you’re looking to dip your toes into the plotting water, this pen plotter is about as simple as you can get.