Misc Hacks

4063 Articles

Advanced Techniques For Using Git With KiCAD

September 19, 2018 by 4 Comments

For most developers “distributed version control” probably means git. But by itself git doesn’t work very well with binary files such as images, zip files and the like because git doesn’t know how to make sense of the structure of an arbitrary blobs of bytes. So when trying to figure out how to track changes in design files created by most EDA tools git doesn’t get the nod and designers can be trapped in SVN hell. It turns out though KiCAD’s design files may not have obvious extensions like .txt, they are fundamentally text files (you might know that if you’ve ever tried to work around some of KiCAD’s limitations). And with a few tweaks from [jean-noël]’s guide you’ll be diffing and merging your .pro’s and .sch’s with aplomb.

There are a couple sections to the document (which is really meant as an on boarding to another tool, which we’ve gotten to in another post). The first chunk describes which files should be tracked by the repo and which the .gitignore can be configured to avoid. If that didn’t make any sense it’s worth the time learning how to keep a clean repo with the magic .gitignore file, which git will look for to see if there are any file types or paths it should avoid staging.

The second section describes how you can use two nifty git features, cleaning and smudging, to dynamically modify files as they are checked in and out of the repo. [jean-noël]’s observation is that certain files get touched by KiCAD even if there are no user facing changes, which can clutter patch sets with irrelevant changes. His suggested filters prevent this by stripping those changes out as files get checked in. Pretty slick.

Visual Schematic Diffs In KiCAD Help Find Changes

September 19, 2018 by 11 Comments

When writing software a key part of the development workflow is looking at changes between files. With version control systems this process can get pretty advanced, letting you see changes between arbitrary files and slices in time. Tooling exists to do this visually in the world of EDA tools but it hasn’t really trickled all the way down to the free hobbyist level yet. But thanks to open and well understood file formats [jean-noël] has written plotgitsch to do it for KiCAD.

In the high(er)-end world of EDA tools like OrCAD and Altium there is a tight integration between the version control system and the design tools, with the VCS is sold as a product to improve the design workflow. But KiCAD doesn’t try to force a version control system on the user so it doesn’t really make sense to bake VCS related tools in directly. You can manage changes in KiCAD projects with git but as [jean-noël] notes reading Git’s textual description of changed X/Y coordinates and paths to library files is much more useful for a computer than for a human. It basically sucks to use. What you really need is a diff tool that can show the user what changed between two versions instead of describe it. And that’s what plotgitsch provides.

plotgitsch’s core function is to generate images of a KiCAD project at arbitrary Git revisions. After that there are two ways to view the output. One is to generate images of each version which can be fed into a generic visual diff tool (UNIX philosophy anyone?). The documentation has an example script to help facilitate setting this up. The other way generates a color coded image in plotgitsch itself and opens it in the user’s viewer of choice. It may not be integrated into the EDA but we’ll take one click visual diffs any day!

Automagic Tool Makes KiCAD Schematic Symbols From PDFs

September 11, 2018 by 28 Comments

Last time we talked about a KiCAD tool it was to describe a way to make the zen-like task of manual assembly more convenient. But what about that most onerous of EE CAD tasks, part creation? Home makers probably don’t have access to expensive part library subscriptions or teams of people to create parts for them, so they are left to the tedium of creating them by hand. What if the dream tool existed that could read the darn PDF by itself and make a part? It turns out [Sébastien] made that tool and it’s called uConfig.

uConfig has a pretty simple premise. It scrapes manufacturer datasheets in PDF form, finds what it thinks are diagrams of parts with pin names, functions, etc, and emits the result as parts in a KiCAD library. To aid in the final conversion [Sébastien] added rules engine which consume his custom KiCAD Style Sheets which specify how to categorize pins. In the simple case the engine can string match or use regex to let you specify things like “all pins named VDD[A-C] should be power pins”. But it can also be used to move everything it thinks belongs to “GPIOB” and stick them on the bottom of the created symbol. We could imagine features like that would be of particular use breaking out gigantic parts like a 400 ball BeagleBone on a chip.

Thanks for the tip [arturo182]!

Better Than Original Pong Using Arduino

September 10, 2018 by 13 Comments

Games like Pong are legendary, not only in the sense that they are classic hours fun but also that they have a great potential for makers in stretching their learning legs. In an attempt at recreating the original paddle games like Pong and Tennis etc, [Grant Searle] has gone into the depths of emulating the AY-2-8500 chip using an Arduino.

For the uninitiated, the AY-3-8500 chip was the original game silicon that powered Ball & Paddle that could be played on the domestic television. Running at 2 MHz, it presented a 500 ns pixel width and operated to a maximum of 12 Volts. The equivalent of the AY-3-8500 is the TMS1965NLA manufactured by Texas Instruments for those who would be interested.

[Grant Searle] does a brilliant job of going into the details of the original chip as well as the PAL and NTSC versions of the device. This analysis will come in handy should anyone choose to make a better version. He talks about the intricacies of redrawing the screen for the static elements as well as the ball that bounces around the screen. The author presents details on ball traversal, resolution, 2K memory limit and its workarounds.

Then there are details on the sound and the breadboard version of the prototype that makes the whole write-up worth one’s time. If you don’t fancy the analog paddles and would rather use a wireless modern-day touch, check out Playing Pong with Micro:bits

Thanks [Keith O] for the tip.

Trebucheting Tennis Balls At 124 MPH

September 10, 2018 by 21 Comments

A trebuchet is one of the older machines of war. It’s basically a sling on a frame, with a weight that you can lift up high and which pulls the sling arm over on release. Making one opens up the doors to backyard mayhem, but optimizing one opens up the wonders of physics.

[Tom Stanton] covers just about everything you need to know about trebuchet building in his four-part video series. Indeed, he sums it up in video two: you’ve got some potential energy in the weight, and you want to transfer as much of that as possible to the ball. This implies that the optimal path for the weight would be straight down, but then there’s the axle in the way.  The rest, as they say, is mechanical engineering.

Video three was the most interesting for us. [Tom] already had some strange arm design that intends to get the weight partially around the axle, but he’s still getting low efficiencies, so he builds a trebuchet on wheels — the classic solution. Along the way, he takes a ton of measurements with Physlets Tracker, which does video analysis to extract physical measurements. That tip alone is worth the price of admission, but when the ball tops out at 124 mph, you gotta cheer.

In video four, [Tom] plays around with the weight of the projectile and discovers that he’s putting spin on his tennis ball, making it curve in flight. Who knew?

Anyway, all four videos are embedded below. You can probably skip video one if you already know what a trebuchet is, or aren’t interested in [Tom] learning that paying extra money for a good CNC mill bit is worth it. Video two and three are must-watch trebucheting.

We’re a sad to report that we couldn’t find any good trebuchet links on Hackaday to dish up. You’re going to have to settle for a decade-old catapult post or this sweet beer-pong-playing robotic arm. You can help. Submit your trebuchet tips.

Thanks [DC] for this one!

Continue reading “Trebucheting Tennis Balls At 124 MPH” →

A DIY Balcony Crane Lifts Groceries For The Lazy But Patient

September 9, 2018 by 39 Comments

If necessity is the mother of invention, then laziness is probably its father. Or at least a close uncle. Who hasn’t thought, “There has to be a better way to do this, one that doesn’t involve me burning precious calories”?

Motivational laziness seems to increase with potential energy, as anyone who needs to haul groceries up four flights of stairs will tell you. This appears to be where this balcony-mounted drill-powered crane came from. Starting with a surplus right-angle gearbox and drum, [geniusz K] fabricated the rest of the crane from steel plate and tubing. We like the quality of fabrication and the tip on making slip couplings from bits of square tubing. The finished product got a nice coat of brown paint to match the balcony railing; keeping the neighbors happy is always important. He tested the crane with a 20-kg weight before installing it on the balcony and put it to work hauling groceries up three stories. Check out the build and the test in the video below.

While it won’t set any speed records, at least the drill is doing the work. But what if you’re impatient as well as lazy? Aside from being two-thirds of the way to programming greatness, you may have to up the game. A heavy-lift quadcopter, perhaps?

Continue reading “A DIY Balcony Crane Lifts Groceries For The Lazy But Patient” →

The E-Waste Apocalypse Looms

September 2, 2018 by 13 Comments

What does post-apocalyptic technology look like? Well, that kind of depends on the apocalypse. Regardless of the cause, we’ll need to be clever and resourceful and re-learn ancient crafts like weaving and pottery-making. After all, the only real apocalyptic constants are the needs of the survivors. Humans need clothing and other textiles. Fortunately, weaving doesn’t require electricity—just simple mechanics, patience, and craftsmanship.

If it turns out the apocalypse is scheduled for tomorrow, we’ll have piles and piles of e-waste as fodder for new-old looms. This adorable loom is a mashup of old and new technologies that [Kati Hyyppä] built at an artist residency in Latvia, a country with a rich historical tapestry of textile-making. It combines a cheerful orange telephone with an old cassette player and some telescoping rods from a radio antenna. [Kati] reused the phone’s hang-up switch to trigger tunes from a deconstructed toddler toy every time the receiver is lifted. Check it out after the beep break.

And yeah, you’re right, it does use batteries. But the looming part doesn’t require power, only the music. In case of apocalypse, just scrounge up a solar panel.

If you’d rather be prepared to have to make your own clothes someday, print this loom beforehand.

Continue reading “The E-Waste Apocalypse Looms” →