Deep in the recesses of a few enterprising hackerspaces, you’ll find old electronic knitting machines modified for use with modern computers. They’re cool, and you can knit colorful designs, but all of these machines are ultimately based on old equipment, and you’ll have a hard time building one for yourself.
For their entry to the Hackaday Prize, [Mar] and [Varvara] is building a knitting machine from scratch. Not only is it a 3D printed knitting machine anyone can build given enough time and plastic, but this machine is a circular knitting machine, something no commercial offering has yet managed.
We saw [Mar] and [Varvara]’s Circular Knitic last January, but this project has quite the pedigree. They originally started on their quest for a modern knitting machine by giving a new brain to old Brother machines. This was an incredible advancement compared to earlier Brother knitting machine hacks; before, everyone was emulating a floppy drive on a computer to push data to the machine. The original Knitic build did away with the old electronics completely, replacing it with a homebrew Arduino shield.
While the Circular Knitic isn’t completely 3D printed, you can make one in just about any reasonably equipped shop. It’s a great example of a project that’s complex and can be replicated by just about anyone, and a perfect example of a project for The Hackaday Prize.
Check out the video of the Circular Knitic below.
Continue reading “Hackaday Prize Entry: Circular Knitting Machines”
Two of the most important prerequisites for successful 3D printing is making sure the bed is level and correctly setting the Z=0 height. Getting both of these right almost guarantees great adhesion since the first print layer is not only at the right distance from the build platform but also at a consistent distance for the entire bottom surface of the part.
Manual bed leveling is tedious, requiring the user to move the print nozzle to different points around the build platform, adjust some screws and make sure the nozzle is a piece-of-paper’s thickness higher than the platform. If you want to get complicated, there is an automatic option that probes the build platform and makes height corrections in the software. The probes come in several flavors, two common methods being a deployed mechanical switch (usually mounted to a servo) or force sensors under the build platform that sense when the nozzle touches the build platform. This method also requires some fancy firmware finesse to get working correctly.
[Jonas] posted a video showing the semi-automatic bed leveling capability of his printer. The build platform is held a bit high by springs that surround each of the 3 screws that support the bed assembly. The nozzle is moved directly over one of the 3 screws and then moved down until it noticeably presses on the build platform, compressing the support spring. A thumb wheel is then tightened at that location, locking the bed in place. The same process is performed for the other 2 support points. The result is a perfectly level build surface. Check the video out after the break to see just how quick this procedure is!
We’ve seen a somewhat similar concept that uses a clever gimbal and lock system under the bed.
Continue reading “Semi-Automatic Bed Leveling Your 3D Printer”
Last Friday we wrapped up a round of community voting for The Hackaday Prize. The theme? Most Likely To Save The Planet. Now it’s time for some results:
The projects voted Most Likely To Save The Planet by the Hackaday.io community are, in order:
Congratulations to everyone who has a project that was voted up to the top. Even though these rounds of community voting don’t decided which projects make it into the Hackaday Prize semifinals, you’ve earned the respect of your peers and a nifty Hackaday Prize t-shirt.
Since NIRGM – Non-Invasive NIR Glucose Meter won last week, we’re moving down the list to #11 and awarding SciPlo a Hackaday Prize t-shirt as well.
A New Round Of Voting!
This week, we’re asking the Hackaday.io community to vote for the most Amazingly Engineered project entered into the Hackaday Prize. To entice everyone to vote, I’m going to pick a random Hackaday.io user next Friday around 22:00 UTC. If that person has voted, they get a $1000 gift card for the Hackaday Store. If that person has not voted, I’ll be giving a few prizes away to people who have voted. Last week, we gave away a SmartMatrix, an Analog Stepper Gauge, and a Simon Says kit. We’ll probably change that up this week; I don’t know what it will be, but someone who votes will get something. Imagine; giving away stuff just for clicking a button. How magnanimous can we get?
As with every community voting update, it is requested that you vote. If you need a nudge to understand how this works, here’s a video tutorial on how to vote.
You’ve just got to go with the hype on this one, because it’s obviously not ready for prime time yet. But a few days ago murmurs started circling the net that an Australian inventor had developed a robot capable of building complicated structure from brick all by itself.
Before you go off your rocker… we’re definitely not calling this real. It’s a proof of concept at best, but that doesn’t prevent us from getting excited. How long have you been waiting for robots that can build entire structures on our behalf? We were excited at the prospect of extruding walls of concrete. But this is more like LEGO buildings in the real world. The beast cuts brick to length, conveys each brick along the telescoping arm, and butters them as it lays them in place. At least that’s what the rendered video after the break shows.
We’re hearing about this now because FastBrick Robotics, the company [Mark Pivac] founded and has spent ten years developing the Hadrian project at, was just sold to a company called DMY Capital Limited. Of course they’re going to want to get some press out of the sale.
There is an image of the brick feeder on an existing excavator that frankly looks photoshopped. And some real images like the one seen here and another of the “print head” holding some bricks. But it’s enough to think there’s potential here.
The idea is that the base of the robot is fixed with the arm long enough to reach any part of the structure being built. Precise positioning is achieved by a fixed marker in a different position from the robot. The head triangulates its position using laser range-finding with the marker (having said that we now assume there needs to be more than one marker).
So what do you think? Are we ever going to see this incredibly complicated bucket of awesome producing structures in our neighborhood which the Big Bad Wolf simply cannot blow down?
Continue reading “Hope it’s real: 3D Printing Houses with Bricks”
There are dozens of circuit board printers out there that lay down traces of conductive ink and ask you to glue down components to a fragile circuit board. This is a far cry from the old way of making PCBs, but these printers are going gangbusters, cashing in on the recent popularity of hardware startups and rapid prototyping.
People who think deeply about a problem are few and far between, but lucky for us [Arvid] is one of them. He’s come up with a way of creating PCBs with any 3D printer and steel rod. The results are better than anything you could make with a circuit board printer, and the technique is very, very cheap.
[Arvid] is using the traditional method of etching away copper, just some ferric chloride and a bit of time. How he’s masking the copper that shouldn’t be etched away is a unique process we haven’t seen before. He’s simply covering a piece of copper clad board with permanent marker, and scribing the parts he wants to be etched with a sharp steel rod attached to a 3D printer.
The G code for the printer was generated by FlatCAM, a piece of software made specifically for cutting PCBs with a mill. [Arvid]’s technique works so well that spindles and mills aren’t needed; only a sufficiently sharp instrument to scrape away permanent marker.
Thanks [Hassi] for sending this one in.
Hackaday loves to spread the message of the hardware hacking lifestyle. That’s only possible where there are hardware hackers willing to spend their time getting together to talk the future of the hardware industry, and to celebrate where we are now. We’re honored that you came out en masse for our Shenzhen Workshop and Meetup!
Zero to Product
[Matt Berggren] has presented his Zero to Product workshop a few times now as part of our Hackaday Prize Worldwide series. This spring that included Los Angeles, San Francisco, and ten days ago it was Shezhen, China.
[Matt] leading Zero to Product workshop
Participants designing a PCB
All levels of experience are welcome
We partnered with MakerCamp, a week-long initiative that pulled in people from all over China to build a Makerspace inside of a shipping container. Successful in their work, the program then hosted workshops. The one caveat, Shenzhen in June is a hot and sticky affair. Luckly our friends at Seeed Studio were kind enough to open their climate-controlled doors to us. The day-long workshop explored circuit board design, using Cadsoft Eagle as the EDA software to lay out a development board for the popular ESP8266 module.
Continue reading “The Spirit of Hackaday Shines in Shenzhen”
How can you generate random bits? Some people think it’s not easy, others will tell you that it’s pretty damn hard, and then there are those who wonder if it is possible at all. Of course, it is easy to create a very long pseudorandom sequence in software, but even the best PRNG (Pseudorandom Number Generator) needs a good random seed, as we don’t want to get the same sequence each time we switch on the unit, do we? That’s why we need a TRNG (True Random Number Generator), but that requires special hardware.
Some high-end microprocessors are equipped with an internal hardware TRNG, but it is, unfortunately, not true for most low-cost microcontrollers. There are a couple of tricks hackers use to compensate. They usually start the internal free running counter and fetch its contents when some external event occurs (user presses a button, or so). This works, but not without disadvantages. First, there is the danger of “locking” those two events, as a timer period may be some derivative of input scan routine timing. Second, the free running time (between switching on and the moment the unit requests a random number) is often too short, resulting in the seed being too close to the sequence start, and thus predictable. In some cases even, there is no external input before the unit needs a random seed!
Despite what has already been discussed, microcontrollers do have a source of true randomness inside them. While it might not be good enough for crypto applications, it still generates high enough entropy for amusement games, simulations, art gadgets, etc.
Continue reading “True Random Number Generator for a True Hacker”