Tag Archives: hacking

Plant Biology is a Gateway

Too many college students have been subject to teachers’ aids who think they are too clever to be stuck teaching mere underclassmen. For that reason, [The Thought Emporium] is important because he approaches learning with gusto and is always ready to learn something new himself and teach anyone who wants to learn. When he released a video about staining and observing plant samples, he avoided the biggest pitfalls often seen in college or high school labs. Instead of calling out the steps by rote, he walks us through them with useful camera angles and close-ups. Rather than just pointing at a bottle and saying, “the blue one,” he tells us what is inside and why it is essential. Instead of telling us precisely what we need to see to get a passing grade, he lets our minds wonder about what we might see and shows us examples that make the experiment seem exciting. The video can also be seen below the break.

The process of staining can be found in a biology textbook, and some people learn best by reading, but we haven’t read a manual that makes a rudimentary lab seem like the wardrobe to Narnia, so he gets credit for that. Admittedly, you have to handle a wicked sharp razor, and the chance of failure is never zero. In fact, he will tell you, the opportunities to fail are everywhere. The road to science isn’t freshly paved, it needs pavers.

If a biology lab isn’t in your personal budget, a hackerspace may have one or need one. If you are wondering where you’ve heard [The Thought Emporium]’s voice before, it is because he is fighting lactose intolerance like a hacker.

from Blog – Hackaday https://ift.tt/2IHu0jm

SpaceX’s Next Giant Leap: Second Stage Recovery

With the successful launch of the Bangabandhu-1 satellite on May 11th, the final version of the Falcon 9 rocket has finally become operational. Referred to as the “Block 5”, this version of the rocket is geared specifically towards reuse. The lessons learned from the recovery and reflight of earlier builds of the F9 have culminated into rocket that SpaceX hopes can go from recovery to its next flight in as few as 24 hours. If any rocket will make good on the dream of spaceflight becoming as routine as air travel, it’s going to be the Falcon 9 Block 5.

While there might still be minor tweaks and improvements made to Block 5 over the coming years, it’s safe to say that first stage recovery of the Falcon 9 has been all but perfected. What was once the fodder of campy science fiction, rockets propulsively lowering themselves down from the sky and coming to rest on spindly landing legs that popped out of the sides, is now a reality. More importantly, not only is SpaceX able to bring the towering first stage back from space reliably, they’re able to refuel it, inspect it, and send it back up without having to build a new one for each mission.

But as incredible a technical accomplishment as this is, SpaceX still isn’t recovering the entire Falcon 9 rocket. At best, they have accomplished the same type of partial reusability that the Space Shuttle demonstrated on its first flight all the way back in 1981. Granted they are doing it much faster and cheaper than it was done on the Shuttle, but it still goes against the classic airplane analogy: if you had to replace a huge chunk of the airliner every time it landed, commercial air travel would be completely impractical.

SpaceX has already started experimenting with recovering and reusing the payload fairings of the Falcon 9, and while they haven’t pulled it off yet, they’ll probably get there. That leaves only one piece of the Falcon 9 unaccounted for: the second stage. Bringing the second stage back to Earth in one piece might well be the most challenging aspect of developing the Falcon 9. But if SpaceX can do it, then they’ll have truly developed humanity’s first fully reusable rocket, capable of delivering payloads to space for little more than the cost of fuel.

Different Stages, Different Challenges

While the first stage is there to get the payload up, it could be said that the second stage is responsible for moving the payload sideways. The second stage absolutely pours on the speed: on the most recent launch it accelerated the payload from 8,019 km/h at stage separation to the 26,967 km/h required to maintain low Earth orbit in just a few minutes. Once the payload separates and continues on with its mission, the second stage is for all intents and purposes its own spacecraft moving at orbital speed and altitude.

Bringing it down to a gentle landing on Earth therefore has all the same challenges of landing any other spacecraft, except for the fact that the second stage has none of the hardware that would traditionally be necessary to pull off such a feat. It’s a bit like trying to land an airplane without landing gear. Or wings.

In early concept videos from SpaceX, the second stage was shown outfitted with a heat shield, landing legs, and even a retractable engine nozzle. All of these features would have worked together to make the stage capable of the same autonomous propulsive landings the first stage performs. But the problem with this “super” second stage is weight.

Every kilogram of recovery gear added to the second stage is one less kilogram of payload delivered to space. For a commercial launch provider like SpaceX, that is a problem. Fortunately, the Falcon 9 tends to be underutilized by most payloads, so there’s some wiggle room to play with. For example, the Bangabandhu-1 satellite weighed approximately 3,700 kg, which is less than half the Falcon’s capability for that particular launch profile: geostationary transfer orbit. So if the recovery hardware can be limited to less than 1,000 kg or so, it shouldn’t have an impact on the kinds of payloads the Falcon is likely to encounter.

An Unexpected Solution

Weight really adds up when building spacecraft. Consider that the landing legs on the Falcon 9 first stage weigh around 2,000 kg on their own. Any attempt at recovering the second stage needs to be done with the absolute minimum of additional hardware. A full heat shield like the Dragon capsule has would likely eat up too much of that mass budget, same with the “grid fins” used to stabilize the first stage as it falls back down to Earth.

So how do get the second stage through the atmosphere and stabilize it? The first hint at the answer comes from a recent Tweet by Elon Musk:

Party balloons and a bouncy house? If anyone else said something like that, we’d just assume it was kind of joke. But we thought it was a joke when he Tweeted about sending his Tesla Roadster to space, and we all know how that turned out. So what does it all mean?

Meet the Ballute

The idea of using an inflatable balloon to slow down high altitude supersonic vehicles was pioneered in 1958 by Goodyear. NASA demonstrated that the so-called ballute (as it’s both a balloon and parachute) concept could be used for spacecraft reentry when a small one was used to safely decelerate a test article from Mach 4.2 in 1968. Unfortunately tests with larger ballutes failed, and ultimately the concept was never used for the space program.

From Elon’s Tweets, it looks like SpaceX is looking to revisit the ballute concept, using it to ease the Falcon 9’s second stage journey through the atmosphere. After performing a de-orbit burn, the second stage could deploy a ballute to help slow and stabilize it as it comes back down from space. But that’s only half the problem. You still have to get it on the ground without damaging it.

Catch a Falling Star Stage

With the second stage at a low enough altitude and speed thanks to the ballute, it could then deploy either a traditional parachute of parafoil to make the final approach towards the recovery point. As the second stage would likely not have any landing gear or legs due to weight constraints, the landing area would apparently be an inflatable structure of some type that can catch the stage without damaging it. In principle this is very similar to the work currently being done to catch the Falcon 9’s fairings with a large ship-mounted net.

Again, this would borrow heavily from earlier NASA research. In 1963, experiments were performed to determine if the first stage of the Saturn rocket could be recovered using an inflatable wing structure.

Iterative Design

At the risk of trivializing the accomplishments of SpaceX, it’s fair to say that very little of their technology is actually new. Rather, they combine Silicon Valley style R&D and modern construction techniques with technology pioneered during the Space Race of the 1960’s to rapidly produce evolutionary improvements. This allowed them to get to orbit in a fraction of the time it would have taken had they started completely from scratch, and now it seems they’ll be turning their attention towards iterating through NASA recovery concepts from the Gemini and Apollo programs to help turn Falcon 9 into the world’s first truly reusable rocket.

That said, it wouldn’t be the first time SpaceX abruptly changed their approach. The final method for second stage recovery could be vastly different from what Elon has been hinting at. It’s also possible that they abandon it entirely. Even with only partial reuse of the Falcon 9, they’re by far the cheapest game in town.

The bottom line is, we just don’t know yet. It’s interesting to theorize like this, but until we watch a live YouTube stream of a Falcon 9 second stage riding down from space under a balloon, anything’s possible. One thing’s for sure though, no matter what their plans are, they’ve got the world’s attention.

from Blog – Hackaday https://ift.tt/2s5xa5n

Blowing Rings With Cannons, Fogs, And Lasers

In today’s healthy lifestyle oriented world, blowing smoke rings won’t impress too many people anymore. Unless of course you are [NightHawkInLight] and blow them with a vortex cannon and add lasers for visual effects. Although, his initial motivation was to build a device that could shoot lost frisbees out off the trees in his backyard disc golf course, and as avid enthusiast of shooting things through the air using a propane torch, he opted for a vortex cannon to avoid the risk of injuries shooting a projectile may cause.

With safety in mind from the beginning, [NightHawkInLight] chose to build the cannon in ways that won’t expose him or people following his footsteps to any toxic fumes. The barrel is formed by securing a roll of terrace board and simply pulling it into a cone. A series of PVC pipes and adapters build the combustion chamber that fits the terrace board barrel on its one end, and the propane torch nozzle on its other end. For easier aim and stability, he also adds a tripod mount.

Since air vortices are, well, air, and therefore not visible by themselves, they don’t offer the most visual excitement. [NightHawkInLight] solved this with a fog machine attached to the barrel, and a laser line module, which you can see for yourself in his build video after the break. In a previous vortex cannon project we could also see a more outdoorsy approach to add visibility to it.

from Blog – Hackaday https://ift.tt/2×52041

Automatic I2C Address Allocation For Daisy-Chained Sensors

Many readers will be familiar with interfacing I2C peripherals. A serial line joins a string of individual I2C devices, and each of the devices has its own address on that line. In most cases when connecting a single device or multiple different ones there is no problem in ensuring that they have different addresses.

What happens though when multiple identical devices share an I2C bus? This was the problem facing [Sam Evans] at Mindtribe, and his solution is both elegant and simple. The temperature sensors he was using across multiple identical boards have three pins upon which can be set a binary address, and his challenge was to differentiate between them without the manufacturing overhead of a set of DIP switches, jumpers, or individual pull-up resistors. Through a clever combination of sense lines between the boards he was able to create a system in which the address would be set depending upon whether the board had a neighbour on one side, the other, or both. A particularly clever hack allows two side-by-side boards that have two neighbours to alternate their least significant bit, allowing four identical boards each with two sensors to be daisy-chained for a total of eight sensors with automatic address allocation.

We aren’t told what the product was in this case, however it’s irrelevant. This is a hardware hack in its purest sense, one of those which readers will take note of and remember when it is their turn to deal with a well-populated I2C bus. Of course, if this method doesn’t appeal, you can always try an LTC4316.

from Blog – Hackaday https://ift.tt/2x56oji

This Computer Is As Quiet As The Mouse

[Tim aka tp69] built a completely silent desktop computer. It can’t be heard – at all. The average desktop will have several fans whirring inside – cooling the CPU, GPU, SMPS, and probably one more for enclosure circulation – all of which end up making quite a racket, decibel wise. Liquid cooling might help make it quieter, but the pump would still be a source of noise. To completely eliminate noise, you have to get rid of all the rotating / moving parts and use passive cooling.

[Tim]’s computer is built from standard, off-the-shelf parts but what’s interesting for us is the detailed build log. Knowing what goes inside such a build, the decisions required while choosing the parts and the various gotchas that you need to be aware of, all make it an engaging read.

It all starts with a cubic aluminum chassis designed to hold a mini-ITX motherboard. The top and side walls are essentially huge extruded heat sinks designed to efficiently carry heat away from inside the case. The heat is extracted and channeled away to the side panels via heat sinks embedded with sealed copper tubing filled with coolant fluid. Every part, from the motherboard onwards, needs to be selected to fit within the mechanical and thermal constraints of the enclosure. Using an upgrade kit available as an enclosure accessory allows [Tim] to use CPUs rated for a power dissipation of almost 100 W. This not only lets him narrow down his choice of motherboards, but also provides enough overhead for future upgrades. The GPU gets a similar heat extractor kit in exchange for the fan cooling assembly. A fanless power supply, selected for its power capacity as well as high-efficiency even under low loads, keeps the computer humming quietly, figuratively.

Once the computer was up and running, he spent some time analysing the thermal profile of his system to check if it was really worth all the effort. The numbers and charts look very promising. At 100% load, the AMD Ryzen 5 1600 CPU levelled off at 60 ºC (40 ºC above ambient) without any performance effect. And the outer enclosure temperature was 42 ºC — warm, but not dangerous. Of course, performance hinges around “ambient temperature”, so you have to start getting careful when that goes up.

Getting such silence comes at a price – some may consider it quite steep. [Tim] spent about A$3000 building this whole system, thanks in part due to high GPU prices because of demand from bitcoin mining. But cost is a relative measure. He’s spent less on this system compared to several of his earlier projects and it let’s him enjoy the sounds of nature instead of whiny cooling fans. Some would suggest a pair of ear buds would have been a super cheap solution, but he wanted a quiet computer, not something to cancel out every other sound in his surroundings.

from Blog – Hackaday https://ift.tt/2IXEpqr

Machine Learning Crash Course From Google

We’ve been talking a lot about machine learning lately. People are using it for speech generation and recognition, computer vision, and even classifying radio signals. If you’ve yet to climb the learning curve, you might be interested in a new free class from Google using TensorFlow.

Of course, we’ve covered tutorials for TensorFlow before, but this is structured as a 15 hour class with 25 lessons and 40 exercises. Of course, it is also from the horse’s mouth, so to speak. Google says the class will answer questions like:

  • How does machine learning differ from traditional programming?
  • What is loss, and how do I measure it?
  • How does gradient descent work?
  • How do I determine whether my model is effective?
  • How do I represent my data so that a program can learn from it?
  • How do I build a deep neural network?

Google says you should be adept at intro level algebra and that higher math could be helpful, although not essential. You should also know something about programming with some familiarity in Python. The exercises run in your browser, so you don’t need any exotic set up. There are also a few tools that have suggested tutorials if you aren’t up to speed on them already. For example, the pandas library and bash are included in that list.

If you get really serious, Google have a lot of educational resources that will take you further. If you learn better by example, you might try getting naked. If you want an even longer class, try this one.

from Blog – Hackaday https://ift.tt/2IIjmJq

Spared No Expense: Cloning The Jurassic Park Explorer

While you’d be hard pressed to find any serious figures on such things, we’d wager there’s never been a vehicle from a TV show or movie that has been duplicated by fans more than the Staff Jeeps from Jurassic Park. Which is no great surprise: not only do they look cool, but it’s a relatively easy build. A decent paint job and some stickers will turn a stock Wrangler into a “JP Jeep” that John Hammond himself would be proud of.

While no less iconic, there are far fewer DIY builds of the highly customized Ford Explorer “Tour Vehicles”. As a rather large stretch of the film takes place within them, the interiors were much more detailed and bears little resemblance to the stock Explorer. Building a truly screen accurate Jurassic Park Tour Vehicle was considered so difficult that nobody has pulled it off since the movie came out in 1993. That is until [Brock Afentul] of PropCulture decided to take on the challenge.

In an epic journey spanning five years, [Brock] has created what he believes is the most accurate Jurassic Park Tour Vehicle ever produced; and looking at the side by side shots he’s done comparing his Explorer to the ones from the movie, it’s hard to disagree. A massive amount of work went into the interior, leaving essentially nothing untouched. While previous builds have tried to modify the stock dashboard to look like the one from the movie, he built a completely new dash from MDF and foam and coated it in fiberglass. The center console featuring the large display was also faithfully reproduced from the movie, and runs screen accurate animations, maps, and tour information. The seats also had to be replaced, multiple times in fact, as he had a considerable amount of trouble getting somebody to upholster them to his standards.

But perhaps the most difficult component of all was the clear acrylic roof bubble. These were critical to filming the movie, as they not only let the viewer see down into the Tour Vehicles but also let the characters see out during the iconic tyrannosaurus attack. But because the roof bubble was created only for the movie and never existed as a real aftermarket product, it usually gets ignored in Tour Vehicle builds. It’s simply too difficult to produce for most people. The omission of the bubble was always considered a case of artistic license; in the same way nobody expects a replica DeLorean from Back to the Future to actually fly or travel through time.

But [Brock] wanted to take his Tour Vehicle all the way, so he partnered up with a local glass shop that let him rent time in their oven so he could heat up acrylic sheets. Once heated to the appropriate temperature, they could be removed and wrapped around a mold to make the bubble. The process took weeks to perfect, but in the end he and a few friends got the hang of it and were able to produce a gorgeous roof bubble that they fitted to the already very impressive Explorer.

While previous Jurassic Park Tour Vehicle replicas were unquestionably awesome, this build really does take it to the next level. Short of equipping the garage with a movie-accurate super computer, it’s hard to see how the bar can get any higher.

from Blog – Hackaday https://ift.tt/2s3ibsE