Dealing with all the wiring can quickly become a challenge on robots, especially the walking variety which have actuators everywhere. [Eric Yufeng Wu] sidestepped the wiring issue by creating Q8bot, a little quadruped where all the components, including the actuators, are mounted directly on the PCB.

[Eric] uses a custom PCB as the spine of the robot, and the eight servos plug directly into connectors on the PCB. With their bottom covers removed, the servos screw neatly into a pair of 3D printed frames on either side of the PCB, which also have integrated 14500 battery holders. The PCB is minimalist, with just the XIAO ESP32C3 module, a boost converter circuit to drive the servos, and a battery fuel gauge. Each SCARA-style leg consists of four SLS 3D printed segments, with press-fit bearings in the joints.

The little one moves quickly, and can even do little jumps. For this prototype, most of the control processing is done on a laptop, which sends raw joint angles to the onboard ESP32 via the ESP-Now protocol. We think this little robot has a lot of development potential, and fortunately [Eric] has made all the hardware and software files available for others to build their own.

The Sun has been remarkably active lately, so much so that it might have set a new sunspot record. According to the sun watchers at the Space Weather Prediction Center, on August 8, the Solar Dynamics Observatory snapped a picture that was positively bedazzled with sunspots. Counting methods vary, but one count put the sunspot number at a whopping 337 that day. That would be the largest number since 2001, during the peak of Solar Cycle 23. The sunspot number is highly correlated with solar storms and coronal mass ejections; more spots mean more magnetic activity and more chance for something to go very, very wrong. We’ve been pretty lucky so far with Solar Cycle 25; despite being much more active than the relatively lazy Cycle 24 and much stronger than predicted, most of this cycle’s outbursts have been directed away from Earth or only dealt us a glancing blow. Seeing all those spots, though, makes us think it’s only a matter of time before we get hit with something that does more than make pretty lights.

Having done our share of roofing, we can safely say it’s a pretty tough job. Everything is heavy, it’s either boiling hot or freezing cold, and one moment’s inattention can make for a very bad day. Plus, the fiberglass shards in your skin at the end of the day can be incredibly annoying. On the other hand, a good roofing job is a thing of beauty, and there’s immense satisfaction in having been the one to do it. But, with apologies to Steve Miller, time keeps on slipping into the future, humans are expensive and unreliable, and someone will eventually try to automate humans out of pretty much every job.

It’s roofing’s turn now with the aptly named Rufus auto roof robot. It’s a hybrid robot with a base unit containing a hopper for shingles and a SCARA arm that positions, aligns, and nails down the shingles. The base unit, in turn, is moved around the surface of the roof by a capstan-drive cable robot, with cables anchored to the corners of the roof.

It’s an interesting idea, but one that’s going to take some development to make it practical. For one thing, we can see safety regulators having a fit over those cables, which will be a tripping hazard for the workers who have to keep the bot fed with shingles and nails. Also, a roof that’s completely free of obstructions like vent stacks, skylights, or chimneys is a rare roof indeed, and it seems like the cable system would foul on such obstructions very easily. Still, you’ve got to start somewhere, and it’ll be interesting to see how this develops. Or maybe we’ll just throw Atlas at it.

Head up, hams — another spectrum land grab appears to be underway. This one is a little hard to follow, but what we see is that a company called NexNav, which is currently licensed for a Location and Monitoring Service in the 900-MHz band, wants to split the 902 MHz to 928 MHz band and start blasting out high-powered signals on the upper part of the band, apparently for a new 5G position system that will serve as a backup to satellite location systems like GPS. The problem is that amateur radio operators have a secondary allocation in that band, which, despite the company’s unsupported claims to the contrary, will most likely be swamped by their 2,000-watt effective radiated power signals. LoRaWAN fans might also take note of the proposed change, which would likely make life difficult for them and other ISM-band users. It might be time to write some strongly worded letters.

And finally, what a time to be alive! While Boeing has figured out how to turn the ISS into a low-earth orbit Gilligan’s Island by unintentionally extending an eight-day visit into an open-ended stay, down here on Earth, we’re tackling the real problems, like how to safely eat Doritos in space. Luckily, the food scientists at Frito-Lay put their top people on the problem and came up with “Cool Ranch Zero Gravity Doritos,” which substitute a flavored oil spray for the normal finger-staining powdered spice blend that would get everywhere in an environment where gravity doesn’t pull it down onto your shirt or into your neck-beard. And to keep cornmeal crumbs from getting loose, they shrunk the triangular chips down to about a third the size of a regular Dorito, so you can just stuff the whole chip in your mouth without biting it first.

We have many thoughts on this, primarily that being unable to stuff at least three regular-size Doritos in your mouth at one time should be grounds for disqualification from spaceflight and that they literally could have chosen any flavor to send to space, but they had to make it Cool Ranch, which raises many questions of its own. But mainly, we’re just sad that this is what has become of spaceflight — and yes, we know about Tang, but this seems a lot worse.

If you’re unfamiliar with SCARA robots, the acronym stands for Selective Compliance Assembly Robot Arm. This refers to the fact that the arms are rigid in the Z axis but somewhat compliant in the X and Y axes, and that they’re often used for assembly tasks. In any case, you can spend a great deal of money equipping your factory with these robots, or you can build your own for the fun of it. If you’re not endowed with a seven-figure investment for opening a production plant, consider exploring [tuenhidiy’s] project instead.

The build enlists an Arduino Mega as the brains of the operation. It’s paired with a RAMPS controller for running a pair of NEMA 17 stepper motors that actually move the arm in the X-Y plane. Additionally, a tray eject mechanism from a CD/DVD drive is enlisted to act as the Z axis. The frame is assembled from PVC plumbing components and a small amount of aluminium T-slot profile.

The resulting arm isn’t fast in the video we see of the build, but it works as a basic plotter without too much complaint. The benefit of the Z-axis in this case is obvious, as it allows the pen to be lifted off the page where necessary.

We’ve seen plenty of good plotter designs around these parts before, too. Video after the break.

One of the limitations of the conventional Cartesian CNC platforms is that the working area will usually be smaller than its footprint. SCARA arms are one of the options to get around this, as demonstrated by [How To Mechatronics], with his SCARA laser engraver.

This robot arm is modified from the original build we featured a while back, which had a gripper mounted. It uses mainly standard 3D printer components with 3D printed frame parts. The arms lengths are sized to fold over the base and take up little table horizontal space when not in use. It can work in a large semi-circular area around itself, and if a proper locating and homing method is implemented, it can be moved around and engrave a large area section by section.

One of the challenges of SCARA arms is rigidity. As the cantilevered arm extends, it tends to lean over under its weight. In [How To Mechatronics]’s case, it showed up as skewed engravings, which he managed to mitigate to some degree in the Marlin firmware.

Another possible solution is to reduce the weight of the arms by moving the motors to the base, as was done with the Pybot or dual-arm SCARA printers like the RepRap Morgan.

We’ve all seen videos of blisteringly fast SCARA arms working on assembly lines, and more than a few of us have fantasied about having that same kind of technology for the home shop. Unfortunately, while the prices for things like 3D printers and oscilloscopes have dropped lower than what many would have believed possible a decade ago, high-performance robotics are still too pricey for the home player.

Unless of course, you’re willing to build it yourself. The PyBot designed by [jjRobots] is an open source robotic arm that should be well within the means of the average hardware hacker. One could argue that this is a project made entirely possible by desktop 3D printing; as not only are most of the structural components printed, but most of the mechanical elements are common 3D printer parts. Smooth rods, linear bearings, lead screws, and NEMA 17 motors are all exceptionally cheap these days thanks to the innumerable 3D printer kits that make use of them.

Those who’ve researched similar projects might notice that the design of this arm has clearly been influenced by the Mostly Printed SCARA (MPSCARA). But while that robot was designed to carry an extruder and act as a 3D printer, [jjRobots] intends for the PyBot to be more of a general purpose platform. By default it features a simple gripper, but that can easily be changed out for whatever tool or gadget you have in mind.

In the base of the arm is a custom control board that combines an Arduino M0, an ESP8266, and a trio of stepper motor drivers. But if you wanted to build your own version from the parts bin, you could certainly wire up all the principle components manually. As the name implies, the PyBot is controlled by Python tools running on the computer, so it should be relatively easy to get this capable arm to do your bidding.

We’ve seen some impressive 3D printed robotic arms over the years, but the simplicity of the PyBot is particularly compelling. This looks like something that you could reasonably assemble and program over a weekend or two, and then put to work in your ad-hoc PPE factory.

One of the side effects of the rise of 3D printers has been the increased availability and low cost of 3D printer components, which are use fill for range of applications. [How To Mechatronics] capitalized on this and built a SCARA robot arm using 3D-printed parts and common 3D-printer components.

The basic SCARA mechanism is a two-link arm, similar to a human arm. The end of the second joint can move through the XY-plane by rotating at the base and elbow of the mechanism. [How To Mechatronics] added Z-motion by moving the base of the first arm on four vertical linear rods with a lead screw. A combination of thrust bearings and ball bearings allow for smooth rotation of each of the joints, which are belt-driven with NEMA17 stepper motors. Each joint has a microswitch at a certain position in its rotation to give it a home position. The jaws of the gripper slide on two parallel linear rods, and are actuated with a servo. For controlling the motors, an Arduino Uno and CNC stepper shield was used.

The arm is operated from a computer with a GUI written in Processing, which sends instructions to the Arduino over serial. The GUI allows for both direct forward kinematic control of the joints, and inverse kinematic control,  which will automatically move the gripper to a specified coordinate. The GUI can also save positions, and then string them together to do complete tasks autonomously.

The base joint is a bit wobbly due to the weight of the rest of the arm, but this could be fixed by using a frame to support it at the top as well. We really like the fact that commonly available components were used, and the link in the first paragraph has detailed instructions and source files for building your own. If the remaining backlash can be solved, it could be a decent light duty CNC platform, especially with the small footprint and large travel area. This is very similar to a wooden SCARA robots we’ve seen before, except that one put the Z-axis at the gripper. We’ve also seen a few 3D printers and pen plotters that used this layout.

With the roughly 20-day wide launch window for the Mars 2020 mission rapidly approaching, the hype train for the next big mission to the Red Planet is really building up steam. And with good reason — the Mars 2020 mission has been in the works for a better part of a decade, and as we reported earlier this year, the rover it’s delivering to the Martian surface, since dubbed Perseverance, will be among the most complex such devices ever fielded.

“Percy” — come on, that nickname’s a natural — is a mobile laboratory, capable of exploring the Martian surface in search of evidence that life ever found a way there, and to do the groundwork needed if we’re ever to go there ourselves. The nuclear-powered rover bristles with scientific instruments, and assuming it survives the “Seven Minutes of Terror” as well as its fraternal twin Curiosity did in 2012, we should start seeing some amazing results come back.

No prior mission to Mars has been better equipped to answer the essential question: “Are we alone?” But no matter how capable Perseverance is, there’s a limit to how much science can be packed into something that costs millions of dollars a kilogram to get to Mars. And so NASA decided to equip Perseverance with the ability to not only collect geological samples, but to package them up and deposit them on the surface of the planet to await a future mission that will pick them up for a return trip to Earth for further study. It’s bold and forward-thinking, and it’s unlike anything that’s ever been tried before. In a lot of ways, Perseverance’s sample handling system is the rover’s raison d’être, and it’s the subject of this deep dive.

Three Robots in One

NASA has done its usual admirable job of communicating with the public about the Mars 2020 mission, and part of the outreach includes this recent video that shows off a little of the engineering that went into the sample handling system. Honestly, though, for as much tech eye candy as that video had, it only served to whet my appetite. There was so much going on that I had to find out more.

To get a bit of the inside story, I turned to Kelly Palm, one of the JPL engineers seen in the video below. As the Integration and Test Lead for the Sample Caching System (SCS), she’s pretty busy these days, but she graciously fielded my questions and helped give me an idea of what went into building and testing such a complex piece of equipment.

First of all, the SCS is really not just one but three separate robots, each with a specific set of jobs. The “business end” of the SCS is the 2-meter-long robot arm mounted to the front of the vehicle. Like Curiosity before it, the arm carries a turret that’s laden with scientific instruments, sensors, and cameras, as well as the tools necessary for boring into Martian rocks and taking samples. But unlike its predecessor, where the rock drill was designed to abrade rocks and produce a powder that could be easily analyzed by onboard instruments, the Perseverance drill is specialized for obtaining core samples, suitable for both on-board study and in terrestrial labs once the samples are returned.

The drill in the robot arm’s turret is a pretty versatile tool. With the help of the bit carousel (more about which is below) the drill can attach bits designed for different jobs. The drill is capable of running in either a simple rotary mode or in a percussive mode, similar to a hammer drill. A small onboard tank of purified nitrogen is used to gently remove dust generated by coring operations.

Coring into rock to a limited depth using a cylindrical bit raises a question: how exactly is the core recovered? On Earth, the answer would be to use a second tool to pry at the cylinder of rock left behind after the coring bit is removed. While something like that could certainly work on Mars too, especially with a robotic arm at your disposal, NASA came up with a far more clever system.

According to design tests run by a company called Honeybee Robotics in 2014, liberating the core from the parent rock and enclosing it the sample tube in which it will live until being reopened in a lab on Earth is a one-step process. Nestled inside the coring bit is a titanium sample tube. During coring, the axis of the sample tube and the coring bit are aligned with each other, so that the tube slips over the rock core as drilling proceeds. At the proper depth, the sample tube is rotated slightly off-axis, exerting enough force on the base of the core sample to break it off from the parent rock. The core is retained by a lip on the inside of the coring bit, allowing it to be removed from the hole, already within the titanium sample tube in which it will remain until the sample return mission.

Sealed with a Ram

The bit carousel is the next robot in the sample caching process. Sitting at the front of the rover chassis, the bit carousel is outwardly simple — just a rotating turret that transports bits to and from storage in the belly of Perseverance. But what it lacks in complexity is more than made up for by its clever design. The body of the carousel is a wheel with stations around the edge. Each station is at a 45° angle relative to the rotor’s axis, which itself is oriented 45° to the long axis of the chassis. The combination of angles means that a tube can transition from vertical to horizontal just by rotating the carousel with a single motor. There are plenty of sensors and actuators that ensure everything is lined up, of course, but the simplicity of the design is really something.

The ability to transfer tools and samples between horizontal and vertical orientations is critical to the sample caching mission, since the robot that takes care of storing everything lives inside the forward section of the rover’s chassis. The Sample Handling Arm, or SHA, looks a little like the SCARA (selective compliance articulated robot arm) robots that are prevalent in semiconductor fabs. The SHA is capable of accessing multiple locations inside the sample caching compartment and transferring between them and the bit carousel presentation area. To clear the instruments and sample tubes that take up most of the space in the bay, the SHA has an additional Z-axis so that the whole thing can drop below the bottom edge of the rover chassis. In addition to 42 storage silos for core and regolith sample tubes, the SHA can reach storage for a number of tools and attachments, plus instruments for doing some preliminary analysis of the samples, such as volume assessment and imaging.

Once a sample tube is filled, it needs to be hermetically sealed to ensure that the contents will survive for an indeterminate amount of time on the Martian surface as well as withstand the rigors of the eventual trip back to Earth. The seal has to be made without contaminating the sample, so no adhesives can be used, and no heat can be used either, lest the sample be subjected to extreme temperatures.

To seal a sample tube, the SHA brings it over to one of seven seal dispensers. A cup-shaped plug is dropped into the open end of the tube by a dispenser. The plugged tube is then moved to a sealing station, which uses a motor-driven ram to drive a tapered ferrule down a guide rod inside of the plug. As the ferrule is pressed downward, the rim of the plug expands, driving a sharp tooth on its outside circumference into the inner wall of the sample tube. The end result is essentially a cold-welded bond between the cap and the sample tube, hermetically sealing the tube and protecting the sample from contamination.

Return to Sender

Once a sample has been sealed in its titanium sarcophagus, it’s ready to be deposited onto the Martian surface. Most mission profiles that I could find refer to the use of “depot caching”, where Perseverance repeatedly returns to a single location from various regions of interest to deposit sample tubes. This makes perfect sense; finding a big pile of 42 titanium tubes is probably a far easier task for a future sample recovery mission than roaming about looking for individual tubes dropped where they were taken.

Still, whatever robot is sent to clean up after Perseverance has its work cut out for it; since the SHA cannot reach down to the surface, the tubes will have to be dropped, which means an orderly stack of sample tubes will likely not be what the recovery robot will find. Whatever follows in Perseverance’s tracks is going to need the agility to pick up and safely stow every single precious sample tube regardless of its orientation, possibly after digging it out of wind-blown regolith, and the intelligence to do it all autonomously.

With some luck, Perseverance will soon be on its way to Mars, and both when it launches and when it lands in February, we’ll be glued to our seats waiting for results. We’ll also be following the development of the return mission, which could prove to be even more challenging and require even cooler engineering to pull off.

Featured images: NASA/JPL-Caltech