Balanced Design And How To Know When To Quit Optimizing

March 13, 2021 by Elliot Williams 59 Comments

I got a relatively inexpensive 6040 CNC machine, and have been spending most weekends making the thing work, and then cutting stuff, learning the toolchain, and making subsequent improvements. Probably 90% of my machine time has been on making improvements. It’s not that the machine was bad — I got the version with ballscrews and a decently solid frame — but it’s that it somehow didn’t work together as a whole. It’s just an incredibly unbalanced design.

Let’s start with the spindle motor. It’s a 2.2 kW water-cooled beast that is capable of putting tons of work into a piece and spinning at very high speed. Yet to keep up with the high speed spindle, the motors that move it around would have to be capable of high speeds as well — it’s a feeds and speeds thing if you’re not a CNC geek. And they can’t. Instead, the stepper motors that came with the kit are designed for maximum force at low speeds. Which can make sense for some machines, but for one with a slightly flexible X-axis like this one, that’s wasted as well. The frame just can’t handle the low-end grunt that the motors are capable of, so it can’t take advantage of the spindle’s power either. The design is all over the place.

Over the last two months’ of weekends, I’ve been going through this iterative procedure of asking “what is my limiting factor right now?”, working on fixing that thing up, running it some, and then asking the question again. And it’s a good general procedure, and I believe that it’s getting me to the machine I want at the minimum cost of time, money, and effort.

At first, it was the driver hardware/software with its emulated USB parallel port, so I swapped out the controller for an Arduino running GRBL, soldered directly to the DB-25 that comes out of the back. At least it can put out pulses fast enough to order the motors around, but they would still stall out at high speeds. Swapping the stepper motors out for a high-speed pair only cost me €40, which makes you wonder why they didn’t just put the right motors on in the first place. The machine now travels fast enough to make use of the high-speed spindle, and I’m flying through plywood and plastics without leaving burn marks. It’s a huge win for not much money.

The final frontier is taking big bites out of aluminum. The spindle can do it, but I fear I’m up against the frame’s rigidity on the X-axis. For whatever reason, they went with unsupported rods on the X, which are significantly more flexible than an axis that’s backed up by more metal. And this is where the limiting factor may actually be my time and patience, rather than money. I just can’t bear to disassemble and reassemble the thing again. So for now, it’s going to be small nibbles, taking advantage of the machine’s speed, if not yet the spindle’s full horsepower.

But it’s odd, because this machine is a bundle of good parts. It’s just that they haven’t been chosen to work together optimally; the frame doesn’t work with the stepper motors, which don’t work with the spindle. If they went through my procedure of saying “what’s the limiting factor?” they could have saved themselves €100 by just shipping it with a wimpier spindle, which would have been a balanced, if anemic, machine. Or they could have built it with the right motors for more speed. Or supported rails for more grunt. Or both!

I’ll never know why they quit optimizing their design when they did. Maybe they never got past the slow USB/parallel port speed? But I’m near the end of my path, and I can tell because the limiting ingredient isn’t a simple upgrade, or even mere money anymore, but my own willpower.

How can you tell when you’re at the top of a mountain in a dense fog? A step you take in any direction would lead you downhill. How can you tell when you’re satisfied with a project’s state? When you don’t have the need, or desire, to undertake the next most obvious improvement.

Waterjet-Cut Precision Pastry

March 6, 2021 by Elliot Williams 20 Comments

We need more high-end, geometric pastry in our lives. This insight is courtesy of a fairly old video, embedded below, demonstrating an extremely clever 2D CNC mechanism that cuts out shapes on a cake pan, opening up a universe of arbitrary cake topologies.

The coolest thing about this machine for us is the drive mechanism. A huge circular gear is trapped between two toothed belts. When the two belts move together the entire thing translates, but when they move in opposite directions, it turns. It seems to be floating on a plastic platform, and because the design allows the water-jet cutting head to remain entirely fixed, only a small hole underneath is necessary, which doubtless simplifies high-pressure water delivery and collection. Rounding the machine out are cake pans make up of vertical slats, like on a laser- or plasma-cutter table, that slip into registration pins and let the water pass through.

The kinematics of this machine are a dream, or perhaps a nightmare. To cut a straight line, it does a cycloid-shaped dance of translation and turning that you simply have to see in motion. Because of this intricate path, the cake-feed speed varies along the way, so this machine won’t be perfect for all applications and relies on a thin kerf. And we can’t help thinking how dizzy the cake must get in the process.

Indeed, the same company put out a relatively pedestrian two-arm motion cutter (another video!) that poses different kinematic problems. It’s essentially a two-arm plotter with a moving table underneath that helps increase the working area. Details are scarce, but it looks like they’re minimizing motion of the moving table, doing the high frequency small stuff with the stiff arms. Presumably someone turned the speed on the previous machine up to 11 and spun a cake off into the room, causing them to rethink the whole move-the-cake-around design.

Of course, watercut pastry isn’t limited to exotic CNC mechanisms. This (third!) video demonstrates that a simple Cartesian XY bot can do the job as well.

If you think about it, using high-pressure pure water to cut foodstuffs is a win on many levels. We’d just miss out on licking the knife. Thanks [Adam G DeMuri] for the awesome comment that lead us to a new world of watercut edibles.

Continue reading “Waterjet-Cut Precision Pastry” →

3D Print A PCB The Hard Way

February 22, 2021 by Al Williams 15 Comments

There’s an old joke about the physics student tasked with finding the height of a building using a barometer. She dropped the barometer from the roof and timed how long it took to hit the ground. Maybe that was a similar inspiration to [Moe_fpv_team’s] response to the challenge: use a 3D printer to create a PC board. The answer in that case? Print a CNC mill.

[Moe] had some leftover 3D printer parts. A $40 ER11 spindle gets control from the 3D printer software as a fan. The X, Y, and Z axis is pretty standard. The machine can’t mill metal, but it does handy on plywood and fiber board and should be sufficient to mill out a PCB from some copper clad board.

Continue reading “3D Print A PCB The Hard Way” →

Hackaday Links: February 21, 2021

February 21, 2021 by Dan Maloney 6 Comments

Well, that was quite a show! The Perseverance rover arrived on Mars Thursday. Don’t tell the boss, but we spent the afternoon watching the coverage in the house on the big TV rather than slaving away in the office. It was worth it; for someone who grew up watching Jules Bergman and Frank Reynolds cover the Apollo program and the sometimes cheesy animations provided by NASA, the current coverage is pretty intense. A replay of the coverage is available – skip to about the 1:15:00 mark to avoid all the filler and fluff preceding the “Seven Minutes of Terror” main event. And not only did they safely deliver the package, but they absolutely nailed the landing. Perseverance is only about 2 km away from the ancient river delta it was sent to explore for signs of life. Nice shooting!

We’re also being treated to early images from Jezero crater. The first lowish-rez shots, from the fore and after hazard cameras, popped up just a few seconds after landing — the dust hadn’t even settled yet! Some wags complained about the image quality, apparently without thinking that the really good camera gear was stowed away and a couple of quick check images with engineering cameras would be a good idea while the rover still had contact with the Mars Reconnaissance Orbiter. Speaking of which, the HiRISE camera on the MRO managed to catch a stunning view of Perseverance’s descent under its parachute; the taking of that photo is an engineering feat all by itself. But all of this pales in comparison to a shot from one of the down-looking cameras in the descent stage, show Perseverance dangling from the skycrane just before touchdown. It was a really good day for engineering.

Would that our Earthly supply chains were as well-engineered as our Martian delivery systems. We’ve been hearing of issues all along the electronics supply chain, impacting a wide range of industries. Some of the problems are related to COVID-19, which has sickened workers staffing production and shipping lines. Some, though, like a fire at the AKM semiconductor plant in Japan, have introduced another pinch point in an already strained system. The fire was in October, but the impact on the manufacturer depending on the plant’s large-scale integration (LSI) and temperature-compensated crystal oscillators (TCXO) products is only just now being felt in the amateur radio market. The impact is likely not limited to that market, though — TCXOs pop up lots of gear, and the AKM plant made LSI chips for all kinds of applications.

What do you get when you combine a 3D-printer, a laser cutter, a CNC router, and a pick-and-place robot? Drones that fly right off the build plate, apparently. Aptly enough, it’s called LaserFactory, and it comes from MITs Computer Science and Artificial Intelligence Lab. By making different “bolt-on” tools for a laser cutter, the CSAIL team has combined multiple next-generation manufacturing methods in one platform. The video below shows a drone frame being laser-cut from acrylic, to which conductive silver paste is added by an extruder. A pick-and-place head puts components on the silver goo, solders everything together with a laser, and away it goes. They also show off ways of building up 3D structures, both by stacking up flat pieces of acrylic and by cutting and bending acrylic in situ. It’s obviously still just a proof of concept, but we really like the ideas presented here.

And finally, as proof that astronomers can both admit when they’re wrong and have fun while doing so, the most remote object in the Solar System has finally received a name. The object, a 400-km diameter object in a highly elliptical orbit that takes it from inside the orbit of Neptune to as far as 175 astronomical units (AU) from the Sun, is officially known as 2018 AG₃₇. Having whimsically dubbed the previous furthest-known object “Farout,” astronomers kept with the theme and named its wayward sister “Farfarout.” Given the rapid gains in technology, chances are good that Farfarout won’t stay the Sun’s remotest outpost for long, and we fear the (Far)_nout trend will eventually collapse under its own weight. We therefore modestly propose a more sensible naming scheme, perhaps something along the lines of “Farthest McFaraway.” It may not scale well, but at least it’s stupid.

3D Printer? Laser Cutter? CNC? Yes, Please

February 10, 2021 by Al Williams 60 Comments

Most of us have, or, would like to have a 3D printer, a laser engraver, and a CNC machine. However, if you think about it naively, these machines are not too different. You need some way to move in the XY plane and, usually, on the Z axis, as well.

Sure, people mount extruders on CNCs, or even lasers or Dremel tools on 3D printers. However, each machine has its own peculiarities. CNCs need rigidity. 3D printers should be fast. Laser engravers and CNCs don’t typically need much Z motion. So common sense would tell you that it would be tough to make a machine to do all three functions work well in each use case. [Stefan] thought that, too, until he got his hands on a Snapmaker 2.0.

As you can see in the video below, the machine uses different tool heads for each function. The motion system stays the same and, curiously, there are three identical linear motion modules, one for each axis.

Continue reading “3D Printer? Laser Cutter? CNC? Yes, Please” →

Pen Plotter Draws Maps Directly On The Wall

January 19, 2021 by Dan Maloney 22 Comments

For map-lovers like [Christopher Getschmann], poring over a quality map can be as satisfying as reading a good book. Good maps can be hard to come by, though, especially at a scale worth looking at, or worth using as adornment on a dull, lifeless wall. The solution is obvious: build a wall-mount CNC plotter to draw maps directly on the wall.

[Christopher] began his map quest by scraping world map data from a number of sources, including OpenStreetMap, Natural Earth, and GEBCO. This gave him data for coastlines, terrain, and bathymetry — enough for a map of the world large enough to fill a wall. Since the scale of the map would preclude the use of even a large-format inkjet printer, [Christopher] set about building a wall-covering pen-plotter to render the map. The CoreXY-style plotter is large, but still light enough to hang on the wall while it works, and to be repositioned to cover a larger area.

The plotter runs on steppers driven by ultra-quiet Trinamic TMC5160 drivers, so the plotter wouldn’t be a nuisance while it worked. The map was plotted on eight pieces of cardboard mounted directly to the wall, filling the 2- x 3-meter space almost entirely. Landmasses and elevation contours were plotted as continuous lines in black ink, while bathymetric data was rendered in blue ink as cross-hatching with variable spacing, to make deeper oceans darker blue.

We find [Christopher]’s map breathtaking, all the more so considering the work that went into making it. It would be interesting to find alternate uses for the plotter, which reminds us a little of a cross between a draw-bot and a Maslow vertical CNC router, now that it’s done with its cartographic duties.

Checking In On Low-Cost CNC Machines

January 2, 2021 by Al Williams 43 Comments

Low cost 3D printers have come a long way in the last few years, but have entry-level CNC machines improved by the same leaps and bounds? That’s what [ModBot] recently set out to find. Despite getting burned pretty badly on a cheap CNC a few years back, he decided to try again with a sub $400 machine from FoxAlien. You can see his full review after the break.

The machine looks very similar to other generic CNC machines you see under many brand names, sometimes for a good bit less. The 3018 number is a giveaway that the work area is 30×18 cm and a quick search pulled up several similar machines for just a bit more than $200. The FoxAlien did have a few nice features, though. It has a good-looking build guide and an acrylic box to keep down the shaving debris in your shop. There are also some other nice touches like a Z-axis probe and end stops. If you add those items to the cut-rate 3018 machines, the FoxAlien machine is pretty price competitive when you buy it from the vendor’s website. The Amazon page in the video shows $350 which is a bit more expensive but does include shipping.

As with most of these cheap CNC machines, one could argue that it’s more of an engraver than a full mill. But on the plus side, you can mount other tools and spindles to get different results. You can even turn one of these into a diode laser cutter, but you might be better off with something purpose-built unless you think you’ll want to switch back and forth often.

This reminded us of a CNC we’ve used a lot, the LinkSprite. It does fine for about the same price but we are jealous of the enclosure. Of course, half the fun of owning something like this is hacking it and there are plenty of upgrades for these cheap machines.