3D Printing Steel Parts At Home Via Special Filaments

Rocket engines are great for producing thrust from fire and fury, but they’re also difficult to make. They require high-strength materials that can withstand the high temperatures involved. [Integza], however, has tried for a long time to 3D print himself a working rocket engine. His latest attempt involves printing an aerospike design out of metal.

Even steel couldn’t hold up to the fury of the rocket exhaust!

The project relies on special metal-impregnated 3D printer filaments. The part can be printed with a regular 3D printer and then fired to leave just the metal behind. The filament can be harsh, so [Integza] uses a ruby nozzle to handle the metal-impregnated material. Processing the material requires a medium-temperature “debinding” stage in a kiln which removes the plastic, before a high-temperature sintering process that bonds the remaining metal particles into a hopefully-contiguous whole. The process worked well for bronze, though was a little trickier for steel.

Armed with a steel aerospike rocket nozzle, [Integza] attempts using the parts with his 3D printed rocket fuel we’ve seen before. The configuration does generate some thrust, and lasts longer than most of [Integza]’s previous efforts, though still succumbs to the intense heat of the rocket exhaust.

Overall, though, it’s a great example of what it takes to print steel parts at home. You’ll need a quality 3D printer, ruby nozzles and a controllable kiln, but it can be done. If you manage to print something awesome, be sure to drop us a line. Video after the break.

Continue reading “3D Printing Steel Parts At Home Via Special Filaments”

RAMPS Rebuild Keeps Robox 3D Printer Out Of Junk Bin

A 3D printer is a wonderful invention, but it needs maintenance like every machine that runs for long hours. [Rob Ward] had a well-used Robox 3D printer that was in need of some repairs, but getting the necessary replacement parts shipped to Australia was cost-prohibitive. Rather than see a beloved printer be scrapped as e-waste, he decided to rebuild it using components that he could more easily source. Unfortunately the proprietary software and design of the Robox made this a bit difficult, so it was decided a brain transplant was the best path forward.

Step one was to deduce how the motors worked. A spare RAMPS 1.4 board and Arduino Mega2560 made short work of the limit switches and XYZ motors. This was largely accomplished by splicing into the PCBs themselves. The Bowden filament driver motor had a filament detector and an optical travel sensor that required a bit of extra tuning, but now the challenging task was next: extruding.

The printer’s new custom hotend.

With a cheap CR10 hot end from an online auction house, [Rob] began modifying the filament feed to feed in a different direction than the Robox was designed for (the filament comes in at a 90-degree angle on the stock Robox). A fan was needed to cool the filament feed line. Initial results were mixed with lots of blockages and clogs in the filament. A better hot end and a machined aluminum bracket for a smoother path made more reliable prints.

The original bed heater was an excellent heater but it was a 240 VAC heater. Reluctant to having high voltages running through his hacked system, he switched them out for 12 VDC adhesive pads. A MOSFET and MOSFET buffer allowed the bed to reach a temperature workable for PLA. [Rob] upgraded to a GT2560 running Marlin 2.x.x.

With a reliable machine, [Rob] stepped back to admire his work. However, the conversion to the feed being perpendicular to the bed surface had reduced his overall build height. With some modeling in OpenSCAD and some clever use of a standard silicone sock, he had a solution that fed the wire into the back of the hot end, allowing to reclaim some of the build height.

It was a long twelves months of work but the write-up is a joy to read. He’s included STL and SCAD files for the replacement parts on the printer. If you’re interested in seeing more machines rebuilt, why not take a look at this knitting machine gifted with a new brain.

Pen Plotter Is About As Simple As It Can Get

Sure, we see quite a few plotters and other motion machines, but the one from [DAZ Projects] has the virtual of looking dead simple. The Arduino and CNC shield are old hat, of course. But some 3D printed pulleys and a very simple-looking core XY arrangement looks like this could be a pretty quick build.

You might ask; if you have a 3D printer, why you wouldn’t just mount a pen on it and call it a day? Well, you could do that, of course, but what fun is that? Besides, that will tie up your printer, too. You can see a video of the project, below.

Continue reading “Pen Plotter Is About As Simple As It Can Get”

Vintage Test Equipment Addiction Justified

Recore 3D printer board developer [Elias Bakken] has posted about the automatic test procedure he developed using a stack-up of four (at least) pieces of vintage HP test equipment. In addition, his test jig and test philosophy is quite interesting.

Besides making a bed-of-nails test jig, he also designed a relay multiplexing board to that selects one of the 23 different voltages for measurement. We like his selection of mechanically latching relays in this application — not only does it save power, but it doesn’t subject the test board to any magnetic fields (except when switching state).

In [Elias]’s setup, the unit under test (UUT) actually orchestrates the testing process itself. This isn’t as crazy as it might sound. The processor is highly integrated in one package plus external DRAM. If the CPUs boot up at all, and pass simple self-test routines, there’s no reason not to utilize the on-board processor as the main test control computer. This might be a questionable decision if your processor was really small with constrained resources and connectivity. But in the case of Recore, the processor is a four-core ARM A53 SoC running Debian Linux — an arrangement that itself could well serve as an automated test computer in other projects.

In the video down below, [Elias] walks us through the basic tests, and then focuses on the heart of the Recore board tests: calibrating the input signal conditioning circuits. Instead of using very expensive precision resistors, [Elias] selected more economical 1% resistors to use in the preamp circuitry. The tradeoff here is the need to calibrate each channel, perhaps at multiple temperature points. This is a situation where using a test jig, automated test scripts, and and stack of programmable test equipment really shines.

[Elias] is still pondering some issues he found trying to calibrate thermocouples, so his adventure is not quite over yet. If you are wondering what Recore is, check out this article from back in June. Have you ever used the microprocessor on a circuit board to test itself, either standalone or in conjunction with an external jig? Let us know in the comments below.

Continue reading “Vintage Test Equipment Addiction Justified”

3D Printer Automated Bed Swapping System Loads From A Magazine

FDM 3D printing has gone beyond prototyping and is being used as a production tool by many companies. However, conventional printers still require an operation to pop the finished part of the bed and start a new print. [Thomas Sandladerer] wanted a way to swap beds without human intervention, so he built an automatic print surface changing system.

The most obvious solution to this problem may appear to be belt printers like the Creality CR-30, but these come with some trade-offs. Bed adhesion can be a problem, and the lack of a rigid print surface causes some parts to come out warped. [Thomas] wanted to be able to use PEI-coated steel beds to avoid these issues. His solution is a system that pulls beds from a “magazine” and pushed out the old bed after a part is finished. It still uses a magnetic heatbed, which lowers out of the way before changing print surfaces. Each print surface is fitted inside a 3D printed frame which rests on the tool changer frame and keeps it in place as the heatbed drops down. The bed frames are printed using ASA, can handle 90 C without problems. The pusher mechanism and the heatbed lowering system are driven by stepper motors which connect to the spare motor outputs on the printer’s control board. The printer in question is a Voron 2.4, which is perfect for this application thanks to its high print speed.

This tool-changing system is only the first prototype, but it still worked very well. [Thomas] plans to make key improvements like a larger print bed and reduced height. This system might be a good fit for small and large print farms. We’ve seen another bed-clearing system that doesn’t require extra build surfaces, but instead scrapes off the completed part.

Continue reading “3D Printer Automated Bed Swapping System Loads From A Magazine”

3D Printing May Disprove Lord Kelvin

If you think 3D printing is only good for benchies, key chains, and printer parts, you might enjoy the paper by two physicists from Wesleyan University and the University of Gothenburg. Lord Kelvin — also known as William Thomson — hypothesized a shape known as an isotropic helicoid. As its name implies, the shape would look the same from any angle. Kelvin predicted that such a shape would spin as it sank in a liquid. Turns out, 3D printing proves it wrong. (The actual paywalled paper is available.)

It might seem strange that scientists are only now getting around to disproving a 150-year old hypothesis. However, the paper’s authors think Kelvin may have built the structures — he provided precise instructions — and simply dropped it when it proved incorrect.

Continue reading “3D Printing May Disprove Lord Kelvin”

Strangest Upside-Down 3D Printer Fits In A Filament Box

It’s rare these days for a new FDM printer to come along that sparks our interest, but the [Kralyn]’s Positron managed to do it. (Video, embedded below.) It prints upside down and packs down into a filament box while still boasting a print volume of 175 mm x 176 mm x 125 mm.

Unlike most 3D printers, the hotend and XY-gantry is mounted below the build plate, directly onto the base. You might assume that a printer needs to extrude plastic with gravity to work properly, but the real action is in the smooshing of the plastic layers. It appears that it might even improve bridging since the hotend is supporting the plastic as it gets extruded. A clear glass build plate is used, with the same heating strips found on the rear windows of most cars. This also allows the user still see the part, and provides the added advantage of being able to quickly spot bed leveling and adhesion problems.

Another interesting side effect of this arrangement is rigidity. There is no need to suspend the XY gantry with the heavy hotend in the air, so it can be mounted directly on the thick aluminum base plate. It uses an H-bot style gantry, with Synchromesh timing cables instead of belts, which eliminates the concern of belt twist. To get the best possible print volume within the size of a filament box, the gantry axes are arranged diagonally across the base plate. The Z-axis can disconnect and lay flat on top of the printer and uses the linear rails to keep it perfectly straight and perpendicular when mounted. Continue reading “Strangest Upside-Down 3D Printer Fits In A Filament Box”