Keep That Engine Running, With A Gassifier

Every now and then in histories of the 20th’s century’s earlier years, you will see pictures of cars and commercial vehicles equipped with bulky drums, contraptions to make their fuel from waste wood. These are portable gas generators known as gasifiers, and to show how they work there’s [Greenhill Forge] with a build video.

A gasifier on a vintage tractor
A gasifier on a vintage tractor. Per Larssons Museum, CC BY 2.5.

When you burn a piece of wood, you expect to see flame. But what you are looking at in that flame are the gaseous products of the wood breaking down under the heat of combustion. The gasifier carefully regulates a burn to avoid that final flame, with the flammable gasses instead being drawn off for use as fuel.

The chemistry is straightforward enough, with exothermic combustion producing heat, water vapour, and carbon dioxide, before a further endothermic reduction stage produces carbon monoxide and hydrogen. He’s running his system from charcoal which is close to pure carbon presumably to avoid dealing with tar, and at this stage he’s not adding any steam, so we’re a little mystified as to where the hydrogen comes from unless there is enough water vapour in the air.

His retort is fabricated from sheets steel, and is followed by a cyclone and a filter drum to remove particulates from the gas. It relies on a forced air draft from a fan or a small internal combustion engine, and we’re surprised both how quickly it ignites and how relatively low a temperature the output gas settles at. The engine runs with a surprisingly simple gas mixer in place of a carburetor, and seems to be quite smooth in operation.

This is one of those devices that has fascinated us for a long time, and we’re grateful for the chance to see it up close. The video is below the break, and we’re promised a series of follow-ups as the design is refined.

Continue reading “Keep That Engine Running, With A Gassifier”

Desert Island Acetylene From Seashells And Driftwood

[MacGyver] would be proud of [Hyperspace Pirate]’s rough and ready method of producing acetylene gas from seashells and driftwood.

Acetylene, made by decomposing calcium carbide with water, is a vitally important industrial gas. Not only as a precursor in many chemical processes, but also as the fuel for the famous “blue wrench,” a tool without which auto mechanics working in the Rust Belt would be reduced to tears. To avoid this, [Hyperspace Pirate] started by beachcombing for the raw materials: shells to make calcium oxide and wood to make charcoal. Charcoal is pretty easy; you just cook chunks of wood in a reducing environment to drive off everything but the carbon. Making calcium oxide from the calcium carbonate in the shells isn’t much harder, with ground seashells heated in a propane-fired furnace to release carbon dioxide.

With the raw ingredients in hand, things get a little tricky. Making calcium carbide requires a lot of heat, far more than a simple propane burner can provide. [Hyperspace Pirate] decided to go with an electric arc furnace, to which end he cannibalized a 120 V to 240 V step-up converter for its toroidal transformer, which with a few extra windings provided the needed current to run an arc through carbon electrodes. This generated the needed heat, and then some, as the ceramic firebrick he was using to contain the inferno melted. After rewinding the melted secondary windings on his makeshift transformer and switching to a stainless steel crucible, he was able to make enough calcium carbide to generate an impressive amount of acetylene. The video below documents the process and the sooty results, as well as details a little of the excitement that metal acetylides offer.

For more about acetylene and its many uses, [This Old Tony] has you covered.

Continue reading “Desert Island Acetylene From Seashells And Driftwood”

Tired With Your Robot? Why Not Eat It?

Have you ever tired of playing with your latest robot invention and wished you could just eat it? Well, that’s exactly what a team of researchers is investigating. There is a fully funded research initiative (not an April Fools’ joke, as far as we know) delving into the possibilities of edible electronics and mechanical systems used in robotics. The team, led by EPFL in Switzerland, combines food process engineering, printed and molecular electronics, and soft robotics to create fully functional and practical robots that can be consumed at the end of their lifespan. While the concept of food-based robots may seem unusual, the potential applications in medicine and reducing waste during food delivery are significant driving factors behind this idea.

The Robofood project (some articles are paywalled!) has clearly made some inroads into the many components needed. Take, for example, batteries. Normally, ingesting a battery would result in a trip to the emergency room, but an edible battery can be made from an anode of riboflavin (found in almonds and egg whites) and a cathode of quercetin, as we covered a while ago. The team proposed another battery using activated charcoal (AC) electrodes on a gelatin substrate. Water is split into its constituent oxygen and hydrogen by applying a voltage to the structure. These gasses adsorb into the AC surface and later recombine back into the water, providing a usable one-volt output for ten minutes with a similar charge time. This simple structure is reusable and, once expired, dissolves harmlessly in (simulated) gastric fluid in twenty minutes. Such a device could potentially power a GI-tract exploratory robot or other sensor devices.

But what use is power without control? (as some car tyre advert once said) Microfluidic control circuits can be created using a stack of edible materials, primarily oleogels, like ethyl cellulose, mixed with an organic oil such as olive oil. A microfluidic NOT gate combines a pressure-controlled switch with a fluid resistor as the ‘pull-up’. The switch has a horizontal flow channel with a blockage that is cleared when a control pressure is applied. As every electronic engineer knows, once you have a controlled switch and a resistor, you can build NOT gates and all the other logic functions, flip-flops, and memories. Although they are very slow, the control components are importantly edible.

Edible electronics don’t feature here often, but we did dig up this simple edible chocolate bunny that screams when you bite it. Who wouldn’t want one of those?

DIY Furnace Smelts Magnetite Sand Into An Impressive Chef’s Knife

Some people order their raw materials from a factory, missing out on 99% of the fun… or suffering, we’re not sure which. To make that call, you need to look in on the process [IllyriaD] used to collect magnetite sand and turn it into a wicked-looking chef’s knife.

This began by collecting 150 pounds (!) of magnetic dirt from dry lake beds while hiking using a magnet pickup tool with release lever that he got from Harbor Freight. Several repeated magnetic refining passes separated the black ore from non-metallic sands ready for the furnace that he built. That is used to fire up the raw materials using 150 pounds of charcoal, changing the chemical composition by adding carbon and resulting in a gnarly lump of iron known as a bloom.

From there, it’s just a matter of beating the iron bloom into submission over at the anvil. [IllyriaD] details the process of flattening it out to a bar shape, then folding it over. Seven total folds are made for 128 layers, and in the gallery there’s a fantastic image that captures the striation when viewed on end. After being sharpened and polished, you can see where the bevel descends through those layers.

It’s delightful to see people working through the old ways and proving you don’t need a factory, as long as your true goal is to explore the process itself. Does this leave you wanting even more? [IllyriaD] left some insight about the process in the comments of the reddit thread. You probably also want to check out the tile-roofed hut built by [PrimitiveTechnology] without any modern tools.

Fail Of The Week: When Good Foundries Go Bad

Like many of us, [Tony] was entranced by the idea of casting metal, and set about building the tools he’d need to melt aluminum for lost-PLA casting. Little did he know that he was about to exceed the limits of his system and melt a hole in his patio.

[Tony]’s tale of woe begins innocently enough, and where it usually begins for wannabe metal casters: with [The King of Random]’s homemade foundry-in-a-bucket. It’s just a steel pail with a homebrew refractory lining poured in place, with a hole near the bottom to act as a nozzle for forced air, or tuyère. [Tony]’s build followed the plans pretty faithfully, but lacking the spent fire extinguisher [The King] used for a crucible in the original build, he improvised and used the bottom of an old propane cylinder. A test firing with barbecue charcoal sort of worked, but it was clear that more heat was needed. So [Tony] got hold of some fine Welsh anthracite coal, which is where the fun began. With the extra heat, the foundry became a mini-blast furnace that melted the thin steel crucible, dumping the molten aluminum into the raging coal fire. The video below shows the near catastrophe, and we hope that once [Tony] changed his pants, he hustled off to buy a cheap graphite or ceramic crucible for the next firing.

All kidding aside, this is a vivid reminder of the stakes when something unexpected (or entirely predictable) goes wrong, and the need to be prepared to deal with it. A bucket of dry sand to smother a fire might be a good idea, and protective clothing is a must. And it pays to manage your work area to minimize potential collateral damage, too — we doubt that patio will ever be the same again.

Continue reading “Fail Of The Week: When Good Foundries Go Bad”

Kill The Exhaust, Not Your Lungs With The Fume Coffin

As if slinging around 40 watts of potentially tattoo-removing or retina-singeing laser beams wasn’t anxiety-inducing enough, now comes a new, scary acronym – LCAGs, or “laser-generated airborne contaminants.” With something that scary floating around your shop, it might be a good idea to build a souped-up laser cutter exhaust fan to save your lungs.

We jest, but taking care of yourself is the responsible way to have a long and fruitful hacking career, and while [patternmusic]’s “Fume Coffin” might seem like overkill, can you go too far to protect your lungs? Plywood and acrylic, the most common materials that come across a laser cutter’s bed, both release quite a witch’s brew of toxins when vaporized by a laser beam. The Fume Coffin clears the air in your shop by venting it to the outdoors after giving it a good scrubbing through an activated charcoal pre-filter and a HEPA polishing element. Both filters are commercially available so replacements won’t be an issue, and the entire thing is housed in a wooden box that gives the device its name.

Since it’s ejecting 200 cubic feet per minute, you’ll have to provide at least that much make-up air, but other than that the Fume Coffin should be a welcome addition to the shop. We’ve seen a few other attempts to handle LCAGs effectively before, including a DIY charcoal and automotive air filter design.

See A Cheap Smoker Get An Automation Power Up

[Jason] learned a lot by successfully automating this meat smoker. This is just the first step in [Jason’s] smoker project. He decided to begin by hacking a cheaper charcoal-fed unit first, before setting his sights on building his own automatic pellet-fed smoker. With a charcoal smoker it’s all about managing the airflow to that hot bed of coals.

automated-meat-smoker-air-valve
Custom mount for servo was actually one of the more challenging things to get just right.

[Jason] started by making sure the bottom was sealed off from stray airflow, then he cut a hole into the charcoal pan and attached a length of steel pipe. The opposite end of the pipe has a fan. Inside the pipe there is a baffle separating the fan from the charcoal pan. The servo motor shown here controls that valve.

The pipe is how air is introduced into the smoker, with the fan and valve to control the flow rate. The more air, the higher the temperature. The hunk of pipe was left uncut and works fine but is much longer than needed; [Jason says] the pipe is perfectly cool to the touch only a foot and a half away from the smoker.

With the actuators in place he needed a feedback loop. A thermocouple installed into the lid of the smoker is monitored by an Arduino running a PID control loop. This predicts the temperature change and adjusts the baffle and fan to avoid overshooting the target temp. The last piece of hardware is a temperature probe inside the meat itself. With the regulation of the smoker’s temperature taken care of and the meat’s internal temperature being monitored, the learning (and cooking) process is well underway.

There are many, many smoker automation projects out there. Some smokers are home-made electric ones using flower pots, and some focus more on modifying off the shelf units. In a way, every PID controlled smoker is the same, yet they end up with different problems to solve during their creation. There is no better way to learn PID than putting it into practice, and this way to you get a tasty treat for your efforts.