Building a Stirling Engine Bike

Go to https://ground.news/stanton for a better way to stay informed. Subscribe for 40% off unlimited access to world-wide coverage through my link.

Second channel: @TimStation

Enjoy my videos? These are made possible due to help from my Patrons. Please consider supporting my efforts: https://www.patreon.com/tomstanton

—————————————————————————————————————————————-
My 3D Printers
Formlabs Form 4: https://formlabs.com/uk/3d-printers/form-4/
Formlabs Form 4L: https://formlabs.com/uk/3d-printers/form-4l/

Prusa Core One: https://www.prusa3d.com/product/prusa-core-one/
Prusa XL: https://www.prusa3d.com/en/product/original-prusa-xl-semi-assembled-5-toolhead-3d-printer/

Instagram: https://www.instagram.com/tom__stanton/
#stirling #engine #cncmill #engineering #machining

This engine has no visible fuel, no valves, no exhaust, and runs almost completely silent. And that’s because its fuel source is heat stored in this large metal block below it. And it’s known as a hot air engine or a sterling engine. glass syringe with a small amount of air inside. And if I heat up the syringe and therefore the air, it expands and pushes the piston outward. Then when the heat is removed, it cools down and the air contracts. This cycle of heating and cooling the air to move the piston back and forth is the base principle on how a sterling engine works. The only issue is this heating and cooling is far too slow. So instead of heating and cooling the entire system, we can introduce another cylinder with a loose fitting piston inside which unlike most pistons that require an airtight seal, air is free to flow around it, allowing the piston to shuttle the air from one end of the cylinder to the other. And this is known as the displacer piston. Now, if we apply heat to one side of this secondary piston and keep the other end cool, we can use a displacer piston to move the air between the hot and cold sides, which rapidly heats and cools the air inside and moves the top piston back and forth much faster than before. And this can be done indefinitely as long as one side is hot and the other is cold, or at least until the 3D printed piston melts. Now, the displacer piston doesn’t produce any output power and requires very little power to move, which can be shown by replacing it with these ball bearings. So, when I gently tilt the cylinder, the ball bearings roll from one end to another and moves the air between the hot and cold sides, which expands and contracts to drive the power piston up top. So, if we link the top piston to a crankshaft and use some of the output power to move the displacer back and forth, we should have a self-sustaining engine. In fact, here is a model engine that does exactly that. Heat is applied to this glass section here, and the other side of the cylinder is cooled by this heat sink. So, as a displacer moves back and forth, the air is heated and cooled, which is then piped through to this top cylinder, which drives the power piston. That then outputs power to the crank, which in turn moves the displacer through the gears, creating a continuous cycle. So, now we need to figure out how we’re going to use one of these Sterling engines to power a bike. And it all starts here with a big block of metal and a very long to-do list. Sterling engines produce very little power for their size. So, to make this work on a bike, we need the largest engine possible that still fits within the frame. That said, this isn’t going to be the next superbike engine. Oh dear. Um, I’m aiming for about 100 to 150 W of power output, which is roughly2 horsepower, just enough to hopefully maintain a cruising speed of about 15 mph. Getting that kind of power from a sterling engine isn’t easy. There are lots of factors that affect the efficiency, like the temperature difference between the hot and cold sides, how fast heat can be transferred into and out of the air, and how much air inside of the engine is actually doing useful work. These are just a handful of what determines the design of the engine. Like for example, to keep the cold side as cool as possible. I originally considered using a computer CPU heat sink as they’re cheap and have tons of surface area. But it turns out trying to funnel all that heat into a small square patch designed for a CPU just wasn’t practical. So we decided to go with internal water cooling, which should work much better for pulling the heat out quickly. And when we’re only trying to get about 100 watts of output power, even tiny amounts of friction can make a huge difference, which isn’t easy when you still need airtight seals around the piston and the displacer shaft. With the main engine block complete, I can mount it to the bike using a temporary 3D printed bracket. Since this section of the engine will be water cooled, so the print should hold up fine for some test runs. Next, this large aluminium bracket attaches to the rear of the engine, which will hold the crankshafts and also clamp to the bike frame for added support. The cranks themselves are machined from aluminium, and I fitted a 3D printed pulley on the back, as I’m using a belt to link the cranks instead of the gears like in the model engine, and a connecting rod with press fit bearings joins the crank to the piston to keep everything spinning smoothly. Because a displacer piston is a loose fit inside its cylinder, it ideally shouldn’t touch the walls at all. So, I’m going to mount it to a stainless steel rod. And to make sure that rod doesn’t wobble, I’m using these linear bearings, the same kind you’d find in a 3D printer. And they allow the displacer rod to move smoothly and accurately, which keeps the displacer centered as it moves. Next is the power piston, which I’ve designed to be as lightweight as possible, and it mounts to another steel rod using a set screw. For the piston rings, I want to use PTFE, also known as Teflon, because of its low friction, though it’s not a springy material. So, I’ve machined some smaller inner rings to act like tensioners, which should push the PTFE outward to create a better seal against the cylinder wall. The rings sit in the piston like this, and I’ve offset the gaps in the two rings by 180° to minimize blowby. And when it’s in the cylinder, the piston slides with very little resistance. The piston rod then connects to the crank. But because the piston is so short, it needs a guide to keep it traveling in a straight line, and that’s the job of this back plate. Though, before mounting it, I need to cut some gaskets. I’m using.5 mm silicon sheet, which should handle the temperature this engine will experience and will give a good seal. So, after punching out the bolt holes, I can clamp down a 3D printed template and trace around it with a knife. Then, loosen the bolts and I’ve got a precise gasket that fits the engine perfectly. With the back plate and gasket mounted, the piston rod still had a bit of a wobble. So, I added this PTFE bushing mounted with a spacer to keep the rod aligned and moving perfectly straight. The next step is to connect the two cranks with a timing pulley, which I chose instead of the gears because it’s easier to connect to an output shaft that can then be used to drive the rear wheel. Seeing this thing finally spin makes me realize how much work I’ve put into this thing. Uh, it’s been about 2 months of calculations, designing, machining. It’s been so much work and I’m starting to get a bit nervous as to whether it’s going to run. Uh, I’m excited, but still nervous. So, what needs to be done now is make the displacer, cylinder, and piston, and then we should be able to give it a test run. I actually started this project planning to build a steam engine, but after lots of research, I realized that sterling engines are a much better option. They’re quieter, more efficient, and safer. And once I understood how they worked, it completely changed the direction of the project. It also Before mounting the displacer cylinder, I need to install this plate, which acts as the heat transfer surface between the water cooling system and the air inside of the engine. And the large surface area from this heat sink should help cool the air quickly during the engine cycle. It also doubles as a top plate for the power cylinder. And all the expanding hot air flows through this port to push the piston. For the display set itself, I need something lightweight and fairly heat resistant. So I bought a load of different stainless steel containers and bottles to see if any would work. And apart from learning that none of them are made to standard sterling engine displacer diameters, I did find this water bottle which should fit well inside an 80 mm diameter cylinder with just a few millm of clearance for air to flow past. I then made a small mounting plate for the displacer that attaches to the steel rod with a set screw and then cut the top off of the bottle and drilled a few holes which allows it to mount to the bottom plate. And when spinning the engine by hand, you can see how both pistons will move once the engine is running, where the displacer piston leads the power piston by about 90° crank angle. Next, I need to make the displacer cylinder, which I decide to build in two halves. The cold side is machined from aluminum like most of the engine, and I added some fins to help it stay cool. It’s then mounted to the main engine block with another gasket which creates a good seal, but it also insulates it from the water cooling. So, if the air cooling fins aren’t sufficient, I may need to add water cooling to it as well. But the hot side of the cylinder needs to handle much higher temperatures. So, ideally, the hot side should be made from steel. Since I’m not confident machining steel just yet, I had the hotend made by JLC CNC. They didn’t sponsor this video, but they definitely saved me here because there’s no way I’d be able to machine this part myself. And this is mounted to the cold part of the cylinder with a thick silicon gasket to hopefully insulate it as we don’t want heat conducting through to the cold side as that will be wasted energy. So, with the hot cap tightened down, the engine is fully sealed. All it needs now is heat. [Applause] That should do the job. H. It’s not quite there. Um, the front of the hot cap is starting to discolor, so it’s getting quite hot on the front there. I’m thinking maybe uh I should add some water cooling to see if it does cool down this cold plate. With the hot cap glowing red hot and the water cooled end holding steady at around 40° C, it was clear that simply increasing the heat wasn’t the issue. So, I added some external water cooling to safely bring the engine down to a temperature where I could take it apart. Inside, it was obvious the engine was getting plenty hot. Even the displacer was slightly discolored. And considering it doesn’t touch the hot cap directly, that means the air inside is definitely heating up well. I checked all the gasket seals and added some grease to the O-ring around the displacer shaft to rule out any leaks. And with the front end of the engine reassembled, I focused on the piston rings since this is the largest moving seal in the engine. It was tricky to move the piston against the trapped air, which is a good sign, but I also noticed it was letting a lot of air leak past, which could be where most of the pressure is being lost. I first tried swapping one of the PTFE rings for a rubber O-ring to improve the seal, but it either didn’t seal properly or created way too much friction. Then I thought, what if I used a piston ring like the one I’ve used in my small compressed air engine where the ring expands under pressure to seal against the cylinder wall? The only problem, those rings are tiny, and I needed something way bigger. So, I tried 3D printing one out of flexible TPU on my Prusser FDM printer, which was a bit of a long shot as it needed to have the right flexibility, hold an airtight seal, and hopefully not melt from the engine heat. But after several prototypes, I eventually found a size and shape that offered minimal resistance, but sealed far better than before. And with the help of the water cooling to keep the cylinder walls cold, hopefully it won’t get hot enough to melt. Right. So, I’ve got the new piston on. So, we should be getting much better compression now. You can see when I turn it, the piston actually pushes back, which means it’s got really good compression. So, fingers crossed this will make the difference because the compression before was absolutely awful compared to this. So, let’s get comfy and heat this thing up. [Applause] I feel like it’s so close, but not quite there. So, the expansion’s still working. It just doesn’t quite have the travel. Right, let’s let it all cool down before we try something else. Not sure what I’m going to try, but after reviewing the footage, I realized something. It just doesn’t quite have the travel. If the crank has 30 mm of travel, but the piston is only moving about 14 mm, then it might be overcompressing and overexpanding the air, which could be wasting a huge amount of energy per cycle. Unfortunately, I can only shorten the crank to 25 mm of travel due to the shaft and bearing sizes. But if I also increase the travel of the displacer piston, I might be able to move more air between the hot and cold sides, which could increase the pressure difference and improve expansion. These new cranks were 3D printed from resin on the Formlabs form 4 and should be plenty strong for this stage of testing, though they’re a bit lighter than the aluminium ones, but for now, engine balance isn’t a top priority. With the new cranks installed and a slightly shorter displacer to avoid hitting the end of the cylinder with its increased stroke, the engine is ready for another test. I really hope this works. Come on. Come on. Is that running by itself? That’s running. It’s working. Come on. More heat. More heat. It’s actually running. Wow. It’s not the fastest, but I reckon there’s a few things that we can change to improve its performance. That’s too cool. That is too cool. It sounds so funny. While I’m incredibly pleased to get this engine finally running, I couldn’t resist trying to make it run just a bit better. So, I swapped out the timing belt for a much thinner one since the thicker belt was wasting energy as it bent and straightened around the pulleys. I also replaced the drive pulley to the rear wheel with a flywheel as the original pulley ratio might have been too high preventing the engine from gaining RPM. There’s still lots to improve with this engine, like adding a regenerator to recover some heat energy during each cycle and increasing the internal pressure, as more pressure means more air inside the engine, and more air means more expansion and more power. I also still need to build a proper burner, finish the water cooling system with a radiator, and design a clutch since sterling engines don’t throttle very well. In fact, I turned off the heat a while ago and it’s still running on the stored heat in the steel hot cap.