Hacker News new | past | comments | ask | show | jobs | submit login
Let's Build a Can Stirling Engine (2000) (bekkoame.ne.jp)
44 points by DanBC on Sept 16, 2020 | hide | past | favorite | 37 comments



One of the things which really, really rustles my jimmies is computer nerds talking about matter the way they talk about Stirling engines. Someone once upon a time told them that Stirling engines were efficient heat engines which work on diffuse heat sources, and suddenly, Mr nerd wants to stick them everywhere. No. Unlike using some sorting or hashing algorithm, there are tradeoffs and complexities in technologies involving matter beyond anything you do in your day to day work. "Stirling engine" isn't some kind of algorithm you can put everywhere like your favorite red-black tree implementation.

Stirling engines, like all heat engines, work more efficiently with large temperature differentials: this means advanced materials. External combustion immediately puts you in a weird and unfamiliar engineering place where you have the firey bits outside the piston where you can do the usual things to keep it under control. Cooling is also different from just farting out exhaust and pumping water through the block in a normal engine. If you don't have a large temperature differential you have shitty efficiency and it's not worth doing. Practical Stirling engines which do useful work, whether for heating or cooling, are usually large, have sealed, pressurized helium as working fluids (oh? you don't know about working fluids in Stirling engines?), complicated gearing mechanisms and sometimes even magnetic coupling to the outside world, and have extremely complex and precision machine parts compared to the slop you see in consumer grade items like a Porsche Engine. Ratios in the engine are often tailored to a very specific use case; and will work for shit if your heat spec is outside that very, very narrow use case. Go have a look at what goes into designing a real world Stirling engine for example in [0].

People don't use Otto/Rankine/Brayton/whatever because of some live-steam or anti-Stirling conspiracy: they use it because Stirling engines, outside of a few mostly unimportant use cases, are usually expensive and shit. The two most widespread industrial uses of the things I can think of were a Chrysler air conditioner[1] and a coleman beer cooler[2], both using the same Japanese built Stirling cooler. Even in these use cases, which, mind you are using a Stirling engine in reverse from what HN weebs want them for, they didn't exactly cover themselves in glory. They were expensive, heavy, fragile and not as efficient as would justify their bad qualities over a more standard cooling unit.

[0]https://ntrs.nasa.gov/citations/19830022057

[1] https://docplayer.net/40169613-The-reliability-development-o... (I think)

[2]https://kk.org/cooltools/coleman-stirlin/


I'm impressed that you managed to get your knickers in such a twist and you failed to point out that in fact this is not a Stirling engine at all.

A Stirling engine must have a regenerator which temporarily stores the heat from one stroke to the next. Otherwise it is just a hot air engine which is altogether more ancient.


I cannot see the relation between your rant and the article which is about a didactic implementation by some teacher.


scroll down


> The Gotland-class submarines of the Swedish Navy [...] are the first submarines in the world to feature a Stirling engine air-independent propulsion (AIP) system [...]

https://en.wikipedia.org/wiki/Gotland-class_submarine


The manufacturer of the Stirling engine for those submarines, Kokums, is the source of the engines used for a dish-Stirling solar energy system I worked on for a licensee of Stirling Energy Systems[1]. It was a concentrating solar energy system consisting of a parabolic mirror focusing sunlight onto a heater head atop the Stirling engine. Quite efficient at around 30% conversion of direct sunlight to electricity - each unit generated about 25kW under clear skies. Quite a simple system - just a big mirror, sun tracker, Stirling engine and generator. Super reliable - no nasty combustion products. And not very expensive - BOM cost about the same as a mid-size car. But dropping PV prices ended up killing it.

[1]https://en.wikipedia.org/wiki/Stirling_Energy_Systems


I've come across those Stirling engine parabolic dish systems before and always thought they looked really impressive. Definitely installed in my dream off grid retirement place somewhere sunny.


Although quieter than ICEs, those engines are not silent. Too noisy to be close to homes. You would need to keep that in mind in your off grid dream.


What was the most expensive part of building that engine?


I don't know the details on the engine as we did not build it; it was sourced from Kokums.


According to Wikipedia, the Stirling engine is particularly well suited here because the engine is near silent and can use the surrounding sea water as a heat sink to increase efficiency.


In a similar vein, if you haven't watched a pop-pop boat circumnavigate your bathtub, you haven't lived.

https://en.wikipedia.org/wiki/Pop_pop_boat


Can anyone point to the reason(s) why Stirling Engines appear to be unobtainable for stationary power generation? I've been wanting a reasonably sized one (say 2-5kW) for quite some time to run off of woodgas, or other high temp heat sources. Size doesn't really matter, I just want something relatively quiet, that needs much less maintenance than an internal combustion engine.


If you're ok with using a parabolic mirror(s) as the heat source instead, then it should be doable. eg:

http://web.archive.org/web/20060508061432/http://solstice.cr...

They specifically mention systems that do 5kw, 25kw, etc (in the next pages).

---

That aside, there's a kickstarter project for a 1kw Stirling engine from about 6 years ago:

https://www.kickstarter.com/projects/672465444/low-cost-ster...

The "updates" tab there seems to have potentially useful info, though the project appears to have stopped updating there.

The project itself still seems to be iterating on new developments, just slowly.

http://volodesigns-sterlingproject.blogspot.com/

More recent video's of their development:

https://www.youtube.com/watch?v=p06HDOlxDMM

Mentions here they're intending on making the design OSS:

https://www.youtube.com/watch?v=IrXltsm-Ujs


The videos don't inspire much confidence in their engineering capabilities. Very rough construction and not pressurized.


The kickstarter project has transformed into Sefton Motors, which says it has engines backordered (but in production): https://seftonmotors.com/collections/all


They've certainly been available commercially at various points. If you search for microCHP, you'll find them. I can't see that any of the big manufacturers are currently making them, though. Honda and Baxi used to have residential stirling engine CHP boiler models that seem to be discontinued.

Currently, I think the Senertec Dachs might be closest to what you're after in size, although it's based on a conventional four-stroke, not a stirling engine. The residential CHP market seems to be shifting to fuel cells.


Heh, I actually came here to take advantage of this post to ask something similar: Could a Stirling engine be used to, say, charge a battery using the exhaust heat from an AC unit?


The exhaust temperature of the AC will be as low as is practically possible to remove a given amount of heat (assuming the AC is properly engineered). To extract energy from this exhaust you have the problem to remove heat from the cold end of your stirling engine while maintaining a lower temperature than the exhaust, which will prove difficult, given the assumption.

OTOH, if the exhaust of your AC is hotter than necessary (the unit wastes energy), then the best you could hope for is to recover some part of (depends on temperature differential) the wasted energy. Better get a better AC unit.


Unlikely that you could get much power the heat from an AC unit as the temps aren't that high. Perhaps you could run the AC's exhaust fans off of a low temp stirling.

You can uses a Stirling engine as a cryocooler, by running it in reverse, and cool your house itself directly. This is only really useful if you're trying to get to low temperatures.


I’m interested in reading about this


https://en.wikipedia.org/wiki/Cryocooler

I have two of the Twinbird FPSC Portable Freezers which utilize this technology, nice and quiet and sips electricity.


The energy generated is related to the difference in temperature. So if you're using AC out as hot and ambient as cold, the generation will be small, and it's necessarily much less than the energy spent to run the AC.


Now you've got me wondering if you could design a reverse-stirling engine that runs off an intensely endothermic reaction of some sort.


The engine only requires a "hot" side and a "cold" side, so there's no reason why the cold side can't be way below ambient temperature, such as an endothermic reaction or a tank of liquid nitrogen. Here's a video of a guy running one of those coffee-cup-top toy engines off liquid nitrogen: https://www.youtube.com/watch?v=O6eEvU5piQg

Some people have proposed this as yet another bulk energy storage system for the grid; giant underground cryotanks of liquid nitrogen. With careful design the same gear could be used in "forwards" or "reverse" mode. The limit is that the Carnot efficiency isn't great.


You could certainly run such an engine, very efficiently at that, from vaporisation of LNG.


A reverse stirling engine takes mechanical energy and produces a temperature differential. These kinf of setups are actually being used for quite efficient cooling.


It would still be a normal Stirling engine and not reverse. The engine only needs a large enough temperature differential. Where that comes from does not matter.


What about "Moves more energy than uses"? I mean those heat pumps which have COP over 2?


Yes, although obviously you can't have a perpetual motion machine; the Stirling engine also moves more energy from hot to cold than it can emit as useful work.

Both the Stirling engine and the AC unit are using the Carnot cycle (simplification), and going around the expansion-heat-transfer cycle in opposite directions.


As I understand, the isothermal transitions in the Stirling cycle need to happen relatively slowly to preserve efficiency, which means the cycle can only run at a low frequency, which generally results in a low power density.


WhisperGen have been making silent stirling cycle generators for many years. For technical details see the user manual for one of their products here: https://www.victronenergy.com/Manuals/WhisperGen/UserManual/...

(I have no connection with this company or its products)


WhisperGen are difficult to find in the States, I looked into it in the past and was quoted $12k for a boat version that ran off of diesel. These were also only producing 800w of electricity.

Talking to the folks that worked at WhisperGen, the wobble yoke design added extra complications and tolerance issues that added to the cost.


iirc panasonic had one generator. small scale internal combustion engines are popular because power to weight ratio allows the same design to be used in other equipment and it has near instantaneous power regulation.


Sadly finding a Philips MP1002C Stirling engine generator is almost impossible. Not to mention that they are loud and only output around 200w, which is nothing compared to a Honda 2200i.

A stand alone generator need not be mobile, so it needn't be that light weight. In fact it probably would benefit being integrated with a home's heating and cooling system to utilize waste heat.

What it does need to do is be relatively maintenance free, and run off a variety of heat sources (such as pellets, woodgas or solar thermal). This is where I think Stirling engines shine...you can gasify all of your waste organic matter into dirty gas that wouldn't normally be fit for an internal combustion engine, but work fine for a swirl burner which can be used on an external combustion engine such as steam or stirling.


A stirling engine might be a perfect part for a bare bones open hardware hybrid car.

A stirling engine might not produce enough power to power the wheels directly, but it could produce electricity quite efficiently at constant power output. It's fairly efficient, compact, relatively quiet and simple to build.

The electricity could be stored in fewer batteries and/or supercapacitors. The electricity from these would then power the wheels.





Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: