Hacker News new | past | comments | ask | show | jobs | submit login
Reinventing the Small Wind Turbine (lowtechmagazine.com)
167 points by Red_Tarsius 3 months ago | hide | past | web | favorite | 55 comments

My favorite Hacker News story of all time, about a guy who builds a small, water turbine power plant:



My favorite quote from the first article: "The varnish I used is intended for oven-drying at 140 to 150 degrees Celsius. I don't have an oven large enough to fit this coil, so I heated it by applying roughly 20A of direct current to the high voltage winding. I monitored the temperature by measuring the resistance change of the wire."

The solution belies a very deep understanding of the principles involved. Both the heating and the thermal regulation were improvised.

Good night I just wasted so much time...

And that makes two of us...

The article is called "reinventing the small wind turbine," by which I think they meant "redesigning from plastic/fiberglass to wood."

What I was hoping is that they'd talk about how common small-scale generation was a hundred years ago.

We're gradually getting back to where things were headed in the 1920's before the Rural Electrification Act essentially killed the local generation industry.

If you're interested, there's a relatively quick primer on that history here: http://www.windcharger.org/Wind_Charger/Farm_Electric.html

Not sure if I'm imagining things but I see this pattern of oscillating between distribution (small scale local) and integration (large remote). In computer we had a bit of the same (mainframe -> desktop -> servers -> smartphones -> consoles -> streaming from datacenter). It seems a 'natural' phenomenon of optimization.

This definitely happens in IT departments of large orgs. Centralization happens under the guise of efficiency / role specialization. That works for a while until clients at the edge can no longer get the service they used to get when their IT was a direct report in their org chart, and a decentralization effort ensues.

It also seems to be a function of the exec level change over. The new C level wants to make their mark by doing things differently from the last person.

Could be seen as an expression of dialectic development. The overall tendency in many (most?) living systems is towards integration at the next scale, which needs a certain amount of centralization. But whenever centralization is imposed it is usually overkill (the pendulum swings too far), which causes a decentralization effort (back to the autonomy of the parts), that however will retain some remnant of the previous centralization drive. At a certain point the decentralization swings too far too, and so centralization takes place again, in a slightly different way this time.

The combined effect of these oscillations is an equilibrium between centralization and decentralization of the organism which is optimally adapted to the environment at that scale of integration. Which is why neither centralization nor decentralization are ideals, they are both needed in an evolutionary process, albeit at different times.

I definitely see some pendulum effect scattered through the field.

Maybe there's a sensory about system pressure (need for more bandwidth, or whatever) that when above a certain threshold trigger that swing change.

There is a tension between complexity and order. A decentralized system like the internet is extraordinarily complex, and with that complexity comes robustness and resiliency. Sometimes people see complex systems and feel they’re messy or “inefficient,” and want to replace the complexity with something simple and orderly. Such systems tend to be centralized, easier to control and understand, at the cost of being less robust and resilient.

I also see two things:

- Investment barriers for innovation: you need something large to motivate funding and overcome research/innovation => centralized. When costs amortized, you start selling small and cheap.

- Maybe also a kind of inertia or fatigue. People will try to find ways to improve. When it's too hard to do so in centralized they'll try to distribute, or the other way around.

I remember a long time ago reading a comment about business schools basically teaching the same: you have a large company which is inefficient, so new management comes in and decides to split it into multiple companies that do different things; then at some point, each minor company is inefficient, so they merge with other companies to achieve scale benefits. Repeat forever.

Just about everything was small scale and decentralized a little over a century ago. Many small scale decentralized solutions don't work with a population of seven billion.

I'm being a bit pedantic here, but they are too, so let's consider:

They argue that wooden blades are less energy intensive to produce, which is technically true if you discount the amount of energy to grow the wood in the first place. What we're actually talking about is additional energy cost over the initial sunk costs of the wood (which don't really matter to us). This part is mostly trivia, but it ties into another point:

What are we actually worried about here? We're worried about unbalancing our ecosystem, and additional energy expenditure is assumed to not be supplied by renewable energy sources (which is generally true today). So our fundamental concern is the release of greenhouse gases (due to burning of fossil fuels in order generate energy for production).

The calculation then becomes: What is the cost in greenhouse gases of cutting down trees to produce turbines? Trees are carbon absorbents and carbon sinks, after all. What is the cost in greenhouse gases of feeding the people who expend their vital energy to produce the wooden turbines? How does that cost compare to production using robots? How does that cost compare to production using synthetic materials? Does that calculation change enough to matter as the source of energy moves more towards renewables?

What I'm getting at is: Where is the actual CO2 bottleneck in production? We should be optimizing the bottlenecks first.

This is kind of neat, but the fascination with wood blades is like wanting something to be "organic".

The article claims getting 30,000 KWh/year from a 10KW wind turbine. That's optimistic. Burgey, one of the larger makers of 10KW wind turbines, only claims 13,800 KWh/year. There's a standardized way of measuring this, with reasonable wind speed values. Few places have an average wind speed of 15 MPH near ground level. (And you don't want to live there.) That guy on the Scottish coast mentioned in the article might get that.

Compare a commercial 10KW wind turbine.[1] Costs about the same, about US$40K, plus installation and tower. 10 year warranty. (Wind speeds above 135mph void the warranty.) "Designed for high reliability, low maintenance, and automatic operation in adverse weather conditions." Which is the real problem with wind turbines - surviving adverse conditions.

[1] http://bergey.com/products/wind-turbines/10kw-bergey-excel

The last time I read about localized power on HN, storage came up, esp batteries and inverters.

The majority of appliances now assume always on, very clean, 60hz (or 50 or X) power, because that's what the turbines at the power company have always made.

What assumptions would have to change for appliances and human life to fit with localized generation? (Ie off-grid living)

Generally most consumer equipment in the last decade or two is pretty flexible on voltage. The modern switched power supplies are both flexible and efficient.

But generally equipment is AC, and solar and batteries are DC. So it's entirely up to your inverter how clean your power is.

But generally off the grid users have to either over build a crazy amount (which is expensive) or adapt a bit. Things like hanging laundry vs using a clothes dryer might not always be necessary, but if the wind/solar is behind it would be a good investment.

Even things like dressing for summer instead of blasting the AC so you can wear multiple layers can make a huge difference. If you have a surplus during the day (which is common with solar), run your most power intensive stuff then. That way your very expensive batteries don't have as much load to make it through the night.

Things like air conditioning are among the hardest loads, high peaks, worst when it's hot, and overall just a huge amount of power. Who wants to pay $40k for batteries instead of just being a bit smarter with insulation, whole house fans, and dressing for the weather? In my city I'm often out in about after midnight, with a light jacket on, and listening to neighbors AC blasting with all their windows closed while their AC fights the thermal mass in their attic. Seems insane to not open the windows when it's under 60F.

Generally minor changes. You can still stay up all night watching tv, still have a comfortable life pretty close to normal. But if you don't adapt you might be out of luck a few days a year or have to spend 2-3x the money.

All the things you mention would save 2-3x the money with a normal grid connection too, it's just that energy is much too cheap for people to care. Thus 10kW of continuous primary power consumption for each American.

Agreed. However on the grid running a dryer instead of hanging clothes isn't much difference in money to most Americans. Off the grid the difference in batteries+solar/wind is huge.

What Faak said. I’m living on off grid solar and my tech (TV, Xbox, NAS, LED lights etc) never misses a beat, even when a big load turns on. It’s the old tech (i.e. motors) that are actually a bit more fussy.

What about running out of power?

Well at that point everything turns off :-)

I’m still running the system in, so the (NiFe) batteries are not at full capacity yet (perhaps only at 30%-50% storage capacity). Even so I can easily stay up until 2am watching TV. And that’s only with 4 (25%) of the panels out (it is summer though, I’ve yet to do a winter.). Oh, and I’m running a fridge/freezer on that too.

Still, probably going to get a UPS for the NAS soon.

Edit: just to rabbit on a bit more - panels are cheap. They only accounted for around 20-25% of the total system cost. Power storage is the expensive part (more like 40% of the cost). It is interesting to try to tailor my systems to draw power at suitable times of day. At 2pm I have more power than I know what to do with, I could happily run 3kw load for a few hours most days around that time. At the batteries theoretical max capacity I have 10kWh stored, but I’m only getting around 3kWh at the moment.

I’ve got a backup generator to cover any extreme weeks in winter.

Figuring out the water pumping/irrigation has been a fun one I’ve been looking at today. The trees will have wood chip moats around them, a side effect of which is they can be watered at any time of day (which is great), but watering the grass must happen when it isn’t sunny (so maybe I can squeeze in a watering when the last rays are hitting the panels, or first thing in the morning).

As a software developer it’s an absolute dream.

extremely cool! How easy was it to disconnect from the grid? What's your city/county? Also, just curious, why did you decide on NiFe? Most off-grid folks I know either use li-ion for density or lead acid for cost/abundance/ease-of-use.

> At 2pm I have more power than I know what to do with, I could happily run 3kw load for a few hours most days around that time.

A fun and interesting problem to have. This is why we need so much more research into energy storage.

I've moved to rural Portugal (from London). I would have to get the grid extended to my place in order to get grid power, and I was quoted at around €30k. I realised that for a fraction of the price I could get a pretty beefy solar system (and – you know – free power thereafter).

I went for NiFe batteries because they they are basically impossible to break. This is my first system, and I wanted to take the reliable and stress-free path. Space wasn't a concern. NiFi batteries also last circa 40 years, so the price per year is actually very cheap. However, my planning horizon is definitely not that long.

Pumped hydro.

Pump the water into header tanks, then it doesn't need electricity to water the grass, and you could generate electricity too.

Or... Just don't water the grass?

> Pumped hydro.

Yes, I did some beer-matt calculations on this. I believe it is 10J to raise 1kg by 1m. So to raise 2000lt of water (2000kg) by 10m it would need 10J * 2000 * 10 = 200,000J. Call it 0.5MJ to take possible losses into account. A kWh is 3.6MJ. Which basically means I would get 0.14kWh storage as a reward for building a 10m high tower which can support 2 tonnes.

When my batteries can store 10kWh (in theory), it just doesn't seem to make sense to me.

Of course if I had a lake on the top of a hill that could be a different matter. It would also be a different matter if I didn't need 10m head (1bar). Sprinklers do require pressure though.

> Or... Just don't water the grass?

A good point. There are two reasons I do this: 1) I'm from the UK, green grass feels homely to me, 2) forest fires are an actual part of life here, and stopping the area around one's house being flamable is important. Paving, gravelling, tarmacking, and watering are all ways to deal with this.

Yeah, I wasn't being entirely serious with the pumped hydro, nice in theory... But if you've done the rest of the infrastructure for it.

What about seeper hoses, or have you looked into hydraulic ram pumps? That might give you enough pressure to drive a sprinkler.

I wonder how viable is to cool down some thermal mass? (similar to HVAC systems taking advantage of cheap electricity at night). Fridge can be "pre-cooled" and cut power usage at night.

Joey Hess (the git-annex developer, also lives off-grid) has done a lot of experimentation with his fridge, including adding a thermal mass: https://fridge0.branchable.com/thermal_mass/

That is very interesting to me, thank you!

In winter I turn up the max temperature of my water heater. This way I can soak more energy into it when it's "off-peak" (cheaper) hours for electricity. You just have to be careful not opening the tap on the hottest temperatures anymore.

I don't see how you picture it with the fridge though. Aren't you limited by the fact that you can't go down much below 0°C ?

Or perhaps you add some new mass? Where do you actually put this mass when the fridge is already full?

Right with fridge there is not much wiggle room - but each fridge contains freezer compartment?

The good news is that sine wave inverters are no longer expensive in the context of an off grid system. The bad news is that they seem to idle at about 3% of maximum power capacity. This can easily be your largest energy consumer if you build for infrequent substantial loads.

Not really true.

Nearly all of new (i.e. Post 2000) switch-mode power supplies accept a wide range of voltage (100V-250V), and a wide range of frequency (even 0Hz, i.e. DC current would work).

Frequency is irrelevant. They would run on DC or 400Hz or 10KHz just fine - they all rectify the supplied current first. Frequency is only limited by the bridge diodes.

And 100V is written on them only because there are no countries with lower than 100V standard. I'll bet a crap chinese usb charger will work off 48VDC just fine.

Converters in universal power supplies tend to fail around 75-85Vdc even at low loads and usually can't deliver full current at 90Vac, in my experience (was using universal power supplies in electric cars; okay for 96V+ cars, but unusuable for 72V and below). 48Vdc is not a reasonable expectation, unfortunately.

The ripple filters will be tuned for a specific frequency.

My family have a 5kw turbine (UK). About 12,000kw a year. £30,000, with the help of heavy government subsidies it makes about 7% return a year, including loss of capital, if it lasts twenty years. As soon as subsidies ended industry shut down. Without subsidies the servicing and repair costs wipe out most of the savings leaving with you with a few hundred a year if lucky. I don't think anyone is going to buy another small one any time soon... Low maintenance solar has won, the wind turbines firms all closed themselves or went bust.

What would transform it? If someone like Honda mass produced cheap turbines with Honda reliability with a proper eco system of servicing, parts and repair knowledge. That would make a big difference. Changing the adequate fibreglass blades for wood? Not really. I can't see getting a hand carved blade made being a cheap repair. Government caused boom and bust is not helpful.

And obviously cheap batteries but that would transform everything too...

My favorite open source hardware project: https://www.otherpower.com

If you're into this stuff look at https://www.otherpower.com/, run by Dan&Dan, two windmill enthusiasts from Ft. Collins, Colorado (They funded the forum and all of their development through the sale of Neodymium magnets!).

Small wind turbines are - when done properly - some of the most complex mechanical devices that you can still make at home and they will serve you well if you are located off-grid. On grid they don't make much sense, neither economically nor from a safety or reliability perspective.

Also keep in mind that you can't just rely on wind or solar for your energy generation, you will need some of both, a battery pack for storage and an inverter to connect to your distribution panel if you intend to power your house.

A practical minimum power level for a home is somewhere between 1 and 2 KW, but you're going to have to forego some luxuries.

lol, some guy called Phil wrote in the comments:

If you hand carve a blade, you still have to get energy from eating food to power the human. So calling it "zero" energy used is bad accounting/ misleading.

Someone called whateveryoneisthinking wrote:

shut up Phil

All comments are moderated...

I was just researching wind turbines for my off grid system today, and then this comes up!

I had been concluding that small wind turbines probably were not worth it. But this addresses some environmental concerns and also makes it into a really nice project (even if it still isn’t too efficient).

First of all, do not rely on that site to provide you accurate, unbiased advice. They have an agenda, and they are pushing it.

Second, there is a huge difference between using small wind turbines in the city, where the alternative is reliable and cheap grid power, and off grid, where the alternative is increased use of expensive and unreliable small diesel generator.

Whether a small wind turbine is worthwhile for an off grid system depends on lots of things, primarily on availability of solar, availability of wind an your usage pattern.

Everyone has an agenda, that doesn't mean you can discount what they say.

Now if you want to point out which bits are inaccurate? Then we can have a discussion on the merits.

>Many commercially available small wind turbines with plastic blades and steel towers are infamous for their low reliability, high embodied energy, and limited power output.

>Building them out of wood addresses these issues.

Well, wood isn't going to improve the reliability or power output. It can be significantly less embodied energy per blade, depending on construction however. A blade hand carved from a single piece of air dried timber has a lot less embodied energy than a blade made from laminated wood covered in resin, which in turn has less embodied energy than a fiberglass blade, though it isn't as big of a jump in percentage terms than it is from air dried timber to laminated timber.

That said, as you go smaller, the blade itself is less of the embodied energy of the wind turbine. On the small turbines that used to be common on yachts, the blades are really a minor part of the construction, the generator and the pole weigh in far higher. Also, given that turbines are a swept area game, I suspect that using hand carved wooden blades only saves you anything in situations where you cannot have larger blades. Also, they will tend to be less efficient blades in terms of power conversion due to the limits of hand carving.

If you are interested in the best thing environmentally, clubbing together with others to install the largest turbine you can is still generally the way forward. If that is not possible then I'd say hand carved wooden blades do have a place on small tubines, though I am doubtful it is that environmentally beneficial, unless you have scavenged the other parts. Scavenging the parts for small turbines is perfectly doable though. Personally, if I was scavenging already, I would be also looking at cutting out blades from scrap metal sheet.

edit - blades from scrap metal being a hell of a lot quicker to fabricate than hand carving wood.

You don't have to hand carve blades. As far as I understand it wooden propellers were pretty close to optimal, I don't see why turbine blades would be different

The article does mention one turbine optimised to run at lower speeds, which might be the increased reliability mentioned, but no it doesn't explain the power output. I'd tend to put that down to poor writing / and artifact of a previous draft, so I still don't think it's enough to write off the entire article, website and writer.

The author's article on small turbines not generating their own embodied energy was what first made me aware of that 'issue'. This is an attempted fix. I agree with you that the best solution is to go bigger, still the proposed options, particularly the pole seem sensible even on a village sized turbine.

Definitely do not write them off. I find the site useful and check in now and again, but I read it with a very large pinch of salt.

True enough, I would recommend the same for any publication though.

The main issue with that site, is that every time numbers are involved they are wrong, often by an order of magnitude... Here one wildly optimistic number is kWh/year on the 10kW turbine, but I suspect the energy cost of building a fiberglass/epoxy blade is wildly off too...

"According to the manufacturer...... produce roughly 30,000 kWh of electricity per year"

That works out to a capacity factor of 35% which doesn't seem shockingly high. https://cleantechnica.com/2012/07/27/wind-turbine-net-capaci...

This is really amazing. I wish these were more practical on single-family homes. I don't think my bay area city would ever allow me to put one up, so it's solar.

I love this. Building a small wind turbine from as scratch as possible has been a fantasy project of mine for years.

Why fiberglass plastic and not steel?

Weight. In order to start the turbine spinning at the lowest possible wind speed (meaning it will spin more often) it helps for the blades to be as light as possible. There are other factors involved in getting the blades to spin, but that is an important one. (other factors include the size of the magnets in the coil, the shape and length of the blades, whether it is initially pointing into the wind or not - and how much it takes to turn it into that breeze - and so on).

Applications are open for YC Winter 2020

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