Hacker News Reader

Is this really that important though? I haven't numbers that indicate that farming machinery is such a big share of the oil consumption.

My thoughts on these things is, if we can cut 90% of gasoline/diesel use with BEVs and PHEVs, then the last 10% could feasibly be fueld by renewable hydrocarbons. The development after that is up to the markets.. probably pure BEV will be preferred over time as batteries get better and we develop new solutions.

Why would you need three to five packs? A pack can generally charge as fast as it can discharge.. you should only need two if you have a fast charging system.

The problem is if the machines is only used part of the year. For the cost of a battery pack to make sense, it should be used as much as possible. If it's idle for large parts of the year and it can't be used for anything else, it'll be hard to justify the cost. Maybe they can double as grid storage?

Farm machinery is not a huge part of agricultural energy use, and agriculture is not a huge contributor overall consumption, but energy's share in food production costs are high.

Petroleum use by industry sector: https://www.eia.gov/energyexplained/index.php?page=us_energy...

How energy is used in agriculture (Sweden): http://www.serina-ahlgren.com/p/research_26.html

How agriculture energy use contributes to the atmosphere: https://www.globalagriculture.org/report-topics/climate-and-...

(I am not an expert in this field.)

Edit: an adjective form

You also need a third in case one of the first two goes pop, and because batteries don't like being left on the shelf fully charged (or discharged), it needs to be actively cycled through. That's N+1 redundancy; N+2 means four in total assuming you can charge as fast as you can discharge.

This may be a stupid idea, but is there any reason that this kind of heavy machinery needs a battery and couldn't just be connected to a power source via a (admittedly, very long) cable?

The very long cable sounds like a good enough reason to me ...

Agricultural land is typically in remote areas, and equipment needs to be driven for miles often to get to where it's being used.

Even if, lets say you do take the hit of the upfront costs introduce a heavy-duty power-grid to service all your land (thousands of hectares on many of the large-scale cases) - then you're still going to have to run the cable from some kind of power point, and without some aerial apparatus you're going to be running it across and through your crops.

Yeah I guess you could take the time to arrange the cable, and have somebody follow it around all day but that's a lot of extra work. Probably a lot more work than just the harvesting.

It's probably pretty dangerous too - trailing this multiphase high-power cable around where you've all this heavy equipment often doing chopping and grinding.

Someone needs to run the numbers for a modular chain of wire-carrying drones: if you get rid of the batteries, what length of cable could a drone carry? How much power would that cable be able to transmit, how much of that power would be consumed by the cable-carrying drones? Just how light could we make the power electronics to feed the drones from the (obviously, for weight reasons) high voltage power lines they would be lifting?

If this is possible at all, it could be an important base technology for a very wide range of new application fields.

That sounds crazy.

The kind of heavy duty cabling you're talking about would be very very heavy.

Now, you're not only talking about powering your combine harvester or whatever but also an army of heavy duty drones?

LOL

Does it need to be drones? Would it be possible to use a very large cable system, similar to those TV cameras in football stadiums?

I like the idea, might be the best method to provide power to a consumer moving on a defined 2D surface.

Good idea. If we could control those drones via a blockchain-based AI, it could be a killer product!

Is there an app for that?

We’ll write one in React Native.

Or a power rail. Fields are already very regular, I would think it possible to embed a power rail somewhere along or across it.

Power rails would be way too expensive and interfere with cultivation.

Very long cables at normal supply voltages (240V/380V) suffer too much voltage drop over any distance (say) >150m. Remember that we're talking of multi-kilometre distances, not a mere hectare or two.

Not only that, but anyone who has used an electric lawnmower on a larger area will tell you that cables are a fucking hazard to navigation when you're trying to move a machine trailing a cable. They constantly get tangled, driven over and cut.

So anything at a reasonable voltage to run heavy machinery at a long distance from the nearest supply would need to be... I don't accurately know -- ask an electrical engineer for a definitive answer, but... something like 1.1kV. Cable that's sufficiently well insulated for moderately high voltage is /very/ expensive and deadly if the insulation gets compromised. Doesn't sound very feasible to me. Maybe someone with more EE knowledge can prove me wrong -- this is just my gut-feel conjecture. I got kicked out of the EE program at university for being bored to tears by this sort of stuff.

We run trains over multiple kilometres with overhead wires why can’t we do it with tractors???

>why can’t we do it with tractors???

Trains run in straight lines, farm equipment has various widths and it's not just 'tractors' and you're going up and down rows of farmland that might be tens or hundreds of acres wide and that aren't always in the shape of a square or rectangle (it's often far easier to work around large glacial boulders or trees for example, as well as ditches/running water/property lines etc). You'd have to string millions and millions, if not billions and billions, of miles of electrical cable across fields.

There's 915 million acres of active farmland in the U.S., that's 1.4 million square miles/3.7 million square kilometers. The larger combines can be about 40ft wide so if every scrap of farmland was in literally-square acre parcels you're going to need 5.2 power lines minimum to cover the width of a literally-square acre so you're easily talking 5 million miles of power lines.

You could, but it would be cost prohibitive in a real economic sense.

You run overhead wires where you know you will repeatedly have a large demand for power, and you pay for it once in the capital outlay to put up the wires and other scaffolding.

In other words, you want to do this where (average demand / (fixed capital cost (which correlates to peak demand)) is relatively high. The middle of the farm field is the exact opposite of that: power demand is bursty and very infrequent, so utilization will be a few orders of magnitude lower.

Trains are linear. Tractors and harvesters are areal.

You square the delivery line requirement.

A 1,000 km electric traction rail line needs 1,000 km of cable. (Or 2,000 km for a two-tracked system.)

A 1,000 km * 1,000 km area of farmland with power gridded every 100m has 10k * 10k or 100,000,000 km of delivery cable.

Plenty of sibling comments spelling out why not.

I want to add that for farm equipment, having a big weatherized battery cube and a crane to swap it is practical. For automobiles, less so.

Not a stupid idea, but it requires whole approach to farming, together with machinery to be reinvented from the ground up. Agrokruh is one example of electrically driven machinery which was already proven to be energetically efficient and profitable: https://vimeo.com/99343531

“Large-scale” is not necessarily large!

Instead of tilling/planting/harvesting 24 rows at a time, how about a fleet of mule-sized drones running one row each, deployed from a semi with a power source, swapping as needed?

I'm occasionally boggling that there's no electric farm equipment, since it at least suffers from one fewer problem than electric cars: you're mostly driving around in a finite area. You might need to plumb a dozen or so charging stations at various fields, and have the equivalent ability to essentially carry a five gallon jug of electricity to a tractor in a field when you run out accidentally, but you're not driving 200 miles to an Ikea on the spur of the moment.

Besides the lack of consumer demand (farmers have other things to worry about), I think the other thing that makes the farm market slow to adopt electric is that the market is, erm, slow. People are constantly buying new cars. Some people drive a new car every year, some people drive a car for 10 or 20 years, but I'd guess that most people get a new car every 3-7 years. That means, firstly, that a lot of people have bought a new car since they first started getting good; that also means that an electric car only has to last 3-7 years to meet most people's expectations, and your car will only be 3-7 years out of date when you get a new one.

I would guess the average farm tractor in the fields today in the US is 20-30 years old. You've got some really new ones with GPS, autopilot and satellite TV, sure, but you've also got some 50-75 year old tractors which are still doing just fine. I used to rake hay with a little Farm-All my great grandfather bought back in his prime, which is still going strong today, and I bought a used Cletrac from '45 to use around my woodlot.

This creates three problems for the prospective electric tractor manufactures: not a lot of tractors are sold each year, so you're fighting for a chunk of a very slow-moving market; prospective buyers are going to be asking themselves "will this tractor outlive me?" so you need to have a compelling story on how those battery packs and motor bearings are going to last 50 years, or be easily replaceable; the technology is still rapidly developing, so you also need to explain to the prospective buyer why it's is worth buying an electric tractor now, instead of waiting 10 years to buy a more mature electric tractor.

> I would guess the average farm tractor in the fields today in the US is 20-30 years old.

It's actually surprisingly hard to find hard numbers on this topic, but the numbers I've seen have generally suggested that the average age is more like 10-20 years old. The hard numbers I've seen do say that about 10-20% of tractors are less than 5 years old.

Realistically, the difficult part of electric tractors isn't the mileage but rather the fact that you're consuming far more power per mile. You're looking at measuring fuel consumption of around 10 gallons an hour, instead of about 1.5 gallons an hour for a car. Also, you're going to want to work the tractor for as much sunlight as you can muster, and you want to charge back up only overnight. So you'd want an equivalent not of a 200 mile range car but more like a 2500 mile range car, that can fully charge overnight.

Eh, not that surprising: you don't need to register a tractor the way you do a car, so there's no central registry of what tractors are still being used.

Re: energy storage, oof, yeah, at 146.5 MJ / gallon for diesel fuel, and 10 gallons / hour, 12 hours of fuel holds 4884 kWh of energy. Assuming you need a similarly sized battery, you're looking at 60 Tesla (car) batteries. That will weigh approximately 36 tons and cost about $700,000, based on Tesla's 141$/kWh stated cost.

Hmmm, I'm guessing you can't put more than 8-10 tons of battery on a tractor that size (the John Deere 7250, which is actually 13g/hr, weighs 13 tons) so you're limited to maybe 3 hours or so of power between charges or swapping out your batteries. For that much battery, you're looking at maybe $175,000; the John Deere 7250 retailed for ~$230,000. Assuming your batteries are swappable and fast-chargeable, you'd need at least two batteries. (Heck, it's a tractor, maybe it can be on some sort of wagon you tow behind, which would make swapping a snap.)

Putting it together, I think we might actually be approaching viability for electric tractors. If you can actually fit 8-10 tons of battery into a 13 ton tractor and build the rest of the tractor for $55,000, you could build a tractor comparable to the John Deere 7250 that would run for about 3 hours; that's not a bad amount of time to run between swapping out the batteries, but you'd need at least another $175,000 battery to swap out, which already makes you ~twice as expensive as a regular tractor. You would probably need the battery weight and cost to drop by half again before you could get a tractor that would run all day (with a battery swap) and not cost more than a conventional tractor.

> Heck, it's a tractor, maybe it can be on some sort of wagon you tow behind, which would make swapping a snap.

Well, if you're towing something behind you on a tractor, you can't use the power take-off anymore, which kind of ruins the point of the tractor for most uses.

Or the three point hitch, or the...

You might be able to set up a convoluted system where you have the power cells being dragged behind whatever equipment you're using, assuming you can get enough tires underneath to not undo whatever work you're doing to begin with (first we plow the field, then we pack it down tight!)

But then you also can't use the ~10 tons to provide extra weight on your wheels.

It just feels like you ought to be able to store your batteries in a way which makes swapping them/carting them around easier, if you don't need to have a sleek aerodynamic package that fits on a highway. Actually requiring a crane to swap your batteries adds a lot of overhead, along with room for comic fuckups.

Agricultural machinery is likely to keep using liquid hydrocarbon fuel even when it's being banned entirely from cities. It's just too much of a technological challenge, as you say.

My forecast is that heavy diesel engines (excavators, farm equipment, ships, maybe trains) will be replaced by hydrogen power. Either fuel cells in the 100-800 kW class, or gas turbines for the megawatt to tens of megawatt applications. For the latter, you might need to blend hydrogen with e.g. ammonia to slow down combustion, but we'll get there.

Everything is easier with methane and even compressed natural gas has better storage density than liquid hydrogen (liquefied natural gas destroys hydrogen in energy per unit volume).

I've read somewhere that bio fuels make a lot of sense for powering agricultural machinery, while they'd be disastrous if we tried to power all cars with them.

Asking out of genuine curiosity, does it actually matter if harvesting takes 5 or 10 days? If electricity is cheaper, it might make sense for the job to take a bit longer but be cheaper? Human labour is not involved (with automated vehicles) so price should only be dependent on actual power consumption.

Yes it most certainly does matter! A crop that's harvested even a few days too late is likely to suffer severe nutrient and quality losses.

eta: Of course this varies per crop, so let me back down just a little, but for most food crops there really is a optimal window outside of which the farmers' rather meagre profits evaporate quite fast.

> does it actually matter if harvesting takes 5 or 10 days?

Crops are perishable, and there's always the risk of weather conditions damaging them while they're in the field.

How much it matters will depend on the crop, but it's definitely important.

Yes, something like letting your crops get rained on just before harvest can ruin them and make them rot in storage. Although purposefully planting crops with slightly different yield times could alleviate the problem. Currently though crop harvests are a huge rush to get all of it collected as soon as possible because it all tends to ripen around the same time, if a few plants ripened or finished it usually causes surrounding plants of the same species to ripen too.

I've been wondering recently if there's some way we could "hijack" this sort of signaling mechanisms that plants seem to have... E.g. artificiall encourage the crops to ripen sooner (i.e. when we want them to).

The next step would be, to figure out how to grow just the parts of the plants that we need. I'm guessing internally, plants use some kind of hormone-like signalling mechanism that lets different parts of the plant know what to do. Why do we need a full tree (in particular, bark) when we could just grow apples directly along with a few leaves to "power" them?

I was thinking about this while driving Wednesday. I'm in a strength sport and have lots of friends in strength sports and a recent conversation with someone made me think "hmmm, we need to figure out the AAS for crops, what's the test-e dbol equivalent".

I think the problem is though, plants are pretty damn finicky and have relatively narrow conditions that will result in proper growth and fruit/vegetable production. The tiniest soil imbalance, too much sun, too little sun, too much water, too little water and blah.

The more important task at hand right now though is proper land management, we're quickly headed for another dust bowl, we are heavily relying on groundwater and depleting groundwater sources here in the U.S. (and especially in countries like India) and farmers are killing themselves at record rates.

We're about to be up shit-creek with a hole in our canoe in the next decade or two.

> plants [...] have relatively narrow conditions

Are you sure about that? Evolution would certainly strive for adaptability... Of course, evolution went through the window when humans started interfering (breeding etc), but still, just looking outside, plants seem to survive (and thrive) in relatively uncontrolled environmental conditions with wild weather swings and varying care (fertilizers, weeding, etc).

On the other hand, I can hardly keep a house plant alive, so there's that.

>Are you sure about that?

I see you've never gardened, or even had a house plant.

Harvesting absolutely matters about time. 1-2 days can make a difference and many farms don't actually own the combines, there are companies (often families) that have fleets of combines and go from farm to farm during harvest season harvesting from sun up til sun down and moving on to the next client as quickly as possible not unlike the companies that do the same with bee hives.

Combines are EXPENSIVE, you can easily drop 600k USD on ONE combine and a class 9 (over 500 hp, basically the biggest) combine with a skilled operator will harvest 10-15 acres an hour depending on the crop, quality of the field, soil moisture etc. With over 900 million acres of active farmland in the U.S. alone...

Transport uses a lot more fuel than farm equipment.

Biodiesel from sawgrass on marginal land or waste product like corn husks etc would work out a lot simpler for powering the equipment, and can be made on site.

Hydrocarbons are so useful, eventually we will generate them using electricity.