The Economics of the Tesla Semi | Hacker News Reader

The Economics of the Tesla Semi

torquenews.com

32 pointscoolmike80 comments

For the calculations:

A semi (33T, 25T of load) is consuming 35l/100km.

In the given calculations, it is consuming 47l/100km (35% more than the reality).

The efficiency of the truck is given as 2kWh/mile, which means 125kWh/100km.

A Tesla S requires 25,8kWh/100km. Basically, moving 33T takes just 5 times more energy than moving a Tesla S. For the "Diesel" case, this is more 7 times, but maybe the power scaling of electrical propulsion is better (and a better wind efficiency of this truck).

The last point is that the rate of 0.07ct/kWh is maybe too low, you want high power voltage to load your semi, this costs more than normal power.

Well done EVs will be cheaper, but these calculations feel a bit too biased into making the Tesla semi unbeatable.

I agree, there's no reason to think large haul electric trucks will have a better efficiency margin over internal combustion variants than in the case of passenger vehicles. On the contrary, there's reasons to doubt it, for example large diesel engines are highly optimized for efficiency unlike small consumer cars they are bought by businesses which think in TCO.

So probably the real margin over diesel is about 50% cheaper in fueling. But you have to take into account route restrictions due to limited range, time lost charging etc. Not to say it won't find its place in the market, but the margin and advantages of electrics are not yet so clear cut, it will take decades to reach dominance.

Also the maintenance TCO at scale. A large fleet operator has a staff of experienced diesel mechanics. Is Tesla planning on making a profit on servicing battery packs for these things?

Battery packs are not designed to be serviced, I would imagine that a Semi has like 4-6 packs and each one has to be fully replaced. The same thing goes for the engine, you just remove the whole engine assembly and put in a new one if one breaks.

You are only taking into account the negatives and ignore the positives. Electric trucks will have cheaper maintenance. TCO is much better in pretty much all electric vehicle and every study looking at vans in cities has shown this. Semis will most likely show the same thing.

Such semis will also be faster exhilarating and more maneuverable and be a much more pleasant vehicle to drive. All those things are worth something as well.

Of course they will not dominate from day 1 simply because can't produce enough but they will eat away at ICE market share every year for a long time.

By the time they can produce enough of them to take on long haul routes, batteries will be 2-3 generation further along and the economics will tell.

> you want high power voltage to load your semi, this costs more than normal power

Does it cost more when you're buying huge amounts? If there's a difference I'd expect that to be cheaper since the power company doesn't have to transform it down.

Correct eventually.

But in today's world there are plenty of companies wanting to charge you a 50x markup on power electronics for a truck charge station just because they're novel. Charge stations will end up costing more than the truck (per charge bay) to begin with.

Oh I'm not worried about charge stations, I'm asking about the big fat electrical supply.

It costs more, because you need a bigger bursty supply behind, bigger hardware, etc. You have "some" economy of scale but not fully. This is why for example, you have many charging stations who are now coupled with batteries, to be able to smooth the load on the grid. At the end you get:

grid -> battery pack at the charging station -> battery of the vehicule

Costs more than what?

I thought we were comparing a big bursty supply at 120 volts and a big bursty supply at higher voltage.

Did I understand your initial comment wrong and you weren't talking about voltage at all?

>The Tesla semi, which is an electric truck, can take that same load and move it 200 miles.

What hasn't been discussed here is weight limits. It won't be possible to transport the "same load" if the semi is using a lot of its weight budget carrying a battery vs a diesel truck. The U.S. federal road gross weight limit is 80,000 pounds (20k per axel).

For reference, a Tesla Model Y's battery weighs 4,416 lbs

It seems like they'll be limited to significantly smaller loads and require more trips or drivers to move the same amount of stuff, unless there's another revolution is battery energy density.

The solution is more axels. Haven’t you seen cement trucks with the deployable 3rd axel?

And the weight limit is 160,000. You forgot the 2nd trailer

Even so, diesel with a 2nd trailer >> electric with a 2nd trailer, which changes the economics in a non-trivial way. There's also the question about how to pay for road wear, which is currently subsidised by consumer vehicles, as the diesel tax falls far short of paying for itself, while there's no tax infrastructure for electric yet — that will also impact the economics of this.

I'm a full proponent of electric vehicles, but full electric doesn't seem quite ready for semis. Perhaps hybrid is the way to go vs full electric, at least until the battery tech catches up.

I don’t know what it costs, but adding an axle doesn’t come for free.

If it did, those axles would always be on the ground. So-called “lift axles” can be raised because keeping multiple sets of wheels on the ground makes cornering harder, increasing tyre wear and energy usage.

Exactly. These calculations make it sound like propelling the vehicle itself from A to B is the goal. Nope - the goal is to transport goods from A to B within practical size and weight limits/restrictions. If the volume and weight capacity are cut in half, that needs to be considered. Not to mention significantly increased wear and tear on the tires and the road infrastructure, and the impact of the mining required to make the battery.

Has anyone done a similar calculation for a real-world scenario?

An entire Tesla Model YT weighs 4,416 lbs, the battery alone is closer to 1000 lbs

I'm kind of skeptical of the "This is what it costs to haul with Electric vs Diesel" because the difference is that I can call a guy up tomorrow and he'll have a diesel truck outside my house. A lot of the hand wavey "Once we have mega chargers, solar panels etc." Well yeah sure, and it'll all be free once we hae nuclear fusion. I'm just sceptical because I've worked at companies where out next product will beat our competitors previous generation. It's a stupid game to play. Also, doing this in the Ukraine war, well... might not be the best time to extrapolate. I'm pretty sure that electric is the answer long term whether it's directly economically beneficial right now or not. But this isn't good maths.

Why would you being able to have a diesel truck outside your house affect the cost?

Once we have mega chargers, solar panels etc

Who is saying that? There are already 250 kilowatt super chargers out there now.

https://cleantechnica.com/2019/07/01/teslas-nanochargers-del...

I'm just sceptical (skeptical) because I've worked at companies where out next product will beat our competitors previous generation

What does that have to do with anything? You think diesel trucks are going to leapfrog electric trucks in efficiency by the time electric semis make it to market?

>Why would you being able to have a diesel truck outside your house affect the cost?

Because the electric truck has an undefined cost, because it doesn't exist, and can't ship any goods.

If they're supposed to ship this year, how certain are you that they don't exist and don't have a cost?

Because they're supposed to have shipped every year since 2019.

Were they promising to ship within the next hundred days before?

> Why would you being able to have a diesel truck outside your house affect the cost?

It affects opex like crazy. Tesla service has been problematic since the very beginning. Parts are highly available for diesels. Being able to have a diesel guy outside your house means you don’t have to worry about Teslas customer service disappearing you for a few weeks or months at a time.

I think the point of these Semis is for delivery from a facility (Pepsi/FedEx). The facilities have the Mega chargers (several already built) that the trucks return to.

Another big part of the win is maintenance costs which are also noticeably lower.

This article is a typo-ridden rewording of this tweet thread: https://twitter.com/alex_gayer/status/1557828064347693057

The author is also a Tesla bull.

Besides the range economics, if the trucks come with automatic emergency braking, lane following, alerts when cars slow, and even a rudimentary self-driving functionality, the insurance might end up being cheaper once the rates of accidents with these trucks is compared to other trucks. And if trucking companies can more easily recruit people who want to drive these trucks, that will also help the shipping / trucking companies as well. Hard to say if the maintenance costs will be lower or just different (if you end up needing to periodically replace battery packs for example) -- but it definitely seems likely that many trucking companies would be interested in trialing these to see how the economics work out. Another detail that could be interesting would be if you own a chunk of land in basically the middle of nowhere but has an interstate going past it. You should buy that land, build a huge solar farm there, put a giant charger station next to the road. You literally are not even connected to the main grid at all, just a little power oasis in the middle of the desert for example for trucks to charge / power some convenience store / maybe even a small hotel.

The significantly higher battery requirements for trucks means that the environmental impact of mineral extraction HAS to be taken into account when putting together these numbers.

> Fuel ends up being about $4.99.

Using all-time-high anomalous pricing as reference cost is not particularly useful.

> in the EU, electric semi trucks are allowed to be 2 tonnes (4,400 lbs) heavier than diesel equivalents, while in the US the allowance is 0.9 tonnes (2,000 lbs). (hard to find exact numbers)

Note that 2x the weight = ~~4x~~ (correction: 16x!) the road wear. This might lead to significantly higher road-maintenance costs.

Besides, this is an apples to oranges comparison. A cheap hybrid setup gives you all the pros of regenerative braking / brake wear. Hydrogen seems to be a much better match for the fast refueling, low road wear and low-pollution constraints of future trucking. Lastly, when your car is on the road 24x7, battery wear is accelerated by a staggering amount and all-season guarantees for battery range become important. Neither of these concerns seem to be taken into account in this analysis.

------- one possible solution-----------------

> There are trials of overhead lines (similar to trains) for trucks on highways. That would greatly decrease the need for energy storage and high speed charging.

Further down the thread, @nicoburns brought this up. A street-car-esque solution would address a lot of the above concerns. (weight, battery capacity, recharge times).

> Note that 2x the weight = 4x the road wear.

I thought it was 2x the weight = 16x the road wear.

Thanks for the correction

I thought the scaling was squared, but you are right ! It is polynomial to the order of 4.

It is 16X ! the road wear. That is very significant, given that base numbers for a heavy truck run 24x7 already quite high.

sources:

[1] https://www.insidescience.org/news/how-much-damage-do-heavy-...

[2] https://3kpnuxym9k04c8ilz2quku1czd-wpengine.netdna-ssl.com/w...

> For example, while a truck axle carrying 18,000 pounds is only 9 times heavier than a 2,000-pound automobile axle, it does 5,000 times more damage.

[3] https://www.gao.gov/products/109954

How does regenerative breaking fair on s multi axle truck & trailer, where traditionally each axle has breaks.

I guess my two main questions are:

* What is the economic impact of trucks needing to stop to charge for X number of hours versus traditional refueling?

* Who - the truck driver or the company they work for - is typically responsible for the capex in purchasing the truck?

Trucks, at least in Europe, are already required to stop every X number of hours for regulated rest. If they time it, so that they charge at the same time - there'll be no economic impact.

Also as I understand it, the Semis aren't really a long haul type of truck, so that won't happen as much. (feel free to correct me, HN)

Rest hours are also required in the US. However, trucks will stop almost anywhere - rest areas, off the the side of highway on-ramps, Walmart, etc. Now the trucks will have to find charging stations to rest, which will certainly impact the routes.

That particular problem is pretty much solved historically: at specific distances between critical logistic locations sleeping spot, chargers, food, mechanic shops etc. will pop up due to an obvious and predictable market pressure. There are whole small cities that exist because they are right distance between actually important places.

Yes, on average every 8 hours, so battery capacity needs to be at least that (probably by a bit of a good margin). And then just the rest stops need to be retrofitted with 100s of charging points that charge this capacity in time (and all at the same time). Semis in Europe are used for long haul.

This will probably replace the last mile delivery for the foreseeable future, baring some drastic shift of cargo to rail.

There are trials of overhead lines (similar to trains) for trucks on highways. That would greatly decrease the need for energy storage and high speed charging.

Yeah, a (relatively) weird mesh of train and trucking technology - for sure will have it's uses, but I would guess for most of the long freight trains are then a better option?

How difficult would it be to use interchangeable/hot-swappable batteries? Stations could just be constantly charging banks of batteries. Would make refueling much faster. Also could make repairs easier as you could take batteries out of service without need to take the truck off the road.

I’m assuming we are talking supercharging stations

If EVs became the primary automobile of drivers, and also became the primary semi-truck form on the road, the economics change significantly. When there's demand for charging from an additional tens of millions or hundreds of millions of vehicles in the US alone, my guess is that prices for electricity are going to increase significantly. Especially prices for any type of fast-charging capability while on the road.

Curious to see what this means in 10-20 years when comparing costs between EVs and ICEs.

That being said, long-term maintenance costs, reduction in road noise, etc. are all a win for EVs.

True, but that also heavily incentivizes getting solar where possible, as using that free energy to also power your car makes the investment pay off quite a bit quicker, after which you'll be enjoying almost free energy for most of your needs.

I don't think it'll play out this way. There's no new tech needed to scale up recharge stations, and they can continue to scale up linearly with demand, so you'd expect the supply will increase to meet demand. Fast-charging stations are most needed near highways where land value isn't too expensive so I think you'll see a lot of solar local to the recharge stations too.

I agree that the supply will increase long term, but power generation capacity tends to have a fair construction lead time and I would not be surprised if it lags demand as EVs scale out.

Most of the extra energy usage will come from cars charging at night when the demand is low. So there is no need to build new power stations, it will be sufficient to run the existing ones at full capacity at night.

The cost increase in electricity is not due to land, it's due to actually needing to buy the electricity from the grid. Sure you can use solar and storage for each charging site, but then you transform the problem into a land one. Otherwise you need to modernize the grid to support vast numbers of charging stations because solar and storage won't be able to charge enough vehicles.

It will be key to understand short term maintenance costs too.

Don't forget about sixty cents worth of each gallon of fuel goes to repairing roads etc

Things that are electrified, on average, only require 44% of the energy of those that run with combustion fuels. The raw cost economics are not a surprise.

The per mile costs of an EV are always going to be much lower than a fuel burning ICE.

It's the TCO that really matters, offsetting the initial purchase price.

That also seems to work according to most TCO calculations I've seen.

What works in the favor of bus fleets, truck fleets is that they often have set routes. You can buy an electric vehicle that has exactly as much battery as you need for your most heavily used route. Then use as it as much as possible to get the maximum cost (and CO2) benefit.

In the EU, almost half (47%) of road freight distances are trips of less than 300 km.

> Statistics from the market analysis group IHS Markit show that during 2021 a total of 346 electric trucks (≥16 tonnes) were registered in Europe – an increase of 193% from 2020. Volvo Trucks has the largest market share at 42%. The countries in Europe with the most electric trucks registered (≥16 tonnes) are Switzerland, Norway, Sweden and The Netherlands.*

So 80% cheaper on fuel, but fuel for long distance trucking at least in Europe - with much more expensive fuel - is just 30% of the mile cost. It makes a difference for sure but there is lots of uncertainty when you factor in the vehicle costs etc.

Time is still more expensive than fuel in this business. I can believe the 2kwh/m figure at 55-60mph but this is not how trucks are used. As soon as you leave the city limits they are typically ripping along at 75-80mph which uses 1.5-2x the energy. Why would these drivers take the extra risk and effort and burn through so much extra fuel if it didn't pay to do so? A 1 hour charge time is probably nowhere near good enough for these people. That's a reduction of ~75 miles per day (assuming the second charge happens after the shift)

83% sounds a bit to good to be true. If that were really the case you'd expect the entire logistic industry to by rushing over to take advantage of the savings. That hasnt happened, at all.

Claims approximately $30 vs previous $170 on a 200 mile trip. I have not checked the math.

Is there a loss of payload capacity? That is the concern I have frequently heard.

Electricity is (sometimes) much cheaper than petrol.

But the devil is in the details.

The Tesla Super charger electricity rate can be x5 more expensive than regular electricity.

The battery may need replacing every 2-3 years according to the mileage guarantee. So now you have a $200K battery amortized over 1000 days which is approx ~$250 daily cost - every day.

And so on...

Still, EV trucks ARE the best options we have. Just beware the cheerleading hand waving bros ;)

There's no mention of payload capacity at all -- which makes the numbers entirely suspect.

Exactly, I wouldn't be surprised if a driver's paycheck is a substantial part of the total cost, so you might end up needing two to four trucks to haul the same payload due to extra weight of the batteries (and there's an 80k combined weight limit afaik). At $25/hour average (googled) trucker pay that 200 mile trip taking 3-4 hours would cost > $75 per driver. Fuel savings are calculated to be $140, every additional driver costs at least $75.

That's maybe the best use-case for FSD I've yet to see.

The average weight of a reasonably new semi is around 20k pounds without fuel which leaves 60k for cargo. Every pound the tesla weighs above 20k is a pound less cargo. Another consideration is the fact that tesla trucks won't be paying for huge amount of road wear they cause. The national average of state/federal tax on a gallon of diesel is around 65 cents. As things stand now an electric semi (just like electric cars) will be getting a free ride.

All depends on if the cargo is volume limited or weight limited.

You don't have to solve every single problem at first even though it would of course be better.

Considering that road damage is proportional to axle weight to the fourth power, I can see weight transport taxation in the near electric future...

Well, a Tesla Model X towed a 100-ton aircraft, albeit a short distance.

Imagine what a 4 or 5 motor-equipped semi would haul.

That is not very special. There have been humans that have towed planes too. Albeit for a short distance.

To encourage electric trucking, the feds allow electric trucks with payload to be 82 tons rather than the 80 tons normally allowed. Tesla claims their truck is only 1 ton heavier than a diesel truck with full tanks, so electric trucks have a slight payload advantage.

the lack of noise with EV compared to diesel is the big win

The acceleration is another big win.

Regen braking is a huge win for semi's.

Or maybe not: https://youtu.be/w__a8EcM2jI

I like Adam Something’s commentary on issues of urbanism, but as an engineer myself I find his engineering criticism lacking.

That video uses a pretty strange calculation to estimate the battery weight. Here is a different way: 500 mile range * 2 kWh/mile means they need a 1000 kWh pack. The most recent packs they are shipping (Texas Model Y) are 543 kg [1] for 67 kWh [2], or 8.1 kg/kWh. If they have the same energy-to-weight ratio for the truck, its should weigh something like 8104 kg. As opposed to 17000 kg in the video.

[1] https://insideevs.com/news/596775/musk-says-model-structural... [2] https://insideevs.com/news/593942/report-mit-tesla-modely-us...

> Jeremy Johnson is a Tesla investor and supporter. He first invested in Tesla in 2017 after years of following Elon Musk and admiring his work ethic and intelligence.

> Since then, he's become a Tesla bull, covering anything about Tesla he can find

Does anyone with an EV that has a KW/h gauge actually believe that a semi will run on the highway at 2KW/h?

To people without EVs, at 2KW in your car is idling forward from a stop. When you press the gas at a normal street at a normal start, you’ll be at 10-18KW/h until you are up to speed. A hard pull is 24KW/h, and on the highway I would assume 4-6KW/h in a light unloaded vehicle. So maybe they had some new amazing Tesla transmissions and Tesla low friction tires too?

EDIT: yes, obviously I’m talking about KWh instantaneous. So /second/minute/hour for the short frame until the next update. If you were to hold this power for the end of time you would use xx KW in an hour. Which is useful for the feeling of torque I described.

I am guessing regen is a big factor

My Tesla Y gets 3 mi/kW. They said a normal diesel truck gets 6 mpg, which is about one fifth a normal passenger car on the highway at 30 mpg. So 2 kW per mile doesn’t sound unreasonable.

You have manhandled your units as badly as is possible.

Model S cruises at approximately 300Wh/mi; Tesla is claiming 2kWh/mi (approximately 6.6x the power consumption) for the Semi.

Diesel cars are rated at approximately 35mpg and diesel semi trucks at approximately 5mpg, a similar 7x factor.

So, yes in fact as the driver of an EV with an energy meter, I do 100% believe the 2kWh/mile number.

What I'm actually skeptical of and where I believe Tesla is burying the bad numbers is with the $/cargo ton/mile rating. Carrying an extra 8 tons of battery is going to limit the types of loads where an electric semi mikes sense.

> Carrying an extra 8 tons of battery is going to limit the types of loads where an electric semi mikes sense.

Tesla claims the Semi is only 1/2 ton heavier than a typical tractor, not sure where you got the 8 tons figure from…

You are right to be confused about that claim, but it because the units of the claim have been mixed up. The actual claim is 2 kWh/mile. That is a measure of energy per distance, kind of like gallons per mile.

If a truck is using 2 kWh/mile while traveling down highway at 70 miles per hour, its power draw will be 140 kW -- that is the one to compare to your 2 kW idling or 24 kW hard pull.

My EV uses 0.2 to 0.3 kWh/mile, so 2 kWh/mile for a truck seems plausible to me.

You’re comparing KW/h to miles per KW - it doesn’t work that way. At 4-6KW/h on the highway, you’re doing 70+mph, so that’s 6KW/70mi, which is <0.1kw/mile.

I think there's a unit issue here. The likely speed of a semi is ~60mph, which conveniently is 1mile/min. So even if you're using 20kW (>25hp) continuously or more likely 120kW (>160hp) you'll travel 60 miles in a hour.

So 2kWh/mi = 120kWh/60mi.

My 2015 Tesla S 70D does roughly between 260 Wh/m and 370 Wh/m depending on terrain, temperature, precipitation, wind, speed, and driving style. I have no idea what your units mean. A hard acceleration is 250 kW. The average power at a constant 60 mph is about 18 kW.

2 kWh/m seems plausible for level terrain for a larger vehicle that is presumably optimized for low rolling resistance. It sounds a little optimistic but not by more than a factor of two.

So my answer is yes I can believe it or something close.

kW/h is not an unit you want to use. Just drop the /h. And math makes sense.

But yes 2 kW is very low power. Around 2.7 horse power. Which is very little at highway speed. Specially with drag of any semi-truck.

Reply on news.ycombinator.com