Tesla Semi, a Deep Dive | Hacker News Reader

Tesla Semi, a Deep Dive

evuniverse.io

181 pointsoxplot497 comments

I work with very large fleets making and selling software to them.

Basically none of them are operationally able to go from 15 minute diesel fill ups on any corner to the infrastructure and load planning EVs require. Like they literally don’t have the software or expertise to plan the loads around electric needs without losing money on every load.

Trucking is a single digit percentage margin business. These trucks are more expensive, require infrastructure that doesn’t exist, charge slower, and don’t go anywhere near as far as diesel trucks. And no fleets in North America are ready to figure out converting a 2000 mile long haul over the road route into 5 400 mile EV relays.

Short haul loads? Sure. In places that demand EV on certain timelines (like California).

But for everything else, these vehicles require a fundamental reshaping of transportation in America. Not saying that’s not possible, but it ain’t happening any time soon.

Europe has mandatory rest stops of at least 45 minutes, so any software with the ability to plan around those should have the ability to convert them easily to recharge stops.

That doesn’t mean the recharge infrastructure exists overnight, but those same stops are being converted to support the rollouts of electric car fleets (e.g. belgium is going to mandate corporate car fleets convert to electric from 2026).

> Europe has mandatory rest stops of at least 45 minutes

I'm not a trucker, but if your source is https://transport.ec.europa.eu/transport-modes/road/social-p... I don't read it as mandatory, more like a recommendation.

It is, and there are devices that make sure you do it.

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

It's encoded in the CE rule 561/2006 article 7:

https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CEL...

  Article 7
  
  After a driving period of four and a half hours a driver shall take an uninterrupted break of not less than 45 minutes, unless he takes a rest period.
  
  This break may be replaced by a break of at least 15 minutes followed by a break of at least 30 minutes each distributed over the period in such a way as to comply with the provisions of the first paragraph.

Does the software actually need to plan the rest stops precisely? Planning to drive X hours over Y hours of real time doesn’t require choosing where the rest stops are in advance, which allows a driver to stop where convenient depending on traffic and other conditions. But if charging is required, then the stop needs to be at a charger, which is a much stronger constraint.

I thought truckers commonly work in teams specifically so the truck can keep moving during mandatory breaks. The driver on break goes to sleep, and their partner takes over.

I think this is a very specific niche of trucking. At least here in Western Europe this is not at all common.

You'd have to pay 2 drivers even though only one works at a time for a small gain in time.

Here in Europe, that is rather an exception. Also, sleeping while the other driver drives isn't considered as legal resting time. Only time outside the truck is considered legal resting time.

No. Because then you're paying someone to sit on their arse for 4 and a half hours doing nothing to get 45 mins back.

It’s not “common” but it’s definitely frequent for critical pieces of hardware where the cost of 2 days of labor for a second trucker is irrelevant. (e.g. Getting a replacement industrial component blocking a business from resuming operations.)

Furthermore, there are now multiple cities that banned diesel vehicles entering cities. This would actually give electric trucks an edge over diesel ones, this could be a great solution for resupplying supermarkets. Could be good for increasing the margin.

I heard about this, but have they banned them today or in 2035?

It looks like Paris will ban them from 2024, which is soon enough that HGV owners must be planning for it.

You might find a city with current regulations on this site.

https://urbanaccessregulations.eu/countries-mainmenu-147/fra...

What's HGV?

Heavy Goods Vehicle (Large / LGV in some places)

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

Tesla already has a very long order book for these. There is no doubt in my mind that they will sell as much as they can make.

The main savings come from fuel costs and maintenance. Eventually from autonomy as well.

>Tesla already has a very long order book for these.

Source?

Short haul trucking makes up about 80 percent of the industry, while long haul trucking only comprises 20 percent. I’m not worried about these EV semis finding their niche and thriving. For a first gen product it’s a good place to start.

Also the pollution from diesel engines has a far greater impact when used in cities instead of long-haul. Not to mention that trucks put out far more pollution when doing stop-start in cities.

Replacing short haul trucks with electric trucks makes many kinds of good sense all at once.

The same argument can be made for making school buses electric.

I'm worried about the pollution argument. Not in the climate change sense, that's clear. But in the "local air is cleaner".

I've discovered recently that tire wear actually creates about as many particles as exhaust fumes, at least for cars. Tire wear is directly affected by vehicle weight.

Won't electric trucks weigh a lot more than classic trucks?

Actually, I checked now, and Volvo trucks for example, weigh somewhere between 10 tons and 24 tons, but 24 tons for the long cabin, sleeper, a ton of extra features, etc.

So the starting point of the Tesla Semi, for example, is 27 tons according to the article. That puts it at double the average Volvo.

And the worst part for the wire tear (and I think road tear), apparently it scales to the POWER OF 4 (!!!).

Now, trucks also have cargo, so I guess electric trucks will carry less cargo and then we're back at the same values?

> I've discovered recently that tire wear actually creates about as many particles as exhaust fumes,

> And the worst part for the wire tear (and I think road tear), apparently it scales to the POWER OF 4 (!!!).

That's very interesting. I'm a default sceptic, but would love to learn more, because if true, that could nullify a lot of the local air arguments. Do you have any sources for these?

It took me a while but I found it: https://news.ycombinator.com/item?id=30501072

Check all the subthreads and links, they're not that many.

The formula was actually for road damage, but my bet is on road damage being strongly correlated with tire wear.

Electric Semi will weigh slightly more, but they will most of the time break with regenerative breaking meaning less tier and break ware.

But you are correct that a modal shift to rail makes a lot of sense.

I imagine most American states have their own limits on axle weight as in Europe. In the UK 44t is the heaviest you can run without additional permitting etc.

It will just be an additional constraint, there are lots of loads which are limited by cubic volume rather than weight. However, it wouldn't work well for say bulk tippers which are almost always at their weight limits (and transporting lots of low value products).

I'm not sure about the braking argument. Trucks spend most of their time cruising along. They also have exhaust brakes and retarders to avoid using the service brakes on hills for example.

> Trucks spend most of their time cruising along.

Not inside of cities and places were humans live.

Regenerative braking will reduce brake wear but not tire wear.

> I've discovered recently that tire wear actually creates about as many particles as exhaust fumes, at least for cars. Tire wear is directly affected by vehicle weight.

Even better: in terms of particulates, tire and brake wear creates much more nowadays thanks to particulate filters on exhausts.

That being said local air quality measurements generally don't follow. The air around highways has an increased amount of PMs, but not by a significant amount.

The main difference in favour of EVs is the NOx emissions - diesel trucks produce a lot of that.

Then R&D budgets should be aimed at finding ways to reduce tire particulates. Perhaps some reasonable regulatory incentives as well.

EV trucks should fix the issue when it comes to braking. They can mostly regenerate kinetic energy back into charge in the batteries.

> ...in terms of particulates, tire and brake wear...

Request to all:

More posts about reducing these particulates (from tire and brake wear) please.

Electric vehicles definitely help with particulates from braking. I don't see why they couldn't do active electric braking if required, as apposed to just regeneration.

Audi has a familiar-looking idea how to tackle this:

https://www.audi-mediacenter.com/en/press-releases/audi-urba...

Internal combustion cars do filter a lot of the ambient air already - a 2.0l engine at 2500RPM filters 150m3 of air per hour - that's as much as a small air purifier. It's just that it takes all this clean air and uses it for combustion, thus putting in more dirt than it filtered out.

Audi Urban Purifier's turns autos into Roombas. Neat.

Also, I'm wondering about tires which pollute less. Both in their production and usage.

To reduce particulates during use, I'm guessing tires will need to be tougher, less grippy. So might require better active suspension and braking systems to maintain current comfort and safety standards.

Assuming that trucks are weighing the same due to cargo weight (and, usually, you're either space or weight limited... usually both if possible), then tire wear should be roughly the same. It'll be a little more since there is a small allocation (2000 lbs?) extra for electric semis, but not enormously so. The big difference though, is if much of the brake particulates are eliminated thanks to regen braking. That could easily outweigh the particulate addition of the small weight factor.

Also, from a pollution argument, this assumes that you have trucks that are all compliant and healthy. With EVs, there is no opportunity for a truck to have its emission control fail or just are grandfathered in due to age. There's PLENTY of semi trucks I see around where I live that seem to belch black smoke when they accelerate. Replace those with EVs (which, the value proposition would make itself for short-haul diesels should make itself) and you have a noticeably cleaner city.

That isn't even accounting for trucks that idle for long periods of time. When talking about local construction jobs, I've seen rows of semis idling downtown waiting for their turn to pick up rubble or deliver materials.

Our local high school already has electric school busses! They look exactly like the traditional American school bus, but with a small green stripe, which was surprising, given how many passenger EVs look futuristic.

The school parking lot has solar panels over it, so I assume it charges at school.

And how difficult is 4-hop route planning versus the actual way ? just trying to assess how far they are from having solutions.

Geo-Arbitraging Fuel is also not easily attainable. Like they do here in Europe, where you can fill up your >1400-Liter-Diesel tanks with cheaper eastern Europe Diesel and bypass all the expensive Western Europe Fuel that drives you 5000KM.

I would think that for trucks it's make sense for the battery to be in the bottom rail of the trailer and for pit stops for fuel where you could drive over a pit and swap the whole battery.

Battery swaps make no sense for trucking (and never really did for cars). The storage required for batteries would be prohibitive, and the charging location would have to keep a decent number of expensive batteries on hand.

But tesla could own all of it. If tesla could make the truck 2 batteries, 1 in the truck one in the trailer. Then you make all trucks flatbed, Refueling could be as simple as hitching a new trailer, swapping a container form 1 flatbed to another. Tesla could own the trailers and essentially rent the space. People could buy the trucks and get access to the trailers. And then it's just shipping with iso containers.

Trailers really wouldn't be the ideal place for a battery since they're often dropped off at a site. And that would also take up space in the trailer, adding weight as well. Two batteries would make the entire deal heavy and expensive, and the idea of Tesla owning them wouldn't fly with most shipping companies. Lots of containers are used in intermodal transportation, and that would rely on 100% Tesla Semis everywhere they were used.

> These trucks are more expensive, require infrastructure that doesn’t exist, charge slower, and don’t go anywhere near as far as diesel trucks. And no fleets in North America are ready to figure out converting a 2000 mile long haul over the road route into 5 400 mile EV relays.

I see this all or nothing argument when discussing new stuff. Same thing was said about passenger cars and quoted as one of the main reasons EVs would never catch on. It looks like a chicken and egg problem until you look at it less idealistically and more pragmatically. If you watch the presentation (and I hope everyone commenting here did that first), it's most likely that the charging infra is part of the package fleet operators buy. They showed battery+solar backed Megachargers that the fleet operators are likely to be using for the near term. With this in mind, those companies that are currently buying these trucks in high numbers, are also definitely buying the charging infra with it. So with this context in mind, charging infra isn't a problem right now. And just like Superchargers for passenger cars, they will only ever increase in numbers.

Slow charging is only a temporary problem and as evolution of Superchargers has made quote evident to date.

From my understanding of US laws and human biology, no one can drive 2000 miles non-stop without taking multiple breaks in tens of minutes each. In fact, Tesla Semi's range is 500 miles because it's right at the edge of how long a driver can safely drive non-stop and right before they are legally required to take a break, during which time they can charge the truck.

> I see this all or nothing argument when discussing new stuff.

When did it become popular to start selling beta products? Did it begin when we all bought into social media?

Apple produced a computer, music player and phone that were a polished step up from what was happening before: build-your-own computer, music with insufficient storage and crap interfaces, and blackberries that only targeted business people.

Underpromise and overimpress me. Enough with this overpromising-because-have-to as if that's the only way to innovate or get investment. It isn't the only way and I wish investors would do a little more due diligence to get behind innovators who know what they are doing, not just throwing money at cool things saying they'll be ready by the end of the year over an 8 year period.

People want EV transportation because the world population is at risk. So they cut some slack to Tesla. Tesla makes compelling vehicles even if flawed. It's ok because it's a step in the right direction. Nothing wrong with that, or is it wrong to have ideals?

People want point of care testing, and everyone is at risk of dying of cancer. Theranos makes compelling equipment even if it is a little rough around the edges. It's ok because it's a step in the right direction. Nothing wrong with that, or is it wrong to have ideals? Oh wait...

What a disingenuous analogy to make, between Tesla that has delivered time and time again and surpassed expectations, led the market for the past 10 years and continues to outsell all of its century old competitors, with Theranos which burnt investor money and never produced a working machine, ever.

> Tesla that has delivered time and time again and surpassed expectations

Tesla has consistently failed to meet the expectations set by its own management.

It's called aiming high!

"We want FSD some day and are working on it" is aiming high.

Calling it's driver assist Autopilot, calling its software FSD, repeatedly saying "FSD is coming next year" is overpromising, misleading marketing and lying

> Calling it's driver assist Autopilot, calling its software FSD, repeatedly saying "FSD is coming next year" is overpromising, misleading marketing and lying

OK, finally an actual example and the one I expected.

TL;DR is that if Tesla was in fact doing anything the regulators found to be hurting the customers, they would be forcing Tesla to make changes (and they have re FSD in some minor ways). It's fun to make a big deal about nothing but no one is buying FSD and thinking, Tesla robbed me $10K. They get it, play with it, it doesn't do what they promised, they can get a refund for the FSD package, and even a full refund for the entire car. People tend to keep their Teslas, so that's a non issue.

Majority of people don't buy Teslas for promise of FSD (in fact, most don't even believe it). They buy it because right now, it's a great car. As far as delivering great cars is concerned (which is the vast majority of cases), Tesla has delivered time and again.

Also, Tesla has never lied about FSD, only missing deadlines. If anything, their progress has shown anything but any attempt to lie to customers. They are moving as fast as they can and no one in the industry comes close.

I'll be happy to see, for once, someone talk about misleading marketing about anything but FSD. You can take FSD away any day and it won't make a blip of difference to Tesla's sales or margins.

I'll do it. The ranges of the cars are exaggerated, in a way that makes them look more attractive in the market compared to other EVs:

https://auto.hindustantimes.com/auto/cars/tesla-revises-exag...

https://www.caranddriver.com/features/a33824052/adjustment-f...

Also, regulators are catching up to "FSD" in many places, but they work slowly. They are also catching up on the promised ranges - unlike "mpg," there is no standard by which the range of an EV is required to be evaluated. All of these rules take a LONG time to put in place.

As far as a refund on the package is concerned, I'm not sure you can actually get one. A friend of mine paid the $10k as a "line jumping fee" to get his car 4 months earlier, and I assume that most informed buyers treat the FSD package the same.

From the first link:

> The Model 3 sedan was being described on Tesla’s Korean website claiming 'can drive more than 528 km on a single charge'. However, on its US website, Tesla describes Model 3's performance as "Go anywhere with "up to" 358 miles of estimated range on a single charge." 358 miles equates to about 576 kms, and much less than what Tesla claimed in South Korean markets. After local media raised the issue and the regulator stepped in, Tesla changed the range from ‘more than’ to a ‘maximum’ 528 kilometres.

What??!! 576km is not much less than 528km, it's more! So Tesla was in fact correctly indicating that the range is more than 528km, because it is! Either this is awful reporting or there is a mistake in there. Also saying "can drive more than" instead of "up to" isn't necessarily an exaggeration, it may simply be a mistake in wording of the website. I give them the benefit of doubt here because I've never seen Tesla advertise range like that. It's often a single number with a testing methodology suffix (like EPA). There is no "up to" or the like qualifiers.

As for the second link, super interesting content. But EPA has certified the range and all other manufacturers are free to do what Tesla does to make their range more appealing. Nothing fishy going on here as far as I'm concerned. Exaggerating range in context of a certification program is moot (unless you're cheating and Tesla is clearly not, according to this article).

> A friend of mine paid the $10k as a "line jumping fee" to get his car 4 months earlier, and I assume that most informed buyers treat the FSD package the same.

Really? You thought this was a compelling example?!!! Have you never bought an "as-is" item from a shop which was on display. You pay cheaper but you can't bring it back for refund if you change your mind. This is common practice in every industry. You wanna be treated differently, you give up some perks.

People who buy FSD package as a way to jump the queue aren't buying FSD, they're buying time and as such doubt they will care about functionality of FSD.

I was referring to people who buy FSD for its current and future capabilities.

The problem with that, is that “Theranos makes compelling equipment even if it is a little rough around the edges. It's ok because it's a step in the right direction.” wasn’t true, and that’s why Elizabeth Holmes was convicted for fraud.

Now, I’m not happy that Tesla is still calling the driving assistance software “autopilot” given some governments are criticising this name choice as misleading, but it’s not like the (limited and not ready) software is completely fictional — it can actually get people from A to B by itself, even if though it really isn’t at the quality level where this is a generally wise replacement for most humans, but only for, e.g. impaired humans in a medical emergency and even then it’s really only sane on highways (old story, but probably still illustrative based on the published safety statistics): https://www.theguardian.com/technology/2016/aug/08/tesla-mod...

However, considering Tesla as an EV company rather than as an AI company: it does what it says on the tin. There’s no secret gasoline tank hiding inside the batteries.

It was true. They had 2-3 working tests running on their machines by the end of the company. They sold over 100.

In my mind, this is the same as advertising "the human is only there for liability reasons" on an L2 ADAS system.

As to what Tesla is, the moment people accept that it's an EV company, I will be happy. The market currently does not, largely due to the insane promises of the company's CEO.

Let's not forget that Elizabeth Holmes was acquitted of defrauding patients. She was convicted of defrauding investors. What's the difference?

Theranos defrauded investors. Lied about their product and NEVER delivered on what was promised.

Outcomes speak for themselves. Tesla has had their timelines set back years and they're still ahead of everyone else in every category and their products sell like hot cakes. People who actually care about innovation and progress aren't pedantic about slipped timelines, panel gaps or hiccups, ESPECIALLY on projects that are on the cutting edge. If you think investors, most of whom worked their ass off to build their capital, are so gullible, over so many years, then you're not giving them enough credit (unless you think Tesla is misleading investors Theranos style - at which point, I'm off).

As for the whole Apple situation. If all companies tried to do what Apple does all the time, we'd be living in the stone age. Apple is an exception to the rule in that they're perfectionists in their tiny little bubble. Perhaps you didn't catch Apple's abondoned plan to build their own car. Did you hear about Dyson? Making an EV you can actually sell for profit is something even Apple, with all their money and talent, can't do and Tesla does like it's child play. Tesla has has the highest per unit margin of any automaker out there by factors of 3x and more. Apple does not hold a single candle to Tesla.

I largely agree with you but if I can be pedantic, they’re not the only manufacturer with high margins, Porsche does too for example.

Teslas are still a luxury vehicle so it’s no surprise they’re able to fetch higher margins than say, Toyota, who’s main sales are economy cars.

Apples to apples and all.

Tesla is about to overtake companies like BMW who have existed for 100 years. They already outsell Porsche.

And without having entered into major markets like Pickups, Semis and a number of others.

And they make their margin on EV, while many other companies simply lose money on EV and simply hide that fact. Or make minimal margin on the cars and lose money on EV overall.

Only a few years ago it was widely believed that you can't make money on EVs.

Another concept to look at for margin comparison is the fact all the other manufacturers are still playing the dealer network game. Lots of the profits on the cars aren't going to the old manufacturers, it's going to the local dealers.

Tesla doesn't have dealers, so all the dealer profits go to Tesla.

Hard disagree on cars. Everyone has caught up. I'll raise you Hyundai/Kia against a model 3 any day of the week.

At Plaid levels, a trimmed up Taycan is gorgeous by comparison and Lucid hit the Plaid where it counts ... and let's not even talk about Rimac.

Sorry, the fan boy belief that Tesla is far ahead of the competition is 18 months out of date now.

> Everyone has caught up

They are caught up when they start outselling Tesla (or heck become available so one can actually buy one).

Great cars (which I don't think Hyundai/Kia are by any stretch of imagination, from piss-poor software to poor charging infrastructure, but let's discount this for now) can exist at great prices all they want, but if I can't get my hand on one and the 500 or so that show up once every few month get gobbled up in seconds, they practically do not exist at all.

And mind you, the disparity in production is not gonna go away any time soon. They're all limited by battery production and unless they make their own batteries, Tesla has long been in the line to buy batteries from anywhere they possibly can, as much as they can. They don't screw around.

BYD is far more of a competitor to Tesla than Kia/Hyandai have been to date.

EDIT: and Tesla is no exemption here - once they have a Cybertruck selling in competitive "numbers", it becomes competitive. Until then, Rivians and Fords are leading the pack (and I must say, beautifully).

>>they're still ahead of everyone else in every category

Your definiton ^ ...

And, in terms of sales, that might be your definition of 'caught up' but it's not mine.

Well-used analagy applies here: if sales are your KPI then I assume McDonalds is your idea of a fine meal. Sorry, no.

And your dismissal of Hyundai/Kia by definition singles you out as far out there in biased land. I've test drove all three (including Tesla) and done the research. The only reason I haven't bought an ioniq 5 is because I simply can't justify it at the moment (my work-life means it'll just sit in the driveway however much I want it). But they are all ... great cars.

I'll say Tesla are great cars with groundbreaking technology, if ugly. But the Hyundai/Kia are now the better cars (interior, build, comfort) and the all around package. Again, unless you count "sales" as the definition of 'in the lead'.

*Disclaimer: I have a driveway where I can charge over night. If you don't have this, perhaps the supercharger network means something to you. That appears to be the only advantage.

> Your definiton

And many others.

> And, in terms of sales, that might be your definition of 'caught up' but it's not mine.

Good for you

> Well-used analagy applies here: if sales are your KPI then I assume McDonalds is your idea of a fine meal. Sorry, no.

Thanks for that analogy. Yes, one can brag all they want about the best Italian cuisine they had at a top bistro. It means nothing to most people if: they can't afford it, if they can't get a booking for months. McDonalds delivers round the clock, food that's safe and does it fast. I take McDonalds over all of so called "fine" food any day. I value my time and don't want to waste it, waiting on food.

> And your dismissal of Hyundai/Kia by definition singles you out as far out there in biased land. I've test drove all three (including Tesla) and done the research. The only reason I haven't bought an ioniq 5 is because I simply can't justify it at the moment (my work-life means it'll just sit in the driveway however much I want it). But they are all ... great cars.

I have driven all the above (some through friends and Ioniq 5 by asking a stranger nicely), in addition to BYD, Polstar and anything that's available to drive in Australia, multiple times for some. I've owned a Tesla for the past 2 years and have test driven every other model as well. The handling, software, ambiance and interior of all but Tesla (and specifically Model 3) is vastly inferior. As for handling, they all feel like they'll run away and off the road any moment. I'm sure I'll get used to them after longer drives, but Tesla's doing something there that gives me more confidence when handling.

If any of these was significantly cheaper as to make up for their shortcomings, I'd happily recommend them to friends and family. But as they sit right now, no frigging way.

> if ugly.

I don't know what you're talking about!

> But the Hyundai/Kia are now the better cars (interior, build, comfort)

Interior of all the cars above make me feel suffocated. Their build qualities are definitely good but virtually no one I know actually cares, unless shit's falling off the car. Tesla's quality is only getting better. As for comfort, Model 3 has the most comfortable seat of any car I've ever owned (out of 3).

>McDonalds delivers round the clock, food that's safe

hmmm only in the short term and at low frequency.

Eat Mc Donalds every day for 50y the analysis is way different.

At least with McDonalds, you know what goes into it. They actually have a nutrition label. Show me the nutrition label of the Italian bistro. If you asked, you'd probably get kicked out because you insulted the chef.

So, actually, I'll be way more confident about my health (that's what it's called, not safety, food safety is about you not getting poisoned) eating at McDonalds, than at some rando uptown joint.

Do you really know though? McDonalds has lied in the past quite a bit. For example, they touted their hamburger meat as 100% beef with no additives for quite awhile, then had to fess up when the truth about Pink Slime came out.

Then when they tried to climb on the vegan/vegetarian bandwagon, they had to quickly admit that their fry shortening had beef flavoring added to it, then switch to an entirely different shortening.

McDonalds is a franchise. And like every franchise there is always a lot of leeway depending on the managers and staff.

If you walk five miles a day and do cardio, I wonder if your body would notice.

> When did it become popular to start selling beta products? Did it begin when we all bought into social media?

Since forever? You're comparing a new product category with already existing and optimized one. To keep close to the topic, think about the history of ICEs and cars in general: the first ones that got popular were all pretty much beta products.

> Apple produced a computer, music player and phone that were a polished step up from what was happening before: build-your-own computer, music with insufficient storage and crap interfaces, and blackberries that only targeted business people.

That's some pretty revisionist history. Apple made solid, polished hardware, in an established market with plenty of competition. About the only large leap they made was with the first iPhone, and it's considered both transformative and very much a beta version.

People forget just how basic (and sometimes bad!) the original of many things was. And everyone has an iPhone now but they didn’t all buy the first one, neither.

The original iPod was Mac only and FireWire, for goodness sake! People made do and it improved over time.

> From my understanding of US laws and human biology, no one can drive 2000 miles non-stop without taking multiple breaks in tens of minutes each. In fact, Tesla Semi's range is 500 miles because it's right at the edge of how long a driver can safely drive non-stop and right before they are legally required to take a break, during which time they can charge the truck.

IIRC the same was the case back when Model S was rolled out - the range was intentionally developed to be a little above the legal limit of non-stop driving, which to me provided a pretty solid argument against range anxiety: there's hardly a way to run out of charge on a trip and be a safe and responsible driver.

I’m only allowed to drive 200 miles at a time? Do you have a source for that?

When I do cross country drives I often drive 8-12 hours in a day. I stop to fill up and eat snacks while driving.

Typically the drivers attention drops after 2-3 hours of driving, so it is recommended to take a break then. The legal limit for truck drivers in Europe is 4.5h of driving, after which there needs to be a break of at least 45 minutes. The maximum total driving time per day is 9 hours.

While these times are of course not enforced for private driving, they are a good measure for what is safe. Yes, you can drive 5 hours or more without a break, but you need to be aware that this is risky behavior.

I don't think you can get 3 hours out of a Tesla Model 3 going 80mph in the mountains in very cold or very warm weather. But maybe I'm wrong.

Well, it's hard to maintain 80mph in the mountains, it's also less likely to be very warm as you go up. A summer trip to Tahoe from where I live in the central valley is about 2.5 hours and it easily makes that. I am even close to making it home on the same charge.

My memory of driving Washington/Iahado/Montana is that holding speed wasn’t that difficult. Lots of opportunities for acceleration and deceleration. I have never driven a Tesla. Curious to know if it can handle my driving style.

Could it make it from San Diego to Anza Borrego and back?

I'm more familiar with central California, but looking up the distances I'd be shocked if it didn't make it round trip easily.

I'd say the trouble with holding speed in predominately mountain roads I'm familiar with is that they tend to be filled with slower drivers.

Commercial truck drivers are subject to different rules. They used to have a bunch of problems, like taking speed and driving for ridiculous stretches. Apparently, the rules in the states are 11 hrs/day max w/ 3 hrs of rest breaks [1].

1. https://www.indeed.com/career-advice/finding-a-job/how-many-...

OP was not referring to commercial trucking.

My mistake, well I certainly don't remember learning anything about legal driving limits- I'd be shocked if those were laws and not some agencies guidelines.

It’s actually kind of crazy that limits don’t exist as it has been demonstrated that lack of sleep is essentially equivalent to driving inebriated at a certain point. I suspect it’s only the logistics of enforcement that prevent it from happening.

There’s a saying about small planes - it’s pointless to have range on the plane that significantly exceeds the range on the pilot.

Meaning that having 12 hour of fuel is pointless when you have to stop to piss/eat every 4-6 hours anyway.

There is no legal limit for non commercial drivers in the US though? Certainly it's a bad idea and impractical to drive for, say, 24 hours without stopping, but not illegal?

Guess it becomes illegal when your eye lids starts lowering ...

Fair, I checked and it turns out I must have misremembered the legality aspect. There are, however, safety recommendations that strongly urge to take a 30-45 minute break every 3-4 hours of continuous driving, and - IIRC, that was many years ago - the model S range was more than enough if you followed those recommendations and recharged during downtime.

Most, probably 40-something, states regulate it either explicitly or implicitly.

As we know from many other subjects, anything short of federal regulation is roughly equivalent to “unregulated” for the purpose of internet hand wringing.

Do you have an example of it being regulated at all?

Virginia, Colorado and Nebraska prohibit more than 13hr, 12hr and 12hr (in respective order) in a 24hr period.

I can keep going...

https://law.lis.virginia.gov/vacode/title46.2/chapter8/secti...

I’ve never heard tell of this, but there it is. I wonder if it ever gets enforced.

Thank you for sharing this. I wasn't aware any states had such a thing. I'm happy to be shown the evidence that some do.

LOL you completely made that up.

For an owner-operator it is indeed all or nothing, and these are more than 50% of all semis. Stake his livelihood on a dubious technology or stick to the proven one?

Only a handful of companies run thousands of trucks and could afford a small scale experiment.

Package delivery is a different beast, the vans are guaranteed to spend the night at the base every night.

It will take a long time to make enough Semis to replace a large fraction of ICE trucks. So this will be a smooth transition where they are deployed first for duties which make it a nobrainer, like scheduled routes which are driven every day. As the market share raises, the usage scenarios will expand.

It’s hard to find exact numbers but it is likely that OTR trucking accounts for more than 50% of all miles but less than 50% of all trucks.

Many smaller companies have a small fleet of trucks for local deliveries to their branches, and some will be willing to take a look (even if just for the publicity, etc).

By "All or nothing" under the 'all' category, I'm assuming you mean huge government handouts for ev companies while poor people get to continue to live constrained to the market of a dwindling supply of used efficient and once cheap petrol engine vehicles.

Inequality has only become worse under the old regime of subsidized private technology and manufacturing.

Sounds like a case of needing to read up on history of innovation and how you don't get overnight mass adoption. I suggest picking up a book on advent of internal combustion engine and the five decades that followed.

Do you have any book recommendations?

Do you have any good book recommendations about this that you've read

> it's most likely that the charging infra is part of the package fleet operators buy

That inherently means short haul trucking which parent says is feasible. With long haul trucking, I’m almost never at a location owned by the fleet operator. Pepsi and Anheuser Busch may have long haul routes that are exceptions, but you can see from the other trucks they are buying they are doing a lot of short haul trucking. They’ve both also bought BYD trucks for the same locations that have a range of 125 miles, so they don’t need the 500 mile range for their activities.

Short-haul EV trucking in an area eventually leads to building enough charging stations that longer-haul becomes feasible on some routes, because charging infrastructure is several areas overlaps, or almost overlaps. This fuels (heh) further expansion.

Who has the financial incentive to build charging stations? Both out and back and point to point operators will build charging infrastructure at their own hubs/terminals and use them for themselves.

You might see truck stops adding "public" chargers after that EV trucks become commonplace (in the same way that truck stops have fuel discount programs for member fleets) but how will EV trucks going to become commonplace if private operators using their own infrastructure are 99.99% of users?

Truck stops are thin margin. I don't see them making the investment without a need. Maybe if a big player (e.g. Swift, Walmart, etc) wants to convert to EV and kicks in a bunch of money to have their fuel partner make the capital investment. But I don't think that's going to happen until EVs are so good that it's a "profitably without nearly any doubt" type investment. You'll probably need to see a decade or two of short haul EV trucks and the accompanying refinement before that happens.

> Who has the financial incentive to build charging stations? Both out and back and point to point operators will build charging infrastructure at their own hubs/terminals and use them for themselves.

Don't forget that Tesla will probably end up building the most number of chargers over time. Charging is its own business with its own margins (e.g. Supercharger network). In other words, it's a source of net income for Tesla so as they build more trucks, they'll build more chargers as they've done to date with passenger cars and Superchargers.

It's interesting that Tesla hasn't made a splashy entrance into generation.

I think in most states within the US, generation is a hard problem to solve because of state regulation. There are some outliers (Texas) but in general its hard to just start your own power plant thats connected to the grid.

> There are some outliers (Texas) but in general its hard to just start your own power plant thats connected to the grid.

What if it's not and solely powered by solar (just wondering - Megachargers most likely need help from the grid).

Yep. Kinda like manufacturing cars is highly regulated. Or selling them without local dealerships. Or manufacturing industrial warehouse-sized batteries. Or building and operating an international network of charging stations.

They have a well-known solar roof program, which is generation, but home-scale.

Maybe a huge charging station which has to have large buffer batteries anyway would benefit from a large solar-panel field nearby / above it, if land price is acceptable. That would be a natural extension of the solar roof business.

Charging stations are cheap compared to diesel pumps, it’s not a huge logistical challenge, just some hardware, space, and an electrician to turn things on.

Chargers at truck stops can only be good for their margins, who would have to cater to people waiting for their trucks to charge on things they actually make money in (hint: it’s not diesel or electrons).

In the UK I imagine electric trucks will first be involved in deliveries between regional distribution centers on routes with known logistics.

I'm not convinced we'll see electric trucks take over the world in the medium term but I'm sure they'll be visible on our roads in the next 10 years, if only in limited roles.

In the UK, and most of Europe for that matter, you're also not dealing with 2000 mile runs like the US.

Sure but even within the UK there are people who "tramp" in trucks. They set off on Monday morning and don't return home until Friday night taking mixed loads between premises and sleeping in lay-bys and industrial estates each night without so much as access to a toilet. They don't do 2000 mile runs but they still cover 2000 miles in a week. It's hard to imagine acceptable infrastructure being added to make this work.

> It's hard to imagine acceptable infrastructure being added to make this work.

This is such a typical 1st nation problem,

While China battles droughts and build world's largest high-speed train network in 10 years.

Apparently maintaining toilets for truckers is beyond our capabilities.

Eastern Europe companies operate for months with the drivers living in the trucks. It’s brutal how money is made there. German highways are full of these trucks. 2 guys in a truck do 1000 miles every day except Sundays when the trucks can’t move on highways.

In 10 years we will be deep into the decarbonization of transportation.

Maybe that doesn't mean electric. But if you can get 500miles today at reasonable efficiency then this is one of the really really easy things to decarbonize early. All you need are truck stops with good charging every couple of hundred miles. There are other sectors that are infinitely harder and almost certainly won't work with electric barring unforeseen breakthroughs (e.g. aviation).

In reality most truck drivers struggle to find an overnight parking space with access to a toilet or shower. Drivers work long hours and don't have any flexibility to just drive onto the next place if they can't plug in at one station, often only having minutes left of their allowed driving hours. The availability of charging stations would need to be far better than even existing parking spots.

It might not be impossible but it's not easy by any stretch.

Easy is relative.

Seems like an opportunity to make your truck parking & charging & shower place able to be booked ahead.

> That inherently means short haul trucking which parent says is feasible.

The parent put a weird disqualification on it that it will only work if demanded by the state. That’s a weasel word word of saying that it will be terrible and nobody will use it unless forced.

I don't spend huge amounts of capital for no reason, so why would a low-margin trucking company?

That said, if the price of diesel keeps up, there is a chance that Tesla trucks plus all the charging infrastructure will be worth it economically for short haul fleets.

I'm honestly surprised they didn't try to launch a truck in Europe first, it seems like a better market.

Rough math on this: $5/gallon diesel with 5mpg = $1/mile fuel cost. 1.7kwh/mile * $0.1/kwh = $0.17/mile. $0.83/mile savings * 400 miles/day * 250 days a year = $83,000/year in fuel savings.

I suspect that charging infrastructure is going to be expensive, but cheap enough that if an operator has multiple trucks, the payback will be pretty fast and the low margins will incentive faster adoption for short haul trucking. Long haul is another beast entirely (don't hold your breath for that industry to come around before well into the 2030s), but short haul should eat this up!

Just in fuel savings over the life of a truck that pays for those whole truck. Not the difference, the entire cost of the truck. Margins in trucking are too tight to ignore that much money.

Getting charging in place is critical, but nobody in trucking can afford to ignore that.

There’s also potentially cheaper maintenance - no engine to keep running, and less brake usage.

That could add up. Companies will be looking at these even if they just buy one to use as a yard shunter at first.

If the economics are there it’ll take off like wildfire.

A little cheaper, but in the parts that are different between an EV and a ICE are very reliable. You miss oil changes, but most routine maintenance is still needed (though few people rotate their tires as often as they should)

Mostly brake maintenance! But that's rounding error. Most of the savings is fuel.

0.1$/kwhr? Superchargers now cost up to 0.5$/kwhr. These cost money to build.

https://electrek.co/2022/09/28/tesla-hikes-supercharger-pric...

That's in CA though - and at retail pricing. I suspect most customers will have captive chargers and will be running at industrial rates (so even less than $0.1/kwh).

> I don't spend huge amounts of capital for no reason, so why would a low-margin trucking company?

Setting aside the cost of diesel for a moment, trucking companies didn't build the entire gasoline infrastructure, either.

They will have to build some electric infrastructure, the way most EV owners do. The best time to charge a short-haul truck is while it is waiting to be loaded or while it is parked overnight, and both of those will need the operator to own chargers.

Operator owning charger also means markedly cheaper rates. .36c vs .08c per kwh as a crappy anecdotal reference near me. in the north east.

The $$ savings may make installing the infrastructure in loading/storage bays worthwhile.

Whats we have today has been built over 100 years. How fast will that convert to ev Charging is an open question.

What we have today followed adoption closely, and that existing infrastructure is ripe for adjusting as EVs become popular. Especially the larger truck/interstate stops, that have Starbucks, Subway, showers, etc.

The nice thing about EVs is we aren't starting from scratch either.

Most distribution centers, factories, etc where trucks load/unload at docks already have 3-phase power from the utility. They may need a transformer and service upgrade or for really large fleets they may even purchase 480 or KV power but when you compare it to the fuel savings its a total no-brainer. If you can save $50-80k per truck per year vs diesel that savings makes the EV trucks free. $1m for electrical upgrades? Sold. Will utilities need to perform distribution upgrades? Yes, but not all at once. They will do the upgrades as demand rises just like they do today.

As many others have noted in this thread it all comes down to your routes and logistics. For local and short-haul where the trucks return to a home base they are an excellent fit. Even if you need 50% more trucks because some of them are always on the charger as I noted the fuel savings make the trucks free so it doesn't matter at all.

It remains to be seen what the maintenance burden is. If it resembles passenger cars in any way I'd expect maintenance costs to be much lower since more components can be permanently sealed with lifetime sealed electric motors so no belts, hoses, bearings that wear out because someone couldn't be bothered to grease them, etc. Some consumables like brakes should also last much much longer.

Short haul, for sure. Long haul the truckers are independent from the place where they deliver to, and so there will be some friction - will the warehouses install charging or - like today - will they expect truckers to take care of charging.

Part of this is how the law catches up. In some states only a utility can legally sell electricity which means billing for charging is weird.

Europe has fairly different truck designs and already has some other (Chinese owned) companies in the market, but I couldn’t say if that’s the reason.

It is pretty common for anybody with even the smallest fleet to own their own gas pump. Say a farmer with one tractor, or a flower shop with two vans.

In a world with gas stations on every corner, your argument would suggest this is foolish.

So perhaps the owners don't consider it a huge amount of capitol. Or perhaps they've found a reason or two.

But if I operated a distribution center where dozens of trucks were parked at my docks pretty much around the clock, I'd do a bit of napkin math. 1000 square meters of rooftop solar feeding giant batteries connected to 10 charging stations selling juice at $.25/kwh to trucks who are stuck here for the next hour regardless.

And I don't have to pay Exxon a penny and I wouldn't need a constant stream of tankers refilling underground tanks?

Hand me another napkin. How much battery do I need? How often would I have to replace the cables that reach from the dock to the cab? How much do those cables cost? Hmmm. Hand me another napkin...

I suspect that they don't need the 500 mile range for all their activities. Local deliveries probably don't need it, but mid-range routes may!

>it's right at the edge of how long a driver can safely drive non-stop and right before they are legally required to take a break

Yup, in the US, truck drivers have to take a 30 min break after 8 hours of continuous driving [0].

[0] https://www.fmcsa.dot.gov/regulations/hours-service/summary-...

Which is one reason for team driving. One person drives while the other person sleeps. E-semis might not have cracked that nut yet.

But they don't have to solve every challenge at once. D-semis have been around forever and still haven't displaced trains. Obviously, they still have value. Where e-semis make sense, people will use them. Same for e-vans and e-bikes and e-tugboats.

And yes, people will still use d-semis and d-vans and gas scooters and horses. And clever people will keep finding ways to get the job done, whatever that job may be.

This. I purchased an EV for the daily commute ~3 years ago. A family member got angry and railed about how it was already a failed technology as they likely couldn't replace their f350 towing 4-wheelers deep into the woods.

Infrastructure takes time; and you don't have to solve every use case at once.

Sure, but you should stop selling your product like it can do all of those things and more. Its outright lying.

Tesla semi was production ready in 2020 - as far as i remember. and was more efficient than train. And whatever other lies musk said back then.

Musk is prolific deceiver, I dont trust a single thing coming out of his mouth nor his websites.

Also car is often an expense - bigger or smaller - its not a tool that needs to make money. A tool that can make or break a business.

Tesla cars are notorious for its expensive repairs. Its a huge cost and risk switching to it.

> Sure, but you should stop selling your product like it can do all of those things and more. Its outright lying.

This is the response I hate. Detractors say that to argue against any adoption. It's not a claim any sane proponent (Musk is not sane and no one should listen to him) will make. Will it work eventually for (most) use cases? Yea! But likely not everywhere. Just like there are edge cases Gas or Diesel still don't work well. That doesn't mean to ignore potential where it makes sense.

I bought a commuter car. That's it. When I bought; was more of a novelty as I didn't know anyone else with anything similar.

I repeatedly got angrily shouted at while driving (in a relatively major city), coal-rolled; and heckled by strangers and family members around the dinner table about how poor of a decision I made. 3 years later; I have a nice/fun/convenient commuter car that hasn't needed any maintenance and the trade in is higher than I paid for it. Is it for everyone? No. But that doesn't mean there is a void in value.

I'd never heard of 'rolling coal' before. It's where you modify a diesel vehicle to dump fuel and emit a lot of smoke from the exhaust. As you might expect, people who do this to their trucks are very proud of it; it's usually a 'statement' against environmentalism.

It purposefully wastes diesel which is already much more expensive than gas. It's dangerous for a variety of reasons. It's illegal in the US and parts of Canada, though apparently not well policed. It's stupid.

Take a visit to any part of the south. Plenty of coal rollers around there. In some places, driving a sedan is the weird thing (instead of a truck).

Picture you're going down the highway not really paying attention to cars around you outside of general safety. Lifted dodge aggressively pulls infront of you; forcing you to spike breaks before it's engine revs and you're completely blind in a giant smog cloud as the truck spews all over you. It's nasty and some seem to hunt EVs to vomit on.

> I have a nice/fun/convenient commuter car that hasn't needed any maintenance and the trade in is higher than I paid for it. Is it for everyone? Is it for everyone? No. But that doesn't mean there is a void in value.

I'm sure you have reasons for downplaying this, but I would follow this with "and told them, in your face losers", or something to that effect.

Saying "saving money and hassle" isn't for everyone is suspicious to say the least. Why wouldn't it be for everyone?

> I'm sure you have reasons for downplaying this, but I would follow this with "and told them, in your face losers", or something to that effect.

Why would I ~ever mention it when it's somehow contentious? Family member in question has a truck that costs ~50% more. Why would I care?

> Saying "saving money and hassle" isn't for everyone is suspicious to say the least. Why wouldn't it be for everyone?

I never told anyone it saved money. I paid ~50% more for an electric Honda Civic. For my area/family group I was a new-tech adopter. I wasn't getting told, "You wasted money!" I was told, "That death trap will explode/catch on fire/not make it through the snow, battery will randomly empty, etc", harangued on the absolute basics on simple fallacies.

You are aware that manufactures face penalties for lying and deception, right? I can't recall Tesla ever being fined for misleading marketing, ever.

> Tesla semi was production ready in 2020

I guess you've never missed a deadline. That's not lying. It's miscalculation.

> and was more efficient than train

A Semi convoy is expected to be cheaper than a train and they re-iterated that in the latest event. No evidence to the contrary so far.

> And whatever other lies musk said back then.

Promised 500 miles. Delivered 500 miles. Turned the impossible (according to pretty much everyone, from Bill Gates, to CEO of truck making companies who joked that Tesla Semi will be breaking laws of physics) into late. Still not lying.

> Musk is prolific deceiver, I dont trust a single thing coming out of his mouth nor his websites.

He doesn't have websites. He has companies that have websites.

> Tesla cars are notorious for its expensive repairs. Its a huge cost and risk switching to it.

Is that why Hertz, that puts the most number of miles on a passenger vehicle has ordered 100000 Teslas, 20% of its global fleet, because it loves to waste money on cost of repair?

   I can't recall Tesla ever being fined for misleading marketing, ever.

https://www.cnbc.com/2022/08/05/california-dmv-says-tesla-fs...

You've linked to an "accusation" without any fines. Thanks for wasting my time.

Tesla hasn't been fined by California because a.) the threat was a stop sale not a fine and b.) the process is still underway. I mean hey, Germany told Tesla to stop advertising autopilot because it was deceptive. But the only punishment that counts is monetary, right?

> I can't recall Tesla ever being fined for misleading marketing, ever.

It's coming:

Tesla Is Sued By Drivers Over Alleged False Autopilot, Full Self-Driving Claims [1]

Tesla faces U.S. criminal probe over self-driving claims [2]

California regulator claims Tesla falsely advertised Autopilot, Full Self-Driving [3]

[1] https://www.carandbike.com/news/tesla-is-sued-by-drivers-ove...

[2] https://archive.ph/zceIz

[3] https://archive.ph/7qYSm

The lawsuits are coming, sure, it's US. People sue each other for sports. I'd be disappointed if they didn't sue Tesla ten times a year.

I'll listen when any of those actually results in Tesla being found at fault after a final decision.

I'm curious, can someone explain why this is getting downvoted? My entire comment was just links to 3 news articles.

Trying to figure out have I broken some rule here? Or is this topic really controversial?

Simple answer: the thread has become a tribal flame war. It's not about the content of your comment, but which side you're on.

That said, negative karma on such comments is usually transient, and for that reason, HN guidelines recommended that you don't complain about downvotes. Such cases typically self-correct after a few hours.

> Still not lying.

In 2017 at the event he said they already had the technology to do truck automated follower convoys today (2017), it was only stopped by regulators. Seems like a complete lie given what we know of how their technology played out and even the single track tesla in tunnels in vegas still use drivers.

On Tesla's own website since around 2016 they said the cars could operate on their own but they only have a driver there for legal reasons (opening scrawl to the video at tesla.com/autopilot). Complete lie, the Nvidia stack and the software Tesla had for it couldn't do it, nor could several later stacks over the years. Regulators weren't what was stopping them unless they are saying they at Tesla are murderous psychopaths and would kill if it weren't illegal.

Tesla making empty claims on this is even more annoying since Toyota, which has a far more conservative approach to AV announcements, thinks this is near-term attainable for their Hino semis.

I am not a fan of Musk but why are so many people like you? So much hate and ignorance instead of logical arguments.

Some people are intimidated by success of others, especially if they don't have much going on in their own lives. And some people get excited and motivated by it and want to do more. Those would be the caliber of people who end up working at Tesla or SpaceX or Apple, etc and join the fun.

I don't think it's as much "Success of others" as it's "Musk is a twatwaffle". I say this as a daily EV driver.

It's really weird isn't it, how at some point Musk became the new focus of Two Minutes of Hate[1]?

Sure, there are plenty of valid and harsh criticisms to make about Musk. I'm sure he'd even agree with a lot of them. But the level of obsession people have with hating him is clearly irrational. There's more anti-Musk sentiment than anti-Putin sentiment!

It's as if people were left feeling empty by Trump's disappearance and then latched onto Musk as the new target for their manic vitriol.

It may be that once these people have become accustomed to hating a public figure in this way that they have trouble letting go of the addiction.

It seems a lot like what happened in ancient Athens when one public figure after another would become the target of a citizen mob and then be banished[2] or executed[3]. It also seemed like an addiction in their case.

Usually the mob would quickly come to regret their decision, which is likely what would happen if Musk went away. Many of these same people would come to realize the value of having Musk around, despite his flaws, to advance space exploration, pro-environment technology, brain injury technology, etc.

1. https://en.wikipedia.org/wiki/Two_Minutes_Hate

2. https://en.wikipedia.org/wiki/Ostracism

3. https://en.wikipedia.org/wiki/Trial_of_Socrates

> It's as if people were left feeling empty by Trump's disappearance and then latched onto Musk as the new target for their manic vitriol.

I've has the exact same thought. The timeline fits, and it's exactly the same mob of people. Not saying I didn't dislike Trump, but not spending all my time hating him on Twitter.

I will speak honestly and say that drawing from my own experience, it is because people feel duped.

When Musk came to public notice he was a fast-talking, confident, seemingly self-made rich tech autodidact who had personally created industries through sheer force of will and a vision.

He talked to a certain type of person who grew up with sci-fi and optimism for the future through technological advancement, only to see a disgusting anti-intellectualism and fatalism take over during the Bush Jr. years. He really spoke to me and I thought he was the real deal.

Then, slowly and over the course of years we started learning that he was a charlatan who's success was the result of luck, connections, or usurpation. He promises and promises and a lot of the big promises turn out to be laughably impractical or unrealistic to the point that no person learned in the applicable fields would take it seriously.

We learn about the subsidies that underpins the success, we learn about the terror that he puts his employees through, we learn about how he treats his family, and how he uses real people's lives and serious events as ways to increase his publicity and image with no concern for any negative effects on others.

The real 'ah-ha' moment for me was when he accused the British diver who risked his life and career to rescue trapped children in an underwater cave of being a 'pedo', because they didn't want to use his useless submarine idea.

And all the while, whenever he announces something, the fans and public rant and rave about a new revolution -- "vacuum tube transportation", "going to Mars in 5 years", "exactly like trains, but instead, cars!", etc. When we mention the problems with these ideas, and that Musk isn't really the best person to be advocating things that he knows nothing about, the fans retaliate with personal insults. "You are just jealous"... etc...

And then when he starts to lose popularity he turns 'free speech' and starts catering to a very unpleasant type of crowd (in many people's opinion). When you start appealing to the right-wing because your reputation took a nose-dive, then you are, um, "not a good person".

After a while it becomes a binary thing -- Musk == bad. I have to stop myself from thinking that way, but every time there is a raving fan talking about the new thing he is a genius at, it rears its head again.

Hope this helps you out a little.

I am very pro EV...but a little concerned that the one thing we do (Hauling a 5th wheel with a diesel) will get legislated away or cot prohibitive before a reasonable alternative is available.

But for the other transportation needs, when I replace the 40mpg commuter, I'm all for it. When we bought the commuter, due to covid math, it was an $8k purchase plus trade. An EV would have been a $25k purchase with trade and I really didn't want to incur that much debt.

You should probably understand that somebody who uses an efficient vehicle for commuting, and a big truck for real big truck things, is utterly an outlier. Most trucks in america have never left the pavement, and the largest thing they carry is a 40 pound bag of dog food. The vast majority of americans buy the biggest truck they can get credit for and drive it everywhere at 12mpg, usually while boasting "It's got a hemi!" as if that even matters after 1980

> the largest thing they carry is a 40 pound bag of dog food

No. Sorry. People do work with trucks. Tens of millions of people work in professions that it is necessary. More still do home projects or have hobbies that require high tow capacity like hauling an RV or a boat or other trailers. I drive an F350 to haul my fifth wheel. It actually gets 20 mpg when I'm not towing the RV.

You're misconstruing his comment. Not 'No one needs a Truck'. But, 'Most truck owners don't need a truck". I also will say I don't think anyone is talking about taking the option for owning a Truck away either. There will always be incredibly valid use cases for Trucks (and for a long time still; gas/dies. powered ones).

As you gave an anecdotal response, I'll respond in kind. I actually traded in a Tacoma for my EV. I miss the bed ~5x/year. I didn't need a truck; but i liked the option value. I'll likely get another Truck in the next vehicle cycle in a few years.

The vast, vast majority of truck owners use it in some way that a passenger vehicle is completely unsuited for. That there exist a handful of urbanites who drive a truck for the looks alone does not discount this fact. I see the trope a lot, however, from people who have never done blue collar labor in their lives. I had a Tacoma before the F350, I used it for camping and carried a roof top tent on a bed mounted rack. It was great, I never went camping more than when I had that truck. It was so easy and carefree to take a trip to the woods with no packing or planning required.

None of you are bringing any stats to the table, just a bunch of personal anecdotes. And there is a lot of people in the US, no way any of you know enough people to make a reliable claim about what 'most people do' across that large land. But there must be some statistics on this, no?

Those camping trips are why I <3'd my Taco and will likely get something similar in the future as the kids get older.

> The vast, vast majority of truck owners use it in some way that a passenger vehicle is completely unsuited for. That there exist a handful of urbanites who drive a truck for the looks alone does not discount this fact.

No one claimed otherwise. I feel like we've become a nation of, "This isn't right for me so it isn't right for anyone" bigotry.

There's a lot of anecdotal stuff when it comes to these conversations, You see the folks in the 7 seat SUVs commuting (that was me...but at the time I was also schlepping cubscouts a couple of times a month)...The SUV often got airport duty or the wife was taking her girlfriends somewhere....but during the week...it was me in one seat, driving 22 miles each way.

Also, the pollution is a massive issue in the cities, but a guy rolling coal in pudunk nebraska isn't a threat to the ecology, because there's so much ecology. I hate it, it's reprehensible, but it's leff of a factor...also, we're down in the trenches complaining about our neighbor, when the top polluters globally aren't the cars in the metroplex.

5th-wheel + truck can be addressed with a giant battery on the 5th-wheel itself. RVs, camper vans, 5th-wheels, etc. are going to have a lot of cool tech and builds here soon.

> It'll cost more...

Yea, everything will. We didn't account for the negative externalities of using fossil fuels properly. Now we're starting to. You also have the rest of the world being lifted out of poverty and there's nothing we can do about it and we'll have to adjust to simply not being as wealthy.

If anyone is interested in the opinion of an actual CDL holder, I think the technology has amazing promise and will eventually be adopted for many kind of CDL driving. I used to work for a public utility company (water/waster water), and we had a fleet of about a dozen or so dump trucks then we used whenever we had to dig something up. These were big machines, full size tractors with a dump bed instead of a fifth wheel (the mechanism that semi tractors normally have installed to tow trailers). They also served to tow flatbed trailers loaded with a backhoe, out to the dig site.

Now these machines never left our service area. They were always parked in the same lot every night. Putting in charging infrastructure in that lot would be trivial, and would almost certainly cost less than maintaining our own diesel storage and pumping infrastructure (which we did do because it saved us so much of our working day to be able to fuel everything up in the morning instead of having to drive to a truck stop before starting the work day).

A lot of trucking is local or regional. Those will be the markets that EV makes the most sense for. Where the trucks normally go back to the same lot every night, and the owners can cheaply put in slow charge infrastructure. Could we get EV trucking working for long-haul (a kind of trucking I've also done in the past)? Sure. It will require fast charge infrastructure though. 500 miles is a lot of range, but in long haul trucking, very often you operate in two driver teams, and your truck is always moving. Fast charging could solve that too, but perhaps the solution is to reduce the number of miles that freight spends on the road. Multi-modal shipping could be expanded. Use trains to cover large distance, and try to keep road transportation of freight to the regional level and local level. This is already an option that exists. Multi-modal is already a sub-industry in trucking. This seems like the most logical option to me, but I'm not an expert. Just somebody who has lived the industry for a few years.

Even slow overnight charging will have to be pretty beefy for these trucks and their batteries though. They're going to expend significantly more energy than your commuter car so they'll have to suck down a lot of juice overnight.

In the areas that companies build their truck lots with diesel depot's there is typically significant power available. The US electrical grid has expanded at a fairly steady rate. I don't see why it wouldn't be able to handle a continued growth rate via electric trucks.

Just to put some numbers out there, the semi battery pack is probably around 850kWh-1MWh. Imagine if this gets drained every day and a company has 10 trucks, the trucks have roughly 10-12 hours to charge (~6pm to 6am). So you'd need 1MW of electrical capacity.

This is equivalent to about 20 modern US homes (200A*240VAC) and not that crazy in the scheme of things. A medium sized factory in the US probably has a grid connection around this size depending on what they do.

It's certainly doable but it's not going to be cheap at all. That's at 1300 amp service of 480v 3-phase just for charging plus the chargers to handle that ~100KW of load per truck, to provide that power you're likely going to have the external charger setup instead of just one on the truck itself and those aren't cheap.

No, but installing an maintaining diesel tanks and pumps ain't cheap either! Especially if you have to deal with any environmental destruction caused by spillage or leaking tanks (happens way, way more than most people think).

If the numbers work, they'll do it. It's not like they're going to get 95% of the way through the design process and then give up because power delivery costs a little more.

If the ROI is there, they'll do it. If it's not, they won't.

Well yeah my point is the ROI is pushed out because of the extra expense of the install and acquiring sufficient electrical service above and beyond the cost of the new fleet.

The grid also has lots of excess overnight capacity. From the California Independent system operator data, the state has 42,000 MW of total capacity. Overnight demand is below 23,000 MW for much of the night. Even if we exclude the top 20% of capacity as too expensive or polluting, we still have 11,000MW of available capacity today in the overnight window in California alone.

I wonder how much longer the excess in CA will be there. As we are moving more and more towards solar we need more storage to handle shadow and the loads at night.

Thanks for your perspective. There are lots of people who like to shoot down technologies simply because they're ignorant of all the scenarios that exist. They see only one possible use (e.g. long haul trucking), correctly identify that the technology won't work in that scenario, then proceed to dismiss it entirely because they're unaware of the myriad of other applications.

The problem with these people is that they will go to great lengths to explain to you why your technology won't work in their specific application and feel smug because they think they've cleverly identified a flaw that you didn't think of.

Battery swapping seems like it circumvents the slowness-of-charge problem, both for short haul (port and local urban service) and later for long haul. (Dealing with detours on the latter could be tricky though.)

5 - 10 minute battery swap vs however many minutes charging. Sure it reduces the load carrying capacity and/or range a bit, but it should still be viable for many heavy goods vehicles. Swapping makes it easier to manage battery condition and vehicle utilisation. Over time battery provision/swapping could be outsourced, so the transport company can focus on trucking, not on batteries and their needs.

>I see this all or nothing argument when discussing new stuff.

That's one way to look at it. The other is that you can't pretend that real, tangible, identifiable problems don't exist. And Tesla has a habit of ignoring those.

>Slow charging is only a temporary problem and as evolution of Superchargers has made quote evident to date.

What have the Superchargers proven? That under increased load the locations can become completely overwhelmed? Even with EVs having a single-digit market share?

I haven't met a single person in my neck of the woods that thinks waiting 30 minutes to "fuel up" is remotely reasonable. And that's state-of-the-art. We have a ways to go.

>In fact, Tesla Semi's range is 500 miles because it's right at the edge of how long a driver can safely drive non-stop and right before they are legally required to take a break, during which time they can charge the truck.

Let's wait for some real-world testing before making claims to Telsa Semi's range.

Most people charge at home

> And Tesla has a habit of ignoring those.

Tesla definitely has a habit of ignoring problems, but in all cases I'm aware of, they're non fundamental issues that can be addressed along the way. They see the bigger picture and aren't focusing on non-crucial details.

> What have the Superchargers proven? That under increased load the locations can become completely overwhelmed? Even with EVs having a single-digit market share?

I'm referring to speed of Superchargers increasing over time. Newer superchargers don't have the drop in charging speed problem when two adjacent stalls are used at the same time.

> I haven't met a single person in my neck of the woods that thinks waiting 30 minutes to "fuel up" is remotely reasonable. And that's state-of-the-art. We have a ways to go.

Well your neck of the wood aren't the current intended audience. Thanks to everyone else who's buying the cars at their current state. They are enabling a future where your neck of the wood can have cars that charge in a few minutes. Oh, and I'm one of them. You're welcome.

> Let's wait for some real-world testing before making claims to Telsa Semi's range.

OK.

> Oh, and I'm one of them. You're welcome.

Holy shit. This is some top-level smug right here.

they make consumer products and should know to stay in their lane, this won't take off, or will be beaten by other players

oooor, they'll ignore the naysayers and make bank eventually, if they don't run out of funding before that. (Or don't go under because their CEO can be a raving nutbar sometimes.)

> Basically none of them are operationally able to go from 15 minute diesel fill ups on any corner to the infrastructure and load planning EVs require. Like they literally don’t have the software or expertise to plan the loads around electric needs without losing money on every load.

I understand you saying this is largely a software problem. Taking this at face value, I trust no other transportation related company but Tesla to deliver the best possible software for route planning (or anything really given they're just as big a software company as are in hardware -- see AI day presentations to get a feel of just how large a software effort they're running) if that is indeed a barrier to companies putting orders in for the Semi. I may even have heard them mention something about that in the presentation.

I think you sorely underestimate how much emphasis large trucking firms put into route optimization.

> I think you sorely underestimate how much emphasis large trucking firms put into route optimization.

I know software and I know Tesla's involvement in it. Based on your comment, I'm going to assume, they put maximum emphasis on route optimization, more than any other industry (which is probably false anyway). With that assumption, Tesla is in as good a position as a software company could ever be to hire the expertise (or outright purchase existing companies) and make the best software for fleet management IF that means they'll sell to significantly larger customer base.

That’s the exact same things you could hear 10 years ago regarding electric cats. And now every single car company is panicking to adopt electric, and many countries are going to make burners illegal.

So give out some years. Production capacity isn’t that high anyway, mostly because of battery production capacities.

Whether they are ready or not, they'll have to deal with competitors that won't have to worry about diesel fuel expenses. Tesla is not going to have any demand issues here. If this thing works as advertised (and that seems to be the case) and they ramp up production to 50k vehicles/year over the next two years, there are going to be two classes of truck companies: those driving electrically and those who don't. My guess is that when the dust settles, there won't be a huge market for ice trucks in a few years. Unlike cars, trucking companies operate based on financials. With razor thin margins, decimating fuel cost is going to matter a lot.

> Tesla is not going to have any demand issues here. If this thing works as advertised (and that seems to be the case) and they ramp up production to 50k vehicles/year over the next two years

Pretty big “if” they’re chief. While they very well may not have any demand issues, I’m going to go out on a limb and say they’ll probably will have supply issues. Launching the production of a new model is no walk in the park and the organization already seems to have problems with QA on the existing models.

Tesla has a lot of experience with this. They managed growth and profit throughout the covid crisis and related supply chain disruptions and invested in expanding production capacity further all while many of their competitors were basically struggling with staying profitable.

So, I would not dismiss their ability to execute on what they've announced here. They are obviously well aware of supply chain challenges and have been heavily investing in solutions for that. And they obviously waited with this until they had the logistics and manufacturing in place. And it seems that they are re-using a lot of components they already have - like their drive train and batteries. And they do have a lot of new production capacity coming online.

I don't think Ford has particularly struggled to stay profitable making the MachE and Lightning truck. Neither has Volvo (Tesla's biggest competition in the Semi market).

Yes - if there is money on the table, someone is going to grab it. Perhaps it's true that "Like they literally don’t have the software or expertise to plan the loads around electric needs without losing money on every load.", but if so those guys are going to end up having their lunches eaten by more nimble competitors that do.

Before Tesla launched model 3, targeted at 35k$, there was another EV launched. Chevrolet bolt launched at the same price range, and at a similar range. It didn't have the sleek look of a Tesla, lacked a supercharger network, and didn't have the lies of self driving, the hype machine and status symbol of a Tesla. As a result, it was a failure. When model 3 was actually launched, and much costlier than originally announced for, it was 50% costlier than the bolt. People still bought the model 3 over the bolt.

So, no, financial scrutiny is not why Tesla sells vehicles. It does because of the hype, the status symbol, the better design, and lies of self driving. And these don't really matter for trucks (except stock market rewarding hype, just like Bitcoin mania). So no, Tesla won't dominate the niche of short haul trucks that EV trucks have opened up.

> And no fleets in North America are ready to figure out converting a 2000 mile long haul over the road route into 5 400 mile EV relays.

This advertises a 500 mile range, which at 60 mph is 8h20m of driving. And a 70% charge in 30 minutes.

Does the US trucking industry routinely have people drive for 8h+ without even a 30 minute break?

Legally, no. Practically, yes.

Aren't tachographs or other logging means mandatory as in EU ?

If you're from the EU, you can probably already guess the answer to that question for the land of "don't tread on me" :-p

This is a rather unhelpful response given that I'm not sure many would guess correctly. Paper books have been used in the US for decades and electronic devices were made mandatory for most drivers (>8 days out of 30, apparently). Canada is now planning to emulate the US regulations on this.

> An ELD is an Electronic Logging Device that became compulsory in 2017 in the US for commercial driving operations required to keep hours of service records. If a driver has eight or more days worth of duty status logs (out of 30) they will be required by law to use an ELD.

https://www.optac.info/uk/eld-vs-tachograph/

https://www.fmcsa.dot.gov/hours-service/elds/eld-fact-sheet-...

The legislation linked is not about enforced breaks for drivers, though. Does the US have this?

Yes.

https://www.fmcsa.dot.gov/regulations/hours-service/summary-...

Cool, didn't know that, kind of surprising that the US would regulate this, knowing their general stances.

The conditions are quite harsh, compare them with this:

https://europa.eu/youreurope/citizens/work/work-abroad/rules...

Well, it's still far from EU, but it's something. Although rest times are VERY short.

US:

  - max 11h driving after 10h of rest (that sums 21h, so I guess you can drive 14h a day)
  - rest 30min every 8h of uninterrupted driving
  - max 60h driving a week

EU:

  - max 9h driving a day (10h driving allowed twice a week)
  - rest 45min every 4.5h of uninterrupted driving
  - max 56h driving a week. max 90 every two weeks.

Once you are switched to ELD's in the US, it is more difficult (compared to paper log books) to cheat enforcement [1].

[1] https://en.wikipedia.org/wiki/Hours_of_service#Enforcement

Yes, electronic logs are mandatory in the US.

> Short haul loads? Sure. In places that demand EV on certain timelines (like California).

> But for everything else, these vehicles require a fundamental reshaping of transportation in America. Not saying that’s not possible, but it ain’t happening any time soon.

Electric trucking doesn’t have to (immediately) take over the entire industry. All that counts is for that first part to be large enough. You see that with electric shipping, too. Short-distance ferries may already be economically feasible. That shipping stuff around the world on battery power isn’t, and maybe never will be, doesn’t matter.

yeah but what I hear is a massive software based business opportunity and you are amazingly well placed for it ... can we chat :-)

> Trucking is a single digit percentage margin business.

Only at HN:

1. Credit card processing is a razor thin margin business

2. Trucking is a low margin business

3. Retail is a ultra low margin business

4. Airlines don't make money

5. Phones is a low margin business

6. Desktops don't make money

7. Cars is a money losing venture

But, then how come all of the above companies built empires?

>But, then how come all of the above companies built empires?

By leveraging credit on those consistent low margins and not being ran as badly as most software companies that could not function without 50+% margins or infininte free credit.

Or by coming up with an adjacent industry that does make profit. (airlines, desktops, phones not made by the korean or fruit companies)

>Or by coming up with an adjacent industry that does make profit.

Or being owned by the company making the profit , after said company realised how much they were spending on 3PL when they already had HR/depots/etc that could be used for trucking companies.

> But, then how come all of the above companies built empires?

Scale based on insane amounts of venture capital and debt. If you're running 450M transactions a day like Mastercard, you only need a few cents per transactions in profit and you have a money making machine.

For example, car companies hand out huge bonuses to employees who figure out even a dollar in lower costs - at the scale of, say, Toyota with 8.6 million cars a year [1], that amounts to 8.6 million dollars more in profit. Give the employee in question a hundred thousand, the employee is happy and the beancounters even more.

Or airlines... when your average plane will always have a number of no-shows (around 5% [2]), it is more profitable to overbook and in the rare case that it does go wrong, pay off someone with 500$ in cash.

[1] https://de.wikipedia.org/wiki/Wirtschaftszahlen_zum_Automobi...

[2] https://techcrunch.com/2017/04/11/overbooking/

Well, do you need profit to make an empire? If your $major_company takes in $100B in revenue and spends it all, that's a lot of money to spend that will pay a lot of people across many industries. I'd say that counts as an empire, and I dare say I like it better than the one where the emperor gets all the ducats.

That's what everyone said about Tesla cars, but here we are with them worth more than the bottom 20 cars makers combine minus toyota.

What does stock price have to do with the quality of the product?

There's a New Jersey deli that was worth $2 billion recently. Theranos, a total fraud, got to $10 billion with nothing.

I think this should be launched in Europe first. Fuel is more expensive. Regulations on how long to drive are strict.

And with 400 miles it's already that far that you're not anymore in the country I live in (the Netherlands).

This is a poor design for Europe which has limits on the length of the tractor and trailer combined. Anybody using this tractor in Europe will have to use a shorter trailer.

If only there was some method for easily modifying software to suite changes to the hardware! /s

Isn't the vast majority of trucking 'short haul'. At least in Europe on a long haul, you have to stop quite frequently.

I do see an issue that these truck stops will require massively more electric power and some investment. (Please lets build more nuclear plants).

And in terms of maintenance and fuel cost these trucks should outperform diesel trucks by a wide margin and make up the capx different multiple times over.

A close friend of ours drove for a food service delivery company.

He picked up a load in the morning, drove a few hundred kilometre highway/city route with a dozen stops during the day, and parked back at the depot at end of day.

Perfect for the Tesla Semi.

There are hundreds of trucks doing similar routes just in our metropolitan area.

There is a huge potential market for these trucks.

And Tesla is not the first company to enter it.

In trucks this size, not just vans?

Trucks this size. Plenty of full-size semis with full trailers doing deliveries for larger businesses, especially in dense urban areas (like downtown regions) where you need that much trailer space for all the stuff you load.

I make and sell software to truck fleets too :)

I've seen some fleets plan their fuel stops using software to take advantage of fuel card discounts they get at certain gas stations, which helps stretch the single digit margins you mentioned. I can see this kind of software being extended to EV charge planning as well.

How many developers can there be making and selling software to truck fleets? Makes me think these are small time operations ripe for disruption a la what Uber did to taxi companies. No offense meant of course! Just wondering :-)

Do you mean the trucking companies or the software developers? The trucking companies aren't small companies by any measure.

Werner Enterprises for example has 13k employees, 8k trucks, and $700M in revenue. And they're a small fry in shipping. JB Hunt has 30K employees and $12B in revenue. Swift Transportation has 22K employees, 23K trucks, and $6B in revenue. Schneider Transportation has 19K employees and revenues of $5.5B.

Many of these companies have in house developers working on custom logistics packages to optimize every single detail of their operations. It's a very small margin business.

I’m in the midst of job search right now and I’m interviewing with two different companies doing shipping logistics. I’ve seen at least one other in the course of this job search. I’m guessing that there’s been a lot of thinking happening in this space already.

While you have compelling arguments, there's others in favor of taking the hit on efficiency; lower fuel costs, higher efficiency / speed, but probably more importantly, environmental impact. Unfortunately, being more environmentally friendly frequently comes at a cost. But the lower maintenance and fuel costs will help offset some of those costs, in theory.

> Basically none of them are operationally able to go from 15 minute diesel fill ups on any corner to the infrastructure and load planning EVs require [... for] electric needs without losing money on every load.

> Trucking is a single digit percentage margin business.

This is a basic economics problem then. The cost of writing the software. The cost of hydrocarbons. The cost of electricity. The cost of time. The cost of maintenance. The cost of breakdowns.

I don't know too much about long haul trucking, but I know a decent amount about vehicles in general, and I have an engineering degree so my knowledge of physics is better than average.

I have yet to see a compelling reason why non-electrified trucks will survive. If you only look at the cost of breakdowns and fuel, and compare it to the cost of time, those breakdowns will pay for themselves. Completely leaving aside self driving technology. Even if we have to pay for drivers for the next thirty years. It does not matter.

Everything else will bend around the basic economics of electrified trucking. The law, the charging stations, the software, the schedules, everything. When an industry has such tiny margins as you correctly observe, an entrance of a technology can that can only serve a niche, but at extremely wide margins compared to existing players, will absolutely dominate.

There is a reason Pepsi is one of their first major buyers. The soft drinks are in the cities. But as these trucks dominate there more research will go into supporting them elsewhere. Fast swappable batteries, MASER charging along highways, more charge points per truck.

It doesn't matter. All these problems will be eaten by the massive wave of cash coming towards them by the early entrants that applied the technology where it was most suited in the short-term.

And then after that we'll get self-driving and it will be truly lights out for the legacy logistics companies. The only question is where in the five to fifty year period that will be.

For trips to very remote places that get shipments once a week or so diesel with the ability to strap on extra tanks seems like a niche that will last for a long time. And collectors will keep a handful of engiens going.

Probably other niches too, but the vast majority seem like ev is the future.

In what case is a long haul really required? Other than specialist superheavy or extra wide/long loads, everything should be on a train for those long runs. Modalohr [1] wagons make it very easy, although I have seen a specialised forklift setting semi-trailers onboard [2] trains in Rotterdam. This sort of handling is probably ripe for automation (there is a standard for the trailers, add some tags for computer vision and work on it from there). I saw something recently about a Modalohr-esque system to put hooklift containers on wagons for local waste collection in Switzerland.

My personal dream is that not only is port to industry fully on rails, but local distribution is rail-oriented also, to the point that a significant proportion of deliveries to shops in cities are made using pallets or trolleys loaded on freight trams. I do understand that the unusual nature of America makes this sort of change nigh-impossible, but maybe they could ship their semi-trailers on trains?

[1] https://en.wikipedia.org/wiki/Modalohr (second picture shows it best) [2] As in this image: https://forwardermagazine.com/wp-content/uploads/2022/03/Pic...

Rail companies don't know how to operate the business model required for that. They can get you a train load every day no problem, but if you want a truck once in a while at times not planned well in advance they are lost.

I have worked in "lower tech"" fields and have worked extensively with trucking companies, so I can somewhat relate to what you are saying and agree it is a bigger obstacle than people in tech give it credit for. However, it is also a nice business opportunity for a software team to build a really important product. Somebody should be able to get VC funding for this. BTW, in my experience, once someone wins this business and a trucking company is using your software, if you don't screw it up badly, you are locked in for a LONG time.

> And no fleets in North America are ready to figure out converting a 2000 mile long haul over the road route into 5 400 mile EV relays.

If there are genuine advantages to EV trucking, then I don't see how any of this matters. This isn't a heavily regulated industry and there's very little barrier to entry. If you can save money on shipping by buying a bunch of Teslas and converting long hauls to relays, then someone will just do it regardless of whether the existing players are "ready to figure it out".

Now, are there advantages? We'll see. A lot depends on what happens in the fuel markets in the coming years. But the long term trends point away from diesel for sure.

No disagreement with what you said but hasn't the whole pitch for Tesla Semi been for short hauls right now? Short hauls make a ton of sense on paper, limited highway miles where energy use can be highest, lighter loads and more stop and go. With a ev semi it seems that you might be able to take a large part fleet maintenance out as well as fueling logistics.

Explain why short hauls make sense. Your post doesn't show it.

Probably just throwing food under the bridge but here we go.

Short hauls:

1) Are at slower speeds which for most EVs means greater efficiency.

2) Have more stop and go driving which is more efficient in EVs compared to ICEs. I don't have the papers handy but IIRC there can be more pollutants generated in the acceleration phase of a vehicle, another reason EV makes sense for short hauls.

3) Park at the same facility at night which can have charging setup. Long hauls, unless specific corporate routes, do not have charging infrastructure.

1 goes for any vehicle. 2 yeah, nope no source. EVs weigh more and therefore pollute more when braking so should not be used in cities. 3 no advantage compared to combustion engines.

Ahhh you misunderstood my original statement. Sorry, thanks for taking care of the bridge though.

"Long haul"

Just add a diesel generator to the back. Problem solved. :)

Well this only create a market opportunity for a new software, or new software version.

Imho the biggest global challenge is electric supply. And I mean globally, not US + Canada. If we want EV trucks and cars globally we need way more nuclear power than we have right now.

It is all fun and nice to fill USA, Canada and some wealthy countries in Asia, Oceania and Europe with EVs. Will that change a lot in term of climaye change if other countries cannot develop and decent charging network, are struggling already with energy demand or still rely mostly on coal to produce electricity?

> And no fleets in North America are ready to figure out converting a 2000 mile long haul over the road route into 5 400 mile EV relays.

Uh... I worked for one of the largest LTL companies in the US for years, and to my knowledge all of their road drivers were "home every night". Several runs out West were "hook and turn" - two trucks drive toward each other half the day, meet in the middle, swap trailers, and drive back home.

Further, a 500mi range is starting to get awfully close to the "11 consecutive hour" DOT Hours of Service limits.

Look, you are missing the big picture here.

Imagine a remote controlled truck and then, later, a fully autonomous truck.

This is what Tesla is aiming for here, they just have to begin somewhere.

This Semi is not a truck, is a data collection tool.

For some customers in some settings this already make sense as it is, but the goal is hit the road and start collecting data as soon as possible.

If that in mind, now imagine a fleet of 20 trucks being remotely monitored by a single driver. Does that make sense for you in terms of logistics and economics?

I can't take you seriously if you only cite the negatives and not the positives. There's $60k/year of fuel savings to be had here. Show why that isn't enough, rather than ignoring it.

The side of the equation you're ignoring is maintenance. Commercial fleets also have garages and FTE mechanics maintaining the vehicles in the fleet. That all costs money and needs to be factored into your cost analysis.

I work for a large public utility generating electricity. We switched our fleet to electric trucks years ago, and not because we're an electric company - frankly our bean counters don't care about that! It all came down to lower operational costs, higher reliability, and vehicle longevity. Being an electric company we don't have to worry about charging infrastructure, but still - the electric vehicles proved to be significantly cheaper to operate, and not just in fuel costs.

If trucking is an no dispute single digit margin wouldn’t a double digit reduction in fuel cost create a massive competitive advantage

Diesel Trucks margin will go down as more and more regions make it financially less attractive to use diesel trucks. Then the prices for hauling will go u, infra will catch up, software will catch up too, and EVs will become the best choice.

That’s how I expect this evolution to happen. Timelines are difficult to predict though. Looking at the current momentum I really expect the poitn where most new trucks are EV to happen within a couple of years at most, not a decade.

Calling it deep dive is questionable, at best.

Tesla only released marketing fluff so far. No price, no payload capacity, no details of range (yeah, it did 500m drive, but how does it handle different speeds, environmental conditions, terrain profiles, etc), no details on the tech, etc.

It has potential to be really cool, but at the same Tesla isn’t exactly known for their marketing fluff being accurate in the real life.

Suggest you read the article. They do a pretty good job of estimating a lot of that and showing their work.

Yeah, exactly. Estimating, based on marketing fluff. Let’s wait for real numbers before we claim deep dive.

Bingo, if its a working real product. Why would then not released concrete numbers on the most important factors?

Its like building a house, and only listing colours that house can be painted with and that it comes with reasonable amount of doors and windows. Buy now!

If you think Tesla's "marketing fluff" is factually wrong, you can make a killing with a lawsuit. Misleading investors is a huge thing. Fluff can lack details, but if it false then Tesla will be in a world of hurt.

BTW can you explain how the height of Pacheco Pass on I-5 is marketing fluff?

Hey all! I'm Jaan, the OP of the Semi article here. Great discussion already!

I'm off the 'net for a few days but if you have questions or suggestions, hit me up on Twitter (@theEVuniverse) or on my email jaan/at/evuniverse.io. I also write a weekly EV industry newsletter which is where I uploaded this from. Cheers!

Engineering Explained made a good breakdown and analysis video on the physics of the truck:

"Does The Tesla Semi Make Any Sense?" https://www.youtube.com/watch?v=Uv44W7xa4IU

The TL;DR of that great video goes against Betteridge's Law: yes it makes sense ("no fatal flaw found"), with the known information at the time Jason made the video. The charging infrastructure is likely not trivial, though; in the comments it was pointed out the infrastructure likely makes heavy use of batteries to avoid stressing out local substations and avoid abusing grid peaking.

> the infrastructure likely makes heavy use of batteries to avoid stressing out local substations and avoid abusing grid peaking

From Tesla's presentation, it looks like that.

Right, and that throws off calculations I've seen so far on Semi's ROI because I haven't been able to find out what that charging infrastructure costs per truck in a single truck scenario, two truck scenario, five truck scenario, and so on. MWh of batteries and the accompanying high-voltage installs do not come cheap, and there is maintenance and capex replacements associated with all that.

As much as all that costs and as much electricity it gulps down, you might as well arrange with the utility providers to put in the marginal additional equipment to secure a discount for always consuming out of nuclear or similar baseload during low utilization periods, and only rarely charging batteries outside those periods, trying as much as possible to avoid tapping peaking plants' output.

The lifetime TCO calculations for the charging infrastructure for full battery replacement included is going to be interesting to peek into. On a simplistic basis, that TCO plus the electricity costs would be put up against the cost of diesel, Diesel Exhaust Fluid (DEF), Diesel Oxidation Catalyst (DOC) catalytic converter, maintenance on the DEF and DOC subsystems, and so on.

With my investor hat on, I still don't have the numbers I'd like to see to evaluate whether the truck is a slam-dunk ROI on short-haul routes.

Most industrial facilities already have 3-phase 240 or 480 power and tend to be fed by higher capacity feeders than you see in residential areas. They also tend to be closer to high voltage feeders because everyone hates HV lines running through residential areas.

People are way over-estimating how much work will be needed and how difficult it will be. In many cases the utility will either swap out or drop another 3-phase transformer in place with CT-based metering. The customer will run some conduit. The more power you anticipate purchasing from the utility the bigger the discount they'll give you on the work. In some cases the existing medium voltage lines will be at capacity so they'll pull a new one from a substation but they aren't going to charge $1m for it. This isn't exactly rocket science... everyone who builds a new factory goes through the same process. Utilities do this work every single day.

EV truck fleets will be rolled out over time. The grid and charging infrastructure will adapt. Everything will be fine.

An important question for me was how do you defeat massive energy density of desal. This is a great video to answer that. TLDR; Semi will need at least 4 tons of batteries. But this works out because it is still probably around just 10% of weight of the fully loaded truck. It occurs to me that economy degrades if you have to haul big but less heavy loads. Cost wise you still make profit as long as truck is loaded and electricity remains cheap. Overall, this would be good solution for short haul heavy load.

Even if the payload is 30k-35k lbs in the conservative case, it still makes it a no brainer for high volume , relatively lower weight applications. Consistent, predictable routes and economies of scale can be achieved on high traffic routes, especially if the terrain is uneven. I just hope the Semi can outperform Diesel semi a lot in the lifetime maintenance costs.

In the presentation, Musk stated that the drivetrain is guaranteed for 1 million miles. I have no data on what sort of guarantees Diesel Semi manufacturers offer. Would be good to work that out. There is battery life and degradation as well to take into account.

Elon's word is worth very little these days, needs to be on paper under scrutiny of a lawyer or it's garbage.

Regardless of whether you believe Elon, this has been shown for the Model 3 drivetrain before

https://electrek.co/2018/10/15/tesla-drive-after-million-mil...

And it's not that hard to believe given that it is a system with very few moving parts and no constant explosions happening in it during operation.

> this has been shown for the Model 3 drivetrain

Is there anything more than a picture released by Tesla to that article? I would have expected test results or at least some data on how it holds up.

I can't tell if this is satire...

Anyone who's visited TSMC can tell you drive unit failures have been happening before 100k miles let alone a entire million.

People have had back to back drive unit failures on new cars.

I recall a MS owner getting the run around about an impending DU failure (which manifests as a whine under acceleration) as their warranty was about to expire, only to have it fail out of warranty.

Now people are trying to DIY DU whine on cars that haven't even made it to 100k.

Running a DU in a simulated 1 million mile test is nowhere near the same as actually lasting 1 million miles in the real world over as many years as that would take.

The only thing that matters is the percentage of those failures before the 1mil miles. And that'll be a while.

If Teslas had nearly any concerningly significant number of drivetrain failures, nobody would buy them and they've been around long enough for people to catch up. I did plenty of research before buying mine and I follow Tesla quite closely. Haven't heard any concerns in this regard.

I expect Tesla to only get better for the time being.

> The only thing that matters is the percentage of those failures before the 1mil miles. And that'll be a while.

Please tell that to someone dealing with back to back DU replacements within a couple of years owning their cars.

Or people being told to scrap their cars by Tesla service advisors because even if they fix their DU something else will just break: https://teslamotorsclub.com/tmc/threads/out-of-warranty-driv...

All cars have reliability issues.

We need to stop pretending like replacing one powerplant with another will change that. An entire generation of drive units are failing because of an issue the same boring issue that kills so many ICEs: Coolant getting where it's not supposed to because of a failed seal, causing lubricants to fail.

If you want a car that will go a million miles today, a "boring" base model Lexus is probably a safer bet than a Model 3.

> Elon's word is worth very little these days

His words are worth fine. Articles and second hand mangled version of what he actually says is worth nothing - I agree.

As far as this topic is concerned, demand for Tesla Semi is enough evidence.

> As far as this topic is concerned, demand for Tesla Semi is enough evidence.

The Tesla Semi is the Tinker Bell of automotive.

IMO, this is Elon not understanding the market he's talking to. This type of warranty sounds great to general consumers.

In commercial trucking, it's more about the downtime. Making it to 1 million miles doesn't really matter if the maintenance along the way isn't easier. These trucks will still require tire changes plus maintenance to shocks/struts/control arms/air pressure systems/etc. Stays might be shorter, but commercial trucks already have processes for maintenance.

The standard warranty on a semi-truck engine is 4 years and 400,000-600,000 miles. A million miles isn't impressive, it screams "we're so far behind, we need to distract them with flashier incentives."

Maybe they can build up their reputation in time to pay for all the warranties. I guess it worked for Kia? IIRC that's around the time that all those "cash back" incentives became popular with the other manufacturers, before they jumped on the 10k bandwagon too.

Yeah - if I were signing a contract for these semis, I would want the '1 million miles' warranty to mean 'if the truck stops working and can't be repaired by Tesla within a week, then Tesla will buy back the truck for the purchase price * miles driven/1 million.'

I'd define 'stops working' as 'stops being road legal or able to drive at the speed limit with the rated load for at least half the advertised range'.

So a broken cab heater or faulty app wouldn't be reason for a warranty return.

I would obviously expect to pay a higher ticket price to get this warranty.

> I would want the '1 million miles' warranty to mean 'if the truck stops working and can't be repaired by Tesla within a week

I want this for everything I buy that if out of order, would cost me time or money. My home coffee machine included.

A new semi truck starts at about $100k on the low end. You can get the same level of service by spending $5k on a coffee machine meant for coffee shops.

Intrigued so did a bunch of searches for commercial machines and can't find any mention of "fix it or take it back and pay me for my time" terms anywhere.

Care to share one?

Oops I thought you were being realistic :)

Edit: SLAs?

In general, if you want this you'd lease the machine. Fairly common for independent cafes and restaurants (and those bean-to-cup machines in offices).

Though, also, the heavy-duty commercial machines are pretty reliable and service-able; you wouldn't generally expect to be out of order for a _week_ even if you owned it.

If Tesla current maintenance track record translates to semis...

The promise of EVs is basically less downtime and lower cost. You'll still have to swap tires of course. Brakes will last a lot longer because they are not used a lot. There is no diesel engine to service. The drive train is the bit that is supposedly going to last a long time without efficiency decrease. All the rest is going to be similar. But it looks like some of the more expensive maintenance is going to be easier.

I'd say Elon Musk has a good product here that will likely sell very well.

>The promise of EVs is basically less downtime and lower cost

I’m guessing that you don’t own a Tesla.

I do. And I like mine, a lot. But it doesn’t deliver on either of those promises, and it’s hardly an outlier.

I also own a Tesla. Over the 6 years I have owned it my automotive costs have decreased by some 80% compared to the ICE vehicles I owned prior. Significantly, the “no oil changes/no weekly fill up” costs immediately saved me around $2800/year. Maintenance costs have dropped to next to nothing, primarily being around tires and windshield wiper fluid.

There’s also no risk of getting a catalytic converter stolen.

What are your expenses that are so dramatically above the norm?

>What are your expenses that are so dramatically above the norm?

My comment about cost was in relation to purchase price, not ongoing cost.

I recently bought a 2022 Model S Plaid, and while there are certainly cheaper Teslas out there (namely, all of them), Tesla doesn't come close to competing with ICE vehicles on cost of acquisition.

There are more expensive cars out there than even your top-of-the-line model. I’m not sure what point you are trying to make.

I don't know what your situation is, but here's mine:

Got a Model 3 back in 2020. The service I've had to get done to date has been what's in the user manual: tire rotation, cabin air filter replacement and brake fluid check. That's it! No down time other than the few hours per service visit (for which they paid me equivalent of $50 USD to travel from and to the service center on each visit).

And I've used 90% Superchargers to charge and despite that, saved a lot on gas.

I'm glad that you haven't had any issues. I bought a new Model S Plaid a few months ago. Since then, it has been in for service three times, nearly all of which was for work that can be accurately categorized as "Tesla didn't finish putting the car together."

Examples:

There was a very large, very obvious gap between the charge port cover and the adjacent brake lamp assembly. Easily over a half-inch. I was dreading the possibility of getting 'in spec'-ed by Tesla, but my concern was misplaced. Turns out, Tesla hadn't finished installing the brake lamp assembly. It wasn't fully seated.

Rattling from under the frunk? Multiple clips were missing from the plastic trim, and of the clips that were present, many weren't actually clipped in.

Thunking sound from the driver side front tire on bumpy roads? The hub bolts and upper shock bolts were all loose.

Rattling from the center console? Tesla hadn't torqued down the fasteners that attach it to the car.

The hood ornament on the front comes with protective plastic that is supposed to be removed at installation. It wasn't. And it's not the kind of plastic that you can just tear or peel off. Tesla had to do that.

Did I mention that I found out - at delivery - that the new car needed a new windshield?

I have another service appointment coming up soon. The trunk floor was slightly crushed in (about an inch) where the floor meets the rear passenger seat. They would have replaced the trunk floor a while ago, but they couldn't due to parts availability. Also, some of the weather stripping is separating from the car, and it can't just be pushed back in place.

Oh yeah, here's one: after my most recent service appointment - for the suspension issue - my driver profile no longer existed. Good times.

There's more.

Somehow, I still like the car — mostly because it's insanely fast. But this is totally bonkers for any new car, let alone one with a $145,000 price tag.

OMG! So sorry to hear that. That's awful! :( but thanks for sticking with it.

>thanks for sticking with it

I don’t understand. Why do you care whether or not I stick with my car?

Jesus, I wouldn't feel safe driving that thing.

Sounds like your car might be eligible for the lemon law. The amount of issues you're having is insane.

My car with 200k+ miles doesn't even have this many issues.

It’s not eligible in my state, though I wouldn’t pursue a claim if it was.

I totally agree on the insane comment.

I also own a 2018 4Runner. It has around 180,000 miles, has been driven over some of the roughest backcountry routes in the United States, and has had no quality or performance issues. Ever. Just routine maintenance.

How do you know they didn't take any of that into account?

In every single field lots of studies have been done on electric vehicles of all kinds, including cars, trucks, vans and planes. And every single time, electric vehicles show that they are cheaper to maintain.

> but commercial trucks already have processes for maintenance.

And I'm sure they would like to get ride of as many of those processes as possible.

Might make a lot of sense up in the PNW where electricity prices are very low (thanks to abundant hydro and low population density) and diesel prices are relatively high.

The electric drive train with regeneration would also be very beneficial in the PNW's more rugged terrain.

And at least Washington and Oregon are quite keen on environmental improvements, so they might get low resistance and even help from government and regulators.

Considering everyone tends to point out the Achilles heel of EV trucks is the high kerb weight limiting the tonnage of useful cargo, I find it really funny that their premier touted customer is Frito-Lay, whose products are universally referred to as bags of air.

Did a quick search. Diesel Semis' curb weight ranges b/w 10000lbs-25000lbs. The 10000lbs one may not be comparable in performance/range to Tesla Semi so keep that in mind.

Tesla Semi based on the estimate in the article, is about 27000lbs.

Though their other launch customer is Pepsi whose products are cans of water.

Just to point out the obvious, Pepsi and Frito-Lay are owned by the same company, PepsiCo.

Wow, we can finally make a meaningful dent in emissions, not to mention pollution. In North America, the trucks are the primary mode of goods transportation, and they, not cars, contribute the bulk of transportation emissions.

> In North America, the trucks are the primary mode of goods transportation, and they, not cars, contribute the bulk of transportation emissions.

ORLY, source?

According to epa.gov [0] light-duty vehicles constitute 57% of transportation sector GHG emissions. Medium-and Heavy-duty trucks: 26%

[0] https://www.epa.gov/greenvehicles/fast-facts-transportation-...

According to the DOE [1]:

Light duty: < 10,000 lbs.

Medium duty: 10,000-26,000 lbs.

Heavy duty: > 26,000 lbs.

Light duty covers most of the 10-20' UPS/Fedex/whatever trucks for last mile urban delivery which is far less efficient than delivery to a warehouse. Without more data on fuel consumption by gross weight, it's impossible to separate industrial delivery from personal travel/commuting.

[1] https://afdc.energy.gov/data/10380

According to search results those delivery trucks are >10,000lbs GVW.

Do you have any better sources backing your UPS/Fedex/Whatever light-duty claim than the random results I find in ddg searches?

You're probably right because the max GVW on a single axle is 20,000 lbs but my source is the USDOT sticker on the license plates of several Fedex and UPS trucks that deliver to my suburban address and my parents' exurban address.

And yet, from [1]:

> In the chart here we see global transport emissions in 2018. This data is sourced from the International Energy Agency (IEA) [2].

> Road travel accounts for three-quarters of transport emissions.

> Most of this comes from passenger vehicles – cars and buses – which contribute 45.1%. The other 29.4% comes from trucks carrying freight.

[1] https://ourworldindata.org/co2-emissions-from-transport

[2] https://www.iea.org/data-and-statistics/charts/transport-sec...

most of these passengers are also just carrying a single person. The car industry has won the culture war and convinced everyone it's normal and sane for every individual to have a car and use it all the time instead of investing in our public transportation infrastructure

We're about to have rail strikes in the UK. I asked a taxi driver in London whether he gets more trips/surge, and he said it gets more than negated by the added traffic.

And that really brought home how a public transport infrastructure also massively benefits drivers. Cities without it (I'm thinking of LA specifically) are gridlocked.

https://www.youtube.com/watch?v=d8RRE2rDw4k

As a driver you want everyone to be walking, cycling, taking the bus/trolley/tram/train/...

Why? Because you'd be alone on the road.

The numbers missing in those statistics is how many cars versus how many trucks?

For example, it's more impactful to swap out 1 million trucks driving 12 hours a day than 10 million cars each being driven for less than an hour. Same as for buses, they're on the road all day long doing multiple trips.

Buses regularly carry 50+ people. Even doing four roundtrips entirely empty then one fully packed is better than 50 cars doing a single roundtrip.

> In North America, the trucks are the primary mode of goods transportation

Umm this is incorrect. The primary mode of goods transportation is freight trains.

Because as we all know, _building more trucks_ will lead to the old trucks just magically disappearing, and not just have twice as many trucks on the road now. And as we all know, the building of that truck definitely doesn't emit a truckload of pollution, making batteries is a process that doesn't destroy the environment to mine the metals it needs, 50% of the electricity in the US doesn't come from fossil fuels, and Tesla doesn't make all of its money by selling carbon credits to other car makers, leading to the exact same amount of pollution anyways.

The way you make a dent in emissions is by increasing public transit for the public side of things, and making more freight trains so that your trucks only need to drive 50 miles, not 2000.

I can't wait till Munro and co take a look into this stuff. The amount of people already crapping on this is astounding.

Those whose opinion actually matter have done the math and put orders in to save them cost in the long run.

Of course, if right now, the load capacity or the range isn't applicable to one's business requirements, then Tesla Semi doesn't make sense. However, cost is often the biggest driver in any business. If a business can change things around to fit within the current constraints of the Semi, then they can do a cost-benefit analysis and Tesla Semi may then make economical sense.

I expect the current orders to be PR motivated.

Hard to tell, but given Tesla's historic record, demand for their products has only ever increased (discounting the current China situation which most likely has nothing to do with disparity between PR and actual product perf).

What's the "China situation"?

Pepsi’s order was placed in 2017.

Not sure what you're implying with this. Specs have remained exactly the same and now, they've actually delivered on the spec they advertised in 2017.

Let's wait and see if that's actually true. Tesla specs are a lot more "optimistic" than specs for other EV manufacturers (see the promised vs actual ranges of their passenger cars, for example).

Delivered? They drove a demo version up on stage at a PR event. Come back when Pepsico has their fleets...

I'm gonna be broke after all the egg I have to buy to throw at your faces.

I don't think they will buy one of these.

Who is "Munro and co"?

The top result on Google is this baby clothing brand: https://www.munroandco.co.uk/

Cute outfits, but they don't look like EV trucking experts.

Sandy Munro's company, that specializes in providing the automotive industry with analysis and optimization.

https://leandesign.com/industries/automotive/

Tesla Semi is not the only contender out there now (or at least in progress):

https://freightliner.com/trucks/ecascadia/

https://www.peterbilt.com/electric-vehicles

https://www.macktrucks.com/trucks/lr-electric/

Those "contenders" are referenced in the article:

- Escadia has less than half the range of Tesla Semi and ~6 times slower charging rate per mile.

- Peterbilt has 3.5x less range that Tesla Semi and ~10 times slower charging rate per mile.

- Mack has 5x less range than Tesla Semi and ~10 times slower charging rate per mile (not in the article — I looked at the spec on the link).

What are they exactly contending on?

You’re assuming that range is actually useful. There’s a reason Tesla is also producing a 300 mile semi, EV’s really only work for short haul trips anyway so a big battery is pointless for many users.

Ideally you have charging infrastructure at a loading dock between shore trips not along the highway at which point a 250 mile Escadia vs a hypothetically faster charging 300 mile Tesla isn’t really worth much without a much larger investment in charging infrastructure.

> You’re assuming that range is actually useful.

I'm not assuming that. The entire trucking industry keeps talking about how they won't buy low range stuff. Also, while batteries are still slow to charge, any extra minute spent waiting to charge multiple times on a trip is money lost.

EDIT: instead of speculating, get hold of someone in those companies that put order down for Tesla Semi and ask them why in their right mind they would do that. Then come back and enlighten us with their response.

Not quite, the long haul industry has ~zero interest in the 500 mile Tesla Semi, but that’s hardly the only market for semi’s.

Safeway, Walmart, etc are using semi’s for relatively short daily deliveries from regional warehouses. That model is very different from companies picking up random freight from LA’s port and shipping it to arbitrary locations across the US.

PS: As to your edit the list of buyers was exactly the kind of companies I described with long haul trucking companies ordering token amounts. Walmart, Pepsi, Anheuser-Busch, etc took a bigger bite.

Sysco: The food distributor has reserved 50 Semis. while Flexport: Ryan Peterson, the freight company's CEO, announced the company has ordered one Semi.

https://www.businessinsider.com/companies-that-ordered-tesla...

Burning diesel with unpredictable prices is also money lost, there would have to be a tradeoff or niche where it makes sense

As a strawman, it should just possible to have a kind of "B-train" with a smaller, detachable battery unit just behind the tractor (apologies to truckers if I am using the wrong terminology). This way it would be possible to swap this out super fast. Maybe it could be at the back for faster swap.

And even if not, you could "pony express" a route across the USA with locations that just have the trucks charged and ready to go.

Pull in, grab dinner while the crew unhooks your trailer and hooks it to another already charged semi, and then off you go.

Wouldn't work for owner-operators but a large enough line could do it if they wanted to for the climate/PR benefits.

Of course, that kind of hauling can be done by railroads, too.

Isn’t having crews doing this hooking/unhooking going to kill the price competitiveness of EV semis?

I think pack at the back designs would be quick to swap out. Going on a long trip - use two.

Theres already designs for doing quick swaps, like always its the challenge of standardization:

https://youtu.be/g4EbWRDXzZw?t=49

Assume an average delivery truck driver's job is 10 hours a day. Assuming no time for unloading, that would be 50 miles an hour at the 500 mile range.

50 miles an hour is probably high for the average speed of an in-town deliver truck. And even if all it is doing is dropping off trailers and picking up new ones, that would be adequate for quite a bit.

And if the truck is stopping and the driver is unloading pallets, etc, then 500 miles is way more than enough; 300 comes into play.

You need some margin for days with headwinds or cold or road closures / detours.

You just assumed a brand new battery, fully charged, and then taken to empty in optimal conditions.

Nobody would actually use a truck like that day to day, you need slack for every thing from aging batteries to making it to charging infrastructure and running the AC.

These trucks can all be charged with 240V whereas the Tesla semi requires a 1000V (mega charger) which isn't available anywhere yet. So these trucks are all contending with being far more practical.

Existing Teslas can accept various voltage for charging. Obviously, the lower the longer to charge. Pretty sure the Tesla Semi CAN charge at up to 1000V, but doesn’t require it, while the competitors CANNOT charge at 1000V.

What's your source that Tesla Semi can't be slow charged with 240V? I'd be pretty shocked if that was the case.

I read it as "requires 1000v to get the charge speeds advertised" - at 240v the trucks would all charge at about the same rate, and 240v may be what everyone goes with (though I doubt it, might as well install a dedicated charger if you're buying a truck).

There's a difference between AC and DC charging. All current Tesla's accept 240 volt AC and there's no reason to expect that the semi won't either but with the such a giant battery pack 240V will simply not be fast enough. The 1000 volt is for DC fast charging. Tesla's current models use 350-375 volt battery packs. 1000 volt will just allow them to charge faster for DC fast charging, and more efficiently.

To add to this, for AC, there is a conversion stage (ie inverter) that needs to convert it to DC to charge the batteries. This unit, at the currents that the Semi requires to be charged in a reasonable amount of time will be wasting too much energy in conversion losses.

Due to its inpracticality, AC charging is unlikely to be an option, but maybe they include it for those exceptional circumstances.

Many trucks do only short hauls anyway. It is easier and less risky to develop a truck whose main usage do not already need a wide and extensive charger network.

I think it is more about opportunity than capability.

Why not Volvo and Scania?

I didn't list all the ones, just the ones I found while looking at "the market".

I am surprised that everyone is going relatively strongly toward semis instead of box trucks; it seems that would fit the "current" capabilities much more.

There are box trucks in the market and in service already. They're definitely further along than semis. Tesla is really good at pushing the public perception of EVs and yet doesn't make one, so it's possible it's just a less visible part of the market right now.

https://www.renault-trucks.com/en/electromobility

There are also European and Chinese manufacturers. E.g. Volvo already has trucks on the road and the Chinese have a huge head start for producing all kinds of electrical vehicles. So companies like Geely, BYD, etc. are also entering this space.

But you are right that Tesla is neither the only one in this market nor the first to market. They would have been five years ago when they first announced their plans. A lot of companies have started building their own trucks since then. However, product announcements are one thing and volume production is another thing. If Tesla manages to get to volume production over the next two years (like they announced), they'll still be early enough to grab a decent chunk of the market. Especially in the North American market which is pretty much wide open still. Tesla seems uniquely positioned with their supply chains to pull this off.

Company close to where I live has also been selling semis with a 1MW battery pack for a while: https://www.designwerk.com/en/high-cab/

I always assumed the future of electric trucking would involve little battery trailers that could be quickly swapped out on a rolling basis by fleets of trucks.

It seems the “refuel” paradigm makes sense for fuels where the majority of the fuel is consumed but maybe not as much for charging. With a network of truckers in the US a floating number of charging centers with batteries wouldn’t be that tough to create/maintain

I worked for a company that builds electric trucks and that have also shown battery trailers with cargo space. The engineers told me that the coupling of the battery on the trailer with the semi caused them a lot of worries. High voltage/amperage combined with mechanical stress and failure scenarios (failing mechanical coupling of the trailer, destroying the battery connection).

However it seems like the European companies heading this way.

Yes, just exchanging batteries is an alternative to recharging. But it requires additional logistics and also the investment into additional batteries. There are of course scenarios, where battery change is the only practical way. Like it was with horse changes on long distance travel long ago. This changes quickly, if you have batteries with enough range to make recharging feasible. With electric cars the industry was quickly at the point and looking at the specs of the Tesla Semi, heavy trucks are basically there too. Which probably will kill any attempt at setting up a network for battery exchange before it was made.

> I always assumed the future of electric trucking would involve little battery trailers that could be quickly swapped out on a rolling basis by fleets of trucks.

Maybe the future, but right now, I suspect the battery energy density is not forgiving enough for the battery to just be a carry load. Increasingly, the battery module is part of the structure of the vehicle and further works as weight to increase traction close to the drivetrain. There is also a lot of nitty-gritty details to do with safety, efficiency and cooling that makes moving the battery away from the Semi itself not a viable option ATM.

Also keep in mind that at least for consumer cars, Tesla did try battery swap stations but nobody seemed to show any interest so they canned it.

This was a thing in the 1970s in parts of Germany. Busses used those battery trailers and swapped them out at charging stations. Diesel got too cheap and batteries where not good enough back then to make it worth the effort.

It's bizarre that they didn't rethink the seat placement in the cab. Every opinion I saw from truck drivers was that they didn't want a centered seat because it impairs visibility, but Tesla's going through with that? Why not put it on the left or right side like virtually every other car? I wonder if they did some studies to convince themselves that it's a good idea.

> I wonder if they did some studies to convince themselves that it's a good idea.

They explicitly call out the research they did in collaboration with the industry and the drivers that informed this decision. All in the presentation.

Anybody who has driven agricultural tractors knows a center position is better.

On farmland maybe, but US roads have lanes, and specific behaviors that will make this transition difficult for some. Thankfully I'm sure there's a new generation that will adapt better. The industry annual turn over rate is over 100% I believe.

Farmers do sometimes drive tractors on roads with lanes.

I've been fascinated watching the progress of Edison Motors (https://www.edisonmotors.ca/), who are making a diesel electric hybrid semi. Their logic and arguments make a lot of sense: the diesel engine gets to run at its peak RPM 100% of the time to charge the batteries - and it gets to take advantage of the perks of electric like regenerative breaking, quieter operation, and immediate torque. Not to mention the easier transition time when it comes to working within existing infrastructure. There's also all sorts of side benefits with what they're doing like the semi itself being a generator on wheels for heavy duty equipment. Lots of fascinating stuff coming from them, and they already have a driving prototype :)

The problem with hybrid vehicles in general is that they still rely on fossil fuels to some extent, so they're not a truly sustainable solution. And they have all the maintenance and reliability baggage of their fully ICE counterparts, something BEVs do away with.

The time and money being spent on developing hybrid technology could be better spent on making fully electric vehicles more affordable and accessible.

EDIT: and keep in mind, trucks are purchased with a long service life in mind (decades)

While not perfect, they do meet real world needs in ways that are better than current ICE options. A fully electric truck isn't really a viable option in the market they're building for right now (heavy duty working trucks for logging, etc). Additionally, getting a working hybrid option out into the wild still benefits the eventual transition to fully EV trucks as it allows for testing and validating of the electrical systems required to make a functional workhorse that can deliver value to its owners / operators as well as pushes forward the viability & benefit of electric vehicles in the eyes of the public / industry.

They are a transition, to be sure, but especially for consumers who do 80% short trip and need the ability to do long trip, it would seem to ideal choice for the 2020s.

Given the critical urgency of climate change, there is no situation where burning oil is the ideal choice.

BP, for one, has been making noises about their fuel being carbon neutral. It's not impossible, e.g. you sell fuel that makes so many tons of CO2, you bury in a hole so many tons of vegetable matter that captured that much CO2.

Doesn't fly in practice, perhaps because of public perception. Net-zero on synthetic fuels is more obvious, although not necessarily less expensive.

That's why transition is in the phrase.

Maybe in 5 years it will be different. Maybe the Chevy equinox will stay in stock and stay around 30k. But I'd bet a lot of money we will have gas stations for another 40 years no matter what we do, barring global collapse.

> The problem with hybrid vehicles in general is that they still rely on fossil fuels to some extent, so they're not a truly sustainable solution

Why not ?

https://www.epa.gov/greenvehicles/fast-facts-transportation-...

Trucks apparently are only about 25% of US emissions. Using hybrid would remove emissions from city centers (where your lungs are the recycling filters)

Freight is one of the only use case that requires long range, private cars are used something like 40km per day on average.

Having an electric drive train increases reliability, and having a diesel engine running at peak efficiency 24/7 will be much more reliable than a regular engine, especially given the abuse they take from trucks. (+ not as much stress on the batteries vs quick charge)

> The time and money being spent on developing hybrid technology could be better spent on making fully electric vehicles more affordable and accessible.

You still have to solve the resource issues, 1.4B vehicles on earth today, total lithium production per year is about 100 000 tonnes, even if we only needed 10kg per car that's going to take a bit of time, and that's not even talking about other uses

> Why not ?

As in, we run out of fossil fuels (or it gets prohibitively expensive) and your truck becomes brick because its internal battery and electric drivetrain is pretty much useless without the ICE part.

Solar, wind and thermal are practically unlimited for duration of the planet's existence.

I wonder what kind of chargers Pepsico has installed.

Sure, the ability to charge at 1MW+ is useful, but you won't want to use that often unless you want to spend all of your fuel cost savings replacing batteries. Just like a car, the battery will last a lot longer if you slow charge. But a slow charge on a 900kW battery is still a lot faster than you can get out of standard 240VAC. Pepsico probably wants to do 50-100kW overnight charging on their vehicles.

I wonder if it's something like 1MW of power split between 8 stalls. So if one truck comes in during the day time, it can get a top up at 1MW, but overnight when there are 8 trucks parked, they all charge at 125kW or less.

Charging a battery of this scale at 1MW likely won't have any more significant effect on life span than charging a Model whatever at a normal supercharger (which is negligible assuming thanks to battery temperature management and charge speed tapering).

If you always charge at a Supercharger you will have a reduced battery life.

Still no answer to the big question: how big and heavy is the battery? This thing has to have a huge battery. That cuts into the payload.

The fan community already has that figured out from the standard margin of EVs at "0" miles of range, combined with the percentage state of charge drop from going 500 miles, and the stated efficiency in the presentation (under 2 KWH/mile) and by Elon on twitter (1.7 KWH/mile). This makes it pretty certain that the battery is 1000 KWHr.

kWh

Which is double the size of the other trucks to which it was compared to in the article. So either they're making a massive tradeoff or have managed a physics breakthrough (battery weight or density).

Efficiency is key, as with cars

Typical "We can do 2x,10x better than others" Musk advertising

That's about 15,000 pounds of battery, correct?

No one knows exactly, but I'd guess closer to about 10000 pounds.

Semi's battery is not going to need as much armor.

Tesla Model Y battery: 81 KWh, 1,700 lbs => 21 lb/KWh. (Lithium-iron phosphate)

Tesla Model 3 battery: 82 KWh, 1,060 lbs => 13 lb/KWh. (Lithium-ion)

1000KWh in lithium-ion: 13,000 lbs.

Teslarati guesses around 15,000 lbs.[1]

A typical semi-truck without trailer weighs 15,000-25,000 pounds. So 15,000 pounds of battery is not hopeless. The Hummer EV weighs 9,000 pounds, after all.

Tesla can, and probably will, sell the "extended range" version for a premium price. There are many semi-trucks that never take long trips. This is mostly a battery cost problem.

[1] https://www.teslarati.com/how-much-tesla-semi-truck-battery-...

you can reverse engineer how big the battery is from stated range and efficiency. around 900kwh plus or minus 50

Some time ago I heard that the major obstacle to this truck is the weight. That is trucks in the US have a strict max tonnage and so the Tesla Semi can only be used to transport lightweight or not full cargo

Similar in the EU and Canada where the limit is 40 metric tons (88,000 lbs) vs the US where it is 80,000 lbs.

Disappointing there is not a word about safety to the outside world. I get they have to market to drivers, calling it a "beast," but large trucks are a great danger in terms of losing tires and, especially in cities, collisions. I see it is built with a scoop design so "things" (people, bikes, etc) hopefully don't get pulled under it so easily, but hopefully it also has advanced collision and awareness features suited to such a large vehicle.

It is unfortunately typical of Musk to market to the aggressive impulse, rather than harmonious, even if his ultimate goals are beneficial.

One of the biggest dangers of trucks is losing their brakes and going out of control. As the article points out, regenerative braking vastly reduces that problem. So that’s a safety win!

Why don’t we just put overhead electric lines on highways for semis to pull electricity from, rather than lose weight on big batteries?

While it is theoretically possible to use overhead electric lines to power vehicles on highways, there are a few challenges that would need to be overcome in order to make this a practical solution. For one, the infrastructure required to support such a system would be quite extensive, and would likely be expensive to install and maintain. Furthermore, the system would only be able to provide power to vehicles while they are on the highway, so they would still need some form of on-board power source for use when they are off the road. Overall, while this idea has some potential, there are currently more practical and cost-effective solutions for reducing the weight of vehicles, such as the use of lightweight materials and more efficient batteries.

Feels like the above comment was written using ChatGPT :)

Lightweight materials doesn't change physics.

In railroading we have long ago learned that long terms power delivery via overhead lines is by far the cheapest and best solution.

I agree - trains make more sense than running overhead wires over highways.

And what if we put steel tracks and replace the rubber wheels with steel ones, for the lesser resistance and thus better traction? We could even then connect multiple semis together and only have a few provide power, that would be even more efficient!

Snark aside, we have those, it's called railways. Trucks should mostly be used for "last mile" style transportation, hence the upfront infrastructure costs might not be worth it

It might be less expensive / faster to deploy overhead lines on existing highways rather than building new rail.

See elektrobooms idea for charging tunnels every 20 KM or so that can make urban vehicles not need a massive battery

A truck that can drive on roads and also hop on rails sounds much more complex than an EV truck that can charge from an electric line above highway when available.

Whole truck doesn't have to hop on rails. This thing exists: https://en.wikipedia.org/wiki/Roadrailer

That's admittedly cool.

However, you then need to have multiple tractor units with drivers in multiple cities + also the train engine, all coordinating for the handovers.

I'm definitely not saying EV trucks with batteries + trolleys are better, but there is some simplicity that's not replicated with the rail solutions.

And I have read sofixa's (somewhat mocking) comment as if they are saying that rail is simpler.

Sorry, can't find anything in English on the quick. But loading whole trucks on trains definitely exist, is logistically not very complex and a major component of North / South (and vice versa) freight transit through the alps.

Here's a picture :

https://www.eurotransport.de/artikel/kombinierter-verkehr-ro...

That's smart, but why isn't it done everywhere?

It's only really sensible for long distance transportation.

I'm making assumptions here. But lets' say loading the trucks and the wait until the train leaves takes 90 minutes. Unloading them at destination takes another 30 minutes (again, I don't know the real figures, but this feels somewhat realistic).

So, there will be a certain not insignificant amount of time being lost before the train moves. Time, which is valuable in the transportation business.

Another reason may be that it's really geared for specific topological environments.

There are few routes through the alps and they universally require tunnels (there are mountain pass routes, but those are closed in winter and partially inaccessible for trucks). Such a situation invites such logistical solutions.

Add to this that Switzerland is one of the main transit countries through the alps and that - with very few exceptions - trucks are not permitted to drive at night (after 10pm) or on Sundays. So, loading your truck onto a train and blitz from Freiburg (Germany) straight to Genoa by train may make a lot of sense.

The situation is not quite as ideal as described here. While it's definitely a goal to move transit onto train infrastructure, especially on the Italian side, still leaves much to be desired.

If it's for last-mile transportation, you probably don't want 500 mile range; at that point you're just paying more to reduce the payload by filling it with batteries. These are presumably targeting long-range.

Germany has a small-scale test of overhead lines for trucks: https://www.nytimes.com/2021/08/03/business/electric-trucks-...

> Germany has a small-scale test of overhead lines for trucks

And the accompanying ridicule for recreating a railroad, just with less efficiency because tires consume a good chunk of the truck's energy in deformation and because you need way more drivers than if you would just put the damn bunch of trucks onto flatbed rail carriages.

The entire project was nothing more than a smokescreen by the automotive industry.

"But the onus would be on the German government to build the overhead cables, which cost an estimated 2.5 million euros per kilometer, or about $5 million per mile."

“Numerous studies have come to the conclusion that overhead cable trucks, despite the high infrastructure costs, are the most cost-effective option,” the ministry said.

Tom Scott made a video about this, it's well worth the five minute watch: https://www.youtube.com/watch?v=_3P_S7pL7Yg

I agree, trains are more efficient for moving cargo.

Actually, Tesla talked a bit about how a Semi convoy can even beat trains in cost.

Tesla talks about a lot of stuff and is great at cherry picking examples.

No obstacles to do the math yourself!

Yes, and also no incentive, since as a general rule rail is cheaper than trucks (and boats are cheaper than rail).

I have no way to determine if their cherry picked example is something that's relevant or some weird edge case that makes their graphs look good.

But we will see over time, based on adoption.

And regardless of what they do, I hope they don't lobby against mass transit (of goods and people). But they probably will, making things worse for all us.

C'est la vie, I guess.

> I hope they don't lobby against mass transit (of goods and people). But they probably will, making things worse for all us.

People in general are not fond of the experience of mass transit. They only do it because it's either (or all):

- cheap (no parking fees, car purchase/maintenance)

- less hassle (no need to find parking in busy areas and no need to have all attention to driving)

If personal transport can be made cheaper, less hassle and hands-free (all of which is possible by autonomous vehicles), then people will vote with their pennies and no lobbying will be required by Tesla.

How will this be worse?

Companies have departments and groups meant to fight, shape, and push legislation in their favor, as well as shape public opinion, even against the public's own interest.

Do you know how carmakers literally paid to have tram lined ripped out of cities, in order to discourage public transport usage and get people to buy their cars?

This is not some weird thing and it's a literal, documented, conspiracy with actual indictments.

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

> If personal transport can be made cheaper, less hassle and hands-free (all of which is possible by autonomous vehicles), then people will vote with their pennies and no lobbying will be required by Tesla.

Secondly, physics and economics: you can never make personal transport as cheap as public transport, because personal transport will always carry fewer passengers per ride.

And personal transport, if you look at it, is going in the wrong direction for efficiency.

https://www.autoweek.com/car-life/classic-cars/a1860276/car-...

A 4.4m long Chevrolet Sonic 2017 weighs 1.3 tons, almost as much as the 5m long Chevrolet Chevelle 1967.

The most lightweight Tesla Model 3 weighs 1.8 tons, that's 50% more, on top of the existing ICE weight gain over the last 50+ years.

Adding more safety and comfort features will only make things <<heavier>>.

And people really want to ride in nice, clean, custom cars, plus they really don't want to ride with other people most of the time.

I guess we'll just keep burning the planet down, pumping out particulate matter, destroying wildlife, etc.

Oh, and fully autonomous driving could be 50 years into the future, banking on it is a very risky move. We're still at the basic research phase, our tech isn't good enough.

¯\_(ツ)_/¯

> you can never make personal transport as cheap as public transport, because personal transport will always carry fewer passengers per ride.

Oh man, another "never" statement. An autonomous car that doesn't have a driver, doesn't need to take breaks to eat or sleep, that doesn't need to be parked, can in fact become cheaper than public transport. I invite you to do some research on that.

> And people really want to ride in nice, clean, custom cars, plus they really don't want to ride with other people most of the time.

That's exactly my point, and why in presence of a price competitive option such as a self driving taxi, no one would want to catch a bus or a train with many people and dirty seats.

> autonomous driving could be 50 years into the future

Not if you're in the know. It's far closer, but I'm not arguing for that. Wait and see.

How does an autonomous car not need to be parked? The vision that seems more realistic to me is that lucrative areas would have traffic jams full of empty self-driving taxis circling around waiting for fares.

  How does an autonomous car not need to be parked?

It doesn't need to be parked because it basically can't park itself.

https://www.youtube.com/watch?v=nsb2XBAIWyA

> How does an autonomous car not need to be parked?

As in, they don't need to be parked after every ride in the current personal transport model.

> The vision that seems more realistic to me is that lucrative areas would have traffic jams full of empty self-driving taxis circling around waiting for fares.

They will be run as part of a coordinated fleet of robo taxis. Why would they be circling around waiting for customers in an area where there is no demand? Answer is they won't. They'll be directed to where people want rides much like Uber, but considerably more efficient, because they don't have constraints of a human driver they have to accommodate.

Say there's an area that will have heavy demand soon but doesn't yet, maybe someplace with a lot of bars on a Friday night. Where do all those cars come from? There has to be a holding garage sort of nearby, but land nearby is expensive, and if there's competition for the faster service, you get a race-to-the-bottom.

> Oh man, another "never" statement. An autonomous car that doesn't have a driver, doesn't need to take breaks to eat or sleep, that doesn't need to be parked, can in fact become cheaper than public transport. I invite you to do some research on that.

I love how you sideskipped my phsyics argument like it was nothing.

You're still moving 1.5 tons (and constantly increasing!) of metal everywhere, to move 1 human being weighing 80kg.

The physics just don't work.

> That's exactly my point, and why in presence of a price competitive option such as a self driving taxi, no one would want to catch a bus or a train with many people and dirty seats.

Price competitive how? It's impossible. The only way this works is by offloading a ton of externalities on everyone else, including people without cars. Roads, parking lots, highways, high capacity bridges, tunnels, charging infrastructure/gas stations, exhaust pollution, tire wear pollution, road degradation pollution, noise pollution, light pollution, manufacturing and construction pollution, ...

By what logic can't you clean a bus for 100 people but you can thoroughly clean a taxi for 1, AND make the taxi cost competitive?

> I love how you sideskipped my phsyics argument like it was nothing.

My bad.

> You're still moving 1.5 tons (and constantly increasing!) of metal everywhere, to move 1 human being weighing 80kg.

True.

> Price competitive how? It's impossible.

OK, you can transport at least a factor of 10x more people with the existing number of cars on the road. People who drive to work and back use their cars for a fraction of time in a day. If those cars were out moving people around the clock, they would add a huge capacity to the taxi fleet.

Now, who pays for the car? The owner. They have paid for it whether it's sitting in the car park or driving people around. The only difference is running cost of the car: electricity and mostly tire wear. That becomes the cost of transporting people with margin added for Tesla and the car owner. That's it!

Now I just realized that perhaps in your hometown, cost of public transport may be so dirt cheap. But where I live, it cost $40 AUD a week to travel 2 stations back and fourth once a day. The same trips over the same period with my car costs me $2.40 AUD in electricity and about $2.80 AUD in tyre wear (assuming $2000 AUD per 25000km which is way too aggressive). That's a total of $5.20 vs $40. Even if you double the cost to account for margin, it's still 4 times less expensive than public transport.

EDIT: even if it's exactly the same as public transport, it'll still be the preferably mode of transport, because it's door to door and you don't have to deal with anyone. Imagine someone who's happy to share with one other person, now you have bonker economics.

> since as a general rule rail is cheaper than trucks

That's not the way to approach claims that a new tech/product is disruptive. General rules may not (and usually do not) apply.

I'm waiting for research that's not from their marketing materials.

I won't do it for obvious reasons (life), but my money's on them lying (through any of the ways listed here, I don't really care which one they use: https://en.wikipedia.org/wiki/Lie#Types_and_associated_terms).

Debunking Tesla's cherry-picked argument will take more time than it took Tesla's marketing team to cherry-pick their truck and train data to make the math work in their favour.

https://en.wikipedia.org/wiki/Brandolini%27s_law

Electric trucks don't pay for road maintenance, trains do. It's not a fair comparison.

Uhm, trains.

Yes, this is something that should have been done decades ago.

That and shifting more to electric rail as well.

But sadly the world doesn't really think logically and systematically about these issues.

Along with not building car depend cities and suburbia that makes delivering things way more costly and worse for the environment.

All of these things are wrapped up in many layers of cultural and historic issues.

Overhead wires are way more expensive than you’d think. They’re also delicate — a tree branch knocks out a line of Boston’s light rail system every couple months.

Electrification of existing train lines is a huge initiative, and one with major benefits, but it’s still only slowly gaining traction. In fact, one of Boston’s train lines that is not electrified is opting for battery powered trains to avoid the costs of overhead wires!

I don't understand why you don't have companies like Tesla building fuel cell (hydrogen) solutions for commercial use cases. Here is an article that explains it - https://www.bmw.com/en/innovation/how-hydrogen-fuel-cell-car... If Tesla can build a super charger network, they can build a hydrogen refilling network.

Here’s a video explaining the pitfalls with hydrogen as a fuel https://youtu.be/f7MzFfuNOtY

From Elon's previous comments, Hydrogen is fundamentally not scalable and uneconomical.

Setting aside Musk’s current plunge in credibility, his argument regarding hydrogen is predicated upon electrolysis. A vast network of transparent tubes and sunlight reflectors feeding genetically engineered microorganisms that release hydrogen would undermine that assumption, but transportation, storage, and energy density would indeed remain a problem.

A charging network just requires some wire and electronics, everything is well understood and easy to do for experts in that field.

Hydrogen on the other hand, is difficult to keep contained and cold enough to remain in liquid form. Even NASA who has worked with it for 50 years with massive budget and engineering expertise constantly has problems with it. For instance one of BMWs hydrogen cars you are not allowed to park in a garage! As hydrogen is expected to leak from it. And current hydrogen gas stations still suffer from this challenge.

So hydrogen fuel cells do offer the advantage of better energy density for more range, but you have the dual difficulties of 1. how do I make the hydrogen efficiently AND cleanly and then 2. the infrastructure build out is actually as hard ad EV doubter think wires are.

> Hydrogen on the other hand, is difficult to keep contained and cold enough to remain in liquid form... For instance one of BMWs hydrogen cars you are not allowed to park in a garage!

That particular BMW effectively one-of-a-kind in that it is 1) internal combustion rather than fuel cell, and 2) stores its fuel as a liquid rather than a compressed gas. The liquid storage is a consequence of being ICE because it is required to get even remotely reasonable range out of something as inefficient as ICE combustion of hydrogen.

fuel cell vehicles like the Mirai still have a number of problems, but they have actually done a very good job with fuel storage and range using compressed hydrogen and don't have any of the limitations of the BMW such as leaking hydrogen or losing massive internal volume to the fuel storage.

> I don't understand why you don't have companies like Tesla building fuel cell (hydrogen) solutions for commercial use cases.

Oh god, can that damn hydrogen hype please die out. Hydrogen has an incredible amount of conversion loss both at generation and in recombination (each step only has about 40-60% efficiency rate) compared to way over 90% for lithium based batteries. Additionally, it's fiendishly difficult to safely transport it because it's highly explosive, and it will slowly leak out wherever it can.

The only place where hydrogen really has its uses is in synth-fuel applications (e.g. for airplanes) to shift these away from fossil-origin fuels, in processes where the heat of a burning flame is required and as raw material in chemical processes. Hydrogen generation capacity is incredibly scarce, we should reserve it for those processes that literally cannot use any alternative.

The reason why hydrogen hype will not die, and in fact will massively increase going forward, because those claims are simply wrong. It is nearly as efficiency as the lithium-based solution, while being drastically less dependent on rare resources.

People are starting to realize that most of the criticisms are literally just made-up bullshit from EV fanatics and were never true. The most notable example is the safety issue. Hydrogen is straight-up safer to deal with than lithium in a car. EV fans were just lying about this subject the entire time, and in fact you want a hydrogen car because it is outright a safer type of car.

I wonder if anyone considered trucking a swappable battery as well. That's not for semi, of course, but road trains do exist?

Nit: It seems backwards to measure efficiency in terms of kwh/mile. One would expect a higher efficiency value to be better.

The best metric I can think of is expected lifetime miles driven over total cost of ownership.

But I don't get why you think it's backwards! You want kwh/mile to be low. Basically, you want to use as little amount of energy stored in your battery to drive a mile. That's what that is.

I agree. The equivalent of mpg is miles per kWh. Using the reciprocal is confusing.

How about MPJ (Meters per Joule) since it is close to MPG - though the numbers would be tiny. Maybe km per GJ is more appropriate.

The distance-per-energy is numerically backwards compared to the energy-per-distance. This is important when e.g. wondering which car to replace, calculating a fleet average consumption and so on and so forth.

For example: "Which saves more gasoline, going from 10 to 20 mpg, or going from 33 to 50 mpg"?

There is good reason that the rest of the world uses the opposite system, and the reason isn't just that no one else uses gallons. It's that it's an objectively worse way of communicating fuel consumption in every single way that matters.

>> "Which saves more gasoline, going from 10 to 20 mpg, or going from 33 to 50 mpg"

I'm surprised such questions come up often. Usually a current vehicle gets x MPG and a new one might get y. We use l/100 km in Canada. I think the primary advantage is it relatively easy to compute number of liters needed on a longer trip. It isn't super important however - better just to fill up!

PS I was mostly kidding about using MPJ. The numbers would be tiny and hard to make sense of.

99% of the world uses L/100km to measure efficiency, so they're used to it.

While the article has a US focus, the most common measurement globally is L/100km. The sensible efficiency measurement is consumption per distance, like L/100km. The electric switchover is This is an opportunity for places like the US to switch its measurement of choice too. I don't expect it to become a metric one, but at least one that works numerically (e.g. for fleet averages).

I don't care that it's more common the other way, it doesn't make any sense in a sentence. When we talk about efficiency verbally, we say that higher or more efficiency is better: but the numerical value is better if it is lower. Miles per kWh seems like the appropriate unit to me.

> The sensible efficiency measurement is consumption per distance

No, that’s a sensible inefficiency measure. A sensible measure of X increases with X.

I think it makes more sense to give figures for consumption rather than distance/energy, as we're usually more interested in how much fuel it will cost us to travel a certain distance. Bit like going into a sweet shop and having all the sweet's price as "100 per dollar" or "35 per dollar" etc..

It also helps to compare costs, as 20mpg is far less efficient to 30pmg than 30mpg is to 40mpg, but the numbers suggest the difference is similar.

They probably do not measure "efficiency" but "power usage" or "cost". Those go up as you would expect. I also have not really heard a l/100km value called an "efficiency" metric in german. It's lmost always fuel usage.

The Link is dead? What happened?

Just use trains for bulk like a normal country.

It's all about economics. Whatever's cheaper wins. To that end, Tesla has mentioned in their presentations that a Tesla Semi convoy of only 3 trucks can be cheaper than a train for the same load.

The problem is that highways are literally not cheaper but massively subsidized and heavy trucks are destroying that public infrastructure because they are so heavy.

So just saying its all economics when the underlying rules of the game are totally unequal is not that useful.

Sorry I should have been more specific. I'm referring to economics from the viewpoint of the business operating a fleet. Given almost all business do not hold back from whatever incentive and subsidies are available to them, it's fairly reasonable boundary to confine the comparison to.

Its a reasonable boundary but not reasonable when discussing the politics and what as a society should be done.

> To that end, Tesla has mentioned in their presentations that a Tesla Semi convoy of only 3 trucks can be cheaper than a train for the same load.

I can carry it on my magic carpet for even cheaper, trust me I am rich so I wouldn't lie... this time.

Trains have replaceable engines, power-lines - no battery. Trucks will not be cheaper option no matter what mumbo jumbo techno jargon you throw at it. And decades of technological optimizations. And they dont need lithium to be made and maintained.

> I can carry it on my magic carpet

Oh I see. You're high out of your mind. Explains your other responses.

Oh I see you make effort to read someone's history of posts, explains a lot about you

The US has way more freight rail than other countries: https://www.nationmaster.com/nmx/ranking/volume-of-rail-frei...

Other countries need to use trains for bulk like the US does.

Rail is just a lot more complex to plan for. You still need trucks for last-mile transport, so for the typical short haul it sometimes makes more sense to just drive to the destination directly. Then transporting by train just takes longer which isn't always acceptable. Sometimes you just need things there today/tomorrow.

So basically, if I'm right, you need a nuclear power plant for 500 to 1000 of these trucks if we assume one charge per day. That looks awfull to me, no?

Watch the delivery event. They address that.

TL;DR they sell solar+battery backed Megachargers to go with the trucks.

You can't drop a comment like this without sketching out your math and base assumptions.

I would guess you're assuming a 1GW nuclear reactor, and each truck ~1MWh of battery. Wouldn't that be ~1000 trucks charged per hour, not per day? Are you off by 24x?

Why do we have traffic jams at busy hours?

So basically, the battery-powertrain efficiency is about 3-4 times better than with diesel. You need 3-4 nuclear plants worth of energy now.

But lets see about your math as well.

One reactor (of which there may be several in a plant) is lets say 1GW.

The semi will charge at say 1MW. So 1000 of them can charge at the same time. It takes about 1 hour to charge the battery so we are already at 24000 charges per day. Each of them can cover ~800km so about 19 million fleet km or 480 trips around the earth or about 25 round trips to the moon.

I assume you don't mind pulling three times the nuclear powerplant's worth of dinosaur juice from earth?

Reply on news.ycombinator.com