108 comments

[ 3.2 ms ] story [ 183 ms ] thread
FOMO is a powerful incentive.
What exactly do you miss out on, going forward its only likely these trucks will be cheaper and more in abundance. If anything its better to wait.
Experience. It's the same reason I'll try out a new language on a throwaway prototype: real experience helps you think more clearly about the future.
PR, or lack thereof, when these ship.
FedEx could be delivering my packages by horse as far as I'm concerned so long as they arrive on time
Except your competition may get a lock on the supply contract because they were early-adopters. Or they figure out how Tesla is doing it, and risk competing with Tesla, outcompeting you.

Companies that don't innovate better be ready to be disrupted.

It's becoming clearer and clearer that the main things preventing total EV takeover are lack of good products on the market and production capacity. I'm excited to hopefully watch both of these problems erode in the next 5 years.

Really really ready to stop breathing fossil fuels all day every day in the city.

And sound!
artificial sounds (because they are too quiet)
From my understanding, the requirement for artificial sounds when reversing already exists for big trucks and haulers, regardless of the type of engine inside, for the safety of pedestrians. It's the quietness of small electric vehicles, like cars and SUVs, that prompted the requirement in consumer vehicles.
as long as they are not powered by coal power plants... (but I guess that those are typically in poor neighborhoods)
It's still a decent improvement if they are - one big coal plant running at relatively constant speed is much more efficient than an equivalent number of individual petrol engines in cars afaik.
More energy efficient, yes. But resulting more CO2 and SOx and NOx emissions even with scrubbers.
Ah I wasn't aware of that. Emissions are definitely the primary concern, at least for me.
Coal makes up only a third of American electrical generation, and it’s dropping rapidly.
only a third? wow, I thought that it was less then 25%
Only 30% of USA power comes from coal, and it’s much lower on the west coast than elsewhere.
The main thing preventing total flying car takeover is the lack of good flying cars on the market and production capacity.
Well, physics, safety, and energy efficiency play a huge role there as well.
*specific energy of batteries and FAA regs
Uh, no. This is one of those really specious arguments that sounds clever until you think about it.

The main thing preventing total flying car takeover is that there is no technology that would make the product viable-- as has been pointed out, it would be EXTREMELY loud, create immense downdrafts, and create massive safety risks far beyond those of terrestrial cars.

Similarly, the main thing preventing robotic takeover of the cleaning industry is that there are not yet any robots capable of generically handling all cleaning tasks like sweeping, dusting furniture, washing windows, organizing a cupboard, mowing the lawn and so on. It's not just that there is no production line-- there isn't even one such robot nor is anyone capable of making one.

Etc.

I dislike this kind of empty reasoning. It derails conversations with baseless non-logic.

"You're derailing the conversation!" said the man who had just finished enumerating the practical issues of flying cars and cleaning robots.
I totally agree with you except on one point:

> there is no technology that would make the product viable

Unless it's only a matter of R&D before we have vehicles that can fly without blowing air around. It's plausible that a technology may exist in the future which makes this viable.

Or teleporters. Might as well mention those if you're speculating about unknown future breakthroughs.
Or commercially-viable and massively-produced electric cars, as long as we're talking about unknown future breakthroughs.
I think you wrote this in jest but if there were really good flying cars on or coming onto the market it wouldn't surprise me if it could make a dent or maybe even a total takeover of certain types of industries, depending on what you're targeting. Cars in general? Probably not plus they'd have to fly themselves or people would be dying far more frequently than regular car crashes.
Fedex has 14,500 trucks [0]. That will put tesla's at .13% of the fleet which sounds just like they are buying some to test the economic viability of them. I would hardly use this as a sign "that the main things preventing total EV takeover are lack of good products on the market and production capacity".

[0]: https://expandedramblings.com/index.php/fedex-statistics/

Fair point, but you have to take this development in context. EVs technologically outpace ICE vehicles on every axis other than refuel time, upfront cost and range. EVs can improve on all of these metrics, whereas ICE vehicles cannot. Therefore a complete switch is only a matter of time—it remains to be seen how quickly it will happen.
ICE still have a decent amount of efficiency headroom, so range and operational cost will continue to improve if auto and truck makers choose to spend the 80% investment to squeeze the last 20% out. Automakers seem pretty split - some are building initiatives to go full electric while others are focused on the ICE endgame. We'll see what large equipment and truck manufacturers do but I think the time horizon is quite long.
Indeed. For plain-jane passenger vehicles, Toyota is rolling out new engines with a different stroke-to-bore ratio, and a wider angle between the intake and exhaust valves to speed up combustion. With some changes to ancillaries they've increased thermal efficiency from 35% to 40% (41% in hybrid vehicles).

Mazda (of which Toyota has a fair stake) is rolling out their second generation SkyActiv which uses spark controlled compression ignition (SPCCI), in conjunction with a lean supercharger, to switch between spark ignition and compression ignition based on engine load. They've not thrown out a thermal efficiency number, but it's likely in the ballpark of 43-45%. They're targeting 56% thermal efficiency for the third generation of the tech, which is already well into its development, and we'll see sometime after 2021.

The current benchmark is Mercedes' Formula 1 engines, which are 50% thermally efficient, but understandably cost an arm and a leg.

This is excluding other things like Koenigsegg's FreeValve which uses pneumatic-hydraulic-electronic actuators (PHEA) to replace the camshaft. The only licensee so far is the Chinese manufacturer Qoros, which is a shame as Koenigsegg got pretty significant adoption of their FlexFuel tech by GM, Ford and Chrysler in the US (> 17 million vehicles sold).

A concern of EVs is also the weight of the fuel, which is a significant fraction of the overall vehicle weight.
The weight is substantial. The Lithium-ion packs in the Model S, for example, weighs over 500KG. One of the largest engines in a current production car, the Dodge 8.4L V10, is 42% lighter. Smaller turbo petrols weigh just a fraction of that.

(I am comparing just the weights of the power plants, not the power plant + drivetrain which would include gearboxes, differentials and electric motors).

Sure, but you wouldn't expect FedEx to replace their entire fleet this year. A somewhat more meaningful number is how many new FedEx freight trucks will be built by Tesla over the next year.

And UPS has placed an order for 125 of them (https://electrek.co/2018/03/26/tesla-semi-electric-truck-ord...) as well.

You're right that this isn't going to radically transform the country's freight fleet overnight, but I think it's a signal that these companies are willing to switch to EV technology if it works. The next hurdle now seems to be Tesla's ability to execute.

Few things transform overnight, most of them take time.
I am not saying that they are buying them for just that, but the PR value of this operation is probably worth a big percentage of the cost of the trucks.
Probably more a proof of concept than PR function. Stress testing these trucks will lend great insight into their practicality.
Exactly. Like How many mechanical and electrical problems will develop a year from now ? Two years ? It’s very different to carry 80,000 lbs for a few test rides than everyday for years.
> Fedex has 14,500 trucks [0]. That will put tesla's at .13% of the fleet

Do we know what fraction of FedEx’s annual truck purchasing this order represents? Fraction of flow is a leading indicator fraction of stock (which you helpfully provided) [1].

[1] https://en.m.wikipedia.org/wiki/Stock_and_flow

I wouldn't get too carried away with this announcement. For FedEx, this is a minuscule pilot project and (more importantly) a cheap way to generate positive publicity.
Great products, but the only problem is (almost) nobody wants them.

Hence we need the State to steal via taxes and spend on whatever the intellectual elite determines is in our best interest.

How is Tesla getting all these companies to put out press releases announcing that they are buying Tesla's products? Brilliant marketing. Maybe they offer discounts?
They clearly get as much value out of it as Tesla. Look investors, we're an innovative company too! When was the last time you saw a positive press cycle about FedEx on their own?
Tesla is the cool thing right now. It's the same thing as saying they use AI for something. Good PR.
Judging by past Tesla products, no discounts for pre-ordering.
the chance to put out one of these press releases definitely factors into the purchasing decision for these companies. They're buying good PR almost as much as they're buying trucks.
And no doubt they have every escape clause in history in their contract. Great marketing for both companies, no risk for FedEx, and Tesla probably values the PR as much as the money at this point. They're not getting paid unless and until they can actually deliver the trucks, so why not get good press in the meantime?
Twenty trucks is really nothing for a company like FedEx. As the article noted, this is a very small pilot project for them.
UPS preordered 120. It’s not the amount that is important, it’s how many conservative logistics companies are interested that matters.
Less conservative than you might think: https://sustainability.ups.com/committed-to-more/fuels-and-f...

At 3 billion miles driven per year, there's a lot of room for experimenting with a few hundred/thousand new vehicles.

Anyway, electric freight works well for this business assuming the range is decent. There are lots of lower range freight routes with well forecasted volume, and controlled environments at both ends allows for installing charging infrastructure.

I agree entirely. I believe Tesla is quoting a two year payback period versus traditional diesel class 8 semis, which means they will be sold out indefinitely until production ramps.

It’s a smart move. Between passenger vehicles, stationary storage, and their semis, the Gigafactory should be fully utilized.

Tesla's usual deposit scheme has an escape clause for everyone, you aren't committed until you actually configure the vehicle. Until then, you can get a deposit refund at any time.
FedEx had over 70,000 trucks 15 years ago. Obviously most of them aren't long-haul, but you'd want to know what percentage of the FedEx fleet this represented to judge how much the impact of this is about PR versus operations.
Local delivery of palletized freight still requires a semi truck in most cases. That's where I would see these being used.
Total isn't a good number to compare against. The number of trucks they buy _per year_ would be a better number.

Even if the Tesla semi is the end all, best semi in the world by orders of magnitude, they still wouldn't be replacing an entire fleet in a year.

I would love to see Tesla demonstrate that their Superchargers are anywhere near super enough. A couple days ago, I supercharged at the “120 kW” Tesla factory in Fremont. For quite a few minutes, I got 23kW. After a while, it went up to 50kW or so. I checked the other car sharing my charger, and the total was well under 120kW. I suspect the station has an overall power limit. Also, there was a line.

For trucks, you need something like a megawatt to recharge on a lunch break, and a line blows the whole thing up.

Solar. Freakin. Roadways.
Have you seen the roads through most of the country? We can't keep tar and stone in decent condition, no way we'd keep those solar panels running.
Of course, it was a reference to a viral video a few years back with the same title. A joke, if you will.
These are like two different problems, though probably compounding. Allowing vehicle charging while moving is an interesting problem, regardless of where the power comes from (well, it’s way easier if it’s attsched to the truck).
I don't think the solar road is meant to charge the cars while driving, it's just a place to install solar panels.
Same thing is it not? Wireless charging would be nice, or the trollybus tech.
Not necessarily. Driving over a solar panel doesn't charge your electric car.
Electric semis for true long haul will never be viable unless there is a major breakthrough in battery tech. 300 gallons of diesel fuel has so much more energy than even the best batteries could pack into a reasonable size and weight.
Major breakthroughs in autonomous driving may change that. If drivers are eliminated, the costs of waiting 30-60 minutes while charging becomes almost 0.
Especially if the truck can hand off its payload to a fully charged & waiting replacement cab.

If you have access to a fleet or don't need the same truck back, it becomes more of a baton relay than a marathon.

With the kind of infrastructure required for that, I can't help but think it would just be a lousy version of an electric railway. At least for long hauls with high throughput.
How would it be lousy? The cars can now go anywhere and dynamically route instead of being stuck on rails and the infrastructure costs are orders of magnitude less.
That's like saying that circuit switched networks are better than packet switched.
They would be if packets had enormous per-packet overhead.
At some point you start wondering about improving intermodal transportation. Using rail for longer distances with trucks handling the last 50-100 miles.
Isn't it the other way around? With drivers, you can fit the charging inside the already mandated breaks, while without driver, you have to introduce longer breaks to get enough charge.

Idle time of EV-rig per day (30min charge every 4 hours): ~11% Idle time of Non-EV-rig per day (5min refueling every 4 hours): ~2%

Sure, for short distance trips where loading/unloading takes as long as charging the EV-rig, it doesn't make a difference, but for longer hauls, the charging pauses of EV-rigs will severely cut into utilization of the fleet.

Would a hybrid system using an electric generator to recharge the batteries be more efficient given the advantages of electric vehicles, such as regenerative braking?
I would think if a series hybrid were more efficient than a conventional mechanical drive and transmission, it would already have been adopted. Trucking companies will spend money even for small gains in fuel efficiency because they drive so many miles.

I also think that regen braking in semis would be trickier because a lot of the braking happens on the trailer wheels so you would need to fit all the trailers with regenerative braking equipment.

Long haul trucks fairly hit the breaks as that ends up costing a lot of money. However, a truck specifically for mountainous areas may benifit from a hybrid transmission.
Semi's actually get ~6.5 MPG so they only need around 5x the batteries to get the same distance as normal electric cars. The huge tanks are over 24 hours driving time and are most there so you can pick the states to refuel in.

Battery volume and weight density is not that bad when you consider the weight savings without an IC engine, transmission etc. Gasoline gallon equivalent tables ignore the vastly lower efficiency of IC engines. Costs are mostly non issue as trucks would charge daily making them break even before cars.

The actual problem is manufacturing capacity which has been ramping up but has a long way to go.

Yes but long haul trucking involves distances much farther then the range of any normal electric car.

Good point about substituting the weight of the engine for batteries. Still I think you're going to be way short in total energy density.

And it's not like you can just pack the trailer with batteries either, you still need to allow for the weight of the cargo and there is a gross weight limit on 18-wheelers.

they are already driving them, they seem to have about the range. you can read about this in tesla newsgroup sites, like tmc.
They don't exceed highway speed so swapping batteries every X miles will be worth it if they reduce total costs by doing so.

As to weight the engine weighs ~3000 lb and 300 gallons of gas weigh ~1,900 gallons. Transmission is another 700lb, include tank weight and verious odds and ends and you can get ~5 hours at highway speed. Assuming a battery swap takes 10 minutes. That's easily worth saving 50+k per year in operating expenses.

2200lb for gas, 3000lb for the engine, 700lb for the transmission. Say 7k lb total. Compare this now to the electric one. The battery alone would weight 33k pounds. Engine and transmission would add another 2k. 5 times more than ICE. The battery would also cost $200k.
Most trucks are driven by a single driver, rather than a team of drivers - this means that they legally must have 30-60 minute breaks every few hours, and can only drive a certain number of hours.

So, lets throw some numbers around:

- Range: 500 miles - Time to recharge 80% (+400 miles) capacity - 30 minutes - Speed Limit (Interstates): 50-75MPH - Maximum Driving Times (USDOT): - 8 hours (followed by a 30min break), - 14 hours (followed by a 10 hour break) - 60/70 hours (in any 7/8 days)

So lets assume 60MPH moving average speed (very high) - gives us over 8.3 hours of driving time.

You add in a 30min break at a megacharger and you've now got yourself another 400 miles of range, which would let you drive for 6.6 hours - at which point you'll be out of hours again.

So you park up and let the truck charge overnight.

I don't see how this is so difficult.

Where are you getting 400 mile range on a semi towing a load coming from 30 minutes of charging?
Tesla was mentioning 400 miles. I _could have sworn_ they specified those numbers were all with an average load but it seems official numbers are a bit annoying to source. So now I'm curious if that's accurate or not.
That is the number Tesla promised for their megachargars at the presentation of the semi. So they seem to have clear plans to deploy megawatt chargers at their charging stations.
For the range - you can order either 300 or 500 mile range options.

The range options were said by Tesla to be the range of the truck plus trailer and maximum load, at highway speeds.

The 400 mile mention is the 80% capacity of a 500 mile range pack you'd get by a 30 minute charge. Also information given during the announcement.

[1] https://www.tesla.com/semi

That is for their “megawatt” chargers not the current infrastructure.

And that is a problem just count the average number of semis in a truck stop they’ll all have to share power.

Also many of these truck stops are in the middle of nowhere and while they have electricity they aren’t connected to anything that can provide that much power.

Say you get 100 500kWh semis in a stop that need to be charged 80% that’s 4mW if we’re taking Tesla’s 30min turn around that’s 8mW which is not an unnoticeable portion of what a power plant produces and hour.

> That is for their “megawatt” chargers not the current infrastructure.

I never mentioned 'current' infrastructure, whatever that means. We're clearly talking about Tesla Semi here, that's pretty clearly not going to fit in your ordinary Tesla supercharger stall.

As for the other 'problems' - they are all solvable. I'm sure Fleet operators and major truck stops will all be jointly cooperating to add MW charging infrastructure along the trial routes. It benefits the fleet operator, it benefits the truck stop operator.

Electrical grid operators won't have an issue with it - if you look back at previous postings on HN you'll find several examples where grid operators are struggling to find demand for the grid, given the rise in household efficiency and offshoring of major production infrastructure.

Yes it'll require some work to ensure there's not huge spikes in demand, but that's solvable too. Tesla (and others) are already implementing local battery storage at some of their superchargers. This lets them have a more consistent draw on the grid.

The ends result of this is that you end up having a huge amount of deployed battery storage at truck stops all over, able to take advantage of spikes in energy production, respond to grid shortages, and smartly manage charging of trucks.

Given it's all integrated, there's no reason a major fleet operator wouldn't sign some kind of contract allowing their fleet charging to be slowed or suspended momentarily in return for lower charging costs.

I also suspect that there will be (at least) two categories of charger - the immediate "I'm on a short break" MW charger, where you're expected to move on after getting 80% or more capacity, and then an overnight charging slot where you park up for your 10+ hour break.

ugh, fixing the formatting, because I keep forgetting this isn't markdown.

- Range: 500 miles

- Time to recharge 80% (+400 miles) capacity - 30 minutes

- Speed Limit (Interstates): 50-75MPH

- Maximum Driving Times (USDOT):

-- 8 hours (followed by a 30min break),

-- 14 hours (followed by a 10 hour break)

-- 60/70 hours (in any 7/8 days)

Are you suggesting that Tesla hasn't done the math or are just flat out lying?
He used the weasel words "true long haul". Tesla was clear about what the truck can do; you're replying to someone who wants to argue with unclear statements. Not a good plan.
I believe the superchargers use on-site batteries to buffer demand to the grid. Perhaps the local battery was depleted, so they backed off charge rates to avoid hitting the grid too hard.
Some do, some don’t. Usually only where Tesla is charged demand charges for peaks by the utility.
As far as I know, none do. The big one in Gilroy has a big battery pack that seems to sell storage and VARs to the grid.

In CA, basically every commercial rate structure includes peak demand charges.

Charge rate is variable based on a huge number of factors, not all of which you can see. It also doesn't charge at the full 120kW the whole time.

Some of the factors that can affect it:

- Whether someone is sharing your charger (iirc every two bays share the same backend infrastructure)

- If you are sharing the charger, then who was plugged in first (the second person to plug in will receive lower priority)

- The State of Charge of the battery pack (the closer to the high/low extremes of charge the slower it will charge)

- Temperature of the battery pack

- Condition of the pack

- Whether there are any restrictions on the pack (some packs if they are supercharged a lot will have caps placed on the charge rate)

I imagine with the 'Megacharger' infrastructure given the sheer amount of power needed, they'll have dedicated infrastructure for the entire bay.

Yup, you're also dealing with companies that may have fixed routes and the capital to develop the infrastructure if it makes financial sense.

Your average consumer isn't going to drop $X0,000 to put in a super/mega-charger but a business might.

Try telling a trucker that their truck didn't finish charging because of complicated reasons.
1. tesla superchargers share a the 'bank of chargers' at 2 spots. look for 2a, 2b, those two spots are shared. pick one that doesn't have a paired spot in use 2. they charger at a higher rate when your car has a lower charge. 3. they cool the battery when charging to prevent harm, if its really hot this can lower charge ability. there's a similar thing with cold temps

They are reliable and predictable with those factors.

There is already extensive discussion of how to charge higher power systems like trucks, tesla has a plan, google for details. the easiest thing is just several superchargers together. by the way, a supercharger spot is just a bank of regular chargers working together, feeding dc directly to your car's battery. its not magic, it's all pretty easy to understand once you get some basic info.

The Fremont factory supercharger has always been the worst one in the network. Be social with some owners, you'll be surprised what you might find out.
The biggest issue with electric semis is that trucks have a maximum weight limit, and the battery is much heavier than fuel for the same mileage. As much as you may get the same range, if you can't carry as much cargo, you lose the benefits of price per mile.

Note that Tesla didn't mention the weight of the battery when announcing the truck. (Though I assume they believe they can get it even lighter before they release production models)

Do you think Fedex (20), Walmart (15), Anheuser-Busch (40), Pepsico (100), and UPS (125) would have put down money to buy hundreds of these without checking if they can actually carry a load?

They're being rated to carry a 80,000lb load at highway speeds for 500 miles - clearly the weight distribution isn't a problem.

The other thing you're forgetting is that it's not a battery vs fuel weight you need to compare it's battery+motor vs engine+transmission+gearbox+fuel+urea+etc.

There's so much less that needs to be in an electric truck.

True points - though I would think for at least FedEx and UPS carry significant amount of light cargo in some trucks (boxes with empty space around the actual contents, and other lighter contents) compared to other trucking groups which carry almost solid cargo (a truck carrying sugar for example). Pepsi specifically is an interesting one - because I assume that most of the cargo space is liquid, though I wonder how much empty space is in their cargo areas due to round bottles
Battery alone (for 500-miles range) will weight more than double of engine+transmission+gearbox+fuel of 2000-miles range ICE semi.
Last week, a few super geniuses in SF were debating Tesla's valuation. This was after the last night of Token Fest.

I've only been in SF since August but what it has taught me is there are super brilliant people in this world. So smart that I think it's possible, dare I say probable, that an amazing infinite world is on the horizon.

Elon's unfathomable intellect and ability to execute is, in fact, a normal basic expression of the human mind. He's just synchronized perfectly, like a superconductor, with the invariant object at the center of the Universe.

What is that object? Total information in hyperspace. Here is a video I found 9-years ago that explains it well-

https://youtu.be/r9UsvGo2ImU

Barack Obama recently said during a speech in Japan that he hopes to create a million young Barack Obamas. I agree with him -- I think the principles that make a person great are fundamental, universal and within everyone's reach.

Elon is proof that anything is possible. Here's to a million future Elons, Baracks and, most importantly, a million new Satoshis. Smart, cypherpunk minded super geniuses conspiring without credit in a great work to build a better world.

Sorry for their loss...

Preordering from a soon to be bankrupt company isn’t a great business idea. But then again the Fed is always there to rescue everyone.

It’d be hilarious if Trump bailed out Tesla. Trump too his his Solyndra or whatever that crony piece of shot company that Obama rescued was called.

Those trucks look like beluga whales. I wonder if that's what their inspiration for the design was.