Home » The Tesla Semi Finally Reaches Customers With Some Promising Figures, Let’s Take A Look

The Tesla Semi Finally Reaches Customers With Some Promising Figures, Let’s Take A Look


Tonight, Elon Musk kicked off the Tesla Semi truck delivery event at its Nevada factory. After years of waiting, Tesla Semi trucks are actually reaching customer hands, and at least on paper, these trucks seem pretty smart. They have the power to accelerate 82,000 pounds up a 6 percent mountain grade and apparently enough range to deliver a load 500 miles away. Let’s check this big rig out.

Just over five years ago (which feels like it’s been a decade, I know) Tesla unveiled its idea for a semi truck of the future. The Tesla Semi promised to revolutionize the trucking industry with an EV semi capable of hauling a full load 500 miles on a single charge. Development of the Semi has been a rollercoaster. Back in 2018, Musk said that the Semi would have four Model 3 motors and release in 2019. That year came and went without a Tesla Semi, with Musk saying that it would release in 2020. Then, the release date of 2020 became 2021, then 2023 was even hinted at as the release. Along the way. Tesla said, at least in an earnings report, that the delays were due to a limited supply of batteries and the supply chain shortage.

But now, it seems that Tesla has pulled it off, so what are truckers getting with a Tesla Semi?

The Meat And Potatoes

Mercedes Streeter

Tesla’s Semi delivery event both answered a lot of questions and frustratingly left a lot completely unanswered. I’ll start with what you’re not about to learn here. Elon did not reveal how much the Tesla Semi weighs. He also did not reveal its actual horsepower, torque numbers, price, or battery capacity. But don’t fret, because Elon did reveal some pretty sweet details.

The biggest one, I think, is its range. Recently, Elon Musk Tweeted that a Semi loaded to 81,000 pounds completed a 500-mile drive. Musk didn’t indicate how many times the truck had to be charged for the trip. But at the event, he clarified that the truck went the whole distance on a single charge. And as proof, he brought a timelapse of the truck doing the whole trip.

Even then, it seemed unbelievable. Freightliner, an established player in the semi truck space, has its eCascadia, which will do 230 miles at best with its 438 kWh battery. Volvo’s VNR Electric can go 275 miles with its 565 kWh battery. The Nikola Tre BEV claims up to 330 miles from its 733 kWh battery. And the rest of the competition doesn’t even come close.

So how did Tesla beat the competition so hard? Musk gave a hint when he said that the truck worked out to just less than 2 kWh/mi on the 500-mile drive. Based on that, our contributing engineer Huibert Mees said:

That works out to 1000 kWh for the trip. That’s 10 Model S batteries. Makes sense.

And, he may be onto something. While Musk did not reveal the batteries, he did reveal other details. The Semi is a parts-bin special. Musk says that it uses inverters, drive units, power electronics, HVAC, and infotainment from other Tesla models. For example, the drive units are said to be “essentially” the same units from Tesla’s Model S and X Plaid cars.

Screenshot (66)

Screenshot (65)

Elaborating further, Musk says that the Semi uses the three Plaid motors. This is one fewer motor from the proposal to power the Semi using Model 3 motors. However, using the more powerful Plaid motors apparently means no range or power loss compared to using four Model 3 motors. Musk went on to say that in the Semi, one motor is always running. However, when more torque is needed for climbing or when more acceleration is needed for passing, the two additional motors have an automatic clutch system to seamlessly integrate into the action. This configuration is said to give power when it’s needed and low consumption when you’re cruising.

One thing that confused me about this is when Musk said that just one of the football-sized Plaid motors is more powerful than a diesel semi engine. Assuming that the Semi is making the same amount of power as a Plaid, it’s making 1,020 horsepower. Class 8 semis commonly run 400 HP to 600 HP or more depending on configuration. Heck, you can buy RVs that make 605 HP. So, that claim doesn’t currently check out unless there’s more power.

On that note, Musk says that loaded, this thing can hit 60 mph in 20 seconds, and there’s enough power to accelerate up mountains.

Screenshot (70)

The Semi is running on 1,000 Volt architecture. To charge the beast, Musk says that Tesla developed a liquid-cooled charging connector capable of delivering 1 megawatt of power. Apparently, it will be good enough to charge whatever battery is in the Semi to 70 percent in just 30 minutes. This architecture, Musk said, will make its way down to the Cybertruck. It also has a traction control system to prevent jackknifing and a regen system so strong that Musk says that the Semi can go down a mountain without its driver needing to touch the brakes.

There’s Other Good Stuff, Too

20221119 220439
Mercedes Streeter

Alright, so Musk didn’t reveal the Semi’s price, weight, horsepower, or other important stuff. But there were more goodies revealed in the event.

For one thing, since there’s no diesel engine in the way, Tesla was able to give the Semi a bigger cab. Despite its looks, this is a day cab. However, this is a day cab that’s big and tall enough for a driver to stand inside of it. Senior Manager of Semi Truck Engineering Dan Priestley said that you’ll be able to do stuff in the safety of the cab like get dressed. And there’s ample room to store tools.

Screenshot (73)

Priestley also noted convenience features like a suspension drop to make hooking up to a trailer easier and an automatic light test to make sure your lights are working. These features, Priestley says, are to make driving the Semi really easy.

And for Tesla, the Semi helps it achieve its mission of pushing sustainable energy. Citing a 2021 Tesla report, Musk said that Class 8 trucks make up just 1 percent of all vehicles in the United States. However, the report continues, they contribute to 20 percent of U.S. vehicle emissions and 36 percent of U.S. vehicle particulate matter. Musk wants to combat this by putting 50,000 Semis on the road in North America. And yes, this graphic seems to hint that there may be a Tesla robotaxi concept in the future?

Screenshot (80)

At any rate, the Semi is finally hitting the road with customers. The first deliveries went out to PepsiCo, and a Pepsi-branded truck made an appearance at the event. Pepsi’s Frito-Lay division also completed its first cargo run. As more Semis roll out, Tesla says that it will use the Semi for its own operations between suppliers and factories.

Thus far, things sound good, but we really want to know more about how much it weighs with those undisclosed batteries. How much the Semi weighs will end up determining how much of the 82,000 pounds allotted to electric semis will actually be used for cargo. And the price may determine how close Tesla will get to hitting its production goals.

[Correction: An earlier version of this story said that Musk was accompanied by Tesla Chief Designer Franz von Holzhausen. He was accompanied by Senior Manager of Semi Truck Engineering Dan Priestley. We regret the error and thank you, readers!]

Screenshot (72)

Share on facebook
Share on whatsapp
Share on twitter
Share on linkedin
Share on reddit

71 Responses

  1. I know this is going to sound insane but hear me out:

    What if we got these upgraded to full self-driving by putting them on some sort of guided surface or track instead of on the open highway? They could have metal wheels instead of rubber as well, which would massively cut down on rubber particulate pollution. And then you could use this guided “track” of sorts to conduct power directly to the motors instead of needing to haul around 12,000 pounds of batteries made of scarce lithium? Hell I bet with the weight savings and more available power you could pull multiple trailers with one cab, maybe even dozens of trailers! Is anyone working on this?

  2. That wasn’t Fritz at the presentation; it was Dan Priestley, who’s in charge of truck engineering.

    On the power point, the Plaid motors make 1000 hp when hooked up to 400V. If this is really a 1000V architecture, then presumably they’re making more power in this application. At one point, they said something about the truck having a total of 3X the power of a regular diesel, so I’d guess each motor here is making about 500 HP, for a total of 1500 or so.

    1. Average power for a 0 to 60 in 20 second with 81,000 lbs is 886 horsepower. But assuming constant acceleration it would be peak power of 1772 hp when at 60 mph. So somewhere around 1500 seems reasonable if you get a higher acceleration between 20-50mph.

    2. Another consideration: I’m finding a claim in an Evo article on the Plaid powertrain that each motor can handle about 400 hp, and the battery is the limiting factor. This is less efficient than exactly matching the motors to the battery, but it gives headroom to torque vector (not enough for negative torque, you need positive torque on all three motors to use all of the battery’s power, but torque vectoring nonetheless) under full power.

      This thing’s going to have ~10 times the battery, which would imply 10 times the battery power. (In practice it’ll probably be less than that, as you don’t need that much power, and may be able to optimize the cell construction for more energy density than the cells used in the Plaid cars, instead of power density, to reduce weight and volume.)

      Combine that with, as you point out, a higher voltage system, which all else being equal, would be optimized for higher RPM operation, and I believe this means that you can get higher power when operating at higher RPM, and you could probably get each motor north of the 600ish hp that US-market semis top out at.

      I’d also guess that the highway motor would be geared such that it’s at peak efficiency (not into field weakening) somewhere around 55-65 MPH, where the acceleration motors would be geared such that they’re at their absolute maximum RPM at the Semi’s top speed (I’d guess that’ll be somewhere around 80-85 MPH just because of the highest speed limits for semis in the US, AFAIK, being 80 MPH).

    3. They’re probably going to be conservative and not greatly exceed the motor’s/inverters’ continuous ratings. This reduces wear and tear and operating cost.

      So 1000V increases the continuous power possible by 2.5x vs 400V, because continuous current should stay the same, assuming the motor windings haven’t changed and that the IGBTs used in the inverter retain the same current rating as those in the car/CUV lineup.

      1. Usually you can’t just push 2.5x more current through the motor, either due to cooling issues or core saturation, or permanent magnet demagnetization. Instead you would assume that the motor copper packing factor and cooling stays the same, but re-wind the motor for a voltage constant 2.5x higher. You then get the same torque rating but at up to a 2.5x higher rotation speed.

  3. We’re going to need to rethink how we generate electricity in America. Imagine 50,000 of these all trying to recharge at non existent electric truck stops.

    1. We already have needed to rethink power gen for decades. It being a survival priority wasn’t enough, but maybe the powers that be will pay attention when it’s a market demand.

    2. But think of the environment..!

      No, but seriously, I totally agree with you.. What no one is talking about is how we will meet the overwhelming demand in providing power for all the EV’s once the brainless legislature is put into place to phase out all ICE vehicles..

      No one is talking about nuclear, and for sure we can’t meet the demand with solar and wind.. “Oh, but these vehicles will be smart and will charge off peak”.. YEs, so will every other EV…

      Another thing is the disposal of all the lithium batteries. Sure the vehicles aren’t dumping emissions into the atmosphere, but what about production, disposal and pollution of all the battery packs. On top of that the associated emissions and pollution that comes with the production of energy for all the EV’s?

      Literally no one is talking about this.. Everyone seems to be hoodwinked with the “It’s good for the environment” nonsense.

      1. You and I. We are talking about nuclear. Wee should be talking about hydrogen, too. They go together like peanut butter and jelly and we are going to spend a whole lot of money on this electric phase before we inevitably end up at nuclear and hydrogen anyway. That lithium could be used, more sensibly, in about a million other things.

    3. I wonder how economically feasible flywheels can be.

      I know, for example, that nuclear fusion experiments use flywheels and motors/generators for the megawatts to gigawatts of power that they need.

  4. If my estimate of 1000 kWh is correct for the battery size, then we’re probably talking about a battery on the order of 12,000 lbs. That’s a lot of weight but on the other hand, a big diesel engine is not light nor are the multiple fuel tanks trucks use so, we might be talking only a few thousand pounds more than a traditional semi since those electrical motors aren’t very heavy. It’s actually quite impressive. And if those recharge times pan out then this will really be a game changer.
    What Tesla has always been very good at is electrical efficiency. I’m not aware of any other EV manufacturer that can match Tesla efficiency as measured in Wh/mile. This is really what we should be talking about when discussing EV’s, not range. It’s similar to talking about fuel efficiency in ICE cars, and this is where Tesla really shines and where everyone else is still playing catch-up.

    1. Engine and fuel come in far lighter than that…..like 6000-8000 lbs depending on fuel load (a lot of day cabs run smaller or single tanks). That’s a 13% to 18% payload loss. That’s massive.

    2. Those recharge times only work if you have access to some serious electrical infrastructure though…which won’t be cheap. Or even practical. Especially when you are trying to charge tens of these trucks at a time…

      In summary, these things need to come with fast swappable battery packs.

    3. The Freightliner is managing better than 2 miles per kWh already. It just has a smaller battery.

      This works against the theory their battery management is better because they’re just running a larger battery.

      Also, while he says it will tow up a 6% grade, I’m going to go out on a limb and guess there wasn’t a 6% grade on the 500 mile test. And I can visually see that none of this happened in anything like normal temperatures for the rest of the country this time of year.

      1. The Freightliner may be less than 2 kWh/mile but under what conditions? The Tesla did it while loaded to 81,000 lbs. If the Freightliner is as efficient at the same weight, then I commend them.
        Based on the video of the route and the tweet JST linked (I’m also very familiar with most of this route since I live in CA), it looks like there was a small grade right before Los Banos (probably the one coming into Santa Nella on CA-152) and then a major grade at the Grapevine north of LA. The Grapevine always feels like it’s never going to end so this is quite an impressive performance. And, as JST noted, there is a long downhill on the other side of the Grapevine which helps recover a lot of energy.

        1. Like Denver to New York in winter. 500 mile range? I don’t think so. In snark mode does this bazillion pound thing have Tesla’s autonomous driving package?

    1. That’s probably asking for a huge amount. Maybe with dedicated fleets, but the average over the road driver doesn’t necessarily empty his trailer every time, they pick up and drop off trailers regularly.

      However, if it was fully integrated, that would guarantee no jack-knifing as the trailer could apply more braking force than the hauler.

      1. Dedicated trailers could in theory also hold some more battery capacity and extend the range. Of course what will probably happen is that there will not be a single charging standard and each company will use its own system until a regulatory body steps in and picks one.

    2. I think where possible, diesel trucks use engine braking. Regenerative braking from just the cab would be the equivalent of that, but a bit better since there’s no noise issues. You wouldn’t be able to recapture hard braking or the last couple mph, but you’d get stuff like long descents and coasting towards red lights where most of the energy is anyways. Adding it to the trailer as well only really gets you a little extra for all that effort.

    1. Perhaps the idea is that the range is limited and therefore these trucks will mostly see short haul service at first. If they had the range to reliably do a full day’s driving on one charge and then charge overnight at a truck stop while the driver sleeps, that would be great.

  5. Wonder if they got an overweight permit? Usually 80,000 is the legal limit. Would like to see how this thing scales with 12,000 lb of batteries. While big diesel engine is heavy, it’s not 12,000 lbs heavy.

  6. Impressive numbers without much of the context needed.
    The biggest issue I see with EV trucks are much the same as EV police cars.
    Turnaround time. Many fleet vehicles are expected to be available 24-7. How long to get them back to 100% range ?

    1. The big thing is that – if you’re not doing team driving – 500 miles is actually enough range for a semi – if the range estimate is either at 65 MPH, or if you’re operating in a 55 MPH area if it’s at 55 MPH, thanks to laws.

      I’ll use 100% to 10% for the first charge, and assume the 30 minute recharge gets from 10% to 70%, not 0% to 70%. And, as Tesla hasn’t stated a speed for their range claims, I’ll run the numbers assuming both 55 MPH (the speed which Tesla used for their 97% to 4% 500 mile run, due to California speed limits) and 65 MPH (a more typical trucking speed in the US outside of California). Important caveat: if their claimed range speed is at 55 MPH, the 65 MPH numbers will be wildly inaccurate.

      So, a bit on trucking laws in the US: In a maximum of 14 hours on-duty, you can drive a maximum of 11 hours. You can drive a maximum of 8 hours without taking a 30 minute break.

      90% (100 to 10%) of a 500 mile range battery is 450 miles, and 60% (70% to 10%) is 300 miles.

      Assuming the range is at 55 MPH, it would take 8h11m to get from 100% to 10% – however, you have to take your break by 440 miles, before getting to 10%. Then, after taking your break and recharging to 70% (in reality, probably a bit over 70%), it would take 5h27m to get back to 10% – you’re out of drive time for the day by 275 miles, again before getting to 10%, at which point you’ll need to take your 10 hour rest before driving again, and can do an overnight charge. Essentially, it has enough range, with a significant buffer, to go an arbitrary single driver distance.

      Assuming 65 MPH, it takes 6h55m to get from 100% to 10%. Then, you take your break, and recharge to 70%. It would take 4h37m to get from 70% back to 10%, but you’d need to stop at 265 miles, before getting to 10%, and you’d need to take your 10 hour rest.

      Now, if you’re doing team driving, all of this is blown out of the window, because you can start your rest period while the other driver drives, and they’ll have to wait to start their shift for 30 minutes to get back up to 70%, and then if the range is at 65 MPH would have to make an additional charging stop that isn’t legally mandated. (If the range is at 55 MPH and that’s the speed they’re doing, they’re only leaving 6 minutes of drive time on the table, at which point it’s fine, except for the shift changeover time.) It’s still not too bad, but it’s not as fast as diesel.

  7. OMG at last!
    Immediate thoughts:

    I like the thinking behind the drive train layout,and the shared components.All very sensible.

    That driver has a cast iron bladder. Only one stop in what,six hours?

    I’m unconvinced the center driving position is as safe.We put drivers on the critical side for a reason! Head on’s are the worst kind of wreck.

    Im super curious how pepsi uses them. Hauling frito lay’s will be almost too easy while cans of drink will always hit the weight limit.

    1. If the Pepsi bottler up the street from me is any indication, they’ll use them to pull out into the road when they see you coming, and then idle there for dozens of minutes at a time while you can’t get by.

      Bladder size is pretty irrelevant in conversations about EVs in general*. But doubly so when it comes to trucks, since the driver can and will just pee in a bottle rather than stopping.

      * (Doesn’t matter if your EV battery’s runtime duration is the size of your bladder when your destination has no charger, ’cause you’re still stopping to pee at some charging station frustratingly close to where you actually wanted to be rather than at your destination. That driver would rather have a 2-bladder sized battery every time.)

  8. There’s some bet hedging in the comments so… Tesla has very good engineers working around the demands of a pretty terrible boss. There are some elements of “drug addled boy king” in the design – namely the center seating position, which I’d argue isn’t ideal – but for the most part this is actually good stuff. After the Cybertruck and the ergonomic dumpster fire that is the new Model S, I’m happy to see something that actually works.

  9. Let’s do some back of napkin math…..

    10 Model S batteries=1200 lbs x 10=12000 lbs
    Semi engine=3000 lbs + 1000 lbs fuel (this is a high estimate for a day cab)= 4000 lbs

    12000 – 4000 = 8000 lbs difference with batteries alone….for a day cab.

    I know I’m not figuring other drivetrain differences (motors are lighter than diffs I’m sure), but that is an insane amount of weight difference. Even if they offset some of that, it will still be insanely heavy, limiting it to only light load operations….

    While I support electric cars, I just don’t see full electric as the solution for semis. Tech and infrastructure has to go a long way before it is widely viable in the industry.

  10. This probably should have been Tesla’s first major product. Most bang for the buck, environmentally, and the profit margin is better which could have then funded the less profitable car lines. Then again, it’s using the tech they already developed for the cars, so maybe the R&D savings change that.

    1. I believe that Tesla was successful mainly due to their product roll-out sequence. Everyone else was toiling away with low-cost EVs for the everyman and taking a big loss on each one. Tesla went after the premium market where there was more margin to be had. They focused on a customer base that was eager to be early adopters and would be more forgiving to EV shortcomings as well. Only after that was established did they go after the middle of the market. Now they are targeting the commercial space using their parts bin. Pretty logical.

  11. You and I. We are talking about nuclear. Wee should be talking about hydrogen, too. They go together like peanut butter and jelly and we are going to spend a whole lot of money on this electric phase before we inevitably end up at nuclear and hydrogen anyway. That lithium could be used, more sensibly, in about a million other things.

  12. I have no doubt Tesla can make the battery and EV driveline work. I feel equally confident that the rest of the truck is a fail waiting to happen. Heavy trucks live and die on their uptime, so reliability and parts availability is everything in the class 8 truck market. Introducing a new cab and chassis is a difficult and risky endeavor for even experienced manufacturers, and Tesla has no experience with this type of product, and a less than stellar record for getting basic things like door handles & suspension parts right.

    If the Tesla semi had come out a decade ago it would have been alone in the market and had time to sort out its infant reliability issues, but at this point there are already EV trucks for sale by companies that know the market and product requirements. Tesla should have developed a driveline to sell to other truck makers, and spent the engineering time and money saved updating their EV car lineup.

    1. Agreed. Fleet owners have ZERO tolerance for the service/quality shenanigans some consumers have reported with their Tesla cars. They will have to drastically step up their game in service to be successful. I would not be an early adopter if I was managing a fleet. Of course, many questioned whether they could scale up production for the Model 3, and they did that. That was a huge accomplishment. We’ll see what they do here….

    2. That’s why their first buyers will be large fleet operators for companies like Pepsi. They know what their routes are and what they’ll be hauling, and it doesn’t change too much. That predictability is huge, in terms of figuring out how many of these trucks they need in order to get the level of service they want. And speaking of service, I have no doubt that the early adopters for the Tesla Semi are going to get rock-solid, gold-plated, white-glove support from Tesla. Tesla has a vested interest in making sure that their product succeeds, so they’ll want their early customers to have good experiences that build industry confidence in their truck.

  13. Looking at the video, it appears traffic conditions were ideal for range. From my experience, urban/heavy traffic highway driving (55-65 mph with frequent regenerative breaking to slow down for traffic) is extremely efficient. I can get ~80 miles out of my Leaf in those situations. However, if I set the cruise at 70 mph, I only get around 55 miles of range. It will be interesting to see how much range these trucks have on rural interstates at higher speeds.

  14. A typical Freightliner day cab with a 53′ box trailer has an unloaded weight of about 30k lbs, which means that it can carry about 50k lbs of cargo and still be at or under the 80k lbs weight limit (the maximum for most states. Trucks that carry more weight require special permits). Trucking companies, and their customers need to maximize the cargo on each load in order to be profitable. If Tesla cant get very close to this, it will be a hard sell (although it seems reasonable that states may make weight allowances for electric vehicles, but as far as I know none do so now).

    Also, most local trucking operations run their trucks in 2 shifts (2 drivers use the same truck on alternating 12 hr shifts). So charging time will be a factor in profitability for trucking companies as well. It takes less than 10 mins to fuel a diesel semi, and they all hold enough to operate for at least 12 hrs. Any down time for charging will be costing trucking companies big money.

  15. Impressive run. It will be interesting to see how these battery electric semis actually get used. If you have several hubs WELL within the trucks operating range, you could make it work with set routes and onsite charging infrastructure. Even that will be a long road considering the planning, investment, and change in operating structure while maintaining margins. Whatever the use case, it will have to be very tightly controlled as the last thing you want in logistics is range anxiety.

    1. They’re perfect for local routes, and should make a decent pollution/GHG impact in that role. Long-haul trucks already operate near maximum fuel efficiency.

  16. Good article Mercedes. I’m no Tesla fan, but good for them. It all comes down to operating cost and up-time. If this is a good solution that meets those needs, the market will respond. I am open to any tech that can compete to reduce transport emissions.

  17. With normal brakes you turn motion into heat, big problem on long stretches of descent because if you do it wrong you can overheat your brakes. So with regenerative braking you turn motion into energy. No problem with overheating brakes. But what actually happens when your battery is full and your still pushing in electrons? Is there a way to dump excess energy?

    1. Well Musk did say the driver didn’t have to use the brakes, insinuating that there are traditional brakes available for use. I guess the main takeaway is just don’t put any charging stations on the tops of mountains and the trucks will be using a big chuck of the stored capacity to climb up one side before going down the other.

    2. It’s less of an issue than you think, unless you’re charging your battery at the top of a hill.

      To get to a speed (or altitude), you need to exert energy. Since nothing is 100% efficient, you use more energy to get to a speed or altitude than you can reclaim.

      That being said, let’s say you charge in Denver and drive downhill toward Chicago. It’s possible that you’d have too much energy to recover (though, again, you’re quickly burning electrons just by driving). In that case, regular brakes do the job.

    3. With my Leaf, if the battery is full, regenerative breaking doesn’t work (it feels like the car is in neutral when you let off the gas pedal). I presume the Tesla semi would be similar. Your battery is rarely 100% full, though. Also, with a long-range EV like a Tesla, you may not charge to 100% to preserve battery longevity. If you routinely charge to 80%, you would have plenty of battery capacity for regenerative breaking.

Leave a Reply