Tesla Semi Truck Battery is How Big?

7 months ago by George Bower 110

Tesla recently gave us a “first sneak peek” at its Semi-Truck project

Asked in another way: What MPGe can we expect from Tesla’s new electric semi truck?

In a previous Inside EVs article there developed an interesting discussion in the comments section relative to Tesla’s semi truck battery size. In that discussion one commenter (@PMPU) posted:

“Our hypothetical BEV semi will have an energy efficiency 2.6 times that of a diesel semi. (An EV car is about 3.5 x as energy efficient as an average gasmobile, but diesel engines are about 30-35% more efficient than gas engines.)”

In an effort to further refine this “napkin math”, we dug a little further and found a very interesting chart made by Cummins relative to the Super Truck program. This super truck program was intended to see what MPG one could get out of a Diesel semi truck if one pulled out all the stops: installed the most efficient advanced tech diesel engine, improved the aerodynamics, lowered the rolling resistance etc.

At current MPG of around 6.5 MPG this super truck program demonstrated around 10 MPG (1.5 times better mileage).

Cummins published a breakdown of all the losses in a semi truck: engine, aerodynamic, rolling, drive train, aux loads and braking and put them in a chart shown in figure 1.

Cummins Loss Breakdown for Diesel Semi Truck

The first thing that jumps out on this chart is the engine losses which Cummins puts at 58%. What Cummins has inadvertently told us with this number is the diesel engine cycle efficiency is 42% as shown in figure 2.

Loss Breakdown for Diesel Semi Truck

In other words, at 100 BTU of input energy to the diesel engine, 58 BTU’s is rejected in waste heat.

The other losses are listed for both an interstate driving cycle and an urban cycle. Note from figure 2 that braking losses are fairly minimal for the interstate driving cycle but are a big loss in an urban cycle where we have lots of starting and stopping.

Aerodynamic losses have the reverse effect. Aero losses are minimal in an urban cycle but a big contributor in an interstate cycle where we run higher speeds.

So Cummins chart seems to make sense.

Now that we have the losses broken down we can make some assumptions for what we could get with an electric semi truck. This electric semi truck obviously includes a much more efficient motor than the diesel and like electric cars it incorporates regenerative braking and extensive aero modifications.

The loss breakdown for the electric truck is shown in figure 3.

Estimated losses for Electric Semi Truck

The biggest improvement occurs in the electric motor and inverter. We just lowered losses from 58 BTU (out of 100) to 14 Btu’s. The next biggest improvements come in the form of improved aerodynamics and regenerative braking. Regen saves us 13 Btu’s in the urban cycle and improved aerodynamics saves us 8 BTU’s in the interstate cycle. Improvements in aerodynamics losses were arbitrarily put at 50%. Regenerative braking assumed we captured 80% of the available energy. Rolling resistance, drive train losses again arbitrarily reduced in half.

Adding up our electric semi truck improvements we have lowered the amount of energy required from 100 for the diesel to 34 in the electric semi truck: a factor of 2.94 better than the diesel.

The results of this mini analysis-call it “napkin Math 1.5”- are shown in figure 4.

Electric Semi Truck requires approximately 1200 kwh for 600 mile range

Napkin Math version 1.5 estimates the ratio of electric to diesel MPG at 2.94 compared to “Napkin math” version 1.0 ratio of 2.6.

Applying the ratio of 2.94 to the diesel’s MPG of 6.5 we have an electric semi truck with a 19.1 MPGe = .503 miles per kwh.

As a reference BYD’s electric bus gets approximately 15.7 MPGe at 14 MPH with A/C on and full passenger load. At 42 MPH the BYD bus achieved 32 MPGe so we feel relatively confident that we are “in the ball park” on what Tesla can achieve with an electric semi.

So the battery is how big?

For a 600 mile range at .503 miles/kwh we need 1193 kwh battery: roughly 12, P100D battery packs: at $100/kwh that’s $120,000 battery.

It should be interesting to see what technical information Tesla reveals about the new semi truck when it is officially unveiled this September.

Tags: ,

110 responses to "Tesla Semi Truck Battery is How Big?"

  1. suresh says:

    1.2mwh battery. how heavy with that be. whats left for load?

    1. Odycuz says:

      Even assuming the batteries added 10,000lbs it wouldn’t have much of an effect on the overall load. To the best of my knowledge not very many semis actually have a “weight issue” (imagine a trailer loaded with bath tissue or lawn mowers). So I doubt it would be a substantial problem for most long haul drivers.
      What I never hear, and what makes sense to me is the idea of swapable trailer packs, wherein a driver can pull into a truck stop and swap batteries if he needs/wants to get going real quick. I don’t know why you’d want to build long haul range into the trucks frame. Just put in 200kwh, and have the other 1mWh come from the trailer. In that instance you could put in more or less depending on need. There’s a sh!t ton of room under those trailer floors.

      1. Sascha says:

        why not have that 1Mw trailer be usable with standard diesels too? The massive fleet of existing tractors would only need to idle to run the brakes and power steering!

        1. Pushmi-Pullyu says:

          A battery pack in the trailer simply isn’t going to be practical. The whole idea of a tractor-trailer rig is that, at least in theory, any tractor can haul any trailer. If you require a BEV tractor to be matched up with a special trailer which has a battery pack, then you’ve severely limited the usefulness of the tractor.

          Besides, there’s no good reason to put the battery pack in the trailer. There will be plenty of room in the tractor once the large diesel engine powertrain (with its large transmission) is removed.

      2. trucking expert says:

        A 10,000lb battery is very significant.

        These trucks are limited to a total loaded weight of 80,000 lbs.

        Today, they typically have 300 gallon fuel tanks. On some heavy loads, they try to keep no more than 150 gallons of fuel to keep under the weight limit.

        I’m hoping Tesla has some other trick up their sleeve. It may make sense for them to go the range extender/generator route as an exception for trucking. Maybe 300-500 kWh battery and an onboard diesel generator. You could keep the diesel generator small. If the truck knows the route, it would be able to predict ahead of time when it needs to start running the generator to extend range.

        1. Mister G says:

          Nope the goal is to destroy and replace diesels…BELIEVE ME lol

          1. trucking expert says:

            I get that. Just pointing out this is the one exception that could make sense until battery weight is much improved

        2. pjwood1 says:

          “Range extender..route”

          Ya think? This exercise pretty much proves it. Need like the i3 REX with hold mode, but with just enough engine to regenerate and keep 70mph in the truck.

        3. Pushmi-Pullyu says:

          “trucking expert” said:

          “A 10,000lb battery is very significant.

          “These trucks are limited to a total loaded weight of 80,000 lbs.”

          And furthermore, some State laws limit heavy trucks to an even lighter load than that.

          The weight problem (and the 80,000 lb. weight limit) were noted in my “napkin math 1.0” analysis, too. We can’t just ignore the very heavy weight of the battery pack, which will certainly cut into the load our hypothetical BEV semi tractor-trailer rig can carry.

      3. jamcl3 says:

        Payload capacity is a top 3 concern, that determines how much money a truck can make. The other two are fuel efficiency and uptime.

      4. All-Purpose Guru says:

        The interconnect between the trailer and cab needed to handle a 1MW battery in the trailer would be troublesome; hell, the interconnect in a NON-severable connection would be challenging enough.

      5. pigsfly1830 says:

        The problem with 10,000 lbs isn’t pulling it, its the legal load limit, being 80,000 lbs for most states. That’s a lot less payload, meaning any benefit from possibly increased efficiency isn’t gonna make a huge difference with that heavy a battery

      6. Lorne R Houle says:

        Dont forget that theres no engine and tranny weight. Thats a good 6000 lbs loss at least… so not much extra added weight.

    2. ItsNotAboutTheMoney says:

      Model S/X 100kWh pack weighs about 1250lb.
      10 of those, 12,500lb.
      Don’t know about motor and cooling weights, but semi engines are quite heavy, and one example weighs 2880lb. Let’s say 2900lb (you’ll see why).
      Also, they’ll carry diesel fuel.
      Diesel weighs about 7lb/gal in the USA, and tanks are typically up to 300 gallons. So that’s 2,100lb of fuel. A truck cannot drive overweight without a permit, so let’s keep things simple (and generous) by using max fuel weight.

      So the loss of cargo capacity is:
      motors + cooling + battery – engine – diesel
      = motors + cooling + 12500 – 2900 – 2100
      = motors + cooling + 7500

      Not good, but probably not _terrible_.

      1. jelloslug says:

        Don’t forget the transmission, drive shaft, and other drivetrain parts. The transmission is roughly 750 lbs and the drive shaft and supporting components are ~200 lbs. Throw in another 50lbs worth of hoses, brackets, fluids, and coolers and you have another 1000lbs to work with there.

      2. arne-nl says:

        I assume they’re using the new 2170 cell which reportedly has a better energy density. Also, they will design a custom pack for the truck which will probably be lighter than 10 Model S packs. A different type of ‘economies of scale’

      3. Brandon says:

        Thanks for that informative weight analysis!

      4. Pushmi-Pullyu says:

        For what it’s worth, here is my “napkin math 1.0″, as the article labels it. I cover space required and weight, and I talk some about cost, altho the latter not comprehensively as I don’t try to estimate the amortized yearly cost of replacement battery packs.

        There are some useful figures there, such as weights for a typical diesel semi tractor motor, the transmission, and the fuel.



        (revised April 14, 2017)


        A modern diesel semi pulling a load gets 6.5 MPG; therefore uses 0.1538 gallons of diesel per mile

        1 gallon of diesel contains 40.7 kWh of energy

        1 gallon of diesel varies in weight between 6.85lbs. and 7.5lbs per U.S. gallon, depending on temperature. (I’m going to use the figure of 7.1 lbs/gallon)

        diesel semi typical engine weight 2880 lbs

        Eaton Fuller 18-speed transmission weight 738 lbs

        Tesla Roadster upgrade battery pack: 70 kWh in ~10 cubic feet

        standard sized semi trailer dimensions: 110″ high x 96” wide, or 9.167′ x 8′

        DOT weight limit for a six-axle semi tractor-trailer: 80,000 lbs

        Typical price of a relatively high-end new semi tractor: $150,000

        Typical trucker may drive as much as 600-700 miles in a day, and can legally drive up to 11 hours per day.

        * * * * *


        What we need is a BEV battery pack for our semi tractor which will allow it to pull a load for ~750 miles. This should allow the trucker to complete a daily run on one charge. We assume at the end of the run either the battery pack is swapped out for one that’s charged up, or the pack is charged during the hours the trucker is sleeping. Either way, we avoid the need for fast charging and very high current.

        Our hypothetical BEV semi will have an energy efficiency 2.6 times that of a diesel semi. (An EV car is about 3.5 x as energy efficient as an average gasmobile, but diesel engines are about 30-35% more efficient than gas engines.)

        Therefore, our BEV semi pulling a load needs (0.1538 x 40.7 / 2.6 =) 2.4 kWh of energy to run 1 mile.

        Estimated weight of a 2016 Tesla battery pack using 18650 cells: 11.5 lbs / kWh

        Estimated price for a Tesla battery pack (not just the cells): $180 / kWh

        * * * * *

        We need to look at three limiting factors for the BEV semi tractor’s large battery pack: Space, weight, and cost.


        The space behind a long-haul trucker’s cab, the space now devoted to storage and sleeping space, is about 4.1 feet long, at least on the diagram I looked at; I’m assuming the height and width are the same as a typical semi trailer. (At least, the dimensions should be close enough for this ballpark estimate.)

        Let’s use that space for the battery pack. I don’t see losing this space as a problem. Since we no longer need a long nose for the diesel engine, which isn’t there, we can shove the cabin forward, and leave room for the battery pack behind. The tractor now looks more like a “cab-over” tractor with an extended space behind the cabin, rather than a long-nose tractor.

        So I estimate that space at 4.145 x 9.167′ x 8′ = 303.977 cu.ft.

        An upgraded Tesla Roadster’s battery pack has 70 kWh and measures ~10 cubic feet.

        Assuming a similar configuration, that gives us (303.977 / 10 * 70 =) 2127.8 kWh.

        At 2.4 kWh per mile, that’s 886.6 miles.

        This is comfortably beyond our needs of ~750 miles.

        Space isn’t an issue.

        * * * * *


        Weight *is* an issue, altho perhaps not a deal-killer.

        At 2.4 kWh per mile, enabling a range of 750 miles requires our BEV semi tractor to carry a (750 x 2.4 =) 1800 kWh battery pack. At an estimated 11.5 pounds per kWh, that’s 20,700 lbs. We save just a bit by losing the diesel drivetrain; maybe 3000 lbs or so. We also save half the weight of the fuel*, which for a trip of 750 miles (using ~115.35 gallons) would be approximately 409.5 lbs. (We could also subtract the weight of the transmission, but then the EV motor, inverter etc. do weigh something, so let’s assume that cancels out the 738 lb transmission weight.) This brings us down to an estimated ~17,300 lbs. That’s 21.6% of our maximum weight limit of 80,000 lbs. And note that various State laws may reduce the maximum weight even further, depending on what States our long-range truck travels through.

        Now, that’s not to say this makes the idea impractical. It may well be worth sacrificing some shipping capacity as a tradeoff for lower cost per mile of moving the freight. But it does limit the market for our BEV semi a bit, or perhaps more than a bit, depending on what the customer’s needs are.

        *We can only deduct half the weight of the diesel fuel, because that weight will disappear over the course of the day’s trip. Now that’s assuming the driver starts with as much fuel as he needs for the entire day’s trip, which may or may not be correct. So far as I can find by Googling, there isn’t any standard size for semi tractor fuel tanks; one source says they range from 100-400 gallon capacity, and I see a reference to 2 x 150 gallons as a typical size. So then, it appears reasonable to assume ~115 gallons is carried at the start, with no stops for refueling for the entire shift.

        * * * * *


        Cost for the battery pack is the real issue here. And that cost is almost certainly why, for example, UPS, FedEx, Wal*Mart, and other companies with large trucking fleets have not already started switching to heavy BEV trucks.

        That 1800 kWh battery pack, at $180 / kWh at the pack level, will cost an estimated $324,000. And that’s Tesla’s estimated cost, not price, so you can likely add another 15-25% to that. Note a reasonably high-end diesel semi tractor costs $150,000, so adding that battery pack is more than tripling the cost. With a 20% markup, it’s $388,000, which is 259% of the diesel semi tractor’s $150,000 price. Sure, Tesla will save some money by using an EV powertrain instead of the much more complex, and more expensive, diesel powertrain. But as a percentage of the price of that battery pack, I doubt losing the diesel engine, exhaust, etc. etc. will make much of an impact on price.

        There is also the matter of battery life. A Tesla battery pack may be expected to last the life of the car, but the typical car is only driven about 5-10% of the hours in a day. Contrariwise, a long range truck is expected to be on the road as much as possible. A truck just sitting around still has to be insured, and the owner still has to pay all those fees for a heavy commercial vehicle. A truck just sitting around is losing money for its owner.

        So we need to ask: Just how many times will that very expensive battery pack have to be replaced, over the lifetime of the truck? A semi truck is expected to last an average of 20 years, significantly longer than the average life of a passenger car. Will the truck save enough on fuel costs to justify the amortized cost of buying replacement packs?

        That’s a subject beyond the scope of this analysis.

        * * * * *


        Several people responding to this “napkin math” analysis have pointed out that if we assume the driver only drives for half a shift, then uses either battery swap or fast recharge while eating lunch, then this will allow us to use a battery only half the size, weight, and cost. This of course allows us rather more optimistic assumptions. We’re not sure how realistic it is to think that a trucker would always stop to eat lunch at a battery swap station or a BEV truck stop, but let’s do the math anyway.

        First, we will assume a 60% battery pack size, not 50%, because there needs to be a least a bit of flexibility in what time the driver eats lunch and how far he drives before stopping in mid-shift.

        60% battery pack size weighs 12,420 lbs. Losing the weight of the diesel engine and the diesel fuel saves approx. (3000 + 204.75) ≈ 3200 lbs, totaling 9220 lbs, which reduces the 80,000 lb. carrying capacity of our hypothetical BEV semi tractor-trailer rig by 11.53%, which makes this a much more optimistic scenario!

        1. Steve says:

          battery will only need to be half that at 600kW allowing it to travel 350miles per charge. Fast charging will allow it to charge back to full in the time it takes the driver to drink a cup a coffee and take a piss.

    3. Alan says:

      6 powerpacks would be 1260 kwh and weigh nearly 24,000lbs !

      1. arne-nl says:

        Powerpacks are for stationary use and not optimised for low weight

    4. jeff kemper says:

      Evan though semi trucks can haul vast loads of 500,000 pounds or more up and down hills.
      The true question is why not just add a charging system so that the vehicle can have evan farther range and can charge when the batteries run low while it is moving the loads.
      Their is a lot of new technology that has been sat on for years so big companies could make lots of money in slowly putting out the new technology.
      Semi trucks at one time used pick up truck motors 6 and 8 cylinders. Which electric cars have 120 Hp electric motors that have more power.

      The real improvement in the technology will be when they decide to use the smallest size motor through a transmission to do so. I believe I have a design that will work to lighten the load of batteries you need to carry. Which can be in stalled In older semi trucks for about $40,000. to $60,000. a conversion kit. Since semi trucks are designed to be rebuilt over and over again this makes it cheaper than Tesla by far. The charging system can be almost unlimited depending on how you design it.
      Electric cars can be converted for about $5,000.00 to $6,000.00. I can prove the technology can work.

    5. Joe S. says:

      You might remember that engines, drive-trains, and transmissions/differentials are pretty heavy as well. Not to say that the Tesla battery isn’t heavier than the cumulative weight of all these components, but you can’t simply rate it as a +10,000lb addition without subtracting what will be replaced and/reduced.

  2. ClarksonCote says:

    WalMart had also experimented with a radical design change for increased semi efficiency:

  3. ClarksonCote says:

    How about some “low power” 100kW AC Level 2 charging overnight? 🙂

    Or, perhaps this is why some of the new Supercharger stalls are of the “pass-through” configuration: They can easily accept Semis charging as well, and an overnight charge will get you a full semi truck battery.

    1. Pushmi-Pullyu says:

      “Supercharger stalls… can easily accept Semis charging as well…”

      I find that unlikely; extremely unlikely. BEV semi trucks will need EV truck stops to recharge, just like diesel semi trucks need existing truck stops to refuel.

      Existing Supercharger stations are not set up to handle BEV semi tractors any more than existing gas stations are set up to handle tractor-trailer rigs. Not even those gas stations which dispense diesel.

  4. David Murray says:

    On the bright side, semis aren’t cheap to begin with. So an expensive battery isn’t a deal-breaker, especially when it will likely save money in the long run, which is what accounting departments are more interested in.

    1. zzzzzzzzzz says:

      1200 kWh at $200/kWh at pack level would mean $240,000 extra. Trucks are not cheap, but less than that, or around that for SuperTruck.

      Cost/benefit flies out of window here. Smaller battery for short distance urban route niche where overnight charging is possible and electricity is cheap would make much more sense. Forget on-road charging of such 1.2 MWh monstrosity.

      1. ClarksonCote says:

        I think if you do some math on how many miles a semi drives in a year, at 6.5MPG, you’ll find there’s still a compelling ROI case.

        1. ClarksonCote says:

          Hint: It translates to roughly $40,000 per year in fuel costs.

          If that is cut to $10,000 per year in electricity costs, they save $30,000 per year or pay for the added cost in 4-6 years. Since many operate for 10 years, the net result is a decent ROI even assuming constant fuel prices.

      2. jelloslug says:

        If there are basically multiple packs that make up the battery, you could charge them all individually at once seriously cutting down on charging time.

        1. Brandon says:

          Very good point. Using 3 300-350 kW fast charger cables would do the job of fast charging a 1200 kWh battery. Simple.

          800-1000 volts and 350-400 amps is the max practically possible in a cable if it is to still be usable, flexible, and control the high temperature.

        2. Pushmi-Pullyu says:

          To be fair, diesel semi tractors often have twin tanks, which are often filled using two hoses from the fuel pumps at the same time, which of course means filling twice as fast.

          I suppose we could see the same thing for charging BEV heavy trucks, but it seems unlikely that the market is going to support a forest of charging cables all converging on one vehicle. At some point, it becomes more trouble than it’s worth to connect and disconnect multiple charging cables.

      3. Nix says:

        It is normal for a class 8 truck to drive well over a million miles over the life of the truck. At 6.5 mpg that will likely be well over half a million dollars worth of gas alone, without accounting for any other costs.

        The cost of fuel is a huge factor for fleet operators.

        “1200 kWh at $200/kWh at pack level would mean $240,000 extra”

        Well, since Tesla announced in a conference call that they were at the $190/kWh pack level cost with 18650 cells over a year ago, and they have announced in early 2017 that 2170 battery prices are 35% lower, and Elon is on the record talking about $100/kWh at the pack level by the time this would realistically go to market, it looks like you are likely quite far off on your estimates. Which isn’t surprising given your long track record of failure regarding anything Tesla related.

        Why do you insist on embarrassing yourself so often? Is it an S/M thing?

      4. ClarksonCote says:

        A super charger for a Model S is a perfect slow charger for a Semi overnight!

      5. Steve says:

        Assuming 1200 kw needed. Assuming $200 per kw

        Try 600kw and $100 per kw

    2. Pushmi-Pullyu says:

      “On the bright side, semis aren’t cheap to begin with. So an expensive battery isn’t a deal-breaker, especially when it will likely save money in the long run, which is what accounting departments are more interested in.”

      I don’t believe it will save money in the long run at current battery prices, once you include the amortized yearly cost for battery pack replacements. Keep in mind that semi tractors get heavy use, and we can expect the battery pack to need to be replaced far more frequently than a BEV passenger car’s pack would.

      And that’s the biggest reason why I feel fairly confident in my prediction that what Tesla will show will only be a concept vehicle, not anything intended for production.

  5. Chris O says:

    My own back of napkin math seemed to make for a pretty good business case for EV trucks:


    battery 1MWH, range 600 miles, cost $130K, lifespan 2000 cycles over 10 years to 80% residual capacity
    Diesel $2.50/gallon; 6MPG
    electricity cost $0.12/Kwh

    EV truck battery delivers 1 million miles over 10 years at a total fuel bill of $216K
    Diesel truck uses 166,000 gallons at $416K

    So total cost battery+electricty=346K vs diesel $416K; savings $70K.

    More savings when factoring in lower maintenance cost for the EV truck, together probably well over $100K of savings over a 10 year service life.

    I can see how competing with that theoretical 10 MPG super efficient diesel truck could be problematic, though that high efficiency would come at extra cost too of course.

    1. zzzzzzzzzz says:

      You are using old diesel trucks as basis.
      Supertrucks can achieve 10-13 mpg, and not in theory, at least they are further ahead in development. And in would cost less than such battery only one. You need to get electricity at much lower cost than $0.12/kWh for it to compete by $/mile. It is possible in some locations, but certainly not everywhere.

      1. Nick says:

        It is in theory in so far as none are in production today.

        It seems insane that they are not on the road right now! Aren’t fleets beating down Daimler’s door for this truck? Fuel cost is a large portion of over the road logistics operations, and we have a truck which is twice as efficient in the wings, but not yet in production.

        This is why Tesla is going to eat the big trucks lunch, they are all asleep at the switch.

      2. jelloslug says:

        Why not just use magic trucks that get 20 MPG while you are at it?

      3. Chris O says:

        The race is on than, Tesla’s fabled EV truck vs your fabled 10-13 MPG diesel trucks…

        My bet is on Tesla.

      4. pjwood1 says:

        12 cents at story’s .5mi/kwh = $.24/mile. That’s available today. At 7mpg, $2.50 diesel costs ~$.36/mile. You started it.

        Overnight off peak rest areas, away from high electricity prices on coasts, could deliver $.05, to .08/kwh juice. $.12/kwh is the EIA national average.

        1. sveno says:

          Also worth noting is Teslas solar and power storage offerings. So if the customer says “but electricity is expensive around here”, they can simply fix that too. Even more business for Tesla!

      5. Mr. M says:

        In lees than 3 years semis will accieve 10-12 MPG. See also http://www.greencarreports.com/news/1097471_daimler-unveils-supertruck-12-mpg-semi-is-more-than-twice-as-fuel-efficient

        If you can have these MPG with diesel you can get to 26-30 MPG for a electric truck. That means up to 0.75 miles/kWh. Meaning you need only 800 kWh for 600 miles range.

    2. Pushmi-Pullyu says:

      “battery 1MWH, range 600 miles, cost $130K, lifespan 2000 cycles over 10 years to 80% residual capacity”

      Sorry, but no. You’re trying to apply the figures for a BEV passenger car, 2000 charge cycles, to a BEV semi truck. Passenger cars are typically used only 5-10% of the hours in a day. Semi tractors are working vehicles, and see much heavier use.

      No semi fleet operator is going to invest all that money in a semi tractor, including yearly road use fees and insurance costs, and use it only 5-10% of the hours in a day. That would be a waste of money, a lot of money.

      Even basic math proves the impossibility: At 6 cycles per week (assuming one day off per week), 2000 cycles would be used up in 6.4 years. In reality, I suspect the pack would need replacing more often than that.

  6. Alan says:

    Would you actually need a 600 mile range when drivers have to take regular breaks ?

    1. Chris O says:

      A smaller battery gets cycled more often and the cells are exposed to higher discharge rates during acceleration. It would not last the life of the truck I think.

    2. ItsNotAboutTheMoney says:

      Drivers can drive a max of 11 hours during a 14 hour period but they can drive for 8 consecutive hours.

      1. Jeff says:

        Not quite…the North American law is
        An operator shall require that each driver follows either a 7-day or a 14-day cycle, as designated by the operator for the driver.
        A driver who is following a 14-day cycle shall not drive again in that cycle after accumulating 120 hours of on-duty time during any period of 14 days or during the period beginning on the day on which the cycle was reset.
        A driver who is following a 7-day cycle shall not drive again in that cycle after accumulating 70 hours of on-duty time during any period of seven days or during the period beginning on the day on which the cycle was reset.
        Closer to 8.5 hours/day. If autonomous, with 6-8 chargers per truck at each Supercharger to speed up charging, Tesla could easily double the miles/revenue produced by each truck. Low fuel costs, low maintenance, low insurance, low/no driver cost.
        I also can’t imagine they wouldn’t interconnect the trucks controls like a virtual train to bring drag way down.
        New napkin required.

  7. Bojan says:

    Calculations are incorrect.

    Motor losses are calculated as 14BTU based on an energy input of 100BTU, however the conclusion is that with all the efficiency improvements an input of only 34BTU is required for the same distance, making the calculated motor losses incorrect.

    Also, 80% (round-trip) efficiency for regen braking seems unlikely, 60% would be a more realistic number. I don’t think 50% aero improvements are likely since the trailer remains a box regardless of the shape of the semi. Same for rolling resistance, there’s no reason for current trucks not to be already fairly optimized in this regard. So, while the motor losses may be lower, all other losses will likely be considerably higher.

    1. Terawatt says:

      I fully agree, and will further add that considerations like aero and rolling resistance should be omitted as they have no inherent relation to using electric propulsion. Tesla may be able to improve over the average truck in these respects, efficiency being a necessity, but it’s not like you couldn’t do exactly the same on any diesel truck.

    2. Pushmi-Pullyu says:

      I agree, some of the assumptions are unrealistic. 80% round-trip efficiency in regenerative braking is not even remotely realistic. As I recall, the real-world figure is about 35%. In theory that might be improved somewhat, but I would be very surprised if it ever exceeds 50% in real-world use in a practical vehicle.

      I also agree that 50% reduction in aerodynamic drag is not a realistic number for a semi tractor towing a typical trailer. Of course Tesla could come up with a special streamlined trailer, but that would make the rig non-competitive with diesel semi rigs.

    3. Karl says:

      Regen on a truck should be better (recover more of the kinetic energy) because they are braked more slowly/sensibly than a car (driver gets fired if they brake the cargo driving like an idiot). Larger drive motors should be over 90%. Round trip could be boosted with a small supplemental capacitor bank but not for around 5yrs.

  8. I hope they use an overhead electric power system on the highway, and a battery to get to and from the highway.

    1. Chris O says:

      Just the interstates is 48,000 miles so that’s quite the commitment…

      1. Mister M says:

        But not a terrible idea. Tesla is already electrifying most of the country with its super charges, so what if they install new supercharges just for these high voltage lines for a few miles long and allow the trucks in that segment to recharge while driving, this would allow them to go on for a few hundred miles before reaching another “super charger segment”. So from 48k miles it would need may be 10%? 5k miles with “super charger segments”?

        1. Chris O says:

          Back of napkin math suggests that ~60 miles of overhead charging could add another 400 miles of range so basically just 15% of the interstates would need to be wired.

          Don’t know about the technical limitations of charging on the move but it’s definitely an interesting concept…

          1. Brandon says:

            I seriously doubt any overhead wire stuff is gonna happen on Federal highways.

        2. Pushmi-Pullyu says:

          “But not a terrible idea.”

          It is absolutely a terrible idea.

          The typical EV should only need one-and-a-fraction chargers; one at home/work, and a fraction of one public charger.

          If you require all highways and major roads to be electrified, then that would require, what? hundreds? thousands? of chargers per vehicle

          The cost/benefit analysis for burying EV chargers in all major roads makes it a wildly absurd idea, and definitely a non-starter. If there is any idea related to EVs even worse than “fool cell” cars powered by compressed hydrogen, it’s that one.

      2. Someone out there says:

        You wouldn’t need to electrify every single mile of highway, just a strategic selection of them and probably not the entire stretch either.

        1. Pushmi-Pullyu says:

          If the concept would be, let’s say, 5000x too expensive (in both construction and maintenance) to ever be practical, then only electrifying 1 mile in 10 would merely make it 500x too expensive to ever be practical.

          Not really an improvement.

          (I haven’t seen anyone attempt even a “napkin math” cost analysis for this idea. There’s probably a good reason for that; any supporter of the idea attempting a somewhat realistic cost analysis would quickly realize how absurd the concept is, and lose interest.)

          1. Karl says:

            Charge via overhead connection when stopped as this would then allow automated charging and at a higher voltage and over 1MW. Cable charging for a rig is probably how the market will be because the charge points will end up multi-use but 1MW at say 600V is 1667A and having crawled under a switchroom and over cables running that level, im not convinced based on cable thickness and heat. Charging via overhead when moving is only good for full train electrification globally.

  9. wraithnot says:

    The 0.5 miles per kWh estimate for an electric class 8 truck seems about right. But why assume it has to have a 600 mile range? A bit of research does indicate that long haul truckers tend to go 500-700 miles per day, but at least some truckers recommend taking breaks every 2-3 hours. By law they can only spend 14 hours a day on duty and can only drive for 11 of those 14 hours. This leaves three hours a day where they can be on duty, but not driving. This would be a great opportunity to charge the battery. One scenario could be four stretches of driving for 2.75 hours each with one hour of downtime in between each stretch of driving. In this scenario, all you would need would be something like 175 miles of range and some way to replenish that 175 of range in an hour (rumored 350+ kW charger?). 175 miles of range would obviously be the bare minimum. But I bet an electric truck with a 250-300 mile range (500-600 kWh battery) with substantially lower fuel costs would convince at least some truckers its worth it to take more regularly spaced breaks.

    1. Terawatt says:

      I think you’re making good points. But Elon already said their semi would beat all ICE offerings not only on torque, but also on range.

      What they’re up to is anyone’s guess at this point!

      1. wraithnot says:

        He might have a sneaky definition of range such as range per day of legal driving. Or maybe they have a metal-air battery range extender that gets swapped at truck stops. Or some other trick up their sleeves. I guess we’ll find out in a couple of months.

        1. BenG says:

          Yeah, with 300 gallon tanks and 6 mpg current trucks have range over 1500 miles. I don’t see how Tesla is going to compete with that on a single charge.

        2. Pushmi-Pullyu says:

          “He might have a sneaky definition of range such as range per day of legal driving.”

          But truckers do actually, commonly, drive 600 miles per day in the 11 hour shift they’re legally allowed to drive, and trips of 700 or more miles per day are sometimes achieved by actual truckers in the real world.

          That’s why my “napkin math 1.0” assumes a range of 750 miles for the hypothetical BEV truck. That allows a slight safety margin for bad weather, loss of capacity as the pack ages, etc.

          If there is any “sneaky math” involved in Tesla’s BEV semi tractor, my guess is it will be battery swapping along the route. That would in theory give the BEV semi tractor unlimited range.

          1. Karl says:

            Big cab. 2 drivers. 1 sleeps, the other drives. 5 minute stops to change drivers. Seafood runs cross country.

    2. Brave Lil Toaster says:

      Because while “some” truckers *say* they only drive 2-3 hours at a time, you will find a boatload who do not, so they can actually make money instead of not making money.

      Those rules on how long you’re legally allowed to push yourself weren’t put there by the drivers, that’s for sure.

      1. Pinewold says:

        I would agree, time is money, rules were put in place because drivers neglected sleep to get in a few extra loads a week. This is why so many drivers are still independent, it lets them fudge the math.

  10. Mister G says:

    If battery packs are located under the trailers (batters), I can see a business model where batters are rented/leased to trucking companies. 600 mile batters, 500 mile batters that plug-in…GO TESLA GO destroy and replace the poison producing diesels on our highways.

    1. Pushmi-Pullyu says:

      It really mystifies me that so many people suggest putting the battery pack into the trailer.

      The advantage of semi trailers is that they are interchangeable and fairly cheap. Put an expensive (and very heavy!) battery pack into one, and it no longer fits either description.

      Battery packs for BEV semi trucks belong in the tractor, period. There are no advantages to putting it into the trailer, and multiple disadvantages to putting it there.

  11. Someone out there says:

    I think either overhead lines along the highways or swappable batteries are the way to go. Or hydrogen but I’m not allowed to say that am I? 🙂

    1. Brave Lil Toaster says:

      Oh, you’re certainly allowed to *say* it.

      And I’m certainly allowed to say it’s roughly the dumbest idea ever. Better to just burn CNG directly instead of pretending that you’re not. It’s way cheaper that way.

      1. ¯\_(ツ)_/¯ sven says:

        We’ll find out soon enough.

        Yesterday, Kenworth announced that they will build both a hydrogen fuel cell tractor and CNG series hybrid-electric tractor. Like the Toyota Project Portal hydrogen fuel cell tractor, these Kenworth T680 tractors will be used for drayage operations at California ports.

        “According to Kenworth, these two T680 tractors will be identical, with the exception of their power generation systems. Each truck will have an electric-only range of approximately 30 miles, and the onboard fuel – hydrogen or natural gas – will provide sufficient range for a full day in regional haul applications.”


        1. sveno says:

          Cost and refuelling aside, H2 fuel cell on highways seems easier for the driver but at a port with a lot of start-stop, low speed driving a battery-only electric 23h a day is very much doable right now. I can’t even imagine how the FCV could compete there in spreadsheets.

          1. Heisenberghtbacktotheroots says:

            You are right. Inside modern ports they already use battery electric autonomous haulage vehicles. That’s already established technology with the outlook of falling battery prices. If it made sense to go electric 10 years ago (that’s about the time they have been using that tech…) it should moreso make sense today.

            Hydrogen? No way!

            Hydrogen will only start to become interesting (for stationary applications (heating )) once we see massive oversupply of renewable energy. Then they can use the existing gas grid and storage capacity to store and distribute clean hydrogen.

            While we already see times with negative electricity prices we still have some way to go until it makes sense to produce h by electrolysis. Norway and Denmark might already start to think about it, but many countries first have to replace coal and gasoline. First things first.

          2. Someone out there says:

            All hydrogen vehicles have a battery for temporary storage and regen braking.

        2. Karl says:

          I think methanol will be the hydrogen feed of the future for fuel cells for trucks and cars (as range extender units). Because it can be distributed and pumped by all the existing filling stations. Methanol can already be synthesized from electric with CO2 from the atmosphere but not yet evonomically. The setup cost for hydrogen refueling stations cross county will be prohibitive. Tesla invested up front in charging stations and i cant see anyone looking like they are doing the same on such a scale for hydrogen. Methanol woul work and may be the Tesla surprise.

  12. Brave Lil Toaster says:

    What I want to know is the current engineering reason for using a diesel engine directly attached to the traction wheels through a 21 speed transmission, when diesel electric drivetrains have been in use for railroad locomotives since the 1940s. Or why nobody has bothered to build hybrid semis ever.

    My guess is that trucking is severely intolerant to even small increases in TCO, and nobody has actually improved that at all.

    1. ¯\_(ツ)_/¯ sven says:

      “Or why nobody has bothered to build hybrid semis ever.”

      See my comment above. Kenworth is building a CNG series-hybrid tractor, however it will have a 4-speed transmission.


    2. Karl says:

      Basically weight and power. Weight is good in a traction unit on rails but not so good if the trade off in efficiency does not work. Plus the elimination of any gears increases reliability at the expense of weight. Take a ride on an old geared passenger train and its just like being in a truck. That was before cheap rare earth magnets arrived which resulted in hybrids with only a single motor/generator due to legacy designs (typically designed to fit into an existig ice model) and new concept development costs when ice is on the way out does not make for a good market.

  13. TimE says:

    For all those recommending batteries on trailers – that’s not the answer.

    You might put a relatively small battery pack on a trailer for purposes of a refrigerated load, but NOT on a dry van trailer.

    Most trucking companies have between 2 and 5 times as many trailers as they do tractors. Trailers tend to sit idle as well – whether at a shipper, a consignee, or a drop yard. Current trailers are relatively inexpensive boxes – between $20 – $30k cost to purchase – you would NOT want a $100k trailer loaded with batteries sitting idle!

    Loaded equipment moving for a customer = $$$. Idle equipment sitting = loss of productivity = depreciation = lost $$$.

    1. Nix says:


      On top of that, lots of independents don’t even have a trailer. They just have a truck and show up to pull whatever trailer they are hired to pull.

    2. Pushmi-Pullyu says:

      Thank you! Your argument against putting battery packs into trailers is even better than the ones I’ve been using.

      1. Mister G says:

        Tim and you are correct batteries do not belong on trailers

    3. Mister G says:

      I agree and reverse my argument on placing batteries into trailers

  14. Nix says:

    For reference, that is about the same as 2 battery packs that BYD already delivers in their 60 foot buses.

    So when looking at the shock value of how big a battery this is, it like it is 2011, and talking like it would be a big deal for some company to come out with a car with double the size battery as the existing Leaf. As if 50 kW would be unimaginable!

    Then having Tesla show up with more than triple the battery size.

    This just isn’t that controversial.

  15. Daniel says:

    Doesn’t seem like there’s actually any truck drivers commenting on this article so allow to enlighten some of the commenters. I do not deny that new technologies are upon us in the trucking industry but the logistics for a vast majority of companies autonomous, and electric are in an uphill battle.

    Firstly, truckers routinely would be maxing out the 600 mile range in good conditions and depending on the companies speed limiters they would exceed this, in Canada this is even more of an issue as we are aloud 13 hours driving (2 more then the US permits). You are only required a 30 minute break after 8 hours of continuous driving and with the way regulations are currently (Electronic Logging) drivers are “under the gun” with a clock literally counting down in front of them they are in a hurry, again in Canada the rules on the books require no break throughout the day. Another issue I see being raised is costs. To a small company the prices that are currently being mentioned are well out of reach for any small company to experiment with profit margins are slim and taking on a truck payment double your current one without any idea what your getting for reliability or service is going to be a tough sell. The big companies although they may dabble in this type of venture it will have to prove myself as the likes of JB Hunt and Swift transport are buying 1000s of trucks at a time they are getting equipment for far less then the regular consumer. 100-120k. Also the 6.5mpg average is maybe an industry wide average but the main consumer base this type of truck will be geared towards the averages are closer to 8. And current trucks coming out this generation exceed that when geared correctly and speed governed (big fleets do this).

    Again I’m not naive to the fact of these technologies coming to the industry but I think those pushing it are a little naive in how “easy” and “quick” they are going to be able to implement them. All of these numbers I’m sure exclude colder climates and the use of bunk heaters in the winter and AC overnight while the driver sleeps.

    1. DangerHV says:

      I appreciate your input, Daniel. As you see, most or all these comments are from people (including myself) who are not in the truck buying business. You have provided a good reality check, at least to me.

    2. Priusmaniac says:

      In Europe it is 8 hour maximum and a pause each 2 h. We have rest parking places all along the main freeways with toilets and often with restaurant and hotels. In case the driver sleeps in his cabin it is easy to install an outlet to supply power to the truck. No need to use diesel for heat or power. Of course Canada may be quite a different world altogether where you need a gun in the cabin just in case a bear comes along and an old CB system in case there is no mobile phone coverage. But then again, they could be smarter and use the Australian road train system where one of the many trailers would simply be a big huge battery. Of course if there is no will, the solution to anything becomes impossible as usual.

    3. Pushmi-Pullyu says:

      @ Daniel:

      Thank you very much for your input! My “napkin math 1.0” was just armchair engineering, so it would be great to have input from an actual trucker on the subject.

      I agree that 600 miles seems too short a range; I used 750 miles for my “napkin math 1.0”, but even that may be a bit too short for at least part of the market.

      As to the issue of price for a hypothetical BEV semi tractor, I’m not really that concerned about the up-front price. Yes, a very expensive BEV semi will certainly put it out of the price range of independent truckers and very small fleets, but what we really need is a “toe in the door” into the industry. Once that is established, we can confidently expect the market to grow, as EV tech improves. So if the long-term cost vs. as diesel semi tractor made sense, we could expect large fleets to pay even a very high price for a BEV semi tractor.

      Unfortunately, I just don’t see that a BEV semi tractor used for long-distance hauling can be economically competitive in any circumstances, with current battery prices and current EV tech. It’s not just the upfront cost of the BEV semi tractor, it’s also the cost of multiple replacement battery packs over years that makes the whole thing impractical.

      Perhaps Tesla can pull off a miracle, but my prediction is that they’ll just show a concept vehicle, too expensive to produce and sell at a profit.

      1. Mister G says:

        Remember Tesla currently delivers their cars using diesel semis, maybe tesla semis will be put in service for tesla deliveries and it would be a huge write off on corporate taxes.

        1. Karl says:

          Use an interconnect to all the model s/3 batteries to supplement the main battery.
          Overhead wires to charge when stopped as the current transfer can be higher and automated.
          Cycle life of the newer chemistry in model3 can be adjusted to get over 3000 cycles. Power pack type chemistry was lab tested to 9000 cycles.

  16. Jonathan Brusco says:

    1.21 Gigawatts! Enough to power the flex capacitor.

  17. Priusmaniac says:

    The range needed to go from the Gigafactory to the Fremont plant is 268 miles, so rounded up that’s 300 miles of range. That is not 600 miles, so I thing the 600 miles value is completely off scale. This makes a huge difference in battery since 600 KWh is much less than 1200 KWh. Perhaps we will have 400 miles but that would still be “only” 800 KWh. So 1200 KWh is way off chart.

    1. Pushmi-Pullyu says:

      Earlier discussions of this issue seemed to have reached a consensus that Tesla was aiming at the “short haul” range of 200-250 miles, which of course would have allowed a much smaller battery pack.

      But then Elon tweeted that the Tesla semi tractor would beat a diesel tractor in torque and range. So that’s why you’re seeing discussion here of ranges of 600 or more miles. 600 miles is actually a shorter range than the market needs for long-haul trucking.

  18. Lowgous says:

    Don’t need no stinking semis; Haul freight via underground hyper loop.

    1. John says:

      I’ve been saying this for a long time.

      Perhaps Elon is just trolling everyone and “The Boring Company” is the biggest piece of the E-Semi puzzle.

      If you can hyperloop standard 40 ft. containers across the country quickly and cheaply, then all you need is a short range semi to get it the last few hundred miles.

      Seems to make sense to me!

  19. Erling Klausen says:

    I’m wondering how much is a Semi truck in maintenance over its lifetime there will have do be some significant costs there right?

    also this significant reduction in maintenance costs for an electric truck needs to be more expanded on.

    1. monkeymaster says:

      when first computer was invented, it took up 680 square feet of space. people said computer won’t work, it’s too big and it won’t fit their home. guess what

  20. Pushmi-Pullyu says:

    Thank you, George Bower!

    Yeah, my “napkin math 1.0” certainly could use some improvement with more accurate figures. I took the actual fuel consumption of a loaded diesel semi tractor-trailer rig at highway speed, and just estimated the amount of energy used by an electric powertrain. Others can probably improve on the accuracy of that.

    I seriously question the 50% reduction in aerodynamic drag in your estimate, but I don’t think we have any fundamental disagreement. My “napkin math 1.0”, as you call it, didn’t assume extreme reduction in coefficient of drag, and I can easily believe Tesla can improve on my estimate of 2.4 kWh/mile by 15-20%. If we take the more optimistic assumption, that would be 1.92 kWh/mile. For a 600 mile trip (your figure), that comes to 1152 kWh, which is pretty close to your estimate of 1200 kWh.

    My “napkin math” assumed a longer range of 750 miles, which is about the longest trip that any trucker in the USA drives in a single day, with perhaps a slight safety margin. 750 miles at 1.92 kWh/mile comes to 1440 kWh, which would be somewhat less than my “napkin math 1.0” estimate of 1800 kWh.

    At any rate, we certainly agree that the battery pack would need to be bigger than 1000 kWh, or 1 MWh, if the semi truck is going to go that distance without recharging or battery swapping along the way.

    P.S. — It would have been nice if you had attributed “napkin math 1.0” to Pushmi-Pullyu, and not “PMPU. Is it possible to get a correction?

  21. georgeS says:

    It is an incredibly difficult problem statement as you point out. You have stated before that you think it is just a concept truck and you could be right. Maybe Elon has run the numbers and put the problem on the shelf until we get a big battery break thru.

    However, the fact that he has said that the truck will use model 3 motors may be signalling that he is totally serious about it and as “trucking expert” said Elon has a trick up his sleeve.

    So just for a moment let’s pretend that he IS totally serious about this project and he does have a way to solve it.

    Use your imagination.

    What could this “trick” be.

  22. Bob Mosby says:

    You made a mistake when you calculated the motor/inverter efficiency losses. You show a loss of 14BTUs on a total of 34BTUs, that’s only 60% efficient.

    To do the calculation correctly, you need to total the load first and then add 15%. That gives a loss of 3BTU and a total battery size of 800kWh for a 600 mile range.

  23. Alexandre says:

    How do the guys from Nikola One get the weight low enough so that it doesn’t reduce the overall capacity?
    They have a largish battery pack and a fuel cell and H2 tanks onboard!

  24. Ron says:

    I would think that Tesla might target the local delivery networks first, with city driving and lower overall mileage.
    There, smaller battery packs would be needed, and the gain in regen for stop and go driving would be greater. As battery energy density improves, the ranges can become longer with similar weights. Meanwhile, the pollution and noise pollution gains in cities would be beneficial.

  25. John T says:

    I think an electric truck operation cost will be much lower than a diesel truck with or without a driver. I do some estimated cost calculation below:

    I also found a p100d battery pack weight is 1200lbs so for 1200kwh battery weight is only 14,400lbs well under 37,000lbs empty semi truck nowadays.

    Diesel truck purchase cost 150k
    Electric truck perchase cost 300k ( some over estimated number here)

    Operation costs:
    Diesel truck:
    – 2 drivers for 16hrs per day 120k/year with 50wks 5 days a week which is $480/day
    – 60 miles per hour which is 480 miles per day at 6.5MPG which has fuel cost of $185/day
    – Total operation cost is $665/day

    Electric truck:
    – Total operation cost at 19.1 MPGe and 480 miles per day is only $63/day
    This is 10x cheaper and I am assuming 8 hrs of rest to charge per day.

    I use human drivers who only work 50 weeks a year. The autonomous truck can work 24/7 and 52 wks out of a year.
    Maintain cost of electric truck also much lower than diesel truck that I did not take into account.

    Even with twice the purchase cost electric truck it is cheaper overall to operate.

    If we are talking about with drivers on both truck than the cost per day is still $120 cheaper which can be recover within 1250 days of operation or only 5 years later.

  26. charlie says:

    think of the following considering only for the cells and assume the trucks are comfigured to be short mid or lange range:

    truck efficiency is 0.5 to 0.75 miles per kWh
    cell weight is > 250 watts-hours/kg
    cell cost is < $150/kWh

    short range 200-300 miles 400 kWh
    buy the cells for $60k
    cell weight 1600 kg or 3520 lbs

    mid range 400-600 miles 800 kWh
    buy the first $60k
    lease next $60k of cells
    cell weight 3200 kg or 7040 lbs

    long range 600-900 miles 1200 kWh
    buy first $60k
    lease next $120k of cells
    cell weight 4800 kg or 10560 lbs

    for long range that leaves 10000 lbs or so and at least $100000 or so to package the batteries and create the truck.

    for jump starting truck stop charging for long range, a genset like the cat 3516e that takes up the space of one trailer could charge 10 trucks at night at the same time in 6 hrs

  27. Tom Andersen says:

    Math is obvious. It won’t work.

    Even if it did work by the math, actually trusting your cargo and livelihood to a new car manufacturer with new technology takes over 20 years.

    So with better batteries (say 2 – 3x improvment) and lots of trust and development over two decades, electric semis might get some small fraction of the business. Will they be Teslas? Seems doubtful.