Saturday, February 6, 2016

Electric Long Haul Trucks

I've delved before into what it would take to convert the world's car fleet to battery power with our current battery technology (spoiler: lots and lots of cobalt).

What about trucks?

Semi trucks (18 wheelers, HGVs, tractor-trailers - whatever your preferred local term for the above pictured vehicle is) spend their lives hauling heavy loads around the highways.  They burn an awful lot of diesel doing it, so a commonly asked question is, "Can we make them electric?"

Let's find out!  Keep reading if you're interested in what it would take to build an electric semi!

Truck Fuel Economy & Energy Use

Looking around the internet, Peterbilt has designed a tech demonstrator that manages just over 10mpg (US gallons) with a 65,000 lb total weight.  Not bad at all, given most of the semis running around will manage 6-7mpg with the same payload!  To be fair, this is a tech demonstrator, with all the aero options, and a lot of special tricks (including exhaust heat recovery), but it shows what's possible.  I'll use 10mpg (US) as my "upper end of reasonably achievable diesel fuel economy" value for this post.

To convert from diesel use to electric, the thermal efficiency of the engine is needed.  A large truck diesel will be somewhere in the 35% thermally efficient range - so about 35% of the thermal energy ends up as useful mechanical energy.  The engine is going to use some amount of this to run it's own loads (coolant pumps, fans, etc), but it's a good enough number that I'll use it.  A few percent each way only changes the final values by a few percent.

A gallon of diesel, on the other hand, is easy to work with.  It contains 139,000 BTU, which (more usefully for this calculation) is 40.7 kWh of energy.  A Model S battery pack contains the equivalent energy of two gallons of diesel, and a Leaf pack is barely over half a gallon.  If you take the ~40mpg a typical diesel car gets, multiply by 3 (since a diesel is only 1/3rd thermally efficient), and figure for two gallons, you get 240 miles - or about what a Model S gets!  I like sanity checking my figures as I go, and it's nice to see things like this work out - it means I'm not an order of magnitude or two off.

So: back to our truck, which is rolling along I-80 through Iowa at 65k lbs, 65mph, and 10mpg (yes, I picked a very flat chunk of highway).

10mpg works out to 4.07 kWh of thermal energy per mile.  At 35% thermal efficiency, this gives a mechanical energy used of 1.43 kWh per mile at highway speeds, and a sustained mechanical power output of about 95kW (~125hp).  Not bad at all for 65k lbs and a slightly aerodynamically shaped box!

Truck driving duty regulations are fairly easy to find.  What it works out to is 11 hours of driving a day for a truck driver (assuming they're obeying the law).  Or, ideally, about 700 miles a day worth of driving.  The 14 hour limit means you can't do this in two shifts across the day, so forget about 5 hours, a nap, then 6 hours.  It's not legal, and this is important.

Battery Pack Size

To do 700 miles at 1.43kWh/mi works out almost perfectly to 1 MWh of energy.

What does a 1MWh pack look like?

The 85kWh Model S pack weighs around 1300 lbs.  Using that as a reference, a 1MWh pack would weigh around 15k lbs.  That's a lot.  Like, most of the weight of a current tractor (they're about 20k lbs).

The weight will come down in the future, and it's probably closer to 10k lbs of pack if you build it big, but it's still going to be damned heavy to do it safely, especially for a pack that can last the million or so miles a tractor typically does.  And even at Tesla's 2020 rates of $100/kWh, that's still $100k of battery cells alone - plus the pack assembly, the motors, and the rest of the tractor.

The concept involves replacing the engine and transmission with a battery pack and electric motor, so there's quite a bit of iron to get rid of, but 10k lbs of battery is still a lot.

Night Time Charging

A huge battery pack is quite useless unless it's charged.

To recharge a 1MWh battery pack in the ~12 hours of driver downtime, the charger has to sustain an average of right around 85kW.  Or, you could do it with current Tesla supercharger rates (120kW) in a bit over 8 hours - if you could fit in a stall.  You might annoy a few people in the process as well.

What if you're team driving?  Then, you've got 11 hours driving and 1 hour downtime, endlessly.  You'd need to charge at over a megawatt to fill this pack up during downtime - and that's hard.  For an 800v pack, that's 1250A, sustained.  Or 2500A on a more standard 400v pack.

The other side of the challenge is powering the charging stations (which will probably be the rest areas and truck stops scattered around the country).  A small rest area with 40 truck stalls would need about 3.5MW of charging capacity, and that doesn't count all the trucks that pull along the exit and entry ramps (which would also need to be charged).

At the other end of scale, you have places like the Iowa 80 Truck Stop in Walcott, IA.  It's got 900 truck parking spots.  If you want to outfit all of those with a 100kW charger, you're looking at 90MW worth of power capacity - and if you throw in a few quick chargers for those team drivers, an even 100MW of power seems reasonable.

That's a LOT of power.  It's certainly doable, but running the power out to the truck stops and rest areas (which are often in the middle of nowhere) is going to be a massive infrastructure project.

Right now, we're in the realm of "technically possible, but unlikely in the near term."  And if trucks can't charge, trucks don't roll.

Let's look at some other options.

Series Hybrid Trucks

A series hybrid (of the "plug in hybrid" variety) solution is radically more interesting for long haul trucking - and more importantly, it's an option that is immediately useful, even while charging stations are being built out.

There are at least two series hybrid systems on the market right now for trucks.

The first is Via Motors.  They heavily modify a Chevy pickup to include a 23kWh battery pack, a 115kW (150hp) generator, and a 190kW (255hp) electric motor.  This is a nice work truck, especially if you need site power.  It has up to 14.4kW of site power output, and will only run the generator as needed.

The second company is WrightSpeed.  They have several platforms, one of which is a heavy chassis designed for garbage trucks and similar heavy vehicles.  The WrightSpeed platforms are pure series hybrids as well, with a variety of generator options including a microturbine designed for dirty methane (say, coming from a landfill).  They're plenty powerful for over the road use (up to 500hp), and can recover huge amounts of energy with regenerative braking (up to 1000hp worth of energy recovery).

Advantages of a Series Hybrid Truck

A series hybrid truck has several very significant advantages over both the current diesel trucks used for hauling, and over a purely electric truck.

I envision a truck with 100-300kWh of battery capacity, and a totally separate generator module that runs on diesel (though CNG could be a possibility as well).  This generator doesn't have to generate useful rotary motion either - only electricity.  So currently exotic engine designs like a free piston linear alternator can be used.  If the generator design is sufficiently modular, the generator can even be replaced with a different version during the truck's life, if the savings are enough to justify it.


A major advantage of an electric drivetrain, especially in cities, is that it's quiet.  A series hybrid truck with a large battery pack can move nearly silently during the start and end of it's trip when it's in a city.  Even a 100kWh pack offers over 60 miles of pure electric range, which is plenty to get out onto the highway and to get into town at the end of a trip.

Additionally, with a large battery pack, the truck won't need an APU.


The key problem with a purely electric over the road truck relates to energy.  A pack large enough for a day's driving is a very large pack, and the infrastructure needed to charge it is significant.

With a series hybrid approach, both of these problems become less significant.  A more reasonably sized battery pack can be installed, keeping the cost and weight down.  If you can charge off shift, you do (it'll be cheaper than diesel), and if not, you run diesel or CNG.

Regenerative Braking

Another major advantage of having a battery pack is that it provides a place to store energy from regenerative braking.  Trucks regularly use engine braking to slow down (the machine gun staccato of a "Jake brake" is loud enough that it's banned in many towns), and both coming down from speed on the flats, and coming down a grade, there's a lot of energy that can be recovered.  Instead of turning it into noise and brake wear, that energy can be recovered for later use.


This is no different from what one would find with a pure electric truck, but a series hybrid truck will require no shifting, even when heavy.  It'll simply pull from low speed up to cruise, and probably be a good bit quicker to speed as well due to the continuous power delivery.


An electric truck operating in pure electric mode doesn't emit NOx, particulate matter, unburned, hydrocarbons, or anything else that upsets air quality boards in cities and states.

California, as of 2013, requires trucks entering their state to have some sort of aerodynamic improvements, and this has driven a major surge in trailers with aero kits.  If some large cities were to require pollution-free systems for final delivery, these trucks would work, but would still be able to get across the country on a schedule.

Microturbines are apparently also capable of burning cleanly with minimal exhaust treatment - so the weight and complexity savings make something like that worth it (this is one of WrightSpeed's options).

Proposed Design

First things first: I'm not experienced in the design of trucks.  I mostly play in much smaller systems.  This is my "back of a napkin" level design.

If I were to design an over the road truck, I'd build a series hybrid system - this makes it usable in 2016, instead of relying on a charging infrastructure that has yet to be built.

I'd put around a 100kWh battery pack in it - possibly more, if it was likely to be working in a mountainous region.  I'd focus on power density over energy density, so something like lithium titanate would be a good option.  The pack needs to be able to provide a lot of power, absorb a lot of power, and it's in a heavy tractor anyway, so weight matters less.

The generator would be a diesel plant of around 250hp.  This is smaller than most truck engines, but provides an excess of power over that needed to cruise, even with wind.  This engine would be optimized for running at a fixed speed, with the intake, valve timing, and exhaust set up for maximum efficiency at the operating RPM.  For size and weight reasons, I'd use a turbodiesel, perhaps with post-turbo power recovery turbines to extract all the energy possible from the diesel.

The drivetrain would be a standard electric motor and reduction gear setup (probably AC induction, though could be permanent magnet based if this makes more sense), driving both drive axles.  It might make sense to use two separate motors - one for the front drive axle, one for the rear.  This allows for lower power motors, and offers redundancy if one motor fails.

Finally, a key element would be an integrated navigation and charge control system.  There's no point in charging to 100% if the next 50 miles will be downhill - if there's nowhere to put the recovered energy, the friction brakes will be used.  Terrain data is easily available, so with knowledge of the route ahead, an intelligent system could control both charging and generator use to optimize fuel consumption.  Ideally, all city driving would be on battery for noise reasons, the truck would never need to use the friction brakes (because there's sufficient battery capacity to sink the energy), and the truck would arrive at an overnight stop with charging with a nearly flat battery (so it can make the most of the charging available).

This design should allow for a significant energy savings over today's trucks, while still keeping the ability to run cross country on diesel.  And, it doesn't require any breakthroughs in technology - you could literally go out and build one today, if you wanted.

Thoughts?  Comments?  Improvements you can think of?  Let me know in the comments!


  1. Excellent article! The series hybrid design makes so much sense that I can't believe it isn't being done today. 100kWh sounds about right. Going over mountain passes you could deplete the battery to help the weak (250hp) diesel engine and then recover most (80% or so) of the energy going down the other side of the mountain.

  2. A lithium titanate 100 kWh pack would be very heavy - figure around 1500 kg. Still may be okay if the tractor weighs 10000 kg, but will have to fit it somewhere.

    Model S (2500 kg) uses somewhere around 0.5 kWh per 100 m elevation change. A 2000 m climb in a 30000 kg trailer would require somewhere around 120 kWh of "additional" energy beyond steady-state travel. A constant 6% grade would run for 33 km.

    At 70 km/h - similar to what diesel trucks run today on steep grades - the truck would need to supply about 35 kWh for steady-state driving (70 kW) and 120 kWh for elevation gain (240 kW). Assuming 80% pack usage and 200 kW generator, a 70 kWh pack would be sufficient, assuming the navigation was aware and charged it to full prior to the ascent.

    On the descent - assuming constant 6% grade - the truck would likely have to use a blend of friction and regenerative braking (absorb ~170 kW to maintain 70 km/h speed = 85 kWh total, ~290 kW to maintain 100 km/h speed = 90 kWh total).

    Since uphill depleted the pack by ~55 kWh, downhill recovery at 70 km/h would completely fill the pack and require the use of friction brakes.

    Worth noting this is an outlier - most interstate elevation gains will be substantially shorter or less steep (6% is max for new interstate construction) which would allow a 70 kWh pack to deliver almost all of the benefits of a 100 kWh pack, with reduced weight and slightly reduced cost ($6000 less expensive at $200/kWh, which isn't a lot).

    1. One thing to think about is that a smaller pack can't supply or accept power as quickly as a larger battery. It may be necessary to have a larger battery just for the charge/discharge rate.

    2. Protomech - thank you, that's an awesome bit of math I'd not done for the post! I sort of handwaved at it, but it sounds like a 100-200kWh pack would handle almost anything in terms of grade power, and a smaller pack would be useful for the vast majority of needs.

      Titan - the power handling is one of the reasons I mentioned lithium titanate by name. It's capable of both providing a lot of power and sinking a lot of power - most of the modern cells don't think twice about discharging and charging all day long at a 10C rate, and the battery can provide a lot more power as needed. A 10C rate on a 100kWh battery pack is 1MW worth of charge or discharge (about 1350hp), which should be plenty for a loaded semi.

      The cycle life of lithium titanate cells is also excellent - a pack should last the lifetime of the truck. The downsides are mostly relating to size and weight (and to a smaller amount cost), but those matter less in a truck that's expected to run a million miles and is already going to be heavy for traction purposes.

      I agree that you'd need a larger pack, especially for soaking regen energy, if you went with a more common lithium chemistry. But LTO is just about perfect for this application.

  3. Instead of a series hybrid with its power conversion losses I would suggest using a direct drive hybrid system similar to the one in the Koenigsegg Regera, it seems like it would be a perfect fit for long haul trucks.

  4. A massive battery pack can be split up for charging
    So a battery only semi can have up to 10 chargers giving it a time equivalent to a tesla
    On the hybrid
    A steady speed motor can be tuned to extreme efficiency. So is a good choice
    Also a battery bank can take over the refer on the trailer
    Good math.

  5. Running off of a battery pack constantly isn't necessarily the best solution.

    The Siemens system I linked to uses overhead wires to provide a way to turn off the generator in a series hybrid.

  6. What do you think of hydraulic hybrids? Eatons had a prototype truck a few years ago, and the EPA had a development program.

    1. They're certainly interesting, but I'd be very surprised at this point if they outperform electric hybrids. The hydraulic system had the advantage of being purely mechanical, which is useful if you don't have control electronics that can handle large amounts of power. However, in 2016, we've got that worked out, and we have battery chemistries that can take a lot of charge and discharge current.

  7. Hi Russell, I have had the exact same basic design rolling around in my head for a long time (including the destination and terrain aware GPS system a critical component) and have come to virtually the same conclusions as you. One thing that could be added to the system for overall power unity are a bank of super capacitors. Super capacitors are light weight and have the ability to absorb charge very quickly, much faster than a battery can. So adding this element can give power somewhere to go when the vehicle needs to slow quickly. Super capacitors also have the ability to return supply big bursts of power, ideal for getting back up to speed after a full stop.

    As for the main power unit there are a lot of new engine technologies that are coming to market some reaching reported thermal efficiency of up to 52%

    Here is another promising engine tech, the opposed piston engine and its light modular design

    And though some will say that this guy and his power amplification technology is a nutter, I think it may be worth looking into as there is a lot of suppressed tech out there. If this tech is real and can be commercialized it solves all power issues out there. Thus the reason to suppress this type of tech.

    Then there are all the Stanley Meyer patents with relation to resonant frequency water fracture technology and the high voltage low amperage production of Hydrogen and Oxygen.

    If these exotic techs are too much to wrap your head around then one can look at waste-heat recovery systems that can be teamed with any type of ICE or turbine engine system to recover some of the lost heat to steam and a secondary power unit. This link does not show a vehicle optimized system but gives an idea of what is possible with development.

    These are just a few ideas that I have for long haul trucks to improve efficiency. Feel free to contact me. I would love to be become more actively involved in this field.


  8. Everything seems to be right with series hybrid trucks. Their consumption of fossil fuels is low and, unlike pure electric vehicles, does not need extensive infrastructural development to become viable. With all these advantages I wonder why their sales are still very low. Currently only 17% of HEV sales are series HEV's. With more marketing I believe they will soon become dominant on our roads.

    Cody Thornton @ Mobilevalley Ltd

    1. Long haul trucks are very long lived, and require levels of proven reliability that are hard to achieve without serious miles on the road.

      I expect we'll see more over time, but the battery technology has only, in the past 5 years or so, hit a point where this is even feasible, and trucks are on the road far, far longer than just 5 years.

  9. You article couldn't have come at a more perfect time Russel! :) I'm doing some research on electric truck for my Writers Hub work. When it comes to trucks, going electric will certainly save money on fuel and maintenance.

    1. I'm fairly certain this is just Writer's Hub spam, but I'll leave it... you're at least on topic.

  10. Would be curious of your thoughts on the Tesla truck :-)

    1. I think they're going to have to prove they can build a reliable, maintainable vehicle for it to have much of an impact. Enough companies have ordered small test fleets that it will be really obvious in a hurry how they hold up and how they're maintained.

      Fleets won't go for the "Well, you have to take it to the Service Center for everything..." style maintenance Tesla has gone with so far, but I don't think that's a huge hurdle (just a mental jump for Tesla).

      But, really, it's going to come down to how reliable the whole vehicle is. I'm not particularly concerned about the drivetrain - Tesla has done a very, very good job with the engineering on the battery pack, power electronics, and motors (the bulk of "motor replacements" were almost always because it was quicker to just swap the unit out for a minor issue than to pull it apart and fix the lubrication issue). It's rare to see an issue with those. The rest of the cars, on the other hand, are a hot mess in terms of things actually working and lasting - and that's with fairly careful owners proud of their $100k toys. A daily use fleet truck gets treated an awful lot rougher, in much worse conditions, and I'm somewhat skeptical that Tesla actually understands that well enough to build a machine for it.

      I will be interested to see what the actual power consumption numbers are, compared to my estimates. One thing I didn't consider in my estimates here is cooling drag - the diesel demonstrator still requires a lot of cooling airflow, and I don't know how that compares with the total drag of the combo.

      Another area that will be interesting is how the Tesla semis handle gusting crosswinds in terms of range - that really, really increases the drag on most vehicles.

  11. I think you have taken the wrong approach here. you need to have the batteries in the trailer with Solar panels on the top of the trailer. This way the trailer can be parked and a charged trailer can be attached and the off the truck goes again.
    The trailers can be charged at the depots or via the solar on the roof (all be it at a much slower pace). If you can get 600 miles out of a charge on the trailer you would be able to have regional depots where fright is handed off to charge and the truck is kept in almost constant motion.

    1. Solar panels on a trailer are mostly a waste of weight, and (worse for stability) weight up high. And they are aimed wrong for everything but peak summer, and even then aren't exactly right unless you're very far south.

      Solar panels on a vehicle are, almost always, the wrong answer. Put them on the ground where they're able to be the lazy lizards they like to be, soaking in the sun.


Comments on older posts are moderated due to spam issues. If you don't see your comment immediately, and you weren't just spamming me with some irrelevant comment and a link to whatever site you're trying to SEO, your comment should show up relatively soon. If you're trying to use my blog for your SEO purposes, your comments will never show up, so don't waste your time.