When Jack Rickard pulled the battery out of a wrecked extended range Model 3 in a previous post, it weighed in at 478 kg. We are talking here of a 75 kilowatt-hour (kWh) battery. If you have a high tolerance for ‘slow video’ then you can see Jack, together with Bill from Ace Hardware, Linden, Tennessee, extract the battery from a Model 3 (I rather like Jack’s vlogging style, but then perhaps I should get out more).
Let us put this battery into some kind of context. If we divide the energy storage of the 75 kWh Tesla battery Jack Rickard extracted by its weight of 478 kg we get 0.157 kWh per 1 kg of weight. Let’s change our unit from kilowatts to watts: so we get 157 watt hours (Wh) per kg. Note we are talking about the entire battery here, not the battery cells.
The measure of the amount of energy per unit of weight is referred to as the specific energy or gravimetric energy density. If you are building EVs you want to have as high as possible specific energy for your battery. Each technological advance then allows you to either store the same amount of energy for less weight, or store more energy for the same amount of weight.
When you compare EVs with internal combustion engines (ICEs), the EV basically wins hands down with respect to the delivery of power. So boosting the specific energy of the battery is about having more energy in your battery (rather than power). This, in turn, gives your car more mileage range. Indeed, a critical goal in battery development is to boost specific energy to such an extent that mileage range becomes a non-issue. Nonetheless, even if your battery can deliver all the possible miles that an owner could wish to drive in one sitting, an incentive still exists to boost specific energy even further because that would allow you to continually shrink the battery.
Keep in mind that the Model 3 Tesla battery Jack extracted weighs 28% of the overall vehicle weight. So let’s do a thought experiment and imagine that Tesla doubles the specific energy of the battery to 300 watt hours per kg (wh/kg).
The current Tesla Model 3 has a range of 310 miles according to the US government’s Environment Protection Agency (EPA); pretty good, but I should think the consumer is greedy for more. Let’s decide to increase the mileage range by 50% to around 450 miles between charges. With our new 100% specific energy improved battery we could deliver 450 miles plus shrink the battery by 25% to about 325 kg. We are, in effect, lightening the car by over 100 kg. And, if we are lightening the car by 100 kg, we are making it more efficient.
The extended-range Model 3 weighs in at 1,730 kg; if we drop 100 kg in weight, that translates into a weight saving of around 6%. And given a chunk of the battery’s energy is used up lugging the battery itself around, every time we shrink the battery we release energy that can be used for moving the rest of the vehicle and its occupants. This, in turn, allows us to shrink the battery further. In other words, we have a nice little positive feedback loop emerging here.
Specific energy is not energy density. The term energy density, or volumetric energy density, refers to the energy stored per unit of volume, or watts per litre (wh/l). In an ideal world, I would have loved it if Jack had also taken the dimensions of the extracted Tesla battery as well. Then we could have worked out a real-time energy density number for the Model 3 battery. In the chart below, however, you can see that a linear relationship exists between specific energy (termed here gravimetric energy density) and volumetric energy density. Therefore, as we achieve specific energy enhancements, we generally get energy density improvements as well.
Put another way, as you lighten the battery you will also reduce its size. And every time you do that, the size of the total drivetrain, including energy source, gets more competitive versus the ICE drivetrain plus fuel tank. Note that the EV has already won in the configuration competition with an ICE vehicle. You have far more latitude in arranging your drivetrain components with an EV since you are just connecting them up with wires. No crankshaft or gear box required.
Let’s talk a bit more about range. We’ve established that the extended range Tesla Model 3 can drive around 310 miles on a fully charged battery according to US government statistics. So we get 3.85 miles per kWh. Speed limits in the USA vary by state, but the highest speed limit is found in states like Texas and comes in at 75 mph. So our 310 mile range for the Tesla translates into roughly 4 hours 10 minutes of driving time in the high-speed-limit states, and that is assuming the entire journey is at the maximum speed limit.
Given the average motorist doesn’t wear adult diapers while driving, he or she will periodically need to get out of the car to urinate or defecate. Moreover, I propose that the need for a wee is the limiting factor here, so the act of urination is our boundary constraint (apologies to irritable bowel sufferers). According to WebMD, the average adult urinates between once every two to four hours. So I will take an ‘iron bladder’ as my example Tesla driver. Bottom line, the car will not run out of energy before the girl or boy has gotta go.
Nonetheless, we have to think about how long it takes to charge the car up again. And then we need to plug that number into an equation that also incorporates the time it takes to exit the car, hook it up to a recharging source, walk over to the toilet facility, urinate, walk back to the car, unhook the car from the charger and then drive off.
So does time to wee equate to time to charge? Curiously (at least for me), the “Law of Urination” states that mammals in general take 21 seconds to pee. So the actual act of urination is not the limiting factor here, rather accessing a place to urinate is the issue. For argument’s sake, I will assume that the toilet break is limited to 20 minutes since we need to walk to and from a toilet cubicle.
Next we need to know how long it takes to charge the Tesla. There are a lot of factors that come into play here including the starting charge of the battery, the desired ending charge of the battery and the quality of the charger. Given we are interested in the dynamics of a long road trip here, the driver is going to be using some kind of supercharger network. To get a sense of the charging experience at a Tesla Supercharger, check out the video here:
In the video, you can see a top charging rate of 100 kW, which translates into 375 miles added per hour. The theoretical top charging rate at a Tesla Supercharger is actually 115 kW, but as the battery gets close to being fully charged, the rate of charging drops off.
What can we deduce from all this? Well, if a long distance driver wants to do two back-to-back journeys of around 300 miles each at the top speed limit allowed in the USA and only needs one toilet break of 20 minutes, then the Model 3 can’t deliver that kind of range performance (but an ICE vehicle can). You currently need a good hour of charging in that scenario for your Model 3, even assuming access to the best possible charging rate at a Supercharger.
At this stage, you may point out that the vast majority of car journeys don’t include driving 600 miles straight in two back-to-back sessions with a 20 minute break in between. But that is not really the point. In an earlier post, I established the conditions needed for the auto market to tip between being ICE dominated and EV dominated. The condition was that an EV needed to match or exceed an ICE vehicle in every car attribute. So if an ICE vehicle is superior for long distance marathon driving, EV penetration will be slower.
Nonetheless, it is only the year 2018, and Tony Seba’s 95% EV penetration-rate target by 2030 is still 12 years away. Further, I think the EV is not that far away from matching an ICE for long-distance driving already for two reasons.
First, as the battery capacity gets bigger, the battery will still have a lot of juice when the driver reaches the first toilet stop. With a 50% bigger battery and a 450 mile initial range (achieved through charging overnight), 150 miles will be left in the battery at the first stop. So for leg number two of the journey, only half the charge time is required since we are only topping it up and not starting from zero. (Yes, we could talk about three back-to-back stages amounting to 900 miles, but the driver will eventually have to eat, and, as we get beyond the 99.9% percentile of typical journeys taken, I think we can view these as true outliers.)
Second, as each Supercharger generation is rolled out, the rate of charging will go up. Tesla has already flagged that the V3 Supercharger will arrive this summer and the rate of charge is likely to be between 200 kW and 250 kW, so a roadmap exists toward halving the best current charging time.
Beyond Tesla, two open standards exist that allow shared usage by vehicles produced by different auto makers. These are the CCS standard (backed by the CharIn consortium) and the CHAdeMO standard. CCS, promoted by the German auto makers amongst others, is working toward a charging rate of 350 kW as well as inductive wireless charging stations. The CHAdeMO consortium, which is principally composed of Japanese automakers and electric power companies, is aiming for a rate of 400 kW in the coming years. A good primer on fast-charging protocols can be founder here.
Both CCS and CHAdeMO are having to play catch-up with Tesla’s Supercharger network, which has just reached 10,000 sites worldwide. Undoubtedly, Tesla’s strategy of “build them and they will come” appears to be far superior to the opposition’s strategy of “we will build them when they ask for them”. You can check out Tesla’s global Supercharger network here.
Summing up, while I think access to fast charging facilities will become a non-issue in a few years time, the need for batteries to get better with respect to specific energy and density is still a pressing need if EVs are going to unseat the current dominance of ICE vehicles. Further, to get the required improvement, do we need a John Goodenough-style leap in battery technology or will a Tesla-style incremental improvement in existing technology suffice? That will be the subject of my next post.
For those of you coming to this series of posts midway, here is a link to the beginning of the series.