The Energy Collective Group

This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.

9,961 Members

Post

Can Battery Electric Vehicles Disrupt the Internal Combustion Engine? Part 1: "No"

Highlights

  • Battery electric vehicles (BEVs) will do well to take more than 10% of global light duty vehicle market share without perpetual subsidies.
  • BEVs with the large battery pack needed for broad consumer acceptance will remain more expensive than internal combustion engine (ICE) cars.
  • This price premium is unlikely to be accepted by the mass market even under optimistic future BEV integration scenarios.
  • Currently emerging data is starting to support this argument.

Introduction

Electric drive has numerous advantages over the internal combustion engine such as high efficiency over a wide range of power output, regenerative braking and no tailpipe emissions. These advantages make electric drive very attractive, particularly when it comes to stop/go city driving. This promise combined with rapid cost declines has led to great optimism about the future of BEVs, spearheaded by the great success of Tesla Motors.

However, BEVs will always have to deal with a large competitive disadvantage: the battery pack. Even under optimistic assumptions of future technological developments, a sufficiently large battery pack will make a BEV substantially more expensive and heavier than a similar ICE or hybrid vehicle.

Ultimately, pure electric drive should be about 2.5x more efficient than an ICE vehicle. As will be illustrated below, this efficiency advantage does not bring significant savings when accounting for real energy provision costs, whereas a sufficiently large battery pack will continue to put BEVs at a cost disadvantage. For this reason, BEVs do not offer a large-scale solution to the global sustainability problems we must (very rapidly) overcome during the 21st century.

Cost analysis

BEVs will have to achieve a range exceeding 200 miles as standard before broad consumer acceptance can be achieved. Another less often stated requirement is that this range will have to be maintained after at least 10 years of driving and through all seasons. Modern ICE vehicles can operate smoothly for 20 years without bringing any range anxiety issues with age or temperature.

As a result, future BEVs will have to come equipped with a battery pack of about 80 kWh which will cost a hefty $10000 even assuming optimistic future Li-ion battery pack costs of $125/kWh (figure below). This $10000 is a good proxy of the expected price difference between an ICE vehicle and a BEV which will be accepted by the mass market.

Battery pack price declines

An argument can be made that the BEV drivetrain (motor, simple transmission, inverter, step-down converter and charger) will be cheaper than an ICE drivetrain (engine, transmission, stop & go system and exhaust). According to numbers in this paper, the total 2013 costs of a 70 kW electric drivetrain is about €2640 while a gasoline drivetrain will cost about €2950. However, the electric drivetrain costs could decline to €1600 with future technological advances. The potential future BEV could therefore enjoy roughly $1500 price advantage over an ICE vehicle due to the simple drivetrain. For most people, however, this advantage will be cancelled out by the fully installed costs of a home charging station, so we will consider the $10000 cost difference in this article.

A high-BEV future will also feature a large number of additional chargers to further reduce range anxiety and enable longer travels. Many parking spots will include public 10 kW level 2 chargers (giving about 30 miles of range per hour) for about $5000/charger. Highways will also require regular 100 kW level 3 chargers (giving about 300 miles per hour) for about $60000/charger. (Costs from this link.) Let’s say that we need 1 public level 2 charger for every 5 BEVs and 1 level 3 charger for every 100 BEVs. This will add another $1600 per vehicle (without charging station maintenance costs).

On the positive side, conventional wisdom states that a BEV should have lower fuel costs than an ICE vehicle because it is so much more efficient. However, ICE vehicles still have a lot of headroom for efficiency improvement and are projected to exceed 50 miles per gallon by 2025 (see below). Further improvements yield steadily diminishing returns (as will be shown in the calculations below).

EIA vehicle cost and efficiency projections

In addition, much of the cost saving hype surrounding electric vehicles is based on oil exceeding $100/barrel with some heavy gasoline taxes added on top. When looking at real energy production and distribution costs (which must be done when considering the disruptive potential of a technology), gasoline is actually surprisingly cheap. As discussed in this article, the actual production cost of oil is about $35/barrel and we can still extract substantially more oil than the human race has extracted to date below this price point. When assuming a rather high value of $1/gallon for refinement and distribution costs, the actual production and distribution cost of gasoline amounts to only $1.83/gallon. Electricity, on the other hand, costs about $0.13/kWh (US residential electricity prices – tax-free), about half of which is transmission and distribution costs. When accounting for 10% charging losses, this amounts to $4.83/e-gallon.

It therefore becomes clear that, when accounting for total direct costs carried by the overall economy, BEVs need to be about 2.6 times more efficient than ICEs to break even – almost exactly the projected situation in 2025 (figure above).

Real fuel cost savings from the BEV of the future are therefore negligible, but the up-front cost difference will remain. In other costs of ownership, lower maintenance costs are cancelled out by higher insurance costs. Furthermore, BEVs may well depreciate significantly faster than ICE vehicles because the battery pack will degrade faster over time than the ICE drivetrain.

The figure below shows the ownership costs (insurance and maintenance excluded) of future ICE, hybrid and BEV technologies (with fuel efficiencies as projected for 2025 in the figure above). Costs assumed were $25000 for the ICE, $27000 for the hybrid and $35000 for the BEV. Capital costs were calculated over a 5 year ownership period (with a 5% discount rate) during which the car depreciates by the percentage indicated in the graph (60-80%). Fuel costs were calculated for 15000 miles driving per year.

Ownership cost comparison ICE hybrid BEV

The graph shows that the yearly ownership costs of a BEV acceptable for the mass market (>200 mile range in all seasons even after 10 years) would cost $1400/year more than an equivalent ICE vehicle under similar depreciation assumptions and as much as $3000/year more if it depreciates faster. The law of diminishing returns with regard to fuel efficiency is also clearly illustrated by the small contribution of fuel costs relative to capital costs.

Lastly, a carbon price will also not have a sustained positive impact on BEV sales. The largest current and future car markets (US, China, India) have electricity mixes where a carbon price will make EV charging more expensive than ICE refuelling, especially if ICE efficiency moves towards 50 MPG. See the map below. It is true that the carbon intensity of electricity will gradually reduce in the future, but this will increase the electricity price faster than the inevitable steady increase in the real extraction cost of oil.

EV carbon footprint comparison

Justifying a BEV price premium

For BEVs to disrupt ICE vehicles, people will have to be willing to pay this substantial price premium. Tesla has shown what can be achieved with electric drive in terms of performance and driving experience and this is something that customers may be willing to pay extra for. Wireless charging also offers a potential BEV future where you never need to think about refuelling or charging (e.g. wireless charging roads).

However, even though the EV driving experience may fetch a price premium, it is doubtful that this will count for much outside of the small luxury/performance vehicle segment. Wireless charging roads and parking spaces sound very cool, but also rather expensive and, if you think about it, it does not offer such a meaningful improvement over two visits to the filling station every month.

In the absence of a very fast and convenient charging solution at almost no additional cost, ICE vehicles will maintain a price premium over BEVs. Even Tesla’s supercharging stations will need to become much faster before they can offer a real solution to this challenge. Just imagine the queues during rush hour at filling stations taking 6x longer to give cars a 3x shorter range than conventional filling stations. Yes, home charging can substantially reduce this burden, but this adds the costs of a home charging station to the costs of a vast supercharger network.

BEVs may also be able to fetch lower fuel prices by charging only during off-peak hours, but, as shown in the above figure, even a substantial reduction in fuel costs for BEVs will not really alter this situation. In addition, a baseload-dominated power system is the only really practical way in which this can be implemented. Smart charging with politically popular, but variable solar/wind will most likely be impractically complex and expensive.

Lastly, in case the self-driving car ideal becomes a reality, ICE vehicles are likely to benefit more than BEVs. As discussed above, actual fuel costs are similar between BEVs and ICEs, thus offering no increasing value with increased use. In fact, much more free-flowing traffic resulting from a fleet of fully autonomous vehicles will significantly boost the efficiency of ICEs relative to BEVs. Furthermore, ICE vehicles will be able to refuel much faster, thus giving them more time on the road.

Evidence to date

The US probably offers the best example of the attractiveness of BEVs in the real world. Gasoline is not taxed at such high levels as most other developed nations and electricity is not taxed, thus providing a fairly good fuel cost comparison. The federal and state incentive programs also combine to cut close to the aforementioned $10000 price disadvantage from the cost of new BEVs (most sales are in states with incentives such as California). BEV sales as a percentage of the total are given below (data available here). The black line is a 12 month moving average.

BEV market share US

As shown above, even though sales are increasing, the current market penetration is low, even with generous incentives. It should also be noted that only about half of BEV sales come from models in a price range targeting the mass market. The other half are up-market offerings from Tesla and BMW which cannot cause significant disruption in the overall auto industry.

The data therefore shows that, when incentives eventually fall away, sub-100 mile BEVs will have to drop $10000 in cost to achieve a fraction of a percentage point of market share. In addition, they will have to contend with much more efficient ICEs finally entering the notoriously inefficient US vehicle fleet. Higher-priced BEVs with a longer range might be able to secure larger market share, but it is difficult to see market penetration exceeding 10% in the affluent US market – let alone the developing world where per-capita GDP is an order of magnitude lower.

Disruption of a different kind

Even though this article paints a bleak picture for the future of BEVs, I’m actually fairly optimistic about this technology. I just think that the greatest potential for disruption comes not from cars, but from smaller vehicles where the advantages of battery electric drive over the internal combustion engine really come to the fore. These vehicles are fully compatible with a future where the global middle-class quadruples in size, while environmental and space constraints force society to do away with blatant inefficiencies like short-distance-single-person-in-car travel. More about this line of thought in part 2 of this article…

Schalk Cloete's picture

Thank Schalk for the Post!

Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.

Discussions

Ed Dodge's picture
Ed Dodge on March 22, 2016

I see hybrid vehicles as the natural transition course between today’s ICE vehicles and the clean vehicles of tomorrow. Electric drive trains offer significant performance improvements on a number of fronts that can not be replicated with mechanical drive trains, and as the public gets used to the improved driving performance demand for electric drive trains will grow.

But, battery storage alone is only suitable for light duty work, like moving people around town. As soon as you move up to medium and heavy duty loads or long distances then batteries alone cannot do the job, they simply drain too quickly. Battery charging generators are easily added to allow for the heaviest duty work and offer the opportunity for the market to experiment with different fuels and architectures since generators can be engines, turbines or fuel cells.

Some of the most powerful machines in the world have moved to electric drive trains with liquid fuels. The new US Navy destroyers and littoral combat ships are fully electric with the power provided by gas turbines. Diesel freight trains have used electric motors since the 1950’s. There are no performance limitations on hybrids, I would say we are only scratching the surface of what high performance vehicles can achieve.

It will also take many years for electric charging infrastructure to reach critical mass and will presumably not be rolled out equally and uniformly. Hybrids that can use both fuels and power offer consumers flexibility and a low risk path to ownership.

As for battery life cycles, I would like to see Tesla and the others switch to a leasing business model where the manufacturer owns the batteries and replaces them as a service. Leaving car owners stuck with useless old batteries (thousand pound boat anchors) will be bad for business I think.

Engineer- Poet's picture
Engineer- Poet on March 21, 2016

PHEV owner and driver of 3 years’ experience here.

Wireless charging roads and parking spaces sound very cool, but also rather expensive and, if you think about it, it does not offer such a meaningful improvement over two visits to the filling station every month.

Wireless charging is not available for my car or in my area, but I have still managed not to buy fuel for my car for more than 2100 miles of driving. I still have 3/4 of a tank left, too. Charging at home is anything but inconvenient.

Just imagine the queues during rush hour at filling stations taking 6x longer to give cars a 3x shorter range than conventional filling stations.

The misconception is that fast-charging stations will be used like filling stations. The error is that electricity is not like gasoline; it can be supplied almost anywhere all the time. People will be charging in their garages, under their carports, in their parking lots at work, at shopping malls. They’ll start every day and maybe even most legs with a full charge; the idea of “tank getting low” won’t figure.

Electricity, on the other hand, costs about $0.13/kWh (US residential electricity prices – tax free), about half of which is transmission and distribution costs. When accounting for 10% charging losses, this amounts to $4.83/e-gallon.

This claim is unsupported by calculations or references. A gallon of gasoline has approximately 115,000 BTU of energy, or 33.7 kWh. Converted to energy at a pretty good efficiency of 30%, it yields about 10 kWh at the crankshaft. Allowing for 10% charging losses the BEV pays about $1.40 per gallon-equivalent at 13¢/kWh, which is still substantially less than today’s record-low inflation adjusted gasoline prices. When liquid fuel goes back into the $3/gallon range it’s a no-brainer.

Lastly, in case the self-driving car ideal becomes a reality, ICE vehicles are likely to benefit more than BEVs.

The self-driving EV taxi can amortize the battery over a lot more miles per year, while paying much lower fuel costs. It may also enjoy outright preferences in noisy and polluted cities. In a carbon-taxed future, the BEV using 450 gCO2/kWh electricity at 300 Wh/mile (135 gCO2/mi or about 85 gCO2/km) is competitive with anything out there.

Nathan Wilson's picture
Nathan Wilson on March 24, 2016

The MIT study, “On the Road Towards 2050”, which came out in Nov 2015, is also surprisingly pessamistic about BEVs. They forecast the light-duty vehicle sales in 2050 to be:

5% Battery Electric Vehicles (BEVs)
6% Fuel Cell Vehicles
12% Plug-in hybrids
17.5% hybrids
60% pure ICE

Regarding BEVs, they said, “One of our specific findings on the use of electricity in transportation is that, without additional technological breakthroughs, pure BEVs are likely to be limited to modest sales volumes. One major reason is the long recharging time for this technology, which better vehicle batteries will not significantly reduce. Drivers are accustomed to refueling gasoline vehicles for more than 400 miles of travel in about five minutes. Gasoline refueling occurs at a rate of chemical energy transfer through the pump outlet of about 10 MW”

They acknowledge much uncertainty in these predictions, but state:

” Today, it is possible to identify a number of potential alternative fuels, including electricity, hydrogen, biofuels, and natural gas. However, it is not yet clear that any one of these can fully assume the dominant position that petroleum has held as the preferred transportation energy source for the past century.”

From summary, chapter 11

Nathan Wilson's picture
Nathan Wilson on March 22, 2016

“… home charging … BEVs may also be able to fetch lower fuel prices by charging only during off-peak hours…”

Interesting problem: given that the evening is the ideal time to recharge and a multi-kW fast rate is desirable (from a user perspective), and assuming that the local utility must pay around $1000/kW to boost system capacity (plus upgrades to the distribution system) it would seem that the utility will have a very strong incentive to guide BEV users to off-peak charging. Perhaps the utilities will provide $500 cash incentives to user who install off-peak chargers.

As a member of a household with two garage-kept vehicles, I think one or two BEVs with 200+ mile range would be great. But I don’t think I’m typical of the US or the world car owners. In my town in general, most homes have the cars stored in the driveway or on the street (neither of which is well suited to charger location). For apartment complexes, the BEV problem is worse. Chargers are too expensive to leave idle, so they’ll install only barely enough. Will there be a wait list to get an assigned parking space with a charger? Will there be a miminum monthly charge to use the parking space, a charger with a credit card reader or a flat monthly fee? Can chargers be affordable and vandalism- resistant?

Another issue is the peculiar American populatity of pickup trucks (and their cousins, the SUV, mini-van, and crossover). These seem to be designed without regard for energy efficiency. See Scientific Proof that Americans are Completely Addicted to Trucks.

For these reasons, I think 20% EV penetration is a realistic long term goal – i.e. not high enough to solve the sustainability problem. Also, for long-haul trucks, batteries don’t make sense at all. Hence, my support for ammonia fuel – made from off-peak electricity or from fossil fuel with CC&S.

Engineer- Poet's picture
Engineer- Poet on March 24, 2016

pure BEVs are likely to be limited to modest sales volumes. One major reason is the long recharging time for this technology, which better vehicle batteries will not significantly reduce.

Color me skeptical. I am regularly seeing announcements of cell chemistries which have demonstrated charging times under 10 minutes. It only takes one to hit the market and upset all the applecarts.

Even without that, though, ponder a Leaf-like car in a world of ubiquitous charging. Suppose the car spends 3 hours a day off the grid, and 8 hours at work. Work provides demand-managed Level 1 charging at 50% duty cycle (performing grid regulation): 720 W average 8 hours/day, 5.76 kWh. The vehicle is plugged in but generally not charging for the 6-8 PM evening demand peak. The remaining 11 hours it generally charges at Level 1 rates (minimizing demand on neighborhood distribution transformers): 15.84 kWh, going up to Level 2 during the off-peak hours to top off the battery by morning. That’s 21.6 kWh per day without even overloading a 14-gauge extension cord and playing really nice with the grid for demand-side management.

Far from being impossible, it is closer to not even being difficult. All we have to do to start is make better use of the wires we’ve already got.

Schalk Cloete's picture
Schalk Cloete on March 21, 2016

Always nice to hear about some direct experience. Just one question. If you almost never need to fill up, will your next car be a BEV? I quite like the idea of PHEVs, but this article is about whether battery electric drive can disrupt the ICE, so PHEVs is not part of this discussion.

To address your comments:

1. I’m not trying to say that home charging is inconvenient. I’m just saying that a future scenario with wireless charging (which is often quite enthusiastically discussed in EV circles) will not be such a highly significant improvement over filling up at the gas station.

2. True. I do state in the article that the demand for charging at fast charging stations will be strongly reduced by home charging. Can you share how much was the fully installed cost of your home charging station and how fast it can charge your PHEV? I just found a good link and will include it in the article now. Will be good to compare.

3. I used 33.41 kWh/gal. The cost per e-gallon is thus 0.13*33.41/0.9=$4.83/gal. The gasoline costs are more up for debate. Oil price fluctuations are obviously an important factor, but it should be noted that current prices are much more normal than oil north of $100/barrel. Tight oil technology should help to keep the price around these normal levels for the foreseeable future. Overall, I think the real energy provision costs for gasoline ($1.83/gallon) and electricity ($4.83/gallon) are fairly accurate.

4. Agreed about the amortization. But in the end the fuel costs are the most important for this point. The ICE vehicle of the future (when battery packs cost $125/kWh), will achieve excellent fuel economy (probably >60 mpg) in free-flowing automomous vehicle traffic (essentially getting highway efficiency in town). When accounting for the real energy provision costs calculated in the article, the ICE vehicle may well be slightly cheaper to fuel. I also think the ICE’s much longer range and ability to “recharge” more than an order of magnitude faster are also important advantages. About CO2 emissions, a 60 mpg future ICE car will have emissions of 169g/mile. Which is competitive with electricity carbon intensity of 563g/kWh (before accounting for higher embodied carbon in BEVs). Electricity in the biggest future car markets (developing world) will be above this mark for a very long time.

Lastly, do you have any comments on the real-world sales of BEVs (and PHEVs which sell about the same as BEVs) which are very slow despite large tax credits? Do you think market share can rise to 10% and beyond when tax credits fall away?

Engineer- Poet's picture
Engineer- Poet on March 24, 2016

Just one question. If you almost never need to fill up, will your next car be a BEV?

The problem is I have short-range personal, long-range personal and load-towing needs. The Fusion Energi could have been 2 cars if I’d been able to get a replacement diesel (long range) plus something like an iMiev (short range) back in 2012. The tow vehicle has to be something like a truck. Perhaps a Tesla could perform all 3 roles, if it was also rated as a tow vehicle. So far, nothing seems to be. The Model X appears to have all the necessary capabilities, but Tesla won’t certify or warrant it for those uses.

I’m just saying that a future scenario with wireless charging (which is often quite enthusiastically discussed in EV circles) will not be such a highly significant improvement over filling up at the gas station.

On the contrary, it will be vastly different. Charging while parked takes a few seconds to plug and unplug; if charging was ubiquitous, it would eliminate most trips for fueling. Ubiquitous wireless charging would even eliminate the plug/unplug steps; it stops being an “event” and becomes seamless.

Can you share how much was the fully installed cost of your home charging station and how fast it can charge your PHEV?

It came free with the vehicle. I’ve been using the “convenience cord” with a standard NEMA outlet for 3 years; a full charge takes about 5 hours. I bought a kit for a Level 2 charger but I haven’t gotten it assembled yet. I will be doing my own wiring. The charger and its cable ran a bit under $400.

I used 33.41 kWh/gal.

Which is grossly misleading if not erroneous, because the PHEV does not convert electricity to work via a heat engine. Put another way, $4.83 of electricity will take you a lot farther than $4.83 of gasoline.

it should be noted that current prices are much more normal than oil north of $100/barrel

Normal for what? Compare pre-1973 prices to post-1979 prices and tell me which should be designated “normal” going forward.

do you have any comments on the real-world sales of BEVs (and PHEVs which sell about the same as BEVs) which are very slow despite large tax credits?

Very slow? Aside from the price-crash induced slump, interest seems strong at the high end (Tesla). Interest in the middle will bounce back when the weak oil producers are shaken out and prices return to something normal.

I think PHEVs are set to explode. At $125/kWh, a 10 kWh battery is just $1250. Silicon carbide power electronics are going to slash the size and cost of inverters. If the engine is cut down to a 2-cylinder turbocharged beast producing 100 kW peak, a large fraction of the parts count and cost disappears from the engine compartment. Fast response to throttle inputs is handled by the electric side, so the engine can be optimized for cost, efficiency and emissions.

The thing that will make or break the PHEV as the agent behind peak gasoline is ubiquitous charging. Electricity is already sent almost everywhere (often to light poles just feet from parking spaces), but it’s not available to vehicles. Change that and the first 20-25 miles of every drive can be gasoline-free. I’m averaging almost 134 MPG (admittedly, it’s a game I’m playing) but if every stop had 240 VAC @ 16 A available it would stop being a challenge and just be the way things are

Schalk Cloete's picture
Schalk Cloete on March 25, 2016

The futue PHEV scenario with ubiquitous charging is also an interesting one, especially for the US where longer travels are quite frequent due to urban sprawl. It should be noted though that having public charging infrastructure on most parking spots will add a non-negligible cost to this scenario both in terms of up-front capital and maintenance (see this link for costs).

About the oil costs, the average inflation adjusted price over the past 50 years is about $50/barrel, hence my statement that current $40/barrel prices are much closer to the norm than >$100/barrel prices. There are several factors which should limit the oil price over coming years such as muted global economic growth, tight oil, efficiency standards like CAFE, changes in driving habits and alternative fuels (e.g. the PHEVs you mention).

Sure, I state clearly that I expect the BEV of the future to be about 2.5x more efficient than the ICE car of the future. Electricity at $4.83/e-gallon is therefore equivalent to gasoline at $1.93/gallon – a bit above the current before-tax price.

I agree that PHEV (and standard hybrid) sales are closely tied to the oil price, hence the fall in US electric drive vehicle sales from 3.5% in 2013 to 2.5% in 2015. We’ll have to wait and see what the oil price does over coming years, but I think it is safe to say that, if peak oil is caused by the attractiveness of PHEV and BEV technology, these technologies will have to be able to dominate the market with an oil price in the $30-40/barrel range. I struggle to see that happening.

Engineer- Poet's picture
Engineer- Poet on March 25, 2016

My time with the crew at The Oil Drum has convinced me that $40/bbl oil is an abberation going forward, below the cost of production. This doesn’t mean we won’t have periods of it, but there won’t be any investment while prices are so low and depletion will contract production until the price goes back up. This won’t take long in the USA; when half of lower-48 oil production comes from wells less than 2 years old, a couple of years of not drilling much is going to shrink the supply quite a bit.

In the mean time, PHEVs will get cheaper for quite a while. I also expect buyer preferences to shift; people will buy PHEVs as “insurance” and a hedge against resale value. This will give them major options when fuel prices rise, because they can literally switch away from petroleum if they haven’t already ingrained plugging in into their habits (the people who already did would be largely insulated from fuel prices).

Schalk Cloete's picture
Schalk Cloete on March 24, 2016

Interesting report with a lot of information…

Personally, I’m quite keen on saving emissions, money and health through smart behavioural/consumer choices, so the part on the influence of factors like driving style, vehicle 0-60 times and vehicle weight on fuel consumption was especially interesting. It is quite impressive how large this effect can be (quantified nicely in several graphs in chapter 8). As an example of these findings, the rented Audi A3 Diesel which I use once or twice per year generally gives me below 4 litres per 100 km (59 mpg). I’ve checked now that its EPA rating is 36 mpg, while the rating on Audi’s Norwegian website is actually exactly 4 litres per 100 km. Just driving style can therefore make the car 64% more efficient. Quite remarkable…

The other interesting graph is the deminishing returns in terms of the fraction of distance driven on electricity when increasing the battery capacity of a PHEV (Fig. 8.13). It would be useful to get such a graph for more data since it gives a nice indication of the battery capacity that 100% electric BEVs will need before broad consumer acceptance is achieved.

Clayton Handleman's picture
Clayton Handleman on March 24, 2016

Nathan,

When you look behind the curtain, what they tell you is they don’t have any good data on behavior. Of course, this is one of the reasons that MIT has a great reputation, they are honest and will tell you when their data is garbage. The relevant part of the study appears to be section 8.3.8 Paraphrasing:

“Prior studies are so poorly done that they cannot be given any validity. So another study was done using 125 3kwhr pluggable hybrid prius’s and then extrapolating using our best guesses as to how people might behave. This is still an awful study but it is the best we have and we will make up a bunch of stuff and add some equations to give it the MIT petina”

From the study:
Charging behavior is an area of even greater uncertainty. Due to a lack of real-world data,
charging behavior in existing work has been largely assumption driven [Khan and Kockelman,2012] or based on small samples.

Then since the data is junk they rely on more recent data from 125 Priuses with 3kwhr batteries.

Simply putting the letters MIT in front of something does not make it fact or even good information. What MIT means is that it is an honest assessment given the avialable data and that they will let you know the quality of the data. In this report they make it abundantly clear that current data on driving habits is slim to no

Schalk Cloete's picture
Schalk Cloete on March 22, 2016

Interesting about the electric drive trains on US Navy destroyers. What is the main motivation for doing this? I would guess that generating electricity in a gas turbine and then using that electricity to drive an electric motor is thermodynamically less efficient than a diesel engine. Why not have a diesel engine for cruising and gas turbines for high speeds?

Bob Meinetz's picture
Bob Meinetz on March 22, 2016

Ed, possibly a leased-battery business model could fly, although for people who avoid monthly payments whenever possible (like me), it will be a hard sell.

I’m currently in a dispute with Nissan over a recent software “upgrade” they performed on my LEAF, which magically added 10 miles to its range – and exempted them from the battery replacement promise in their warranty. It seems they assumed rewriting software to adjust the numbers which appear on the vehicle’s dash display would “fix” the problem, a la Volkswagen. But battery capacity is a bit trickier to hide than emissions data – for $30, one can buy a plug which fits into the car’s diagnostic port and a matching phone app which provides the capacity data for each of the 96 cells in the battery pack.

Disappointing. I’m hoping Nissan steps up to the plate.

Hops Gegangen's picture
Hops Gegangen on March 21, 2016

I would not ignore the potential for charging while driving on a freeway, thus allowing a smaller battery pack.

http://mashable.com/2015/08/17/electric-car-charging-uk/#5609Ji8vs8qF

I also would not assume the current car ownship model, which Uber is disrupting.

And I would not assume that we won’t see tailpipes outlawed in urban areas. A city with all BEV would be far more liveable. Some cities are already aiming for that in the long term.

BEVs can also be more fun to drive because that battery pack is low to the ground and centered.

That said, biofuels are also making slow but steady progress, so not being 100% BEV is not the end of the world.

Schalk Cloete's picture
Schalk Cloete on March 21, 2016

All good points. Have you seen any cost estimates for such wireless charging highways? In the end it will come down to costs of large battery packs + superchargers vs. costs of building and maintaining many miles of wireless charging roadway.

Agreed on the car ownership model. This is discussed in the next part of this article which hopefully comes out today.

Personally, I think the likelihood of seeing carless city centres is much higher than the likelihood of seeing tailpipe-less cities. Completely removing cars from the equation, instead relying on publically available e-bikes and a few neighbourhood electric vehicles will be a big boost in making cities more livable.

True that EVs can be more fun to drive, but this is more applicable to the small luxury/performance market segment.

Agreed about biofuels. As mentioned by Robert below, synfuels from electricity is also an important longer-term option which will bring great flexibility while minimizing complexity of the energy system.

Hops Gegangen's picture
Hops Gegangen on March 21, 2016

I don’t know the cost — expensive, I’m sure.

http://mashable.com/2015/08/17/electric-car-charging-uk/#5609Ji8vs8qF

I also expect to see a drop in price and rise in performance of super capacitors.

So, you give up some battery for a capacitor. You pull the car over a pair of electrodes that zap a massive high-voltage charge into the capacitor, and drive away. The capacitor then charges the battery.

Clayton Handleman's picture
Clayton Handleman on March 21, 2016

“given that the evening is the ideal time to recharge and a multi-kW fast rate is desirable (from a user perspective)”

Evening is the ideal time to plug it in. But you can easily set it to start charging later. My suburban experience is that a lot of people have a vehicle that is primarily used for commuting and errands. The people I know with EVs love to plug them in at night and wake up to a fully charged vehicle. I can see plenty of demand for 240 Vac but that is no different than an electric stove or a large AC unit. Is there really special utility infrastructure required? I think the burning demand for fast chargers is something made up by people that are simply looking for debate points.

Here’s a quick thought experiment, in a post I linked to in another part of this thread there is data showing that roughly 70% of households have 2 or more cars. 90% of the days the cars drive less than 90 miles, about the range of a 25 kwhr Leaf. A 20A 120V circuit could recharge it fully in 12 hours. A 240V 40 A circuit could get it charged in roughly 3 hours. I am not seeing a problem here. At least for the 70% of homes with 2+ cars, I don’t see it being a major inconvenience to take the ICE for long distances and the EV for most everything else.

Nathan Wilson's picture
Nathan Wilson on March 22, 2016

Regarding fast charging, standard US 120V plugs are only rated for 15 Amps intermittent, or 12 Amps continuous, so around 1.2 kW worst case with line tolerance. So even a 100 mile range EV (24 KWh) is not guaranteed to recharge overnight with one.

Furthermore, any additional loads on the same electrical circuit would likely trip the breaker while the charger is in use. So as the industry matures, we can expect garages to have dedicated circuits for EV chargers. And 240 V circuits cost almost the same to install as 120V circuits.

For example, Leviton sells after-market EV chargers. They have discontinued their 120V charger (the Evr-Green_120). Now their only offerings are 4.8, 7.7, and 9.6 kWatts. So yes, this is a substantial increase in the household evening demand, probably double an average house (since kitchen loads are only on for an hour or so per house, and the charger can run all evening).

Leviton says their 4.8KW chargers (such as the unit below which is $400 from Home Depot) provides 20 miles of range for each hour of charging. So get in the habit of charging it during dinner, and you’re ready to go out again afterward. Some green early adopters might volunteer for off-peak charging, but mainstream users will need a big incentive to go that route.

Clayton Handleman's picture
Clayton Handleman on March 22, 2016

“but mainstream users will need a big incentive to go that route.”

So I have posted data in this thread that argues to the contrary, here it is again. Would be interesing to push the conversation forward addressing that. Load shifting is already demonstrated and setting the charging timer on a Tesla is pretty much a zero learning curve (it took me about 2 minutes to fully master it on my test drive), no doubt others will emulate.

Clayton Handleman's picture
Clayton Handleman on March 21, 2016

Schalk, Sorry I don’t have much time, could have a lot of fun with this one. But let me make a couple of quick points:

– First, glad you have moved closer to Navigant projections, I think it was as recently as last spring that you were pained to concede that battery prices would go below $400 / kwhr. I think that your assumption that battery prices will be at $125 / kwhr by mid century is deeply flawed. Notice that the graph you are using for projections shows that the top EV makers are already well below the “experts / published” projections. And the makers of the graph apparently don’t consider Musk an ‘expert’ (I know, it says published but Musk doesn’t generally say numbers he doesn’t mean so they need to give him the credibility of publications or give him his own mark on the graph). I find that a little strange because most would agree that he is a pretty smart guy. Given that he is building the largest battery factory in the world, one could reasonably assume that his team likely is second to none in their understanding of battery pricing dynamics and he has indicated they think they will hit the $100 / kwhr point before 2025. It is also worth pointing out that experience curves are based upon volume and not time. The gigafactory along with the Model 3 will see a step change in volume and so, a likely step change in the graph that you have posted as that production ramps. Battery prices have been on the optimistic side of the Mckinsey and Navigant projections which are already ahead of yours and they don’t go much past 2025 – i.e. nobody dares go there – except me and while my track record is pretty good on projections, I don’t have the institutional affiliations or resources to be a primary source. So my suggestion is that Musk’s prediction of $100 / kwhr by 2025 should be your starting point given your timeline. Since enormous resources are being poured into battery tech, it would seem absurdly conservative not to go to $62.5 / kwhr by 2035, still long before your mid century date.

– On range anxiety, looks like Bolt and Model 3 are both planning 200 mile range by 2017. After 70% degradation that takes them to 140 miles. Still plenty of range for many applications. While I would agree that a relatively small percentage will still want / need the longer range, I would argue that a 200 mile range that drops to 140 miles near end of life, will be great for far more than10% of drivers. I dug up some driving data and put together this piece that addresses this point. Let me be clear, I agree that 200 miles is not sufficient for all applications, but 10% of auto sales by 2050 is a strangely low number to put it charitably. Something that many forget when comparing EVs to ICE is that it is very convenient for EVs to leave home every day with a full tank. ICE vehicles are not topped off every day so they actually can run low on fuel at the most inconvenient times. IOW, they have their own range anxiety and filling up when supper is getting cold and the family is eating without you is no fun and won’t happen to an EV commuter.

Clayton Handleman's picture
Clayton Handleman on March 21, 2016

Looked as well, looks like my mistake. We had gone back and forth about current prices in a $2014 thread and you were in the $400 to $1000 range when Nissan was selling battery packs for $270 / kwhr.

In any event, the conscensus work you are doing troubles me. It highly favors the incumbents.

I think it would have been a lot more useful if it had also included 5 years, 10 years and 15 years out.

Schalk Cloete's picture
Schalk Cloete on March 21, 2016

Oh yes, that was for current costs.

Sure, the Seeking Consensus work will eventually arrive at future internalized and externalized costs (after the current chapter on current externalized costs is done). It has been a bit slow going lately due to a lot of other responsibilities, but it will hopefully happen within a year or so.

Robert Hargraves's picture
Robert Hargraves on March 20, 2016

Let’s not exclude zero-net-CO2 internal combustion engines. Synthesized hydrocarbon fuels are possible. The US Navy research lab extracts CO2 and H2O from seawater to make C-(H-C-H)….-C hydrocarbon chains. Cost was about $5 gallon, still a bit expensive. Ammonia, NH3, can be made from air and water. These become economically feasible as the costs for new nuclear power drop; 3 cents/kWh is possible. We’ll be able to continue to benefit from the well-engineered internal combustion engines.

Schalk Cloete's picture
Schalk Cloete on March 20, 2016

Thanks Robert, that is a good point. I was asked to participate in a survey about the future of energy storage about two years ago and also expressed much more faith in synfuels than battery technology. Large scale synfuel production whenever electricity is cheap will give the energy system a great deal of flexibility while keeping things simple enough to actually work in practice.

However, I doubt that this is going to happen on any meaningful scale before mid-century. Many people think that battery electric vehicles will seriously disrupt gasoline cars long before mid-century. This is the notion I’m trying to address in these two articles.

Willem Post's picture
Willem Post on March 20, 2016

Schalk,

With 80 kWh battery, range is about 270 miles for a compact EV.

At a future 125/kWh, battery cost would be about $10,000.

For mass market EVs, there cannot be mass subsidies, as that would bankrupt the country.

Thus, compact EV prices must not exceed $30,000 by about 2020, as that would be what families could afford.

BTW. Fast charging on the road reduces batter life. Slow charging at home preserves battery life.

US primary energy for transportation was 27.1 quad in 2014, of which 21.4 quad was rejected as heat and 5.68 quad performed services to users. The energy categories are as shown in the below table. Air and Ships would require syn- and biofuels; most of Rail could be electric; some of Hv Truck could be electric battery; all of Lt Truck and LDVs could be battery. A quad = 10 ^15 Btu.

By 2050, 18.5 quad is assumed to be replaced by 5.68/27.1 x 18.5 x 1.2 (battery loss) = 4.65 quad of electricity, or 1363.7 TWh. About 27.1 – 18.5 = 8.6 quad would be syn- and biofuels, which would provide to services to users of 5.68/27.1 x 8.6 = 1.8 quad, or 528 TWh.

………………………………..quad

Air…………………………….2.5

Ships…………………………0.7

Rail…………………………..0.5

Hv Truck…………………..5.0

Lt Truck…………………….0.5

LDVs……………………….15.5

Misc………………………….2.4

Total………………………..27.1

http://energy.gov/sites/prod/files/2015/09/f26/Quadrennial-Technology-Re...

Energy per Mile: In 2013, 38.4044 quad was used to generate 4065.965 TWh; less self use of 164.78 netted 3901.185 TWh; plus imports of 46.73 yielded 3947.915 TWh to the grid; less T & D of 253.580 netted 3694.335 TWh to user meters, or 12.6056 quad, resulting in an energy in/out ratio of 0.328.

An EV requires about 0.30 kWh/mile, or 1024 Btu/mile, or 1024/0.328 = 3119 Btu/mile on a primary energy basis. Gasohol (10% ethanol/90% gasoline) contains about 120,900 Btu/gal. An ICE vehicle, @ 38.8 MPG, would use 120900/38.8 = 3116 Btu/mile.

EPA MPG-Equivalent: For the EPA to claim the EV mileage is about 38.8/0.328 = “118 MPG-equivalent” is misleading, to say the least. The EPA-invented mileages are used to help manufacturers meet the federal CAFE requirement of 54.5 MPG, EPA-Combined, by 2025.

Worldwide CAFE Standards: The three largest passenger car markets representing two-thirds of global sales have strong fuel economy standards in place: US, 54.5 mpg by 2025; EU, 56.9 mpg by 2021; China, 47.7 mpg by 2020.

Clayton Handleman's picture
Clayton Handleman on March 21, 2016

“Can you share how much was the fully installed cost of your home charging station and how fast it can charge your PHEV?”

You are doing your projections over 35 years. Todays price for an in-home charging station is really only a crude starting point for an infant industry. As new construction, garage remodels etc. are done they will go in for much lower cost. Hard to imagine that building codes won’t develop that mandate EV charging outlets. Also, assuming the outlets standardize as has the common 120V plug, the charging station can / should be amortized over a much longer time than the car. Outlets installed in the 60’s and 70’s are going strong today.

“Do you think market share can rise to 10% and beyond when tax credits fall away?”

Hmmm, If your projections are right and we stay in the single digits of penetration, is it really possible that oil prices will remain this low as the developing world continues to add cars to the road? If so then you might get me on board for something closer to your scenario. That and if there are no carbon disincentives. However it is kind of looking like the stupid people have won. They said we won’t trust scientists and the scientific method, we shouldn’t have to pay anything until it is patently obvious to everyone. And so they have had there way. Its looking like the lobster is finally realizing the water in the pot is pretty damn hot. Climate change is beginning to have real impacts, from CA drought to record after record arctic ice loss.

So to answer your question with a question: Do you think that for the next 35 years, things will be so stagnent as to not have either or both of the following: ongoing and increasing incentives for low carbon solutions and / or ramping disincentives for use of carbon energy sources.

Schalk Cloete's picture
Schalk Cloete on March 21, 2016

It will be hard to standardize EV chargers as part of all new construction before most cars sold are BEVs. If BEVs eventually peak out at ~20% of the market, this will result in a lot of underutilized infrastructure.

Fuel efficiency standards, slower economic growth and, most importantly, very un-American vehicle usage habits will restrict global oil demand over coming decades. In addition, as discussed in the next part of this article, I think that small electric vehicles will have displaced many hundreds of millions of cars come mid-century.

That being said, light-duty vehicles are responsible for less than 15% of global CO2 emissions and this number will further decline with improvements in user habits and efficiency, so even a perfect BEV future is not going to save the world. In addition, CO2 reduction potential from BEVs is limited, especially in the developing world where electricity is very carbon intensive. I therefore agree that policy will increasingly favour cleaner technology in the future, but it is not going to help BEVs much.

Clayton Handleman's picture
Clayton Handleman on March 21, 2016

“That being said, light-duty vehicles are responsible for less than 15% of global CO2 emissions and this number will further decline with improvements in user habits and efficiency, so even a perfect BEV future is not going to save the world.”

OK, then why did you write the article?

Schalk Cloete's picture
Schalk Cloete on March 21, 2016

Good question. It was primarily a lead up to the second part where I argue for the environmentally sound socio-economic development value of small electric vehicles. I just don’t think the capital-heavy wind-solar-BEV pathway is compatible with the billions of global citizens that still need to increase their standard of living 10-fold before we can realistically expect them to pay for climate change mitigation. Very cheap and efficient personal transportation is essential for achieving the rapid economic development required and I think that small electric vehicles in collaboration with efficient ICE cars/trucks will serve us better here than BEVs.

Clayton Handleman's picture
Clayton Handleman on March 20, 2016

Great, will read this weekend. My knee jerk reaction is it may work for the developing world where they are not already accustomed to bigger cars but will be a very hard sell in the OECD countries.

Bob Meinetz's picture
Bob Meinetz on March 20, 2016

Schalk, there are so many faulty assumptions in Part 1: “No”, should I wait until you address them in Part 2: “Yes”?

Although syntactically, I’m not sure how battery electrics could disrupt the internal combustion engine anyway.

Schalk Cloete's picture
Schalk Cloete on March 20, 2016

Sure Bob. Fire away. Part 2 has a bit of a different angle, so it is best to discuss all my faulty assumptions here.

About the syntax, I get your point. But I’ve seen the term “battery electrics” used before and it seemed like a good term for all forms of motorized transport relying only on an electric motor powered by a battery (no ICE required).

Erich J. Knight's picture
Erich J. Knight on March 26, 2016

The light duty market will do the best, however companies like Oakridge Global Energy Solutions, Inc. (OGES), have signed on Tractor-Trailer companies with fast charging/500 mile range batteries.

Engineer- Poet's picture
Engineer- Poet on March 28, 2016

Don’t be a tease! When you bring such things up, at least give a relevant hotlink.

Erich J. Knight's picture
Erich J. Knight on March 28, 2016

Also on the medium duty market, Tesla is designing a pick-up truck.
Now I’m totally sold! (Because ya ain’t a F#@k without a truck).
I imagine it will be front wheel drive and without a rear differential will allow a very low load bed and better performance both loaded & particularly when not loaded with the majority of weight over the drive wheels.

Mark Heslep's picture
Mark Heslep on March 26, 2016

“..Gasoline refueling occurs at a rate of chemical energy transfer through the pump outlet of about 10 MW…”

Yes. At the US pump standard of 10 gal per minute I calculate nearly 20 MW, so that a large highway-side station with 36 pumps requires 720 MW of stationcapacity, or perhaps half as much capacity provides equivalent mileage per refill time for the more efficient BEVs. The highway filling station implausibly becomes a good sized power plant in the all-BEV future. Also, the vehicle-charger connection moves outside of any plausible conducting cable solution at these power levels.

Some of the usual objections:

“Most charging will be done at home”. Yes in the US, but not so much on weekends and especially not on holidays, when most US auto travel simultaneously goes long distance, and not in the like of China where the millions live in 30 story high rises.

“Battery swap”. See the defunct Better Place.

Despite the difficulties with recharging electric vehicles, to me they seem trivial compared to the non-existant state of hydrogen distribution for FCVs. MIT forsees more FCVs than BEVs? I dont see how they smooth over mass distribution of 10 MW electrolyzers and compressors, or twenty 3K psi tube trucks to carry the energy of one standard gasoline tractor trailer.

Nathan Wilson's picture
Nathan Wilson on March 26, 2016

“All we have to do to start is make better use of the wires we’ve already got [for EV charging].”

There’s a big hump partway down the road. Getting chargers into the garages of EV owners (i.e. to support 5-20% EV penetration with 200+ mile range EVs) is much much easier than getting chargers into a substantial percentage of all the parking spacing in the country (to support higher penetration with sub-100 mile EVs).

I think one option that should seriously be discussed is the “Too Cheap to Meter Model” for public charging. In this concept, the $4000 internet-connected credit-card-reading Level II techno-chargers would be replaced with simpler 1kW safety outlets which could be cheap enough to provide in all workplace parking lots. Users would provide their own cords, and instead of paying a metered rate for electricity, they’d pay a fixed annual fee and display a window sticker. Some people would get $0.25 worth of electricity each day; others would get $1 worth (some would leave the A/C or heat on); but that’s something we’d ignore under this model.

If you add smarts to such a system (e.g. to provide metered billing, or demand-response for grid support), the whole thing gets more expensive, and frankly degrades the EV ownership experience, thus discouraging widespread acceptance. If a user plugs-in all day, but charging doesn’t happen because of unfavorable weather, that’s a bad outcome.

Note that Tesla’s Supercharger network uses the Too Cheap to Meter model, and is paid-for using a one-time fee as part of the vehicle purchase. Surely it would also have a mix of heavy and light users.

Engineer- Poet's picture
Engineer- Poet on March 26, 2016

There’s a big hump partway down the road.

Key phrase, “down the road”. This will only hit once the easily-converted existing assets have become fully subscribed.

Getting chargers into the garages of EV owners (i.e. to support 5-20% EV penetration with 200+ mile range EVs) is much much easier than getting chargers into a substantial percentage of all the parking spacing in the country (to support higher penetration with sub-100 mile EVs).

I don’t think it will be that difficult. Consider the parking lot that’s already wired for sodium lamps: a 240 VAC 30 A circuit for every few poles. Convert the lamps to LED and the bulk of the circuit capacity is surplus. Leave the circuit powered during business hours and switch the lights by carrier-current command (e.g. X10) and each light circuit can charge 4 EVs at Level 1 rates of 120 VAC 12 A (possibly dropping to 10 A when the lights are on). Power already goes to poles in the parking lot, all it needs is outlets/cords for the last yard.

Once you’ve got enough subscribers to fill e.g. 20 charger-equipped spaces in a 200-space parking lot, you have enough demand to justify a dedicated infrastructure effort without having made a major outlay to get the ball rolling. The whole chicken/egg problem is avoided. There may even be a revenue stream if DSM can be rolled into the first implementation and aggregated by a company selling grid regulation services.

I think one option that should seriously be discussed is the “Too Cheap to Meter Model” for public charging.

$CITY next to me has this model for their ChargePoint units; they put a heap of PV panels on top of one parking structure and the electricity is provided to EVs gratis. This may have been paid for by a DOE or NREL grant. However, these ChargePoint things are big, fancy (vacuum-fluorescent status displays) and no doubt expensive. It should be possible to make a J1772 unit much cheaper than anything on the market.

There are going to be problems with the “safety outlet” model. What do you do if you have a transient ground-fault condition? How do you manage overloads; do you automatically reset circuit breakers? How do you protect customer gear from theft or vandalism? What about demand-side management and the revenue stream you lose if you can’t support it? There’s a bunch of stuff there that is going to affect acceptance and uptake. The J1772 connector on a cable costs more, but it is a known quantity.

Nathan Wilson's picture
Nathan Wilson on March 28, 2016

I still feel like you’re proposing a complicated and expensive set of infrastructure, where something cheap would be adequate in most cases. I don’t see why the cord needs to cost more than $10; let the customer lock it to their car.

I have not seen a rigorous analysis of consumer demand side management (DMS), but my gut feel is that it does not have much to contribute. I think there is value to incentivising people not to do evening fast charging, but for 1kW (“level 1” slow) charging, the grid should simply grow to accomodate it (as a public good). I suspect that as long as we have a big chunk of our power coming from fossil gas, hydro, or grid energy storage (or we have power going to fuel synthesis), the claimed benefits for DSM for frequency regulation (and other services that can be delivered without increasing charging time very much) are probably are not real (i.e. such a revue stream, if it were mandated into existence, would not elimiante any grid infrastructure that was not needed for other purposes).

…problems with the “safety outlet” model:

the management is not responsible for loss or damage to customer equipment.
the management is not responsible for failure to charge due to equipment problems.
if the outlet doesn’t work ( e.g. breaker trips), send a text message to the number on the label, and someone will fix it in a few weeks.

Ubiquitous charging can’t be upscale charging. There will still be room in the marketplace for a few Level 2 chargers in public places. But charges that are worth more than the average car will never serve more than a small fraction of all cars (remember that cars last 15-20 years, a no car that is over 8 years old is worth more than a couple thousand dollars).

Engineer- Poet's picture
Engineer- Poet on March 28, 2016

I still feel like you’re proposing a complicated and expensive set of infrastructure, where something cheap would be adequate in most cases.

High-volume electronics are extremely cheap; look at microwave ovens.  The major infrastructure costs are for digging, laying conduit, pulling wire and re-paving over cuts.  This is most easily done en masse when things are being torn up for other reasons.  If you can re-purpose existing wire in already-laid conduit to charge vehicles, you’ve immediately cut out the majority of the cost of infrastructure for just about anything that isn’t just bolted to an existing electric pole.

I don’t see why the cord needs to cost more than $10

You need vehicle detection (so you aren’t applying power to something other than a vehicle), ground-fault detection, overload interruption, preferably reclosure after faults… generally the basics of the J1772 feature set.  Rather than have two interfaces, car-to-cable and cable-to-fixed (the latter which doesn’t have a standard yet), it is going to be much quicker and less duplicative to have just one.

Ubiquitous charging can’t be upscale charging.

You’re defining “upscale” differently from me.  J1772 is a glorified ground-fault interruptor; if you read the specs you’ll see it’s a very well thought-out but very basic protocol that supports a few features that can be leveraged.  Most of this leverage takes only software, and software is a good with extremely low unit cost.

The basic “brain” of J1772 EVSE is a single-board computer with a few sensors, one PWM output and one A/D input per charging connector, and its associated power supply (also a small circuit board); status displays can be as simple as a single multi-color LED.  One board can control several charging connectors.  The microcontrollers available today are so powerful that a great deal of intelligence can be packed into the control of those outputs, and they’re so cheap the capability is close to free.  We might as well take advantage.  If all you do is build a unit that “speaks” J1772 to vehicles and “speaks” something like BSR X10 to whatever’s at the breaker box, you’ve got more than a charger: you have an applications framework.

charges that are worth more than the average car will never serve more than a small fraction of all cars

Aside from weatherproof enclosures, I expect such things to make the average microwave oven look up-scale.  No high-voltage supplies, no magnetrons, no motors, no moving parts save the relay contacts… says “cheap” to me.  Just order them a million at a time and see how the quotes come in.

Clayton Handleman's picture
Clayton Handleman on March 26, 2016

“If you add smarts to such a system (e.g. to provide metered billing, or demand-response for grid support), the whole thing gets more expensive, and frankly degrades the EV ownership experience, thus discouraging widespread acceptance. ”

In the 35 year timeline that is being discussed it is hard for me to imagine that there will be any sub 100 mile batteries as Li-experience curve will bring down costs so much. But if there are, the experience need not be cumbersome. Look no further than the FastLane transponders used to manage tolls. It is transactionally very similar. It is being implimented at low cost and minimal inconvenience to the driver.

BTW, if cars can be made to drive themselves, they undoubtedly can plug themselves in. I think that will take all of the pain out of this.

Nathan Wilson's picture
Nathan Wilson on March 26, 2016

Yes, I agree that with plausible reductions in battery cost, BEVs with 200+ mile range will be much more popular than Leaf-class vehicles. Even if public chargers become ubiquitous, the complete unsuitability of the short-range EVs for road trips will sharply reduce the perception of “freedom of travel” that cars normally provide. I think road trips start getting viable when a car (with fast charging) can drive 100 miles between stops and still have a 50+ mile reserve (for an old vehicle).

But even with a 200 mile battery, I think users will expect that after being plugged-in for 6+ hours, they’ll have a full charge (that’s how phones etc work), unless it’s marketed as a special low-tier Too Cheap To Meter service (backed-up by gas station style fast chargers). Conversely, users with bigger batteries may decide that for the sake of convenience, they won’t bother plugging-in as often. This reduces the number of public chargers that society must buy and maintain.

Regarding the comparison to electronic toll readers, note that one reader can serve thousands of cars per day. A metered charger can serve only one. Similarly, cell-towers and wifi hotspots are also multi-user systems, with very low cost to serve the incremental user. Transporting information is much easier than anything else.

Did you see the video of Tesla’s prototype robotic charging tentacle? It’s creepy and expensive looking.

Willem Post's picture
Willem Post on March 28, 2016

EP,

Here are some numbers of interest.

TESLA MODEL S EXPERIENCE

Travel, miles/y…………….20000…………..20000

Energy from battery………0.29……………….0.29

Energy charged……………..5800…………….5800

Vampire loss…………………..847………………847 @ 8 mile/d

Charging loss………………….997………………997 @ 15%

Total through meter……….7644…………….7644

Elect. rate, $/kWh…………..0.20……………..0.10

Elect. Cost, $…………………1529………………764

Gasoline, $/gal……………….2.50……………..3.50

Gasoline, gal/y………………..612………………218, equivalent

Mileage, MPGe………………32.7……………..91.6

If electric rates are high and gas prices are low, BEVs are not a good deal.

Bob Meinetz's picture
Bob Meinetz on March 28, 2016

Willem, what’s the source of your data?

Engineer- Poet's picture
Engineer- Poet on March 28, 2016

At the US pump standard of 10 gal per minute I calculate nearly 20 MW, so that a large highway-side station with 36 pumps requires 720 MW of stationcapacity, or perhaps half as much capacity provides equivalent mileage per refill time for the more efficient BEVs.

That doesn’t account for the conversion losses which are upstream of the charging station.

A Tesla allegedly uses about 380 Wh/mi. Using Nate’s figure of 20 vehicles per lane-mile, 4 lanes, and 20 miles of road per station for traffic moving 70 MPH, I get 20 v/lane-mi * 4 lane * 20 mi * 70 MPH * .38 kWh/v-mi = 42.6 MW per station. That’s a lot, but a far cry from 720 MW.
Then there’s the detail that almost everyone will be leaving home with a full battery, rather than stopping to fill up at the freeway. Sometimes, these problems are a whole lot smaller than they look at first blush.

Nathan Wilson's picture
Nathan Wilson on March 28, 2016

720 MW for a BEV service station sounds like a lot.

I’ll try from a different angle. This reference says each freeway lane holds 20 vehicles per mile, in uncongested conditions. If each vehicle uses 20 kWatts on average, and the road has 4 lanes (both directions combined), how much power must be provided to a service stop that handles 20 miles of roadway?

P= 4 lanes * 20 cars/mile/lane * 20 miles * 20kWatt/car = 32 MWatts

If the charging station has a 100 kWatt connection for each car, then it needs 32MW/100kW = 320 chargers!

This is quite different than the gasoline dominated world, but it is still plausible.

I agree H2 FCVs seem unlikely. The most compelling argument for me is that fuel synthesis provides the most satisfying fix for the load balancing problem of zero-carbon grids.

So rather than hoping consumers will buy the H2 (which is a bad fit for cars), I say synthesize ammonia (NH3) instead, and use it for electric peaking and long-haul trucking as a diesel fuel replacement. Also, since the total annual diesel energy demand is smaller than the electric grid demand, we’ll need to push down the need for load balancing by having plenty of nuclear and energy storage on the grid.

Ammonia is a stinky & hazardous fuel, but truckers are professionals, so we can teach them to wear goggles while refueling, so the safety will be fine. Ammonia has triple the energy density of 5000 psi hydrogen; this means that we can deliver it to stations by truck. Plus, with refrigeration, it can be stored in warehouse-sized unpressurized tanks, so it works well for seasonal energy storage.

Mark Heslep's picture
Mark Heslep on March 28, 2016

36 pumps x 10 gallons per minute x 119 MJ per gallon is indeed a lot.

The notional charging station calculation I laid out was for the *peak* power required with all chargers simultaneously in use, as was the multi-megawatt, single gasoline pump equivalent peak power figure referenced from the MIT report (thank you for the pointer). I also adopted the refuel/recharge time assumption MIT cites, that mass market drivers expect the process to take no more than a couple minutes, not 50 minutes to top off an 80 kWh battery from a 100 kW source.

Though *average* power would be considerably lower for a multi-charger station as you indicate, I don’t see how to escape the power requirements in an attempt to replace the large chemical energy delivery rate in existing liquid fueling stations (corrected for the greater efficiency of EVs).

UK filling station, 36 pumps

Willem Post's picture
Willem Post on March 28, 2016

Bob,

The Tesla Q&A site.

Here are the revised numbers:

TESLA MODEL S EXPERIENCE

Travel, miles/y…………….20000…………..20000, assumed

Energy from battery………..0.29……………..0.29

Energy charged……………..5800…………….5800

Vampire loss…………………..847………………847 @ 8 mile/d

Charging loss………………….997………………997 @ 15%

Total through meter……….7644…………….7644

Elect. rate, $/kWh…………..0.20……………..0.10, assumed

Elect. cost, $/y……………….1529………………764

Elect. cost, $/mile…………0.076……………0.038

Gasoline, $/gal……………….2.50……………..3.50, assumed

Mileage, miles/gal……………..28………………..28, assumed

Gasoline, gal/y…………………714……………..714

Gasoline cost, $/y……………1786……………2500

Gasoline cost, $/mile……..0.089…………..0.125

If electric rates are high, and gasoline prices are low, EVs are not a good deal.

Bob Meinetz's picture
Bob Meinetz on March 28, 2016

Do you have a link? I’m not finding it.

Pages

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »