EV economics updated and a Catch-22

After posting my December 6 article about EVs Economics are getting interesting I’ve received numerous comments and I’ve had discussions with utility executives and board members.   Based on this input I’ve refined the economic analysis of the Leaf vs. Camry and I’ve addressed a potential regulatory Catch-22 concern that utilities might run up against if they aggressively go after the EV market.


The more I discuss EV usage, and considering how I would use an EV, I’m increasingly convinced the 100 mile class of EV will be used like a cell phone.   At the start of the day, the EV is unplugged and driven.  At night, the car is parked in or near its home garage and charged up.  The whole discussion about public charging, or changing out batteries, will be irrelevant.  These vehicles will be short range around town cars.  For drivers that go on long trips the EV won’t be used, the owner’s other, gas power car, will be.    Range anxiety can be addressed when necessary with a little high cost topping off from a 120 VAC outlet at the destination.

Since the car will be charged at night this represents an opportunity for utilities to market power in a new way.  Namely the utility can offer to sell electricity via a 220 VAC outlet at very low cost during the night off-peak power block period.   During the rest of the day, power would be unavailable from the 220 VAC outlet and a customer would have to rely on 120 VAC charging.  This avoids potential overloading of distribution transformers and aligns a cheap tariff with cheap power while placing no cost burden on other utility customers.

As part of updating the economics I double checked wholesale market outlook (thanks to www.weccterm.com) and found the outlook continues for very low off-peak prices that easily allow provision of electricity to the EV at 5 cents/kWh and allows for some utility margin:

Quarterly forecast prepared 1/18/2010, Off-peak prices

period        NP15 (Northern California)
2011-1      2.9 cents/kWh
2011-2      2.5 cents/kWh
2011-3      3.6 cents/kWh
2011-4      3.7 cents/kWh

Previously I used 15,000 miles per year as the average annual mileage driven by Americans in a year.  While the DOE/EPA Model Year 2011 Fuel Economy Guide bases its annual fuel costs on 15,000 mile per year, EPA’s transportation and air quality group  peg the average miles driven at 12,000 miles per year.  And the Federal Highway Administration shows drivers in the 20 – 54 age range averaging over 15,000 per year.  Taking this all in I’ve decided to conservatively base the average analysis on 12,000 miles per year.

Some comments noted that EV will have lower maintenance costs than gas power cars.  I think it’s fair to credit the EV with avoided oil changes.  This isn’t a big factor but does improve the EV economics a bit.

I also used an estimated 10 cents/kWh for a nationwide average for retail electric prices.   For the average charging analysis I’m now using 11.5 cents/ kWh which according to the EIA is the actual August 2010 nationwide average.

In my first analysis I included 10% losses in the charging equipment because I was in a hurry.  This is probably a little high and I revised the loss figure to 5% consistent with losses in a couple of thyristors.

Previously I based my analysis on $3.50 gasoline.  This still seems a fair estimate and I’ve continued to use it.   Of course, if gasoline prices spike EVs will get a boost.

All these changes taken together erode the EV economics a bit but not enough to change my previous conclusion that EVs can be a hit given some creative utility rates.  But at my current rates, I’ll wait on the EV, sigh.

Scenario                 Break-even years      IRR at 96,000 miles (8 years)
1  (5 cent/kWh)                 4.6                                14 %
2  (11.5 cent/kWh)            5.6                                  9%
3  (17.6* cent/kWh)          6.9                                 4%

* this is what I pay today to SMUD for each incremental kWh


The Catch-22

Many utilities, and certainly those in California, are facing Renewable Portfolio Standard (RPS) requirements.  So if a utility added significant new load, say 50,000 EVs charging at 3kW at night, the utility would need to provide an additional 150 MW of power.  In all likelihood this power would come from fossil sources, at least for some period of time.   In the western US the marginal generating resource at night is almost always very efficient natural gas fired combined-cycle power (thank you again www.weccterm.com) .  Essentially we would be running domestic natural gas through a high efficiency conversion and displacing imported crude or gasoline.  But a utility may be penalized by RPS requirements for this very sensible activity – the Catch-22.

The RPS standards have been enacted in large part to address climate change resulting from burning fossil fuels.  To test whether the RPS standards are counter productive to their own purpose in the case of EV charging, I dove into carbon calculating.

I first calculated the annual pounds of CO2 the Camry would produce.  Using EPA mileage, 12,000 miles per year, and the EPA’s CO2/gallon of gasoline figure, I computed the Camry would produce 9,502 lb/CO2 per year.

Next, to calculate the Leaf’s CO2 production, I adjusted the Leaf’s kWh consumption back to the generation level by adding back transmission system losses.   Then I determined the amount of natural gas consumed using night time combined cycle heat rates of 7,500 Btu/kWh.  Finally I applied EPA’s latest C-factors (including the 2 oxygens) for natural gas to compute the CO2 produced. The result: 4,049 lb/CO2 produced per year, less then one-half that produced by a gasoline engine.

This result makes sense: (1) combined cycle power plants, even after transmission and charging losses are really efficient, and (2) natural gas produces a lot less CO2 than an equivalent amount of gasoline.

The conclusion is straightforward, EV charging should be exempt from RPS requirements and the EPA should be gung-ho for EV charging.   And at the end of the day I don’t see any way a utility will ultimately be penalized for encouraging EVs.

6 replies
  1. Bilsko
    Bilsko says:

    One question on the Charging Voltage issue. My understanding is that, to the utility, the difference between supplying 120VAC and 220VAC power is negligible. All of that transformation is going to be taken care of behind the meter or at the service drop.
    Remember, we get 220V power in our homes today with a 'two-phase' 110/120V line from the service drop – you just create 220V by running the two 110V completely out-of-phase in order to create the 220V circuit.
    Even if the utility did provide a dedicated 220V drop to the customer, the power would still be sent over the primary feeders at voltages anywhere from a couple kV up to a dozen or so kV. So as far as the utilitiy is concerned – the voltage difference (even when magnified over 50k users) would seem pretty negligible.

    • Mark Henwood
      Mark Henwood says:

      A typical utility would deliver something like 12kV to a street and then step it down to 220/110 single phase for delivery into residences.

      Many newer residences will have single phase 200 amp 220/110 VAC services. If the existing service is 200 amp this should be sufficient to supply the EV charger and the rest of the nightime load. The service will provide 200*220/1000 = 4.4 kW of power and if the owner isn't running anything big at night this should work.

  2. Andrew Neilson
    Andrew Neilson says:

    Thanks for the insights and the helpful data. Agree that home-based charging makes sense. Also agree that using CCGT is a good start. Even better – homeowners could install a natural gas powered fuel cell generator and then generate low emission power for their home and car, 24/7/365. Ceramic Fuel Cells has installed its BlueGen 2kW co-gen system to power a free EV charging station in Adelaide, Australia. Peak electrical efficiency of 60 percent, no distribution losses, plus hot water for the home. Home generator could be bundled with a EV and sold to homes as a "home energy solution"…

    • Mark Henwood
      Mark Henwood says:

      I'm not sure about the overall economics of a home based generation system. Almost certainly this solution will have a big headwind due to how natural gas is priced at wholesale and at retail in the US. Large scale generators, like CCGT, typically connect to the gas system for bulk deliveries. Consequently their gas cost, in the US at least, is something like Henry Hub plus a transmission adder. Yesterday (1/31/2011) in the area where I live the PG&E Citygate price was $4.42/mmBtu which is a blend of the Henry Hub (LA) price and the Canadian border price at Sumas. To move the gas to the powerplant requires only another $0.18/mmBtu for a burnertip price of $4.60/mmBtu (data courtesy of http://www.weccterm.com) . At home, I pay a whopping $12.40/mmBtu which is 2.7 times the price paid by a CCGT in our area. Even at 60% efficiency this price differential strongly favors just buying power from a CCGT…that is assuming a utility will flow this through the brand new EV load.

  3. Felix Hoenikker
    Felix Hoenikker says:

    Great article and explanation of analysis. Another way I would be interested in seeing your analysis presented would be to have one chart with gasoline $/gal on one axis and battery cost $/kWh hour on the other axis and each curve representing a different payback target 1,2,3,4 years etc. Keep up the good work!

Trackbacks & Pingbacks

  1. […] This post was mentioned on Twitter by Neal Dikeman, clean_tech_org. clean_tech_org said: 【Blog Comments】 Comment on EV economics updated and a Catch-22 by Andrew Neilson: Thanks for the insights and th… http://bit.ly/ehkId9 […]

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply