Range, Batteries and Some Musings

One of the design goals of our conversion is to be able to get 70 to 80 miles on a charge. This will let us make most of the local trips we'd ever need to make in the CR-V. With a range of 70 miles, trips like picking someone up from the airport, or going to a client site in Washington PA will be possible, even without charging at the far end of the trip. We'd only need a gas car when taking longer trips like visiting our parents, where the drives are 80+ miles.

To achieve the 70 mile range, we'll be using between 35 and 38 large format Lithium Iron Phosphate (LiFePO4) batteries. These cells will be rated at 200 Ah, or amp-hours. The cells will operate nominally at 3.2 volts, between 3.4 volts at full charge and 2 volts when fully depleted. This configuration will provide between 22.4 and 24.3 Kilowatt-hours, or kWh of power. The math used here is simple. You take the nominal voltage x Ah x the number of cells and divide that number by 1000 to get kWh, so (3.2v x 200Ah x 38 cells)/1000 = 24.32 kWh.

These batteries are gaining a lot of popularity in the DIY community.  Their rated cycle life to 80% depth of discharge (DoD) is an order of magnitude (at least 10 times) higher than deep cycle or marine batteries.  Lead acid batteries may get 200 cycles before they degrade to the point where they have less than 80% of their rated capacity.  Once they reach this point, my own experiences show they degrade even faster.  The LiFePO4's coming out now also have Yttrium as part of the chemistry (sometimes seen as LiFeYPO4) which has improved the already great cycle life of LiFePO4.  With the newer LiFeYPO4's by companies like Winston Battery (aka Thundersky), now you can expect to get 3000 cycles to 80% DoD before the battery degrades to 80% of it's rated capacity.  If you build your car with enough reserve capacity, these batteries should provide ten years of daily driving before wear starts to be dramatically apparent.

Lithium also has a much higher specific energy than lead acid.  Specific energy is the amount of power that can be stored per unit of mass, usually measured in Wh/kg.  Think of it as 'power to weight ratio', although that term technically means something else.  A lead acid battery capable of storing 24 kWh, which is the capacity of our target battery would weigh over 1800 pounds.  Our lithium battery will weigh only about 610 pounds.  The lighter battery helps to stay within the GVWR of the vehicle.  GVWR or gross vehicle weight rating is basically the weight that the vehicle was designed to carry added to the weight of the vehicle itself.  Adding one ton of weight, as would be the case of using a lead acid battery, would not be a safe modification to a vehicle.  Plus, the added weight would disrupt the ride quality, reduce the range and impede the acceleration ability of the car.  Lithium batteries help make medium range battery electric vehicles (BEVs) possible.

LiFePO4 is also a very safe chemistry.  If you read any of the tech sheets that happen to be published in English, such as http://www.batteryspace.com/prod-specs/5493.pdf, you can see that they are quite safe to use.  Even if overcharged, physically destroyed or even shorted out, they will not catch fire or explode.   Now, Tesla is using either Lithium-Nickel or Lithium-Cobalt cells in their Roadster.  The specific energy of batteries like these can be much higher, but the downside is they are much more dangerous and require sophisticated engineering to use safely.  This totally rules them out for DIY-class projects.

Now, back to range.  When we drive a gas car, we think in terms of miles per gallon. When driving an electric vehicle, we use the metric of miles per kWh. A common figure achieved in an AC driven car conversions is around 4 miles per kWh. Using this information we can estimate the range of our CR-V. Even if we use 3 mi/kWh as a pessimistic figure, the CR-V will still get almost 73 miles on a charge. If we can approach the magical figure of 4 mi/kWh, it would get an amazing 97 miles on a charge.

Using the comparison of MPG to mi/kWh, we can also calculate the cost to drive per mile. If the gas car is getting 27 MPG on average, and gas is $3.50 USD per gallon, it will cost 12.96 cents per mile of driving. This is calculated simply by dividing the price per gallon by the miles per gallon. In the electric CR-V, it will go between 3 and 4 miles per kWh. In my region, a kWh costs about 12 cents. To find the cents per mile figure, we'll divide $0.12 by 3 (and 4) to find that we'll be paying between 3 and 4 cents per mile. This means using gas to get around would cost between 3.2 and 4.3 times as much to drive a gas car. Put another way, for 10000 miles at 27 MPG at $3.50 gas, it will cost $1296 in gas, but it will only cost between $300 and $400 in electricity. Over the 10 year expected life of the car (and battery) this saves no less than $8900 assuming current prices.

You may say that gas and electricity will both go up.  That is very likely.  You might be interested to look at the information at http://www.eia.doe.gov/aer/txt/ptb0810.html , http://www1.eere.energy.gov/vehiclesandfuels/facts/2005/fcvt_fotw364.html , and http://www.randomuseless.info/gasprice/gasprice.html .  You will learn that while gas is fairly high right now, it has actually been on a downward slope in cost over the last 90 years.  Prices now are going up due to increases in global demand, political unrest, etc.  These factors simply don't affect electricity production since it is generally produced domestically, using whichever fuels are economically available domestically.  You'll also notice that inflation adjusted gasoline prices have been much more volatile during any time period than electricity has been.  Gasoline is currently trading at three times the inflation adjusted price of it's historical lows between 1960 to today.  Meanwhile, electricity has never varied more than 60% of its own inflation adjusted price in the same period, and now it is trading on the low side of the historical price range.

The Controller: Curtis 1238 Series

In my own independent research, I also chose a similar motor/controller combo as ChargeCar's setup.  Some of you know, I have been planning on building an EV for years and the reason that I started in 2009 with an electric bike because I didn't have the funds at the time to do a car. Plus I figured if I screwed up royally, it would be on a much smaller, and less costly scale. Fortunately the ebike project worked very well and got me 5500 miles or so before I needed to do any maintenance. And that maintenance was directly related to the fact that the original SLA battery was shot. I'm currently wrapping up the upgrade to LiFePO4 [which will last many years] and I'll be on the ebike again in the Spring for my daily 19 mile round trip commute.

I only found out about ChargeCar last summer. This was back when I was still going to convert my black Z24. At the time I was looking at the AC-30 / Curtis 1238 combo, but now that I decided on a CR-V, I figure the motor should be sized up a bit due to the heavier weighted vehicle, but the controller's still fine. When I found out ChargeCar was going to use the same controller and motor that I planned to use, I thought it was cool. It kind of gave me a lot of confidence that I had done my homework, and done it correctly :)

Now, let's talk about this controller [finally, right?]. This controller has a lot of awesome features. The coolest is that it's fully programmable using Curtis's Vehicle Control Language (VCL). Not only does it let you set parameters for how to control current, it also can handle two way communication, so if you want, you can program the controller to make decisions based on internal or external sensor data, GPS information, dash mounted tablet pc... and more [with some programming and electronics work of course.] In some future posts, I'll talk about some ideas I have for this, but suffice it to say that this car could provide a decade of entertainment for me as I refine and invent new things to do with this functionality.

Of course, the controller offers regenerative braking, which converts the kinetic energy [the motion of the car] back to electricity and charges the battery, extending your range. Normally, this kinetic energy is transferred by your brake pads to your rotors for dissipation as waste heat. The controller is also quite efficient at powering the motor which will make the car go even further on a charge. We are expecting to get between 60 (on the low end, with A/C or heat running) and 85 miles per charge with this system. This is similar to the real life expected range of the Nissan Leaf.

You can read the data sheet on the 1238R controller if you like. It gives a lot more basic information that I don't really feel like regurgitating here. Read it though, because it is worth checking out.

In my next post I'll talk about the battery we plan to use in the CR-V.