EnviroKarma.org - Electric Vehicle (EV) Conversion '...it's (NOT) hard to be Green' - what Kermit should have said

 

Click to receive email when updates are posted to this page:
Powered by ChangeDetection.com

Defining project goals for an all-electric commuter vehicle

    In my particular case in Santa Fe, NM, my average daily commute needs are less than 20 miles per day... and I often don't use a car at all on summer days when I can use a bike+trailer for errands or bike+tag-along to haul kids around. Santa Fe is fairly 'rolling', and most city streets have 25 or 35mph speed limits. There are a few major arterials with 45-55 mph speed limits. In my case I also need to transport two children, so I need rear seats, and not just a small truck. Since 'smaller is better' when it comes to range and speed with electric motors, I focused my search for a small car as a donor vehicle.

    It's important to set a realistic range goal, as that plays into what kind of batteries you will use. A small car with 12 x 8volt lead-acid deep-cycle (Golf Cart) batteries will probably get a maximum of 40 mile range when new and in warm weather, whereas you could probably get a little more loading up a truck with 6 volt batteries. Its best to make the choice between small car and truck early on as it will determine the size of other components. Trucks are easier in some ways because they have higher load capacity and easy access for battery racks; but if you need to haul kids, you probably should use a car. Today's (2010) LiFePO4 large format prismatic cells can easily offer 50 to 90 mile range with various capacity cells just costing more... Turns out that the Lithium is a better deal than lead-acid in the long run, but a bigger investment up-front; 3x the cost for 4x or maybe 5x the life with better range and performance because of lower weight and less internal resistance.

Deciding on a donor vehicle

    I decided on 'small car' early on, and then added the parameters of 5 to 10 years old with a max purchase price of $5000, and hopefully less than $2000... I had trouble finding listings of curb weight and average trade-in prices until I picked my way through Kelly Blue Book. I made up a bit of a spreadsheet of my choices in preferred order and started watching the paper.As it turned out, more cheap local cars show up in Craigslist.org than the paper if you know what you are looking for... The important thing is to get something you won't mind driving for 10 years!

    Lucky me, I found a 10 year old Suzuki Swift (Geo Metro, Pontiac Firefly) for $500. Still running, still getting 40 mpg on gasoline... I bought it, registered and insured it, and drove around for about a month trying to burn through one tank of gas!

Battery technology - updated 8/2010

    There are lots of choices with batteries... Flooded Lead, Lead Gel Mat, NiCad, and various forms of Lithium with the most popular, affordable and available being the large format prismatic cells. I spend about 20 hours sifting through information on the web, writing the Li manufacturers for specs, asking questions in forums. My conclusion in 2007 was that Lead was more cost effective and still is the least expensive way to get started, although Li is less expensive over the life of the batteries (in 2010).

    There are lots of choices even when it comes to Lead, but the basic deal is you have to look at 'deep cycle' batteries which are designed to be 80% discharged and re-charged without damage. These are NOT Auto or Marine batteries, but are commonly used in Golf Carts. There are flooded cell, and Glass-mat, or Gel-mat designs; the flooded lead acid (FLA) gets you the most amp-hours for your money, but you have to maintain them, checking water levels, and be careful with the acid...

    I took a while to decide between 12 volt or 8 volt... I liked the idea of less weight and cost for the 12v battery system at the same pack energy. But, the smaller class 31 12v batteries had pretty low aHr ratings, and the total energy wasn't looking better than the (12)8v batteries. US battery offers a larger 12v sweeper battery that would give higher total pack energy with 12 batteries, but just would not fit in the space available in the Swift as they are about 3" longer than the 8v batteries. So, for FLA batteries, the most I could get into the Swift and retain the back seat was (12) 8v batteries having a footprint of 10.25x7.125x11.25 and 183 Ah at 20hr rate, or 95 minutes at 75 amps, which translates to about 100ah at a one-hour use rate which is more typical 'real' use with EVs.

    I did look very closely at the LiFePO4 batteries like Thundersky, Nilar, and some others during 2007 and 2008 while researching my project. I was even ready to pay a lot more because of the reduced size, weight, and longer life. The problems were that nobody could really tell me exactly how much longer the life would be, and the BMS required to manage the batteries is not available 'off the shelf' in 2007. Things have changed fast, and now (in 2010) prices on the Li cells have come down, more people have been using them so they are getting to be more of a standard if you can ante up the initial cost. The advantages of Li are great... 1/3 the weight, 4x or 5x the life, very low internal resistance and almost no voltage sag during normal use.

    (8/2010) - I have revised my cost analysis to fit actual use parameters better now that I have some experience with using a pack of FLA through their life cycle. To compare apples to apples I think the best way is to come up with the cost/mile over the life of the battery. Both FLA and Li have longest life if 'normal use' doesn't exceed 50% Depth of Discharge (DOD) per cycle and average amp draw is no more than 2C or 3C for brief accelerations. Comparable capacities in my case would be the 8v FLA, and the 100ah LiFePO4 cells.

    If you start with the basic parameter of a 96 volt system using 8v batteries like the usbattery 8vgchcxc , which has a 1-hour a rating of a little over 100ah. This would give a total pack capacity of about 12 x 8v x 100 ah = 9.6kWhr and the cost (delivered) would be around $1600. They can be expected to last 500-700 cycles at 50% depth of discharge with good maintenance before they start losing capacity. FLA don't die completely all of a sudden, but the capacity (range) does start dropping noticibly after 18-24 months around 20%-30% less than when new. At 50% DOD I was getting about 20 mile range when new. This gives the raw lifetime cost/mile = $1600/(700*20) = $0.1143/mile . Keep in mind this is a rough number, and highly variable with how hard you push batteries, if you can continue driving as range drops, etc. Its possible these batteries would go for 1000 cycles although with reduced performance, which would bring the cost down to $0.08/mile. But I am now convinced you also need to add a one-time 'extra' for a centralized watering system, or you go crazy with maintenance.... So add another $400 or so to the lead-acid equation for the initial build for a watering system, but this is comparible with what you might need to allocate for a BMS with Lithium over its life. If you are lucky, you might even be able to sell 'dying' floodies to an off-grid PV person for $20 each, or salvage yard for $7-$10 each.

    Working through the numbers for Li has gotten very interesting... life cycles are still somewhat unknown as people have not really been using them in EVs long enough to prove they are going to last closer to 3000 or 5000 cycles; but for my purposes I feel pretty safe using the more conservative number of 3000 cycles at 50% DOD, which would get the same car a little further because of the lower weight and internal resistance, probably closer to 25 miles at 50% DOD which gives (conservative) lifetime miles of 75000 at a current cost around $4500 for 30 x 100ah cells (96v nominal in this example) delivered. So the bottom line is lifetime cost/mile = $4500/75000 = $0.06/mile, or if you believe spec sheets these batteries may last 5000 cycles giving a cost of $0.036/mile! Lots of people think you need to have a $1000-$2000 BatteryManagemetSystem (BMS) to keep the Li cells balanced and prevent over/under charging, but I think the jury is still out on that topic.

    How does this compare to gasoline? Well, 'Zappy' used to get 40mpg. Using gas price of about $2.75/gal gives $0.0688/mile ....and the price of gas will likely return to the $3/gallon or above level before too long. To compare to electric you have to add in the cost of electricity. My little car uses about 200 Whr/mile when running on lead, but closer to 400whr/mile 'at the plug' with charging losses since there is a built-in over-voltage gassing cycle; at current retail price of $0.09/kWhr in NM, thats only an extra $0.018 per mile added to the cost of the batteries. To be really fair, you'd have to add in the considerable 'other' costs for a gasoline engine such as the oil changes ($800 over 60,000 = $.013/mile) and average maintenance averaging $1000 ($.016/mile) for the replacement of serpentine belt, filters, plugs, etc. that electric motors simply do not require. The cost of electricity on Lithium I expect to be lower both because of lower vehicle weight, and because there is no energy wasted in a 'gassing' or balancing cycle at the end of charge cycles the energy used at the plug will be much closer to that measured at the motor.

    I'll update the electric use number on lithium after I gather some more data, but estimate is lowering the electric cost from .018 (on lead) to .015 or less per mile on lithium.

    As of 8/2010 my rough numbers for amortised battery cost plus electricity are that an Electric car (a 'small' one like my Swift/Metro) w/
    FLA Lead: $.08-.114 + .018 = $.098 to .132/mi batteries+elec,
    LiFePO4: $.036-.06 + .015 = $.051-.075/mi batteries+elec,
    gasoline: $.07/mi gas + .03/mile for oil and maint = .10/mi (at $2.75/gal w/ 40mpg)

    So you can see that while the life-cost of a FLA build is close to the operational cost of gasoline (at $2.75/gallon), Li actually makes financial sense on top of all the usual ecological 'feel good' benefits (like -19.6 pounds of CO2 released per gallon of gasoline!) even with today's artificially low subsidized cost of gasoline. I researched possible Vendors for quite a while, got burned by non-delivery from EVComponents.com (still in litigation), and finally re-ordered 38 x Thundersky 100ah cells from EVolveElectrics.com; they delivered reasonably on-time from stock in the US. Another solid dealer is a new company run by Dave Kois at www.currentevtech.com, but they may not have stock in the US at any given moment, so you have to contact them to find out current stock...

Motor technology - AC or DC?

    There are pros and cons to AC vs DC... AC motors give you regen braking to recover *some* energy when braking but cost considerably more in part because you also need a more expensive inverter/controller. For example, a 8" brushed DC motor with modest 400amp controller might run around $2500 or less, where a comparible AC motor+controller would be closer to $4500. On the plus side, with an AC motor you can get 'regenerative braking' basically built-in to some degree that can extend range in city traffic and mountainous terrian. Realistically, you might extend your range about 10%-20% in average conditions, but add lots to the cost of the electronics and complexity of the system. Just a rough estimate at the electricity 'saved' with regen indicates on the order of 20 years to pay back the difference in cost between DC and AC. If long range is needed, it is more cost effective to buy more batteries at this point. As a first-time home DIY guy, I decided to stick with the simple inexpensive DC solution, at least for now.

Putting together a 'shopping list'

    There are several great suppliers of EV kits... which I have listed in the Links page on this site. After poking around the web for a while, looking at some other EVs online, looking through the Kit catalogs... I settled on using a kit based on the popular 8" Advanced DC motor; seemed to be a good match for my donor car, range, and highly reliable. Once the motor and battery pack voltage is selected, the EV kit people can pick matching controller, charger, and other important components.

    I was orginally considering a 120 or even 144 volt system, but opted for 12 x 8volt batteries to get a little more range than I would have gotten with 10 x 12v class 31 batteries, and it was all I could physically fit into the Swift. With a 96 volt system you can save quite a bit by using lighter duty controller and charger. Price between suppliers will vary as they include slightly different components. I ended up going with KTA Services and got all my 'stuff' for a grand total of $4956 including S&H, plus another $1096 for the adaptor plate and heavy duty rear springs from CanEV.

    The major components include:

    • 12 x US battery 8vgchcx 8 volt deep cycle batteries
      this is all you can fit in a swift without giving up cargo space or back seat.
      might be possible to go with smaller 12v for more zip, but you'd lose range.
      larger 12v are about 3" longer, and I really don't think they would fit.

    • 8" Advanced DC motor - 203-6-4001
      a tailshaft WOULD fit if you wanted to try regen or use it for AC,
      but a 9" motor would NOT fit without some re-fabrication of the mid transmission/motor/CV mount. It would be possible to chop off the leg that used to go to the ICE engine block, but then you'd need a different place to mount the vacuum system, and support the CV half-shaft bearing... better to stick with 8" motor.

    • Curtis 1221c-7401 controller

    • Zivan NG-1 charger

    • KTA 'complete' vacuum system kit

    • 1540 watt solid state heater core

    • Curtis 1400e72/96-1201 DC/DC convertor

    • pre-fab motor adaptor plate and heavy duty rear springs from CanEV.com
      in retrospect, the heavier rear springs are pretty optional... I haven't installed them yet, and the car sits pretty level, and doesn't bottom out even on the big speed humps that are all over town. I might return them, or sell them.

    • ...and lots of misc 'stuff', meters, connectors, etc

Total time for this phase was not really tracked, but I would guess I spent at least 50 hours poking around the net looking at various EV parts stores, batteries, and reading Forums. I'm not going to count this time in the construction total though...

 

Web Site Design and Maintenance By Emerging Technology Services

Disclaimer of Liability and Assumption of Risk: Working with high voltages in electric vehicles can be dangerous. Working with welding equipment or plasma torch or sawzall and other power tools necessary to fabricate battery racks or create holes necessary to lay the battery racks within an electric vehicle can be dangerous. The purchaser/builder solely assumes all risk. Further, the viewer of this site, and builder of the vehicle, agree to hold harmless Daniel T. Baker and any associates from all loss, liability, or damage resulting from any failures or defects in any project completed with these instructions. It is suggested that the builder follow common-sense safety procedures, such as the utilization of gloves where working with high voltages, and ear protection and goggles when working with power tools.