Prometheus Unbound

Custom Search

Project Overview

This EV conversion project will take a stock gasoline-powered car (a 1999 GMC Sonoma pickup) and convert it to an all-electric vehicle using lead-acid batteries. The vehicle will be a road-worthy (and road-legal) commuter car for short trips. Since the car is charged using electricity from a wall outlet instead of being fueled by gasoline, and since the electric system is highly efficient compared to a traditional internal combustion engine, the car will be cheaper to operate and more environmentally-friendly than a gasoline vehicle. In addition, the design of the electric drivetrain and propulsion system is such that maintenance costs are much lower than its internal combustion counterpart.

Project Goals

Plan

This vehicle has been in the works for several years. Fleshing out the initial details of the project occurred with the help of Bob Brant's excellent book Build Your Own Electric Vehicle. At that point, the basic plan was to get a used light truck, such as a Chevrolet S10 or Ford Ranger. This would have enough cargo area and heavy-duty suspension to handle the battery pack without modification, and the rear-wheel drive design would make attaching the electric motor simpler. Preliminary calculations showed that an electric motor of 20-30 horsepower would be adequate.

After doing some more complex calculations to determine the torque required versus the torque available for a given motor/vehicle combination at a given speed and gear ratio, and the battery pack size required to support the amperage draw represented by a given amount of torque, I determined that converting a truck was going to be more expensive than I had originally thought, due to the heavier stock weight and poorer aerodynamic properties. Instead, my ideal conversion vehicle became a compact car, such as a Hyundai Accent or Toyota Corolla with aftermarket suspension to handle the added weight of the batteries.

Many conversion enthusiasts regard small-car conversions like these as more difficult due to the constraints imposed by the size of the car, and potentially more expensive if sealed batteries are used to fit into oddly-shaped spaces, instead of the much cheaper flooded batteries. However, I calculated that cost savings could be a third or more of the total cost of the project due to the smaller size of the conversion, and battery replacement costs would be likewise significantly lowered. I could still use a flooded-battery design if I removed the rear seat and installed a battery compartment in its place, instead of splitting the pack between the trunk and engine compartment like I had seen other hobbyists do. (Since the battery pack needs to be insulated and heated in winter, vented of hydrogen when charging, and routinely maintained, keeping it contiguous simplifies the pack design. A contiguous pack also reduces the costs of heavy wiring by shortening the total distance, and reduces resistive losses in the wiring, making the vehicle more efficient.)

After learning about aeromodding, I decided that range and efficiency could be improved even further with some aerodynamic additions, and that this vehicle could potentially be highway-capable. This breed of electric vehicle will never be particularly light due to the constraints of the lead-acid battery pack (1200 to 1600 lbs), and that combined with the relatively low power output of the electric motor sharply limits high-speed acceleration, but low aerodynamic drag combined with improvements in rolling resistance (low-rolling-resistance tires, worn-in drivetrain, manual transmission with no engine braking) and the lack of fuel consumption at idle could make for a vehicle with the capability to cruise at constant speed very efficiently. This efficiency in cruising also extends the rather poor range of the vehicle, necessary because lead-acid batteries have only about 1% of the power density of gasoline.

Unfortunately, this particular conversion won't be a high-performance compact car. On Dec 28th, 2008, we bought a 1999 GMC Sonoma to use in the conversion. Vehicle cost was a bit under $2000. The reason for buying this vehicle was simple: we needed a truck. It came with a functioning but elderly (160,000 mi) engine in it, which gave us the opportunity to use it for a while before we started the conversion in earnest. Aeromodding is possible with a pickup truck, but the results won't be as good as with a car that starts with a lower frontal area and lower coefficient of drag, so the choice of a pickup will reduce total range as well as increasing cost. Since the primary purpose of this vehicle is as a commuter car for in-town driving, the reduced range shouldn't be a problem.

The increased weight of the truck frame requires a powerful motor. After looking at torque curves on quite a few different motors, I chose the 19.5 HP continuous (approximately 100 HP peak) Advanced DC FB1-4001A motor, the most powerful motor available from Advanced DC. Other EV hobbyists have used this motor successfully in light truck frames and had plenty of peak acceleration. A 144 volt system will yield the most power I can get with this motor.

I had looked at several controllers for this motor, and determined from reports of failure by other hobbyists that I should keep with the tried and true Curtis brand of controllers. The Curtis 1231C-8601 is the highest-capacity controller that Curtis makes, with 144 volt and 500 amp limits. This 500 amp limit would reduce my peak acceleration at lower speeds, and this particular controller is very expensive, so I decided to build my own, after determining that I could make a controller with 1000 amps maximum capacity with only about $300 worth of parts.

The amperage requirements of this motor require a battery pack with a capacity of 200 to 250 amp-hours at 144 volts for reasonable battery life with lead-acid batteries. Currently, I'm looking at 24x US 125 XC (about $3600 and 1600 lbs for 242 Ah), or 12x US 185HC XC (about $3400 and 1450 lbs for 220 Ah), but I'm still looking for bulk discounts and other pricing breaks that could change the cost-performance equation.

In addition to the major components mentioned above, I'll also need a few other key components: