For many years, I have been interested in electric vehicles and gasoline-to-electric conversions. In November 2007, I attended a weeklong electric vehicle (EV) conversion workshop, where the class converted two Volkswagen Rabbits. After this, I became even more determined to find a way to put this technology to work for my family.
Our initial goal was to use an EV to drive to town (2 miles), run errands, then return. Speed limits along the route are 25, 30, and 45 mph. Although Washington state limits neighborhood electric vehicles to 35 mph, we could probably have sneaked by with one. However, eventually we decided that we wanted a car with enough range and speed to travel to the nearest large town, 36 miles away, and be able to recharge before returning. Speed limits there and back are 60 mph.
We considered converting our Ford Escape hybrid to a plug-in hybrid electric vehicle (PHEV), but the conversion would not allow “electric only” driving for the short distance to town and back. As with the stock hybrid, the PHEV would require the internal combustion engine (ICE) to run until the catalytic converter was warmed up. In winter, this often requires the entire trip to town and back. Unlike Toyota Prius PHEV conversions, which can use electric propulsion for up to 20 miles without starting the ICE, the Escape PHEV would require some use of fossil fuel for the short trip.
There are no production all-electric vehicles that would meet our needs and budget, so converting an ICE vehicle to electric drive was our only option. We decided to find a lightweight, relatively late-model car to convert—one with four doors and a more efficient standard transmission.
After considering several vehicles, we identified the Toyota Echo as the “donor” car. The Echo has the same body as the 2001 Prius—light and aerodynamic. A good donor vehicle weighs 2,000 to 2,500 pounds; the Echo’s curb weight is 2,055 pounds. In December 2007, we test-drove a 2002 Echo four-door sedan with a five-speed, manual transmission. It was a comfortable, quiet, peppy car, and we bought it on the spot.
I cleaned out the garage, built a new workbench, and put the Echo inside just as snow began to fall. Armed with the Echo’s official repair and electrical system manuals, I started planning my “winter project.”
One of the first things I did was sign up for the Seattle Electric Vehicle Association e-mail list, and sent an e-mail to the list with general plans, asking for suggestions. Overnight, I had replies to several of my questions. I found the Internet to be a terrific resource for researching EV questions and components, and I bought several components online, including the conversion kit.
In January 2008, my son Greg and I started the project by removing the ICE components and mapping out battery storage. While we waited for the conversion kit to be delivered, the deconstruction continued with removal of the power steering rack and the starter gear ring from the flywheel.
The kit arrived six months later—so much for the “winter project” idea—and we started assembling the EV components. Attaching the electric motor to the transmission and reinstalling it in the car just took a few hours, with help from friends, and we starting putting the car back together, installing a manual steering rack.
In September, I started planning battery placement. I downloaded a spec sheet for the battery I had chosen and made a cardboard battery model to test mounting options. I printed a scale drawing of the battery on my computer, pasted it to architectural foam board, cut it out, and then glued it together. From that, I drew up plans for the battery boxes—one would go beneath the rear seats and one would go under the trunk.
Fourteen Odyssey 12-volt AGM batteries were delivered in October. Thirteen were to be used (156 V nominal), with one spare. A friend helped cut out the floor pan under the rear seat and trunk to prepare for the battery boxes. The locally made battery boxes arrived in December, fitting the holes we made in the floor pan with only a little trimming on the floor pan. The batteries fit into the boxes perfectly, leaving a little room for battery expansion as they warm under charge and discharge.
Once the batteries were installed and secure, I started working on the wiring, running cabling from the batteries to the components in the cab of the car, and installing the Link 10 battery monitor. Once that was complete, I secured the rear seat battery box top and reinstalled the rear floor carpet, seat belts, seat backs, and seat cushion.
At the end of January, I tested the Link 10 and set it for my battery’s amp-hour capacity. I also tested the AC battery charger. With the new batteries at above 20% state of charge, the charger draws 3 to 9 amps AC.
Next, I installed the Azure digital motor controller (DMOC). In early February, I gave the car—still on jack stands—its first “test drive.” I switched the forward/reverse switch to forward, made sure the transmission was in neutral, and advanced the throttle. The motor turned! I put the car in first gear and slowly advanced the throttle again. The drive wheels turned in the right direction!
Greg and I took the car off the jack stands and checked the fender well height at the rear wheels. It was 1 inch lower than the stock height before the conversion. I added 12 psi to the rear tires (Air Lift recommends adding 2 psi per 100 pounds of added weight, and we have 650 pounds of batteries), which reached the maximum 44 psi recommended on the tire sidewall. I also added 15 psi to the air shocks I had installed and could feel and see the rear of the car rise.
Then, the moment of truth—we drove the car out of the garage, for the first time in more than a year. But when I tried to back up by switching the forward/reverse switch to reverse and applying the accelerator, the controller cut out. The controller reset immediately when I released the throttle. I then switched to forward, and shifted the transmission to reverse, and the same thing happened. I suspected that the controller cutout was a max torque setting issue, which goes into error mode when the motor is starting to turn under load.
The next day, I took the car to town, avoiding the problem by driving the car using the clutch and first gear to start moving, and shifting up and down as needed. The EV drove just like an ICE car, with the regenerative braking acting like engine compression braking when downshifting. It was really nice to see the Link 10 meter voltage climb when decelerating, adding that energy back into the battery pack.
But there was still the issue of the controller cutout to solve. Canadian Electric Vehicles Ltd. e-mailed me instructions for using a laptop computer to capture motor controller data for Azure Dynamics to review. I set up the laptop in the car and recorded data with the car in first gear. As expected, the controller cut out. But this time, I had the data, which I e-mailed to Canadian EV. I received instructions back in just a few hours to change a setting in the controller. On another drive, we captured data on two high-speed controller cutouts, and sent that data to Canadian EV. In the meantime, I tried easing up on the throttle, which eliminated controller cutouts.
I finally figured out that I’d been shifting up too quickly. Based on power curves I’d seen, the optimum motor speed is about 4,000 rpm. On another test drive, with a friend reading out motor rpm and amps from the PC display, I stayed in second gear until 45 mph (4,000 rpm), and stayed in third until 60 mph (4,000 rpm). The car cruised easily on a flat highway at 70 mph with no problem. There was throttle left, but I didn’t try to go faster than that. The return route home from the highway includes a steep, 2-mile-long hill. I kept the car in second gear and came up the hill with no problem at 40 to 45 mph, the speed limit.
Later, I received more advice from Azure to change another setting. After making the change, I drove to town, ran errands, and came home without any controller cutouts. Success!
Once the car was road-worthy, I began monitoring mileage and charging kWh to calculate the car’s energy use and cost. At our electricity rate of $0.028 per kWh, it costs about 25 cents to charge the car—about 1 cent per mile. That’s pretty good, even compared to the original Echo’s 40 mpg. For comparison, at the current $2 per gallon cost for gasoline, it would cost 5 cents per mile to run the ICE car. Last summer’s gasoline price of $4 per gallon would have cost me 10 cents per mile.
On average, we use about 0.27 kWh per mile from the battery pack (0.3 kWh per mile from the outlet, due to charger inefficiency). The battery has a capacity of about 9.7 kW (156 V nominal x 62 Ah). If I discharge to 20% state of charge (SOC), I’d use 7.7 kWh. At 0.27 kWh per mile, the car should be able to travel 28 miles. Although this won’t get us to the nearest large town, it’s an acceptable range. (Most Americans drive 30 miles or less daily.) If I abuse the batteries to 0% SOC, the car might be able to travel 35 miles. But so far, our longest drives have been about 22 miles with 40% SOC left, and this meets our needs very well.
In the five months since we put the car on the road, we’ve put more than 2,000 miles on it. We use it exclusively for local driving and don’t cringe when we make multiple trips to town in a day, since the cost per mile is minimal and we’re using clean, renewable, hydro grid power for our energy source. We also have a 1.9 kW grid-tied PV system that typically produces more power each day than we use in the car. So you could say our car is solar-powered.
We’ve tweaked the DMOC settings a couple times to get optimum performance, and will install an electric air-conditioning compressor and electric cabin heat next. Overall, we are quite pleased with the car—you can tell by our “EV grin” every time we drive it!
Randy Brooks and his wife Anne operate Brooks Solar, a renewable energy business in Chelan, Washington.