Category Archives: Technical

Generators on the wheels

One of the reasons we built Trev was to get people thinking about the energy they use for mobility. Most people appreciate that using tonnes of machinery to transport a 75 kg person uses a lot more energy than necessary. But some people get a little too inspired: Why don’t you put generators on the wheels?

Generators on the wheels is not a bad idea. The electric motors used in most electric and hybrid cars can act as motors converting electrical power to mechanical power, or as generators converting mechanical power to electrical power. Working as a generator, the motor can convert the kinetic (movement) energy of the car into electrical energy, and this energy can be used to partly recharge the battery. But converting the kinetic energy of the car to electrical energy slows the car. It is called regenerative braking, and the braking force that can be achieved is about the same as the driving force that can be achieved by converting electricity to motion.

Regenerative braking cannot convert all of the kinetic energy of the car into electrical energy—some of the energy is dissipated by resistance forces in the tyres and bearings and by aerodynamic drag, and some is dissipated as heat in the motor/generator and in the electronic controllers. But some of the kinetic energy can be converted to electricity and stored for later use—which is better than occurs with normal friction braking, where all of the kinetic energy is dissipated and none of it can be recovered.

But the proponents of generators on the wheels often want to go beyond regenerative braking to generate electricity without slowing the car. Not necessarily perpetual motion, where the generators generate more than enough electricity to power the motors, but enough to reduce the power required from the battery.

It doesn’t work—at least, not in our universe.

To see why it doesn’t work, try writing power values in each of the empty boxes in the diagram below. Suppose the car is travelling on a flat road at constant speed, so the power values are not changing. The total power into the motor must be the same as the total power out of the motor, and the total power into the generator must be the same as the total power out of the generator. It is like Power Sodoku—everything has to add up.


  1. Write in the power required for propulsion. This is the power required to overcome rolling resistance and aerodynamic drag. Trev uses about 4000 W at 70-80 km/h.
  2. Write in the motor losses. Losses in an electric drive system are about 10% of the propulsion power.
  3. Write in how much power will be transferred from the motor to the generator, and generator losses (about 10% of the power into the generator).
  4. Calculate the ‘battery out’ power and the ‘battery in’ power.

The total energy from the battery is (battery out – battery in). What do you have to do to minimise (battery out – battery in)?

Lessons learnt

On 20 July we gave a presentation on lessons learnt to the Adelaide Branch of the Australian Electric Vehicle Association:

Lessons learnt

More video from Berlin

Trev has just arrived by ship in Vancouver, and should hopefully clear customs tomorrow. The next leg of Zero Race starts on Friday.

In the meantime, here is some more video of Trev in Berlin, being repaired and being driven.

Trev being repaired in Berlin

Trev, the remote control car

Zero Race is moving across Europe at a challenging pace—see the Zero Race blog for details. We have had no time to replace our temporary suspension fix with a permanent repair, or to fix the communications problems in the battery management system. So while the rest of the race is in Brussels, we have moved ahead to Sven’s workshop in Berlin to give us a few days to fix these problems before rejoining the event when the other teams arrive on Tuesday. (Sven is the main rider of Team Vectrix, and one of our benefactors. He has also booked to drive Trev between Moscow and Shanghai, so is keen to have the car running reliably.)

Trev in the Berlin workshop

Trev in the Berlin workshop

Communications between the crew in Europe and the rest of the team in Adelaide is improving. In fact, last night we achieved a major breakthrough in remote automotive diagnosis:

Peter, in Adelaide, hooted the horn, in Berlin.

Here is how we did it. The horn button in Trev, along with all the other driver buttons and controls, is connected to a microcontroller under the dash. When the horn button is pressed, the microcontroller sends a ‘horn on’ message to the rest of the car via the Controller Area Network (CAN) bus—a pair of communications wires which connects all of the electrical devices in the car. Another microcontroller, the front right lighting controller, receives this message and turns on the horn. When the driver releases the horn button, the driver controls box sends a ‘horn off’ message and the front right lighting controller turns off the horn. Easy. Connections are made in software, and there are no large looms of wires running around the car.

The CAN protocols automatically handle message priorities and arbitration between devices wanting to send messages at the same time.

Any device can listen in on the communications. To diagnose problems, we connect a netbook computer to the car and log the CAN messages. But we can also put messages on the CAN bus from the netbook. Last night, Nick had the netbook connected to the car, and Peter (in Adelaide) was operating the netbook via the internet using TeamViewer software.

Hooting—via wireless modem, internet, Android phone, netbook computer and CAN bus—was inevitable.

Technical difficulties…

We have experienced a few technical difficulties over the past few days. Most frustrating have been problems establishing reliable communications between the crew in Switzerland and the rest of the team back in Australia. Getting a good data plan is not easy if you are only going to be in a country for a few days. But the information and photos started flowing today (thanks Keith!), and we have started a gallery.

Nick on the start line, Geneva, looking for an Australian socket

We have also had some problems with the car:

  • Moisture in the battery box caused communication problems in the battery management system. This did not stop the car from driving, but Nick and Jason put in long hours in the two days leading up to the start of the race trying to fix the problem. The crew will pick up replacement parts in Brussels.
  • There was some damage to the lower suspension mounts in transit. With the help of the Zerotracer team, the crew has overcome the problem by running reinforcing beams across the underside of the car. Meanwhile, the team back in Australia has worked out a permanent repair.
  • The installation of the GPS tracker in our car has not been reliable; consequently, our location on the Zero Race web site has been missing or delayed. This should be fixed soon.

The crew is working very hard—Zero Race events, driving between cities, and maintaining the car. Mic joined the crew yesterday, and provided some welcome respite. But they are in good spirits, and getting an enthusiastic response to the car wherever they go. We will post some of their stories and photographs over the next few days.

Trev. It’s registered.

A milestone for the logistics team

The logistics team have completed their first major challenge, successfully importing our new batteries from Korea to the Team Trev workshop in Adelaide.

With no experience importing goods (especially dangerous and expensive goods!), we relied on the experts. We thank Chris Sergeant and Ben Poprawski from Customs Agency Services and Chris Donnelly from Donnellys Insurance Brokers for their great service in organising many aspects of the task and providing us with quick advice.

The arrival of the batteries was cause to stop and celebrate the progress of the team in preparing for Zero Race which starts in Geneva on 15 August, 2010.

If you look closely at the photo, you can see that work is still continuing on the car. Coordinating the arrival of the batteries is only one of the tasks which must be done to ensure that Trev will be ready for Zero Race in time. The technical team is working very hard and putting in an enormous effort.

Outside the workshop, there are still many logistical jobs to complete, including organising insurance, car registration, freighting the car to Europe, driver uniforms, driver rosters, visas, travel, and so on. If you want to help, contact us through the website.

Folding a battery tray

Trev’s main tub structure was made by folding and bonding honeycomb boards. Our new battery tray, which will fit under the floor of the car, was made using the same technique.

For the tub we used aluminium foil honeycomb. The battery tray is made from cheaper polyproplyene honeycomb, as shown in the images below.

The first step is to fold an edge. Secure the honeycomb sheet to a table with old lead acid batteries, and clamp a square steel tube along the fold line. To form a right angle fold in 10 mm thick honeycomb, heat a 10 × π / 2 = 16 mm strip using hot air guns.

Use hot air guns to heat the fold line.

When the polypropylene softens, fold and clamp the sheet. The softened honeycomb will crush, and the outside edge of the fold  will have a nice 10 mm radius.

Fold and clamp. This photo shows the second fold.

The fold will retain its shape when the honeycomb cools, after a few minutes.

Next, fold up the front of the tray. For a fold angle α (in radians) and a honeycomb thickness t, the strip that is heated and crushed has width α t.

Design the fold.

Fold the front edge of the tray.

Our honeycomb sheet was not long enough to include the rear end of the tray, so a separate rear panel was simply glued on. At this stage our 2300 × 540 × 80 mm tray had a mass of 2.4 kg.

Next, apply two layers of Kevlar to the inside of the tray. The first layer has the fibres at ±45° to the tray, to resist twisting. The second layer has the fibres at 0° and 90° to the tray, to resist bending. Kevlar is more expensive than fibreglass, but gives a low-mass structure with the stiffness and toughness we require.

Apply Kevlar to the inside of the tray.

Go home. The next day, when the resin has cured, trim the excess Kevlar, turn the tray over (you can use the same four old batteries to support it), and apply Kevlar to the outside of the tray.

Apply Kevlar to the outside of the tray.

A day later, trim the excess Kevlar and the tray is ready for a battery. The mass of the tray is 5.3 kg.

The (almost) finished tray.

Low mass, low energy

Using tonnes of machinery to move one or two people around a city seems ridiculous. But how important is low mass?

Many years ago, before we started designing Trev, I was driving a Solectria electric car to work when my path became blocked by a large “SUV” that had broken down in the middle of an intersection. Pushing the SUV to the side of the road made me realise just how much energy is required to move these massive machines. My electric car was better because it was using energy generated from clean, renewable sources. But it still needed a lot of it.

I sometimes demonstrate the importance of low mass by tying a small child to Trev and another to a conventional car, then asking them to race across the yard. Trev moves, the other car doesn’t. It usually takes five kids towing a conventional car to keep up with one towing Trev. A large SUV took ten kids.

In 2008 I had two high-school students investigate the relationship between vehicle mass and CO2 emissions. They used emissions data from the 2007 Australian Green Vehicle Guide and looked up vehicle masses from manufacturer’s web sites. The results are shown in the following graph.

CO2 emissions vs vehicle mass

The red dots show CO2 emissions for petrol cars, the blue dots for diesel, and the green dots for hybrids. The trends are obvious—halving the mass halves the emissions. The reduction is not entirely due to the reduction in mass; lower mass cars are generally smaller, with smaller engines and less aerodynamic drag. But mass is the dominant factor.

The graph also shows that for a given car mass (and size), there is wide variation in emissions. Improving vehicle technologies is one way to reduce emissions, but choosing an appropriate vehicle with low emissions can be a lot more effective.

Trev has a mass of just over 300 kg, and so it takes a lot less energy to push it along the road. In 2007 we drove Trev from Darwin to Adelaide, cruising at 80-90 km/h. We used 187 kWh of electricity, worth $33, to drive 3000 km. Petrol costs for a conventional car would be ten times this amount.

Electric cars are coming. Most will be based on conventional cars, and so will be heavy and require a lot of energy. A 1300 kg electric car with a range of 120 km might have 200 kg of batteries. Trev has the same range with only 45 kg of batteries. (We are going to increase this to 80 kg for Zero Race, because the charging points are up to 250 km apart.) Reducing the mass of a car is a very effective way of reducing the energy required to move the car, and the amount of materials required to build the car.

Trev. Not only does it use clean energy, it also uses a lot less energy.

Does it get hot in there?

We always get a lot of interest when we have Trev on display. The questions we were asked at our recent day in Rundle Mall were typical. Here are some of them.

Does it get hot in there?

In October 2007, two UniSA students drove Trev from Darwin to Adelaide in ambient temperatures around 35°C. Both survived. One of them has come back for more, and will be driving Trev for part of its tour around the world.

Like any car, Trev can get hot. When it is moving, air flowing through the car from an inlet in the front provides some relief. We also had a small fan in the car for our trip across the Australian outback. (We took it away from the driver to provide additional cooling for the motor controller, but we will put it back.)

The air conditioning systems used in conventional cars are not suitable for low-energy vehicles—they use more power than Trev driving at 100 km/h. We are still looking for efficient, effective ways to keep the driver comfortable.

How fast does it go?

Trev was designed to fit in with normal urban traffic, including on freeways. It has a top speed of over 100 km/h, and accelerates smoothly up to 100 km/h in around 10 seconds.

How far will it go?

When we drove from Darwin to Adelaide, we could travel up to 120 km at 90 km/h before we had to stop and recharge. Recharging took about an hour, and we were able to travel about 500 km per day. For Zero Race, we are increasing the range to over 250 km so that we don’t have to stop so often. For urban use, however, a range of 100-150 km is plenty.

Where are the solar panels?

Our experience with solar racing cars inspired us to build Trev—if you can drive across Australia without using fossil fuels, you should be able to drive to work and back. But if you have a photovoltaic panel, it will be more effective on the roof of your house than on the roof of your car. Trev is a pure electric car, and can be recharged using clean electricity from solar, wind or other renewable energy sources.

What type of battery does it use?

We are using lithium ion polymer cells. There are thirty-six large cells connected in series, giving a battery voltage around 130 V. The estimated life of the batteries is 250000 km.

How safe is it?

You will be less vulnerable in Trev than on a bicycle or motorcycle. Trev is also less ‘aggressive’ towards other road users than most conventional cars. The occupants sit within a rigid tub structure that will provide some protection during a crash. But heavy vehicles will have a greater impact on Trev than Trev will have on them.

Did you see where my husband went?


Our next appearance will be at the final stage of the Tour Down Under. Come and see us, and ask your own questions.