Mad dogs and engineers…

According to that old Noel Coward ditty, only mad dogs and Englishmen go out in the midday sun. He never mentioned solar-powered cars…

According to that old Noel Coward ditty, only mad dogs and Englishmen go out in the midday sun.

Not strictly true. Add some engineering students from Germany, Malaysia, and the United Arab Emirates – all studying mechatronics or CAE in south London – and you have the makings of a team that is shaping up to take on the world at solar car racing. They’ll be out in the sun not only at midday, but from dawn to dusk too. And they will probably spend a good deal of the night carrying out running repairs on the vehicle, which is named, appropriately enough, Mad Dog.

On 14 July this small team of lecturers and students from the Sustainable Transport Research Centre of London’s South Bank University will set out to motor west on the legendary US Route 66. Their latest solar-powered car ‘Mad Dog III’ will follow the road as, in the immortal words of Bobby Troup, ‘it winds from Chicago to LA, more than 2,000 miles all the way’.

Though solar racing has a large, committed following in Australia, the US and Japan, the South Bank team will be the only UK entry aiming to get their kicks in the American Solar Challenge’s field of up to 70 teams. It is expected that it will take 10 days to complete the course.

South Bank’s team leaders, senior lecturers Mike Duke, Nigel Burgess and Deborah Andrews, are veterans at this sort of thing. Duke and Burgess first competed in the World Solar Challenge in Australia with Mad Dog I back in 1996. The route was the Stuart highway from Darwin to Adelaide.

They returned to compete in Australia and Japan with successively more sophisticated cars, and will mount their fourth attempt at the Australian challenge in November, with the most high-tech version yet (see sidebar below). They have acquitted themselves honourably against teams put together on a shoestring by other universities and private individuals, as well as against companies such as Honda and GM, with millions in funding and vast support teams.

Before then, though, comes the challenge of the US. And Route 66 will be tricky. The Australian route is pretty flat but in the States, they’ll have to tackle the Rocky Mountains. ‘In Australia there’s one hill, at Hayes Creek on the first or second day,’ says Burgess. ‘That’s the only hill till Adelaide.’

The Rockies, by comparison, come about two-thirds of the way through the American Solar Challenge. ‘That section could take a day, and there’s one 16% gradient 200m long. I’m a bit concerned about it,’ says Burgess.

Then there is the question of negotiating cities such as Chicago and Oklahoma en route. ‘The car’s easy to drive on the open road – but not in heavy traffic. It’s 6m long and 2m wide, the turning circle isn’t very tight and rear vision is poor – it’s like a small bus.’

Burgess, who says he is ‘not optimistic by nature’, has spent the past couple of months worrying about one missing component, which plays a crucial part in the whole enterprise: the electric motor.

Two weeks before departure, the new motor, due to be delivered in January, had still not arrived – forcing the team to re-engineer the car to work with last year’s toroidal flux unit, originally designed to power ceiling fans.

Made by Australian firm T-Flux, it gradually became the power plant of choice for solar racers because of its compactness and efficiency. But like many competitors, the Mad Dog team was planning to switch to a more efficient hub motor, made by New Generation Motors of the US.

With no NGM motor to work with, the team had resigned itself to having to run a car optimised for the new motor put powered with the old one. This has had quite a few other ramifications. The suspension had to be redesigned to accommodate the 20kg weight of the motor on the hub. ‘We designed the new suspension from photos and diagrams on the internet,’ Burgess says. The change in the ratio of sprung to unsprung weight could have a marked effect on the car’s handling and ride: ‘We’ve lost 50kg of batteries and 10kg from the old motor from the sprung weight, but hugely increased the unsprung weight.’

The electrical system had also been adapted for the NGM motor, which runs at 100V, whereas the old T-Flux ran at 72V. While it is possible to run the T-Flux at 100V, it rotates 1000rpm faster. But shaft size limitations meant the team could not reduce the gearing any further. So running with the old motor would also hit the performance of the car on the dreaded 16% gradient.

There was, though, a Plan B, which involved taking a spare Lynch motor, which was used to power the first Mad Dog car, and fitting it just for the encounter with that awkward gradient.

In fact, Burgess needn’t have worried. The NGM motor did arrive – in true ‘just-in-time’ style, less that a week before the team set off to the US.But there are still plenty of other pre-race worries. The first big hurdle is scrutineering. The cars have to have indicators and a horn and meet requirements regarding field of view and turning circle. There is also a test to show that they are stable in the turbulent airstreams of passing trucks. ‘It’s a nightmare. It’s like taking your car for its MOT,’ says Burgess.

The US event is organised slightly differently from the Australian challenge. In Australia the teams race between 8am and 5pm and set up camp wherever they finish. In the US the teams all stop at the end of the day’s stage, whatever time they arrive. This should make it more sociable, but it also means more jockeying for position at the start of each day, which brings its own problems. Overtaking in a solar car is difficult, mainly because the speed differential is not very great.

Some help is provided by the two support vehicles which precede and follow each solar car and monitor telemetry data. The support vehicle will radio to say the road looks clear, but an overtaking car can still be on the wrong side of the road for two or three minutes as it eases by – a nerve-wracking prospect.

Evenings are taken up with repairs and maintenance. This may be as simple as changing a tyre – the special Michelin items designed for solar racing are built for low rolling resistance but use a soft compound, and may need changing every couple of days.

Some repairs have been more drastic, such as that required by a team from Berkley and Stanford Universities in the WSC 99 in Darwin. Racing for the first corner at the start, they hit a kerb with enough force to push the suspension through the solar array.

Some teams are forced to carry out running repairs through the night. Burgess reckons his team has been lucky so far: ‘We had a bad night in Australia in 1996. We were running at 32V when we left Darwin. Mike and I decided it wasn’t running as efficiently as it should be so we decided to drop to 24V. The rewiring took four hours.

‘And then, in Australia in ’99, we lost a drive sprocket. My feeling was that one of the students hadn’t tightened it up properly because he kept putting off the job until he ended up having to do it in the dark. We lost an hour, but the batteries got overcharged and were never the same again. We dropped from 14th to 20th place.’Creating a cohesive team from a disparate group of engineering students is all part of the challenge. Some find the relentless race regime a shock to the system. ‘You sleep in a tent and get up at 4 or 5 in the morning. It’s a long, hot, tiring day – and then someone chirps up ‘You’ve got half an hour to eat something but then we’ve got to strip the rear suspension’,’ says one.

But there is the upside: ‘Just being there is amazing,’ says Deborah Andrews. ‘There’s a real buzz, with a lot of strange and wonderful people.’

Sidebar 1: Evolution of a modest ‘nothing to lose’ project whose success took the build team by surprise

The very first Mad Dog was put together by Mike Duke and Nigel Burgess and a handful of students in a spirit of ‘build it and see if it works’ on a very limited budget for the 1996 World Solar Challenge. Despite the limited resources, the vehicle surpassed expectations by completing the 3,010km course across Australia in 10 days, compared with four for the fastest cars.

Mad Dog II incorporated a number of improvements and raced in the World Solar Rally at Akita, Japan, in 1998. With lead-acid batteries and an 8m2 solar array, it won its class. It is now on show in the Science Museum, as part of an exhibition of ’20th century icons’.

Mad Dog III completed the 1999 World Solar Challenge in seven days, gaining second place in its class.

For the American Solar Challenge, Mad Dog III has been substantially improved to make it the most advanced version of the car ever. The old-tech lead acid batteries have been replaced by advanced lithium-ion cells from Saft of France, which are 95%-97% efficient. They weigh half as much as the lead-acid items while storing nearly twice as much charge. (The output from the solar array is too erratic to power the electric motor directly, so the batteries have a dual storage and management function).

‘We are the first solar car to have this type of cell. They have the potential for a huge step change in efficiency,’ says Patrick Sonntag, one of the mechatronics students on the project.

The solar array, provided by Gochermann Soar Technology of Hamburg, consists of 760 cells, covering 8m2, with a special non-reflective coating giving it a relatively high efficiency of 16.5%.

Data on solar radiation intensity, battery condition, solar voltage, battery temperature and speed are all gathered on a sophisticated telemetry system that uses a tiny 486, 66MHz PC chip connected to a Canbus which sends the information to the support vehicle for analysis via a radio modem, as well as recording the data on a flash card.

Sophisticated electronics adjust the constantly varying voltage from the array to suit that of the batteries. Last year’s 92% efficient, 4kW T-Flux toroidal flux motor will this year year be replaced by a 97% efficient 4.5kW hub motor. Top speed is 73km/h.

The car’s structure, apart from the glass fibre nose cone, is made from Fibrelam by Hexcel Composites. Fibrelam is a lightweight honeycomb material, more advanced than the Nomex used in Mad Dog II.

But, as senior lecturer Deborah Andrews points out: ‘We’re reaching a point where the weight is stripped down to a minimum. Success is now down to the technology you can afford.’