Today the Nuon Solar Team sliced off a large chunk of their gap to Tokai, while Tokai had to carefully and slowly nurse their Challenger to their camping site.
How did this happen? Yesterday I pointed out that a solar car’s performance depends on improving small bits in every area. Did the Nuon Solar Team find such improvements overnight? No.
The answer is strategy and here’s how it played out over the past 2 days.
Solar racing is a strategy game
Imagine 3 identical solar cars, aptly named A, B and C.
Let’s put these cars on the Stuart Highway and let them run for 8 hours, a typical amount for a WSC day.
Car A drives 95 kph every hour, so that’s his average for the day.
Car B drives 90 for the first 4 hours and 100 for the second part of the day. His average: 95 kph.
Car C has a more exotic profile: it starts the day at high speed, but slows down tremendously towards the end, but his average is also 95 kph.
Let’s turn to the power consumption of these cars during their run.
One important thing to note: the faster you drive, the bigger influence the aerodynamic variables have. As an example, at 80 kph, 40% of your total power consumption is due to rolling resistance, and 60% is aerodynamics. However at 120 kph, the ratio is 20:80.
Power consumption for each of the speeds of the 3 cars are given below:
Speed and Consumption for cars A, B and C
Since 95 kph yields a power consumption of 1170 W, Car A consumes a total of 8 h x 1170 W = 9.4 kWh.
From this table however it becomes clear that although car B and C average at 95 kph, they consume more power. The story of car B says that if you fluctuate within a few percentages of the target speed, energy consumption will only be a few percentages off.
Car C however gives a different and not so attractive picture: with this speed profile it spent 10 % more energy than car A while covering the same distance.
This is mainly due to the aforementioned influence of aerodynamics, that grows bigger as speed goes up.
Here comes the real catch.
Suppose car A and C are in close competition. In all solar race cases, target speeds are carefully selected by team strategists because they want to end up with a certain amount of energy left in the batteries for the next race day. If headwind or clouds are expected, they may want some more energy than in the case of soft blue sky weather.
Car A’s strategist determined based on weather reports that they can end up with a little less energy so car A accelerates in the final 3 hours of the race to 100 kph. At that speed the car would consume 1345 W. Let’s assume Car C maintains the same profile in this example.
Speed, Consumption and Distance for cars A and C
As seen in the table above, Car A now spends a total of 9.9 kWh which is still less than Car C but manages to end up 15 km further!
This is critical: Car A finished ahead of C with less energy spent, because they initially selected a target speed and did not deviate much from it. Not only is Car A ahead of C, it has more energy now to start the next race day with.
The big challenge in solar racing is selecting that right target speed based on knowledge about the car’s consumption, weather reports and solar energy intake. Weather reports may vary by the hour so it is of the greatest importance to keep an eye on the sky.
The question now remains: why does Car C show a such a wide speed profile? One reason could be that C-team expected to retrieve enough solar power during the day to sustain the speed and consumption they start out with – but Mother Nature’s habits gave them cloud coverage where they did not anticipate it.
What happened in the past 2 days in the World Solar Challenge might to be exactly that.
The Nuon Solar Team fluctuated between 90-100 kph for most of these 2 days while Tokai went wild with averaging over 100 yesterday, and just a bit slower for the greater part of today. The Japanese however were forced to slow down tremendously after 14:00 as soon as the first signs of cloud coverage appeared. Their last half hour saw speeds dropping dramatically – first to 80, then to 73 kph before pulling over to camp at 17:00.
It seems Tokai was confident the energy intake throughout Day 4 would have been sufficient to uphold these speeds but the clouds might have changed their minds.
Was the Nuon Solar Team better informed of the approaching weather and acted accordingly? Or does Tokai have more up its sleeve?
If Nuna 6 appears to have driven more efficiently up until now, we’re expecting them to have more battery energy left than Tokai, who may be forced to start tomorrow’s stint on the defensive.
At this point we are left to guess, but the above proves one thing: solar racing is difficult to learn, but a real challenge to master.
The only thing that really speaks the truth here is the finish flag: once it drops, the speculation stops – and we will know who is the true Master of energy strategy.
Another one of Diederik’s insightful articles. I hope it’ll spark a lot of discussion!
As far as I know the teams are allowed to charge their batteries after the race day. How much energy can they charge? If the solar panel can output 1 kW*, they can charge the battery in about 5 hours (if its completely depleted). They normally stop at about 5 o’clock and sunset is at 19:30 [url]http://www.timeanddate.com/worldclock/astronomy.html?n=5[/url] they would be able to charge about 50% of the battery. The next day the sun rises at about 6:30, and with race start at 8:00 they can charge for 1,5 hours, equaling 30% of the battery.
So if the battery was depleted on arrival, it would have charged to about 80% again before departure. So draining the battery during the day wouldn’t have a major effect on the charge levels.
Does the above make sense?
* Just an assumption, I have no idea how much power the panel can deliver.
Jaap: you assume it’s always sunny during static charging but take a look at this video of the Nuna 6 campsite this very morning.
And while you can get full panel power at 5 o’clock, the sun drops lower and lower, giving less and less solar power. The last half hour of sunset (and the first of sunrise) is pretty weak.
With these Si-panels, we are no longer able to fully charge a battery in static charging.
Sorry here is that vid: http://yfrog.com/1szrzsz
Diederik: Assumption is the mother of all fuckups, I know
. It’s true that you can’t control the weather. The angle of the sun should not be much of a problem as you can tilt the panel, so its angular to the sun. The sun does have to travel to more of the atmosphere and will lose a lot of it’s power.
Do you have any insight in how much power the panel generates during the race and while charging statically?
Power consumption at 65 km/h seems to be inconsistent with those at 60 km/h (and 85 km/h).
Power consumption at 120 km/h seems to be too high, because 1345W * (120/100)^3 = 2325W is the upper bound. Did you not use the power calculation site you mentioned in the “How good is Tokai really?” article?
Of course these minor details does not affect the outcome of your research at all
And a next step might be how to take localized cloudy areas in account. Maybe Team C had a good reason for driving fast in the first three hours because of clouds, so it might actually end up with more energy than Team A.
VincentG: you are absolutely right about the short term strategy – that’s an article for next time
And in fact, I used an simplified excel-sheet for these calcs this time… I guess they are ballpark figures 
It seems that VincentG is a very good stratigist. Is it the same person that pulled N4 through Day 4 to the finish in 2007?
BTW D, nice article! i like
@Rabih, not Vincent de Geus but Vincent Groenhuizen from Twente
And thx 
@Jaap:
Have a look at this:
http://www.pveducation.org/pvcdrom/properties-of-sunlight/atmospheric-effects
And especially this effect is important:
http://www.pveducation.org/pvcdrom/properties-of-sunlight/air-mass
The Air-mass is the length the light travels through the atmosphere. In the evening the sun is low at maybe 20 degrees above the horizon when they start charging. This gives an Air-mass of about 3 reducing the insolation by about 30%. At 10 degrees above the horizon the air-mass has risen to 5.75 and insolation has dropped to <50%.
This is at clear sky conditions, but you should also consider that the chance of having a cloud between you and the sun also rises and there are probably more effects
————–
"The graph shows the intensity of direct radiation in kW/m² through out the day. It is the amount of power that would be received by a tracking concentrator in the absence of cloud. The time is the local solar time. "
Hour and direct radiation
5.25 0.0
5.5 5.060218839888843E-10
5.75 0.029211322364389454
6.0 0.1657604907209817
6.25 0.31885082575210105
6.5 0.4489733926738679
6.75 0.5537372020343405
7.0 0.6376912810456109
7.25 0.7055303310452954
7.5 0.760968848940809
7.75 0.8067786971433879
8.0 0.8450081820415042
….
17.0 0.6376912810456109
17.25 0.5537372020343405
17.5 0.4489733926738679
17.75 0.31885082575210105
18.0 0.1657604907209817
18.25 0.029211322364389454
18.5 0.0000000005060218
18.75 0.0
Calculated using:
http://www.pveducation.org/pvcdrom/properties-of-sunlight/calculation-fo-solar-insolation
And I am also not Vincent Groenhuizen, but I am Vincent Groenhuis from Team Twente 2005. I did the telemetry and assisted Ceriel with the strategy.
Hi Vincent nice to see you here
I once spoke to somebody who did research in solar energy production (ECN Netherlands), he explained that due to all kinds of dust effect and so on, the effective Airmass can be 30 or 50 in the morning/evening, leaving very little power to charge batteries.
For triple junction GaAs cells AM (Airmass) is also important because more AM will also influence the solar spectrum, taking out parts of the spectrum, and can deteriorate the performance even more. Silicon cells don’t have this problem as much.
I think that in really cloudy weather (Clouds add a lot of AM), you will get very little extra charge in your battery in the morning and evening. And as a side note, concentrating arrays are totally worthless without direct sunlight..
There is one team that uses GaAs Cells, and concentrators. I think they are in a bit of trouble right now.
Great to have articles like these in there; gives all of us some more insight in the complexity and strategy of the race, thx
And yet somehow the managed to have the winning strategy
But i agree… they could have been faster with a more consistent speed.
The flag has dropped so the speculation has stopped