European Solar Challenge: Further Analysis

I updated my previous post to include visualisations of the official results of the iLumen European Solar Challenge. However, a helpful commenter shared a link to the detailed lap data, so above and below is some further analysis on lap times. Colours are the same as in the previous charts. As always, click to zoom.

The chart at the top shows pit time (including recharging time) as almost-horizontal lines, and pit time is also visible as gaps in the chart below. The histograms show that Top Dutch (6), the two cars from Sonnenwagen Aachen (7 and 70), Istanbul Technical University (34), and the experienced drivers in BOSolarCar e.V. (11) were particularly good at driving consistent lap times.

Measured by median lap times, Stella Vita from Eindhoven (41), Covestro Photon from Aachen (7), SunRiser from Bochum (11), RED E from Twente (42), and BluePoint from Agoria (8) were the fastest cars.


European Solar Challenge Lap Data

All too soon, the 24-hour iLumen European Solar Challenge is over. The charts above and below (click to zoom) show results from the live timing board. The winners of the two classes (Challenger and CR = Cruiser) will be decided on points scores, which are still to come. Meanwhile, however, the German teams from Aachen and Bochum deserve special congratulations. For updates, see also the race social media at    

Update 1: there are apparently no official results yet. All that has been released has been the podiums:

  • Challenger: 1. Covestro Photon (Aachen), 2. Covestro Sonnenwagen (Aachen), 3. RED E (Twente)
  • Cruiser: 1. Lodz, 2. Onda Solare, 3. Stella Era (Eindhoven)

Using the guidelines in the official regulations, I can calculate the Challenger class scores (see below).

Update 2: the official results have now been released, and I have added the chart below. Bochum seems to have lost out from having a two-seater car. For the rest, practicality judging makes up 40% of the final Cruiser score at iESC, so that Lodz essentially won on practicality (although I note that four of the Cruisers raced in Australia in 2019, and I would have expected both practicality and efficiency scores at Zolder to have been somewhat similar).


Solar racing car numbers

As in all races, solar racing cars are identified by number. Some solar car numbers are simply traditional, like the 8 for Agoria Solar Team from Belgium (above). Others have a specific meaning, as shown in the chart below.

Some numbers are lucky in some way, such as 21 = “Twente-One.” Some are coded references to solar technology, such as 55 = the year that Western Electric began to sell licenses for silicon PV technology. Alternatively, numbers indicate the team’s home base. This can be done by specifying a road, such as the the Interstate 35 or Strade Statali 9 = the Via Aemilia. More commonly, telephone country or area codes are used, such as 40 = Eindhoven, 46 = Sweden, 82 = South Korea, or 828 = western North Carolina. Not shown in the chart is 34 = the vehicle license plate prefix for Istanbul.

Mobile phone picture by Rafael Fernandez


World Solar Challenge Cruiser scoring

Having participated in some recent discussions about the World Solar Challenge Cruiser Class, I thought I would explore the scoring again. Scoring in the WSC is based on a multiplicative formula (see reg 4.4.7), but as we all learned in high school, multiplying is equivalent to adding logarithms.

By appropriate scaling of logarithms, the chart above breaks down the various components of the multiplicative formula into additive points (black bars are negative numbers).

For example, on this system, in 2019 Eindhoven received a total of 67 points:

  • 100 points for completing all three stages
  • +12 points for an average of 2.6 people in the car
  • −53 points for 71.2 kWH of external energy
  • −0 points for no lateness
  • +8 points for a practicality of 93.1%

Lateness refers to arrival at stage stops after the “soft cutoffs,” which in 2021 will be Sat 15:30 in Tennant Creek, Mon 16:30 in Coober Pedy, and 11:30 Wed in Adelaide (there are also “hard cutoffs” leading to elimination, of 17:00, 17:00, and 14:00 respectively). According to reg 4.4.8, teams are ranked by the number of completed stages, and then by score.

In 2019 Minnesota received a total of 39 points:

  • 86 points for completing only one stage (thus also ranking after all the finishers)
  • +9 points for an average of 2 people in the car
  • −40.5 points for 25.7 kWH of external energy
  • −21 points for 165 minutes of lateness
  • +5.5 points for a practicality of 76.3%

Suppose a larger battery and a longer race had increased Minnesota’s external energy use to 128 kWH (an extra −20 points), but this had removed the lateness factor (an extra +21 points) and allowed achieving all three stages (an extra +14 points). Minnesota would then have been 15 points ahead, for a total of 54. This would have put them neck-and-neck for second place with Sunswift.

This example makes clear how the rules create an incentive for large batteries. It also highlights the difficulty of the second stage from Tennant Creek to Coober Pedy – I wonder if an extension to the “hard cutoff” is possible there?

Technical note: I am multiplying natural logarithms by 12.48 (so that 3020 gives 100), and I have also doubled practicality, and thus the total score (this doesn’t, of course, give practicality any extra weight). Sanity check for Eindhoven: 12.48 × ln(104×2) = 67.


Solar racing teams: the US and Dutch models


Stanford at the finish of the World Solar Challenge in 2015

Everybody knows that I’m a big solar racing fan. Today I wanted to talk about solar car team models, comparing what I call the “US model” (although most other countries also use it) with what I call the “Dutch model” (also used by the Belgian team). In the “US model,” students work part-time on a solar car team, and new members are added each year. As an example of this, I will look at the Stanford Solar Car Project, and specifically at one team member: Rachel Abril, who is forever famous for her May 2014 TEDx talk.

Rachel Abril did a 4-year Bachelor degree in Mechanical Engineering (the blue blocks in the chart below show Stanford’s academic years) followed by a Masters degree. The hashed region on the chart shows her extensive involvement with the Stanford Solar Car Project, first as a junior Mechanical Team and Aerodynamics Team member, and later as Suspension Lead and Aerodynamics Lead. She did not, I believe, attend the 2013 World Solar Challenge, but she did attend the 2015 and 2017 races (Stanford was improving during this period, but so were the other top-twelve teams!).

Rachel Abril’s story highlights one great advantage of the “US model,” namely that long-serving team members develop enormous experience in the design, construction, and racing of solar cars. They can take the lessons of one race, and apply them to the next one (and Rachel’s TEDx talk mentions some lessons that Stanford learned).

There are a number of disadvantages to the “US model,” however. New recruits often have limited knowledge of relevant physics (especially in the US, where high school graduates are educationally about a year behind their European or Australian counterparts). What work can new recruits be given that is both interesting to them and useful to the team? How can they be properly integrated into the team, and feel that they are genuinely part of the group? How can the team stop new recruits from feeling “cheesed-off” and dropping out? Answering these questions well is the key to success for US teams. One of the answers lies in running internal training courses for new recruits (there is also the IEF Solar Car Conference), but teams do not always include “Education Lead” or “New Member Coordinator” as one of the key team roles.

Another disadvantage of the “US model” is that the mix of people with varying lengths of experience creates a power structure. It can be difficult for a new recruit to disagree with someone that has been on the team for many years (even if, objectively, the new recruit is right). This can be a trap.

A final difficulty with the “US model” lies in balancing solar car construction, academic study, and personal life. Conventional wisdom is that you can hope for at most two out of three. Privately, team alumni sometimes suggest that one out of three might be more realistic. I don’t know what support mechanisms might help with this.


Solar Team Twente at the finish of the World Solar Challenge in 2019

In contrast, in the “Dutch model,” a smaller group of people gives up a little over a year of their life to work full-time on a solar car. This is quite a sacrifice. The Belgian team’s recruitment page explains the return on investment for the year like this (my translation):

  1. A project filled with experiences that you won’t find in your regular studies;
  2. Discovering a genuine engineering project and its various phases: concept, design, production,
    and test;
  3. Connecting and collaborating with the largest companies in relevant industries;
  4. A close-knit group and a racing adventure never to be forgotten;
  5. The experience of a lifetime and so much more!

Essentially, the year on the solar car team functions as an unpaid internship (speaking as someone who has helped arrange engineering internships in the past, I can’t think of an internship where you would learn more). One positive feature of industry internships is normally industry networking; this is also worked into the Dutch/Belgian solar car experience (as #3 on that list indicates). Of course, the need to set up those industry connections is one more reason to have a really professional sponsorship team.

As an example of the “Dutch model,” I will focus specifically on the 2018–19 “edition” of Solar Team Twente. Behind this team sits a part-time organisation (mostly of alumni) which handles recruitment and provides technical advice. This organisation began recruiting in February 2018, and a new team was announced on 9 June 2018. All these people were complete solar car novices, of course. The new team began work at the start of the 2018–19 academic year (with the aerodynamic and management subteams starting a little earlier). In the chart below, coloured blocks show academic years, and the hashed region shows the typical duration of full-time team involvement:

One of the first activities of the novice Twente team was to race the previous car, Red Shift, at the European Solar Challenge (iESC) on 21–23 September 2018. Team alumni raced the even older Red One, so that this was not only a training activity for the novice team, but an opportunity for knowledge transfer from alumni. Building on their iESC experience, the novice team then began designing and building their new car, RED E. The new car was revealed on 21 June 2019. After a test race on 17–18 August, the car was shipped to Australia on 30 August (a tragic crash due to wind gusts put RED E out of the race, but it was in the lead when that happened).

Engineering education in the Netherlands is traditionally a 5-year Ingenieur degree. Because of EU regulations, this is nowadays packaged as a 3-year Bachelor degree plus a 2-year Masters, but local students generally take the full package (because of the superior Dutch high school system, the 3-year Bachelor degree reaches at least the same standard as the 4-year US equivalent). As a result, the novice Twente team would have had substantially more formal education under their belts than new solar car recruits in the US. Dutch engineering schools also benefit from a close connection to industry, which drives a practical focus. The Eindhoven University of Technology, for example, is traditionally a feeder school for Philips, DAF Trucks, and other engineering companies in the Eindhoven area.

Of course, not every university teaches every skill needed for solar car design and construction. Dutch engineering schools typically teach agile project management, for example, but this does not seem to be the case in Belgium. The Belgian team therefore arranged industry training on the subject from their sponsor Delaware Consulting. Dutch teams also often benefit from industry-based “team building” activities (this video shows such an activity for Top Dutch). Practice races (including the European Solar Challenge) compensate for the fact that team members have never attended the World Solar Challenge before.

Because of team-building, educational initiatives, and good knowledge management, the “Dutch model” consistently produces top solar cars (Vattenfall/Delft has won the World Solar Challenge repeatedly, the Belgians won in 2019, Twente was on the podium in 2013 and 2015, Top Dutch came 4th in their first race, and Eindhoven has won the Cruiser Class every time). While the “Dutch model” relies partly on specific features of engineering education in the Netherlands and Belgium, I think there are several Dutch/Belgian practices that teams in other countries can learn from.


Nuon (now Vattenfall) at the finish of the World Solar Challenge in 2017

I should finish with a note on Vattenfall (Delft) Solar Team, which runs a variation of the “Dutch model.” Vattenfall (Delft) alternates what I call “big build” teams with “small build” teams. The “big build” teams design and construct new cars for the World Solar Challenge, while the “small build” teams modify existing cars for other events. For example, Nuna9 was a “big build” for the 2017 World Solar Challenge, while Nuna9S was a “small build” modification of the same car for the 2018 South African race (it included a clever radar system). Likewise, Nuna Phoenix was the same car modified again for the 2020 American Solar Challenge (that event was sadly cancelled, but Nuna Phoenix did set a world record). As part of providing a return on investment for the “small build” teams, Vattenfall (Delft) is careful to give these modified cars their own identity.


European Solar Challenge: Results (2)

This will be the second of two posts on the results of the just-completed iLumen European Solar Challenge (iESC), following on from this post.

To begin with, consider this chart of lap counts over time (using the same colours as the previous charts). Problems, recharges, and even (if you look closely) driver changes are clearly visible (click to zoom). The dotted red line shows the champion performance of Solar Team Twente in 2016, surpassed by 5 cars this year. The dotted blue line shows the fastest Cruiser, Bochum’s 2013 PowerCore SunCruiser, in that same race (bear in mind, though, that it drove with no passengers, while this year both Stella Era and Stella Vie ran with 3 passengers each!). Particularly notable is the Belgian decision to allow BluePoint only one recharge stop (the sunny weather and the superb efficiency of the car made that feasible).

This chart shows the calculation of the final scores, incorporating points for most laps, for dynamic parcour timing, and for fastest lap. The left-hand coloured bars (total scores) are each the sum of the other bars of the same colour. So Agoria 1st, Twente 2nd, and Top Dutch 3rd:

Scoring in the Cruiser class was simplified to be based on laps, but with bonus laps for the dynamic parcour timing and fastest lap won by Stella Vie. Nevertheless, Stella Era still won: Eindhoven 1st and 2nd. I won’t include a chart for that.


European Solar Challenge: Results (1)

This will be the first of two posts on the results of the just-completed iLumen European Solar Challenge (iESC). If anyone doubted that BluePoint is the fastest solar car in the world, Agoria Solar Team from Belgium proved them wrong. In 24 hours, they drove almost the distance from Darwin to Alice Springs on the very challenging Zolder track (click the image to zoom):

All five of the top cars broke the record set by Solar Team Twente in the 2016 race.

After some battles for the fastest-lap title, Top Dutch won that, reprising their performance in the 2019 World Solar challenge qualifier (the Covestro Sonnenwagen from Sonnenwagen Aachen came second, and Stella Vie from Solar Team Eindhoven third):

I also looked at lap time distributions (I have re-done this analysis since I posted on social media, using the official laptimes rather than my sample). The data includes time spent in the pits, so I drew an arbitrary cutoff at 600 seconds (10 minutes) for a lap. The histogram below shows these lap time distributions. Note the very consistent driving speed of the two Belgian subteams, driving BluePoint and Punch 2. Equally consistent was RED Shift, driven by experienced alumni of Solar Team Twente:

The chart below shows the same data temporally (with an 8-minute cutoff this time). The two fast laps by Top Dutch can be seen, as can the fast lap by the Covestro Sonnenwagen. Car problems show up clearly in this view, as does the consistent driving of the Belgians:


European Solar Challenge: Dynamic Parcour

The iLumen European Solar Challenge (iESC) is ongoing. All cars have passed scrutineering, either yesterday or this morning, and all but SER got some test laps in this morning as well.

Today (Friday) saw the rather nail-biting “dynamic parcour,” which counts for 22.2% of the final score (and which at times came down to milliseconds!). Results are shown above (stars mark Cruiser-class cars), and some snapshots below (click to zoom). Well done, Green Lightning!

Credits: 1. RED Shift shows how it’s done; 2. Spotting Stella Era; 3. The winning driver; 4. Aachen’s Covestro Sonnenwagen begins its run (notice the timing device on the side of the car).


European Solar Challenge: Thursday morning

In the iLumen European Solar Challenge (iESC), scrutineering will take place today (Thursday), followed by the “dynamic parcour” tomorrow (Friday) and the main 24-hour track race on the weekend.

Because of Covid, the iESC is not open to the public. Fortunately, there is a plethora of technology options for following the race remotely:

Here are some snapshots from Zolder (click to zoom):

Credits: 1. Aachen settles in at Zolder, 2. SER heads out on a 700-km drive early on Thursday, 3. Agoria prepares for scrutineering, 4. aerial photograph by Circuit Zolder.


European Solar Challenge: the teams arrive

Below (click to zoom) is my final chart of cars for the 24-hour iLumen European Solar Challenge to be held at Circuit Zolder in Belgium on 18–20 September (all car photographs except the Swiss one are mine, taken at WSC). Turkish team ITU have made it to Belgium, I am happy to say, so I’ve added them back to the chart.

Agoria, Aachen, Eindhoven, Top Dutch, Twente, and ITU are already at Circuit Zolder (as at the end of the day). SER plan to travel very early on Thursday. Durham is not attending, but in the “solar car family” spirit, will run their personal 24-hour race in the UK. The weather for Zolder is looking good, but with a chance of a tiny bit of rain on Sunday.

Agoria will do a live Q&A on their Facebook on Friday 18:45 (17:45 in London; 12:45 in New York; Sat 02:45 in Sydney), as well as a live start on Saturday 12:55 (11:55 in London; 06:55 in New York; 20:55 in Sydney). Zolder will provide a live tracker.