I decided to place this article here at this time because I see an incomplete version placed elsewhere. I am not sure precisely why that is being done, though I have some ideas. So I decided the best way to proceed is to first document the information at my site as it was submitted to the solar car team more than 10 years ago. There were several drafts, and the version placed at my site is what I considered the complete version (minus the Abstract to avoid redundancy), which was dated 16 January 1994. I certainly could not document all this during Sunrayce 93 - I was much too busy for that those 7 days - though did gather all the required data.

I did not attempt to place any specific information regarding my solar car activities at my site prior to this, because it wasn't long before it became apparent there would be little benefit from my participation in this activity other than for the teams I helped. Many people in the solar car community highly respect me, and many team members appreciate my great efforts. Yet I received very little tangible benefit personally. So I conclude that if I am not exclusively responsible for something, such as other material I place at my site - forecasts for Mount Pocono and Ann Arbor, articles about the effect of the forest on local climate, etc. - it is best not to mention it. I make an exception for this because I feel I must.

My comments below require some familiarity with solar car racing and how Sunrayce 93 was run to completely understand them, but I feel I should state the following:

The importance of weather forecasting for solar car racing is probably overemphasized. The first long-distance solar car race was the 1987 World Solar Challenge in Australia, basically run along the Stuart Highway from Darwin to Adelaide (a part of which was not even paved at that time S of Darwin). General Motors (GM) decided to enter a car on short notice, and spared little expense. They received consulting from high-tech corporations, constructed a solar array (for the car) of very expensive and highly-efficient gallium arsenide cells, etc. It is not my purpose here to tell their story, so I suffice to say that during that race GM's Sunraycer took advantage of a favorable weather situation, which allowed them to escape what became an intense desert thunderstorm complex when all other teams were caught in it further behind. It is questionable if they would've needed that - they were well ahead at that point and their car was much faster than any other. Yet because of that, they won the race by more than 24 racing hours (finished early day 6, when the next team I think finished day 9).

During most races, nothing will happen such that weather can be used so advantageously - though it can be used to optimize performance in many situations. That was the purpose of the weather group providing information for our race strategists and providing them with the daily solar radiation forecasts I include and mention which were used for race simulation. Though our team did many things to optimize vehicle performance also, there was so much as I am aware no strategy or tactic which greatly improved our position over other teams. We won the race to a large extent because of everything we did before racing - designing and building a reliable and efficient car which would collect and utilize solar energy well - so it could race longer at a particular speed than others. I helped with the design of the solar array, modeling many geometries for solar energy collection under various conditions. Of course, there are many other aspects of racing a car - funding, logistics, etc., - which is not the scope of this discussion.

Our race strategy was pretty aggressive. We went from 10th position (where we qualified prior to the race) to 1st during a primarily cloudy racing period day 1. Doing that is not easy when you consider 3 vehicles of one team's caravan must pass at least 3 of another's in traffic. The 1990 team would not think of doing that, but the extended charging periods for this race tended to reward going fast. Because we got plenty of charge that evening, it was the right thing to do. Though array position does not matter much around midday, being able to tilt the array toward the sun to charge makes a huge difference approaching sunrise & sunset. Thus forecasts for the charging periods were most important. Yet I only can include data I have available for verification. The radiometers were not always on after racing nor could the vehicle it was atop be parked in a position to get accurate readings.

Examining (our positions in the standings), you can see that we raced a quite consistent speed days 1-3 - averaging right around 50 km/h (including traffic lights, etc.). We began day 4 with plenty of charge in our batteries, so the strategy to that point was pretty close to perfect. Then we had 2 consecutive primarily cloudy days. If a person could not only forecast but see the future - knowing days 4 & 5 would be as cloudy as they were and day 6 as clear - the most efficient thing to do would be racing at a slower constant speed days 4 & 5, nearly draining our batteries at the end of day 5. That is more efficient than racing fast one day and slow the next, because the power required to race a specific speed is approximately proportional to the speed cubed (speedł). Note that we averaged about 49 km/h day 4 and 28 km/h day 5 - to a large extent racing the same speed day 4 as the first 3 days, and much slower day 5. That is not close to optimal, but it was better energy management than Pomona which averaged 48 & 23 those days (who were basically shadowing us because they had a small lead because of a couple penalties we took), and much better than GWU which averaged 50 & 17. If we could be certain we'd have poor charging periods the evening of day 4 and the morning of day 5 - then we could've maybe averaged a constant 40 km/h both days 4 & 5 instead of the 35.5 we did. That is simply using the cube of the speed - the strategy group would really be needed to make an accurate simulation - though this would perhaps save 90 minutes of racing time. Yet uncertainty in the weather as it is, we had to allow for the possibility of better charging periods. I.e., better to use all of the solar energy available inefficiently than not be able to use a large amount of it.

So basically what I am saying is that if there was any large gain to be made using weather in Sunrayce 93, it would've been going slower day 4 to conserve more charge for day 5. This is why I spent so much time analyzing data the previous night, though I ultimately had no control over what the strategists would do. Some people seem to imply the time of clearing day 5 made such a difference. That was nice and certainly made everyone happy, but didn't matter much strategically because we had the better car. No matter when skies cleared, we'd be further down the road and finish much faster, take a smaller penalty, etc. (penalty time for distance not completed was only 2 minutes per mile - about the same speed as we raced the first 4 days). Pomona raced the same speed as us until they simply couldn't. If that was 40 km/h days 4 & 5, we'd probably be much further ahead. Conversely, if days 4-7 were all clear, they might've been able to travel the speed limit and we could do nothing about it. Note that even after virtually draining our batteries day 5, 2-3 hours of good charging during the evening and subsequent morning was all the contenders needed to basically race at normal speeds day 6. So you have to be careful if you're gonna travel slow one day to save energy. If you think you have a better car, the safe thing is to race with the lead cars and let them eventually lose pace if conditions become bad - which is exactly what we did. So this is not meant as any sort of criticism of the fine job that our strategists and team members did. It is stated because some people seem to be trying to imply things regarding this which simply aren't so - perhaps in a way as a criticism of a small bad portion of my forecasts. Considering the detail I tried to provide and the uncertainties being considered, I was very happy to provide the numbers for simulation so well - they could've been much worse. Nor do I particularly care about the opinions regarding this of people who do not understand race strategy.

Regarding the religious reference in the paper - though it is possible, I doubt we were really that important. Whatever the case, the skies did clear in time for Maize & Blue to finish day 5 under its own power.

Text and embedded images are copyright of Joseph Bartlo, though may be used with proper crediting.