Tuesday, September 12, 2017

WSC 2017 Challenger Summary (Part 2)

Our first post on the 2017 Challenger class was a broad overview, but we didn't talk too much about specific cars, or what our favorites are. So, who do we here at MostDece think are the teams to watch in 2017?

First, lets quickly discuss the array. The 2017 regs allow for 4sqm of silicon, or 2.64sqm of multi-junction GaAs cells. We've seen teams quote around 24% for Si this year, and 35% for GaAs, which works out to 960W and 924W, respectively - not a huge difference! I think that cars with the smaller GaAs array should be able to be designed in a more aerodynamic manner to more than make up for this small power deficit... And when you account for the fact that GaAs arrays don't suffer as much efficiency losses in higher temperatures, and cooling the arrays with water has been banned this year***, I think a GaAs array should have been a clear choice this year! But it seems that few teams came to this conclusion - or perhaps only a few could afford it. Only Nuon, Punch, and Michigan switched from silicon in 2015 to multi-junction in 2017. See this post from Scientific Gems for some more details and graphs on this topic.

*** Reg 3.18.3 states "Spraying of water from external sources is no longer allowed" (Emphasis mine). So I think if you wanted to put a water tank in the solar car, and carry all the water that you are going to use for spraying for the entire race, all the way from Darwin to Adelaide, maybe you would be allowed to do that - as it's no longer water from an external source. But if WSC allows a team to do this, they'd better be extra careful to make sure the team can't stealthily refill it along the way.


Top Picks

The array discussion out of the way, let's discuss who our favorite cars and teams are this year. Looking over the field, I think there's a reasonable chance that the top five finishers from last year will be our top five again - they're all coming back with great cars.

Nuon

Nuon burst onto the solar car racing scene in 2001, winning their first WSC - and the next three in a row, as well. After 2nd-place finishes to Tokai in 2009 and 2011, they won again in both 2013 and 2015. They're the champs, what else can we say?

Well, a lot, actually. This year, Nuon is one of the few teams building a car with a multi-junction array, and it's a diminutive catamaran. And despite the new rules with a much larger bounding box than array size, most teams this year have stuck to a rectangular array (or close to it). Not Nuon - Nuna9 has an array with dramatically curved sides that conform to the sides of the wheel fairings, and the array doesn't extend forward or aft of the wheel fairings at all. The leading edge of one wheel fairing blends smoothly across the leading edge of the array and into the other wheel fairing - and the same can be seen on the rear of the car. They've distilled the asymmetric catamaran into it's simplest form, devoid of any and all unnecessary protrusions.

Photo: Jorrit Lousberg
(image source)
Photo: Hans Peter van Velthoven
(crop of this photo)

As aesthetically pleasing as this is, it serves a purpose. By eliminating the front overhang and the rear "winglets" that we usually see extending laterally from the fairings, Nuon has eliminated a lot of junctions on the car - junctions where flow velocity can slow, and energy-robbing vortices can form. It's clear that Nuon has put a lot of careful thought into their car this year, and they're going to be a tough team to beat.

Punch

Despite having what may have been the best car at the last WSC, Punch finished in 5th place in 2015. They simply didn't have their race operations perfected as much Nuon, Twente, Tokai, and Michigan; a few small breakdowns and a race conduct penalty (against their media crew) put them an hour and 25 minutes back from the winner. Talking with the team after the race, I think the team understood what their shortcomings were, and I hope they've worked on correcting that for 2017.

(image source)
(image source)

With respect to the car itself, Punch Two looks great. It's a very small car - overall, very similar to Nuna9 in dimensions and proportions. Punch has eliminated the front overhang and blended the front edges of the array into the wheel fairings in a manner similar to Nuon, although they've kept a rear overhang and the array "winglets" protruding sideways as the wheel fairings narrow to the rear. The multi-junction array shows that they're thinking out of the box rather than just building what they have before, and that they have deep pockets/resources to tap. I expect Punch Two to place very well in October.

Michigan

Michigan has been coming to WSC since 1990, finishing in 3rd place five time - but never higher. Most recently, Aurum finished in 4th place in 2015, slipping behind Tokai on the last day. This year, Michigan has made some bold moves - not only are they one of the few teams taking the multi-junction array option, they're also one of the few that didn't construct an asymmetric catamaran. Instead, they've built a long, narrow bullet car - or "monohull".

(image source)
(Photo: Evan Dougherty, Michigan Engineering)

Novum is long, low, and sleek; it's right up against the 5m length limit. The track width appears to be as wide as is practical for a car that is only 1m wide overall, and the rounded nose looks like it will do quite well in crosswinds. Looking at the design, I really don't have any stability alarm bells going off in my head. Similarly to Nuon, this design eliminates a pair of long array-fairing junctions on the car, although in this case they've eliminated tunnel rather than the junctions on the sides of the car.

Any time the rules change, the chance is ripe for someone to see something that everyone else missed, and gain an advantage from it. I don't think Michigan would have made such a radical change for change's sake; they must think there's an advantage to this over a catamaran. Eyeing the rest of the field, Michigan might have a good shot at breaking their 3rd-place WSC curse with Novum.

Tokai

Tokai won WSC back-to-back in 2009 and 2011, then placed 2nd in 2015 and 3rd in 2015. For 2017, they're taking a similar strategy to Michigan: a monohull.

(image source)

Unlike Novum, the 2017 Tokai Challenger has a silicon array. It's a little wider than Novum at 1.2m wide, but it may have similar frontal area: Tokai's car has a very narrow track width (0.6m), and the bottom of the car is noticeably narrower than the top. It also has a quite pointy nose (and I wonder what sort of drama this is going to cause around reg 2.7.4).


What I said about Michigan goes for Tokai as well: Tokai is a smart team, and I don't think they would make such a radical departure from the now-near-ubiquitous catamaran if they didn't think there was an advantage. Tokai is certainly a team to watch this year, and may be a contender for the victory.

Twente

Since first attending WSC in 2005, Twente has consistently finished in the top ten. In their early years, they finished 9th in 2005, 6th in 2007, and 8th in 2009 (managing to complete the race after a nasty crash!). Over the past three WSC events, they've been steadily marching upward: they finished 5th in 2011, then 3rd in 2013, and finally a nail-bitingly-close 2nd-place finishing 2015; only 8 minutes and 20 seconds behind Nuon.

This year, they're bringing a long, narrow, silicon array catamaran named Red Shift.

Red Shift
(image source)

I've gotta say, I'm... not super impressed. It's not that I don't like the car - it build quality looks great and I'm sure it'll do well at WSC, but I expect more of Twente. The team has always built unique cars. From their tilting array cars in 2007 and 2009, to The Red Engine in 2013 (still one of my favorite solar cars from the upright-seating era), Twente has always flaunted convention. Even 2015's Red One was fairly unique in a field of nearly identical catamarans - the 3-fairing layout and the short front, long back, driver forward proportions stood out from a largely-similar field. When I think of a team that consistently thinks outside the box, I think of Twente, and given the regs changes this year I had high expectations.

But Red Shift, uh...

...

...it looks like a 2/3 scale Nuna 8.

Knowing the rivalry between the two schools, I fully expect Dutch assassins to arrive at my house for saying that, but dang. Every time I look at Red Shift, I just see a smaller Nuna 8. I can't unsee it. Seriously, here's a photo of Nuna 8 from nearly the same angle:

Nuna 8 at WSC 2015. Photo by MostDece

The other four top finishers from 2015 have all made significant and divergent changes to their designs, but Twente hasn't even taken advantage of the larger bounding box to break out of the (nearly) rectangular array planform.

I expect that Twente will do well; I'd be shocked if Red Shift is not a top-ten finisher, and I think it has a better-than-most chance of finishing in the top five. But given how the rules changed this year, I'll be really surprised if the winning formula is "2015-but-smaller". Of these five cars, it's the only one that I would be genuinely surprised to see crossing the finish line first.

Rising to the Challenge

In 2015, the rules hadn't changed too much from the previous WSC and I was pretty certain with my predictions for the Challenger class. I correctly called four out of the top five, the one I called but missed came in 6th, and my sole "dark horse" pick - a rookie - came in 7th. Not bad, right? But with the rule changes this year, there's potentially a lot more uncertainty. Particularly, if the monohull cars have stability issues, we could see some shakeups near the top. Here are the other teams that I think we should keep an eye on:

WSU

Western Sydney University is a relatively new team - their first WSC was only in 2013, and like most rookie teams they failed to finish the race the first time around. But they've aggressively built their program, and the team finished an impressive 10th at their second WSC. Depending on how you look at it, I think that they were the team that improved the most since 2013, and their 2017 car appears to be another large step up.

(image source, Photo Credit: Sally Tsoutas)
(image source)

Unlimited 2.0 is a long, narrow, silicon array catamaran. WSU has pushed the wheelbase much shorter this year, resulting in shorter wheel fairings and less whetted area. The trailing edge of the car is one clean sweep from the wheel fairings to the array, mimicking the aft end of Nuon's entry - no protruding array winglets here! Super smooth. It's clear the team has put a ton of effort into the car, and they've taken advantage of the new regs to optimize the aerodynamics of a silicon catamaran as much as possible.

Like we mentioned above, WSU finished 10th in 2015 - but three of the strong teams that finished ahead won't be challenging them this year. EAFIT-EPM and MegaLux aren't attending WSC in 2017, and Team Arrow has shifted to the Cruiser class. Soooo... can WSU do better than 7th? They'll need to beat Stanford to do so. Can they break into the top five??? I think it's plausible, especially if any of the monohulls turns out to have stability problems out on the Stuart Highway.

Stanford

After a nasty wreck knocked them out of the 2007 event, Stanford finished 10th in 2009 and 11th in 2011, then rallied to a 4th place finish in 2013. Then the team slipped to 6th in 2015 - not because the team got worse, but because they didn't get as much better as other teams did. 2015 was a great year with a a ton of quality teams and stiff competition; in particular both Punch and Michigan improved dramatically from 2013 to 2015, and passed Stanford in the process.

But Stanford is still a strong team, and has had a particularly good strategy team over the past two races. In this post about the 2015 race from Scientific Gems, note how level Stanford's line is - they knew what speed they they could drive from day 1, and they stuck to the plan even when they were basically neck-and-neck with MegaLux through Kulgera.

(image source)
(image source)

This year, Sundae abandons the rectangular array entirely. The front overhangs the wheels a bit, but at the rear the array becomes dramatically narrower, and it has a trailing edge treatment similar to Nuon and WSU. I expect that Stanford will run a good race again this year, but can they best their WSC 4th-place record and finish on the podium for the first time? They have a lot of tough competition to beat in order to achieve that goal.

NIT

I haven't seen many folks in the western solar car community paying attention to NIT this year, but I think this is a team to keep an eye on. NIT finished pretty far back in 16th in 2015... but they did so with a car designed for Suzuka! Their new car Horizon 17 is their first car that they've specifically designed for WSC, and they claim that it's the first one that they've done aerodynamic analysis on.

(image source)
(image source)

Just like Nuon, NIT has also dramatically curved the sides of their array to conform to the wheel fairings, although the narrowness of the car and the larger area of the silicon array require it to have some array overhang on the front and rear. Still, the way that NIT has blended the overhangs into the body is incredibly clean. The only thing I'm not a fan of is the length of the wheel fairings; they seem overly long compared to other cars this year and result in a lot of extra surface area. Regardless, Horizon 17 is a huge step step up from 2015's Horizon Z, and I expect the team to do extremely well.

Wild Cards

These two teams are both teams that have shown in the past that they can run a good race, and are fielding pretty radical cars this year. In contrast to the teams above (that I'm fairly sure will do well), I'm not sure if these are going to perform well, or disastrously. Whatever happens, they're worth keeping an eye on.

Kogakuin

Kogakuin unveiled Wing back in late June, and it's... wild, unique, and bold. But risky.

(image source)
(image source)
(image source)
(image source)

Conceptually, it's similar to Michigan and Tokai - a long, narrow, low monohull, and it's a silicon array car like Tokai. But rather than integrate the array into the body, Kogakuin has chose to elevate the array over the driver as a separate unit. It looks neat, but I'm really scratching my head about the aerodynamics. I think the separate solar array results in a very poor array-area-to-surface-area ratio, which is a pretty basic rule of thumb for guesstimating how a solar car design will perform. On the other hand, if you can maintain laminar flow, surface area doesn't matter as much... I'm also concerned about crosswind performance with that weird flowing wing shape, as well as how the wing will hold to 3022km of vibration across the outback. Kogakuin is doing all sorts of other weird stuff under the hood, from the high trail multi-link front suspension to the mono-shock, rigid tube axle, hydraulically steered, multi-link rear suspension.

It's all just completely weird. There's not a single conventional thing about this car. Ordinarily I'd dismiss this as some sort of engineers-gone-wild monstrosity or an art project masquerading as a car, but... I can't help but wonder. Last year's Cruiser was wicked fast. The fit & finish on Wing appears exemplary. They've been doing a lot of testing, and the team seems to have a TON of Bridgestone support and sponsorship. I don't think they built a car that looks like this for whimsical reasons; I think they are very seriously gunning for the overall victory, but I really have no idea what their chances are.

NWU

The team from North West University in South Africa ran a solid race in 2015, finishing 11th. This year, the car they're entering is kind of a blast from the past: a narrow driver pod with outrigger wheels, and a tilting array. It's a really large car, too. Due to the fact that the array is short and wide, but placed entirely aft of the driver compartment, NWU's car is 4.98m x 2.05m. It's right up against the 5m length limit, and the only 2017 car I'm aware of that is wider than the old 1.8m limit from the previous decade of regulations.

(image source)

This style of car largely fell out of favor in solar car racing's infancy; the extra power from the tilting array just didn't make up for the aerodynamic shortcomings. But maaaaybe it makes sense again with a smaller array? Due to the small 4sqm array regs this year, NWU's tilting array is much more tightly integrated with the driver pod. It's able to be located lower and entirely aft of the driver, rather than elevated on support pylons and extending over the driver's head as seen with the huge tilting arrays on the '87 Ford/Aurora entry, the '87 Spirit of Biel, the '93 Solar Kiwi, etc... And Twente drove tilting-array cars to reasonably successful finishes in 2007 and 2009.

I think NWU has a good shot of breaking into the top ten this year, but a top five finish is unlikely.

Wrap Up

Alright, I've typed a whole lot but it's time for me to finish this post up. Here's my prediction:

I think Nuon, Punch, and Michigan are the most likely top three. The teams that chose multi-junction arrays are on to something; they're getting nearly the same power (or perhaps more, depending on heat!) out of a smaller, lighter, more aerodynamic car. That's a winning formula. Nuon is obviously the favorite for 1st, but I'm also going to guess Michigan will not finish 2nd. Assuming all three cars run a clean race, I expect Michigan to be clearly ahead or clearly behind Nuon and Punch; I'll be surprised if the performance of such wildly divergent designs ends up close enough that Michigan places between two very similar catamarans.

Tokai and Twente are good bets for 4th or 5th, but may face tough competition from WSU, Stanford, and NIT. Particularly, I wouldn't be surprised if WSU beats Twente. Kogakuin is so weird that I don't have a good instinct for how they'll do, but they could be very, very fast.

I also expect JU to do quite well - I almost placed them up under the heading of "Rising to the Challenge", but they just didn't quite make the cut for a top-five contender in my mind. I think they're a good bet for a top ten finish though! Blue Sky and ITU are other teams that I think have a good chance of making the top ten.

Some final bet-hedging: If the monohull designs are a lot better than I think, expect Michigan and Tokai (and perhaps Kogakuin?) to place very highly - possibly edging Nuon further back than they've ever finished. On the other hand, if the monohull designs aren't as good (or have stability issues/accidents on the road), they'll end up further back and will give teams like WSU, Stanford, and NIT the opportunity to place much better than in 2015.

Of course, it's a race, and anything can happen - we'll just have to wait and see.

16 comments:

  1. Is the driver allowed to spray whatever is remaining of the drinking water on the panels, if only to remove sand and dust?

    ReplyDelete
    Replies
    1. That's a good question and I would suggest probably not. It does raise the question though of when it is acceptable to clean the array and how. Presumably if no charging advantage is to be gained the array can only be cleaned with a liquid after dark and any cleaning during daylight hours would need to be done with a dry cloth.

      Nigel

      Delete
    2. Unless the team carried the driver's drinking water in the solar car all the way from Darwin, the water came from an external source. So no.

      But after dark when the battery is out of the car, they can clean dust off however they want.

      Delete
  2. That's another great post Mostdece.

    I sincerely hope that you are wrong in your initial prediction(before you hedge your bets)of the top three.

    If the race is dominated by teams using GaAs arrays it will make a mockery of the competition. It will send out the message that only teams that can afford to spend c.$250,000 on cells need contemplate the possibility of winning.

    It also sends the wrong message to the watching public about the viability of solar power. Yes we can make a vehicle that runs on the power of the sun but you'll never be able to afford one.

    My preference would be for the rules to stipulate a reasonable budget limit for arrays (retail price, not the cost to teams)so that advantage is gained through engineering skill rather than through deep pockets.

    Returning to the race, it is always difficult to see anyone beating Nuon but the 2015 race proved that there is always a chance.

    One conceivable chink in their armour this time might just be their driver cockpit. It worries me when teams say that their drivers are the only people who are small enough to fit in the driver seat. Now, if the three drivers were chosen at the formation of the team because of their driving abilities combined with their size that is one thing. If however they were chosen from within the team after the car was built simply because they were small enough to fit then that is a concern. Looking at how close the race was last time, if their drivers had been even 0.5% less competent, confident or committed Nuon might well have lost the race to Twente.

    I'm sure that the team's training programme will have ensured that the drivers are all able to proceed safely but I'm not sure that they can be taught what is needed if it comes down to a battle, for that you need the best driver - not the smallest one.

    Nigel

    ReplyDelete
    Replies
    1. People have talked about budget limits for years, but I've never seen a reasonable proposal for how it would be monitored and enforced in a way that A) doesn't involve a lot of team oversight by WSC between races, and B) doesn't cause tons of controversy and drama at the race.

      Agreed on the driver, but that's a losing battle. Aerodynamics is king, and most teams pick their drivers almost exclusively based on size. I don't think Nuon's cockpit is particularly worse than any of the top teams.

      Delete
    2. I agree that an overall budget limit would be almost impossible to implement, especially with so many donations in kind, but with the huge difference in cost between silicon and gallium cells a limit, for the array alone, which excludes the latter would be pretty simple, i.e just ban gallium unless or until it becomes as cheap as silicon.

      Nigel

      Delete
    3. >just ban gallium unless or until it becomes as cheap as silicon.

      Won't ever happen. Too process intensive compared to monocrystaline silicon. The price gap is growing, not shrinking.

      Mutli-junction was effectively banned 2009-2015 due to the severe area restrictions. Sure you could do it, but you were allowed such a small aperture compared to silicon that you'd be shooting yourself in the foot to do so.

      I'm just happy that there are competitive array options other than whatever Sunpower happens to be selling this year; the solar array portion of the event had been monotonous and boring for years.

      Delete
    4. My personal opinion is that the current rules are just very slightly biased in favour of GaAs, and a very small adjustment in areas should be made.

      Delete
    5. As to Nuon and drivers, my understanding is that they picked three drivers very early in the design phase, and then used their dimensions as design constraints for the cockpit.

      Delete
    6. Not normally one for responding on forums such as these (though that takes away my ability to compliment on well written and very informative blogs!), but as a former Nuon Solar Team member I would like to confirm that WSC solar car drivers are chosen from within the team at an early stage (unless the team consists of only Dutch giants, but that has never happened) and the car is then indeed built "around them", optimised for aerodynamics (amongst others) but also constrained to fit the driver for 3-4 hours in Australian heat.

      The testing phase (in which each of the drivers drives well over the 12 mandatory training hours) is then used to gain confidence in the fact that all solar car drivers possess the necessary driving and car handling skills for the Australian outback at target speeds.

      (As a final note, it is not unheard of for our team to switch around the driver order for a day/stage given weather forecasts, driver health, team confidence or any other factors.)

      Delete
    7. Thanks for your reply. I should have known that Nuon would have every angle covered. It is re-assuring to know that drivers are chosen for their ability first and then their size. Having been to concerts in Holland I know that it not easy to find small people there.

      Nigel

      Delete
  3. Great overview and analysis indeed Mostdece.
    Predicting the top 6 or 10 in the Challenger class this year is definitely more difficult this year compared to 2015. I looked up our 2015 predictions on Scientific Gems (https://scientificgems.wordpress.com/2015/10/10/world-solar-challenge-the-favourites/#comments) and we were not only agreeing a lot but also very close to the actual results. This time the variety of designs allow for much more speculations. I have been trying to find out how much energy the Si and MJ arrays actually produce at cell temperatures between 30 and 75 Celsius but have not been able to find or calculate that (yet). Then you have to also know or guess what the actual expected cell temperatures will be during the race from North to South for "typical" conditions this time of the year. I know that at least some teams have given (extra) consideration to the cooling during driving, but I have no idea how effective that will be compared to the air flowing over the array when driving at 90 kms/hr.
    I am also curious about the aerodynamic advantages of the monohulls compared to the catamarans. And how much energy or time does 10 or 20 kg more weight cost for the slightly bigger cars with the Si arrays?
    I expect that the qualification at Hidden Valley will be a bit easier for the catamarans than the monohulls. When the teams are close like in 2015, 10 or 15 mins advantage at the start may well be worth fighting for this year.
    In 2015 some of the top teams have suffered damage (from cattle grids or other) costing them important time for repairs. Avoiding unplanned road side work could mean the difference between 1st and 5th place.
    Overall I agree with your expectations and hope the monohulls can put up a good fight without fear of being turned over or even take to the air with a Kogakuin Wing.

    Erik

    ReplyDelete
  4. Thanks for the fantastic analysis. It's good that MostDece is written with a team. For one person alone this detailed research is more than a full time job.
    Following your results, weather will play -once again- an important role. How perform monohulls with crosswinds? How perform GaAs at cloudy weather? Due to the smaller arrays the average speed of the race might be well below maximum speed again. Therefore the use and the performance of the battery could gain relevance.
    As a German I also hope that Sonnenwagen Aachen is a candidate for the top ten.
    Dietrich

    ReplyDelete
    Replies
    1. Yes, weather always, always matters at solar car events. I don't think I've ever seen a race that wasn't significantly affected by it.

      Crosswind performance is key at WSC because there are almost constant crosswinds across the Stuart Highway. Tokai and Kogakuin concern me the most... And although I'm not worried about the stability of NWU's car, I'm doubtful of its crosswind performance when they tilt the array.

      A cloudy race will mean cooler temperatures, so GaAs arrays won't have their temperature advantage, and any aerodynamic advantage the smaller cars have will be reduced if clouds keep the race slower. So it's possible that a cloudy race could actually favor silicon cars.

      On Sonnenwagan Aachen: It's often challenging for a new team to simply finish the race! Teams have finished well on their first race in the past, but the competition at the top will be fierce this year.

      Delete
  5. How do you think the monohull designs will perform on the track at Hidden Valley? If those teams are conservative to prevent a rollover, how will a start further back in the pack impact their ability to get out of Darwin with minimal energy use (passing the slower teams)?

    ReplyDelete
    Replies
    1. My 2 cents as someone who has rayced there twice is that it is better to keep safe on the track, there are a couple incidents every year at Hidden Valley, and the cream always rises to the top by the time they reach Adelaide anyways.

      Delete