1st revision of 2015 rules released

Kev, at least they're discouraged on site now (which they really always have been unwritten-like). Ever been to a dirt kart track? All that is is tire prep. Learning what combination of chemicals to rub on the tires and how long to take the torch to them to get them right. There's definitely an art, and if someone were to have individual sets of tires for accel and skidpad they could probably pickup a tenth or two, with the same compound (we had a good 40 Shore A swing between 3 different sets of the same compound of tire).

Other than the bafflingly complicated ETC rules and sound limits that have already been discussed, the only one that really confused me was the 45 degree tube bracing rule. Could someone diagram that? It really doesn't make any sense. Which plane are we talking here that it can't be out of skew with?

Example of why brace must at an angled no more than 45 degrees from the plane of the bent tube.

Out of interest I thought I would show the page count for some of the rules releases. This is from 2001 (the first year I competed as a student). I am missing 2003 and 2005. By the way it might be worth creating a place to access the history of FSAE (i.e. previous rules) to see how it has evolved. Here is an image of the page count vs. year:

http://i.imgur.com/Pa1yg9i.jpg

2001 - 86
2002 - 93
2004 - 104
2006 - 125
2007 - 124
2008 - 118
2009 - 105
2010 - 108
2011 - 130 (Alternative frame rules added)
2012 - 131
2013 - 163 (EV rules added to current rule set)
2014 - 163
2015 - 176

By the way this is my first image post since joining here in 2002. Even with a com sci degree it takes a little while to get around to some things :)

Kev

Hey Kev,

I'll quote first, so it's easier to explain:

"This looks like a violation of A6.1 and A6.3. For other difficult design areas (such as custom engine) it has been proven to be in the capability of a team to do the complete design and fabrication inhouse."

I don't think custom tires are a violation of this rule. Since its intent is teams basically buying their whole car. That's not what happens with custom tire design (at least within our partnership with Apollo or Eindhovens)(I don't know the details of Darmstadt, so I can't speak for them). It's not like we go to our sponsor and say we want a tire. We actually design everything about it, they just deliver the raw material and production process. Just like someone designing a titanium upright or a printed steering wheel. If you can't have those production processes sponsord, a lot of teams wouldn't build the cars they are building now. No validation on aerodynamics in a windtunnel for example or stock bought ECU's.

You should see a tire not so much as one part, but an assembly of a lot of different parts, with each its own function. If you've never seen the actual internals of a tire I suggest you cut open an old one. The tread compound is only a small part of the tire itself ;)

Daniel,

I have found these clauses very poorly written. I don't think the intent is for the whole car at all. I have heard a rules committee member state about these sort of items is that they should be available for all, or none. I think the intent is shown in the description for the faculty advisors:

"Faculty Advisors may not design, build or repair any part of the car."

Already it is very poorly defined, but the intent is clear. What if a university employees engineers not as faculty advisors? What if I was to not be the nominated faculty advisor for ECU, and instead that role was given to someone else. In my external design company I then started designing parts (or co-designing) for use on the FSAE car. I am acting within your definition of the allowances made to professional engineers. That means any alumni etc are free to be heavily involved in the design of components and subsystems as long as they do not consitute the whole car.

So me, a bunch of ex UWA mates, the ECU alumni design and build considerable sub assemblies of the car and deliver them to ECU for assembly.

A clear violation of intent. In a completely different class to parts such as professionally designed steering racks or the Drexler diff which, while designed externally for fsae they are made readily available at a fair market rate.

...

I was in a similar situation at UWA when we developed the first hydraulic interconnected system. It was a Kinetics system (using their patented system), but we designed and manufactured every component in the system. This included writing our own software for the calculations. Some converstaions were had, but no person at Kinetics had any engineering design input into of any part of the system. Quite frustrating to be asked at comp by a design judge (Oz) for the catalogue we ordered it out of. I do not mean the same offence to you.

At the end of the day we were able to design, build and test the system based on nothing more than publically available knowledge. We also did not request any provision to be the only team able to use the system.

My question is could the same be asked about the tyre design. You made the comment that the design of these compounds required highly specialised help.

I do not think there is any violation on the basis of manufacturing of the tyres, only on the basis that from your comments indicate professionals were required for the design of the tyres.

I also think there are a lot of teams violating the intent of the rules. There have been examples in engines, chassis and probably just about every system. For example you can design a carbon chassis including all tooling etc and have it manufactured, but there have been clear cases where there has been significant professional involvement in the design process. I think this is a breach of the intent of the rule as well.

Unfortunately there has been no case I am aware of where this rule has been enforced, or any idea of what the penalty would be. Maybe it shouldn't be in there at all, or maybe there could be a better definition.

Kev

I have attached two pictures to this. One that shows the increase in teams over the years (from the world rankings), as well as the increase of number of rules pages. Secondly I have attached a picture that shows the increase in the number of teams each year, vs the increase in the number of rules.

The total of teams is a little off towards the end as there are a few teams listed both in electric and combustion rankings. It is probably fair to say that the rate of increase in the number of teams has begun to drop after around 2010.

Some interesting points:

- First up the rate of growth of the competition is incredible. I am not aware of how many universities there are in the world, but almost continual growth at around 36 teams every year for 13 years is impressive. I wonder what the sturation point is. It is helpful to note that at the start of the data there was only 1 comp, by the end there were 11. There is also a bit of a pattern where once the number of teams per comp hits close to 60 a new comp will be formed. The average number of teams per comp tends to range between around 50-60 (except for the period with only detroit). This indicates over time the growth has come from rolling out the competition to new places, rather than an increase in the number of mega-comps. Obvious next steps could be official comps in India and China. I am aware of interest in Malaysia. Maybe extra consideration should be given to improving the situation of comps world-wide rather than a particular focus on the issues of the larger comps.

- In the period of decreasing rules (2006-2009, although 2006 is the local peak) there were 174 teams added, and an additional 4 comps started. For comparison in 2002 to 2005 (the previous four years) there were 2 comps added and an extra 119 teams. In the following four years (2010-2013) there were 2 comps added and an extra 139 teams. The compeition had it highest gross increase during the period of rules reduction.

- The two best years for growth were 2007 and 2008, both of which were in years where the rulebook decreased in size. 2006 was the next best year and had a larger rulebook than 2005, however there were 2 big comps added in that year (Germany and California)

- The two worst years for growth (2011 and 2013) coincide with the largest increase in the size of the rulebook. 2011 was the previous big aero change, 2013 added Percy legs, and a host of small changes. These were also the years of alternative frames and then EV inclusion into the main rulebook.

- The next 3 worst growth years were 2003, 2005, and 2009. I don't have the rules for 2003, but 2009 was the year chassis templates were introduced. Excluding the 2009 template rules addittion and looking at the 3 other years with the decreasing rules pages and you see the 3 best years for growth in FSAE (40 more teams than the three best remaining years)

- The addittion of EV's has not increased the overall growth rate, more clearly it appears that there is a conversion of old teams to EV.

...

Make your own mind up, but is that decent enough data or evidence to at least suggest that simplifying rules, and avoiding too many big changes helps the health of the competition?

At least if your goal is to provide good practical engineering training to as many students as possible.

Kev

http://i.imgur.com/6zmrJ6e.jpg

http://i.imgur.com/dxGfKTk.jpg

Quote:

---

Originally Posted by DMuusers

Another thing about the development of compounds. It's not something that you just 'do'. I don't think any team will be able to do this without a significant and highly specialized sponsor. There's a lot of reasons highly trained specialists design these compounds so therefore I'm not afraid that a team will make their own tire without a sponsor. You can always add a rule that the tires must be produced by an official tire company if you really think students can do this on their own.

How do you know that teams that used tire softeners have never had a rupture? It's something most teams wouldn't like to share on facebook anyway ;)

---

I assumed that "making a tire" is really difficult. Thanks for clearing that up. But as we have seen motors (Zwickau, Zürich) or even engines (OxBrooks, Auckland, ECU) developed by students, at some points, maybe even something like that will happen. Of course with an input of a tire company and their machines but well, at least if we let you guys do it "properly", we have to accept guys doing it by themselves...

True that a team would not write "Yeah, killed 3 tires today due to tire softening-tests, gonna repair the suspension and go on #softsoftsoft". But as I said, something like that has never happened at a competition in the last 5 years that I can think of.
If we would forbid the use of softeners in the name of safety, there is a lot that should be restricted as well...


Kevin,

I am all for deregulation - let the teams do everything that is safe.

But I think your linkage between "more teams if rule book is slim / less teams if rule book gets bigger" is imaginary. (In fact there is a study showing there is a correlation between the number of storks and the birth rates of countries...(http://www.researchgate.net/publicat...bies_(p_0.008)))

If a new team starts the competition I can't imagine following scenario:
"Yeah we wanna build a FSAE car. Look here are the 2015 rules: Uff 176 pages. Last year it was only 135 pages.. nah let's not do it"

In my opinion there is a lot of different reasons why there is an increase in teams or not.. Like Hungary joining the WRL a couple of years ago "opening" the list for Eastern European teams. Now with Russia / India starting we will see a new push in team numbers with probably the same rules...

Julian,

I clearly state that the strongest correlation is between growth and the number of competitions. I also should note that the number of competitions is as per the world rankings site. In 2000 there were actually 3 competitions. Formula SAE / student has clearly grown on the back of more comps.

I did not make any statement that there was a clear correlation between regulation length and the amount of growth (although I admit I implied it). I pointed out particular data points, and asked an honest question as to whether this data suggests there could be a correlation between simplicity and stability of the rules and the rate of competition growth. I happen to think that there is and that your stork analogy is a little misleading. In that case there is no clear or expected relationship. Here we would assume that the regulations will have some impact on the competition, and one that may be able to be measured. There are very few data points to go by, which severly limits the assumptions that you can draw.

What I think you can state without bias is:

- There is no evidence that increasing regulation does anything to increase growth, if there is a correlation it would appear the other way.
- There was only one sustained period (i.e. more than 1 year) where ther rules were reduced in length. I now have all the rules as far back as 1999, and to 1979 sparodically, which confirms this further into comp history.
- In the period of international comps (i.e. 1998 onwards) there have been only 3 years (out of 18) where the regulations were made more compact (2007, 2008, 2009)

The obvious question is that if you don't use data like this what do you use to test the effectiveness of the rules documents? I am all for tracking number of incidents etc at comp, but I do not know of any stored non-anecdotal evidence. What about reliability as a measure to see whether teams are coping well with the rules? Hasn't got better over time. What about average number of rules queries?

We are encouraged to practice data-driven design practices, but collect and analyse no data regarding the effectiveness of the regulations. On another post it was suggested that the rules committee be seen as a steering committee. If that is so what KPIs do they use? I would say growth rate is a pretty good place to start.

If you take the stork/people analogy and assume that growth (or other improvement) is decoupled from the regulations, then why bother changing the rules? It would save a lot of people a lot of time. Alternatively if you say the changes to the rules have improved or hindered the competition, on what basis do you make that judgement?

I think this data should be shown. When the old hands talk about the increasing regulation and complexity there is often no data shown to back it up. This at least shows how the rules document has been increasing in size with data that can be peer reviewed. This is the first time I have seen this presented on the forum, and it at least shows a clear trend of increasing size of the rules documents. Between 2000 and 2010 there was an average of 2.7 pages per year added to the rules (7 increasing years, 3 decreasing). This in an increase of 33%. The five following years have seen an additional 68 pages added at an average of 13.6 pages a year, with no reduction in length in any year. An increase of 63%.

A reversal of this trend or at least some stabilisation may be a good thing for the rules committee to consider. I was speaking to a few design judges about the rules around the period of the rules reduction. There was a specific push from a number of the rules committee members to both clarify the regulations and reduce the number of regulations. What is very impressive is that under that committee they managed to add the whole chassis template sections while still reducing the overall size of the rules documentation. Their efforts for improving the rules probably have not been given as much kudos as they deserve. They fought against the trend and achieved their objectives of returning the rules to a more compact document. During that time the number of competitions and teams increased at a higher gross rate than any other period. This may or may not be linked to their efforts.

The lego company does provide a good example here. You may or may not be aware that as time went on Lego got into pretty big financial trouble. They had a product that was well sought after, but their product line had got overly diversified. In a restructure they consolidated and trimmed the fat. Non-profitable product lines were cut and the number of unique pieces sold drastically reduced. They have been going from strength to strength since that consolidation. Now "everything is awesome". Periods of consolidating what is important to driving the success of a venture is very important for the long term growth of organisations.

I will ask some simple questions about the data presented here. In 2011 the alternative frame rules were added to the documentation. In that year the regulations increased by 22 pages. Has the introduction of the alternative frame rules been a success? If it has what are the measures that show the success? How many teams have benefited by this change? What is the cost benefit relationship? On review of something like that you might find that some of the elements were successful and some were not. Maybe they should be kept as-is, maybe the useful parts could be folded into a more compact rules set, maybe they should be eliminated entirely.

Now that the aero rules have changed, what are the KPIs that will be used to measure success or failure of the changes? For the previous change there was an attempt to encourage teams to investigate aerodynamics. There has been mixed data back from that and it is nearly all anecdotal. On one hand more teams ran wings, but the story goes that many did so without implementing good design practice.

Turbo and supercharger rules have changed. What KPIs will be used to measure success / failure of the changes?

ETC rules have changed. What KPIs will be used to measure success / failure of the changes?

These are honest questions. I am legitimately interested in both the desired goals and how you can determine whether they have been met for a given rule change. I am also interested in what data we may want to collect as a community / organisation in order to provide evidence to show whether targets have been met, exceeded, or failed.

Kev

Kevin, it will be very hard to differentiate between correlation and causation with the teams vs rules lines debate: Correlation does not imply causation. If we rename the vertical axis to 'number of pirates' it will still show correlation but implies nothing. You also need to take into account the dampening effect of Student Familiarity with the FSAE projects as more people know about it, less will be new entries. But this is solely something which relates to the number of total teams.

A better metric would be, assuming there is a correlation somewhere to be found, to take the new entries (actual participation) per year and re-entries as separate graphs. This way you can see the overall trend of the FSAE competition (getting new and more universities and colleges to participate) or the stagnation of the trend (more the same teams participate).

Quote:

---

"- The addittion of EV's has not increased the overall growth rate, more clearly it appears that there is a conversion of old teams to EV. "

---

As building a EV car can be quite hard, as there (were/are) a lot more rules and regulations, it might not be the thing you would chose for from the start. But then, you can always do the "Challenge accepted" part of FSAE.

Wouldn't it be a good idea to take the final 2015 rules if they are there, sit down with a lot of (former) FS(AE) team members and reduce the rules to the minimal needed for safe racing, challenging builds and equal opportunities who do not solely rely on a big sack of money? Post those rules here and afterwards give them to the rules committee for their consideration?

Tristan,

I agree entirely regarding being able to confirm causation. I don't think the data is clear enough to make anything like that claim with any confidence. Not enough data points and too many other factors are included. Better metrics may help, but I doubt we would be able to confirm or deny causation with the data available. At this point it is presented for interest only, and maybe to stir a few pots. It could be that the rule instability has helped maintain growth, rather than see it fall over time (although I doubt it though).

I used the teams as per the world ranking for two reasons:

1 - It is readily available and able to be checked
2 - It is a measure of "how many institutions / students reached"

I think the total number of students reached is more important than how many of the teams are first timers. Some of us long termers may get fatigued by the comp, but at every university there is a constant stream of new students.

I think a metric along the lines of:

Quality of Education x Number of Students / Total cost

Quality x number as some sort of measure of total education benefit contributed by the competition. The equation becomes total benefit / total cost. I'm not sure where the girlfriends / boyfriends lost would fall into the equation. Maybe a negative term for social unrest :)

The number of students is reasonably easy to estimate and can be based on the number of teams. The total cost needs to have some input from institutions. The quality of education is much harder to quantify objectively. Maybe it can be collected by a brief student / university / industry survey at every comp. Offer a decent reward for entries; a small cost for valuable data.

Comp changes can be then measured on whether they reduce cost, increase participation, and improve quality. Participation has definitely been shown to increase, but has it outstripped cost, and has education quality improved?

This is a very similar metric to what Universities use to measure effectiveness. We monitor student satisfaction, cost per student, number of students, and benchmark learning outcomes, . Universities are also regularly reviewed by external professional bodies that assess quality of education based on work samples and accepted standards. Quality is the hardest measure to quantify, and to that end it is approached from various directions to try and eliminate subjectivity. Cost and numbers are relatively easy.

For all the problems Universities have they can be reasonably effective at measuring how well they do their job. Formula SAE / Student is advertised as an educational competition. Maybe some practices adopted by learning institutes could be applied to meet the same means. Anything that helps ensure that such a great learning opportunity is still around and thriving in another 25 years.

...

I like the idea of taking a razor to the rules, and doing so collectively. You mentioned some overall goals for the competition:

- Safe racing
- Challenging builds
- Diversity in builds
- No necessity for large budgets to be competitive

Are there any missing, can everyone agree on these?

Do we have clear agreed upon goals for the competition in general. Without those goals and targets to measure by, any attempt to alter the rules would likely degenerate into personal preferences.

Maybe a topic for another thread.

Kev