Building and prepare a FSAE car in one year.

[Claude Rouelle]

Nick Roberts did post the following message in the ".. what the heck is going on at Goodyear" part of the forum

*****************
Claude - I'm curious. If you were a team leader in charge of managing a team, how would you structure the year? Which aspect of this engineering project would you prioritize? Assuming we have 10 months to design, build, and test a vehicle what would your target timeline look like?
Ours, regrettably, looked like this last year:
Design - 5 months
Build - 4.5 months
Test - .5 months
****************

Nick’s question touches a specific issue of FSAE team success and I think it deserves a discussion on its own.

My answer would be
- 5 months designing
- 6 weeks building
- 4 months testing

Here is my advice: Out of 50 weeks I believe success in FS / FSAE is achieved by spending the same amount of time, energy, people, focus and money in 2 main chapters: A design and B testing.
A1 Concept 7 weeks
A2 Simulation 4 weeks
A3 Drawings 10 weeks
A4 Manufacturing and Assembling 7 weeks
Total of A 28 weeks = a bit less than 200 days

B: Testing, breaking, redesigning, re-simulate, retesting, analyzing, driver training and improvement etc... About 22 weeks = about 150 days

Comments:

1. One advice: Do not start A2 before A1 is finished. Similarly do not start A3 before A2 is finished.

2. Another advice: Do not look at other cars on pictures on or the internet before you finish your own concept. Adopt the form follow function method. Make first a list of the function your car is supposed to achieve. As I said in another part of the forum for each part you design imagine that you are that part.

3. About the car manufacturing. Here is a story…. In 2010 I judged a new car from University of Bologna at FS Spain. Too heavy but amazing design for a first year car. In fact they won the design and finished second overall. There were only 17 cars (first FS Spain) but honestly not bad at all for a first year car. The Bologna students told me they had a faculty adviser who refused to have them accessing any machine shop until ALL the drawings were finished. When he meant ALL, he meant ALL: not only every important structural part like chassis, uprights, wishbones etc.. but EVERYTHING; chassis and data acquisition electronic loom, dashboard, gear lever, throttle cable bracket , bracket for the radio, every single nut and washer was in the drawings.
I have to say I like this approach: I see too often too many cars where it is obvious that some parts (like the ARB, or the steering column just to name a few) were not included in the initial design. I can hear thinks like “Ho s**t, there is no room for the ARB drop link or the bodywork did not include the water radiator” being screamed in workshop…

Well the Bologna guys told me that once all drawings were finished the cars and the jigs were built in 7 weeks. 7 weeks! The other advantage, they said, was that there were no big surprises in the parts order list and in the manufacturing / purchasing budget. It is a bit as when you paint a house. The painting process is very short compared to the preparation

A racing touring car team we work with in Argentina also spent a long time on Catia and it made the team owner very nervous but they also built all the jigs and the whole car in 5 weeks. An no surprises.

4. The 2 main challenges that students face is lack of car reliability and driver training. The best way to train drivers and improve reliability is to test, test and test.

Drivers. I have seen once or twice a FSAE car driven by a student who was also a go-kart or single seater (formula ford or similar) professional racing driver. On a circuit of 50 seconds lap time, these rare and talented drivers were consistently 1 to 2 second faster than any other driver in the team. I also have seen professional drivers making lap time a FSAE car 1 to 2 second faster that students lap time in just a few laps of testing. Let’s face it; if the team you try to beat has a driver who is 1.5 second/lap faster than yours, you will not be able to compensate it with smart car design. So you need to test, compare train and improve your drivers. You need time for that.

Reliability. Formula One teams with 300 to 400 million $ budget and 300 to 400 people still manage to break things. How could you think a car produced by inexperienced students would finish well if finished 3 days (if not worse) before the competition. Yeah I know there are rare exceptions of cars finishing endurance but these guys did not win and they were lucky. Once.

5. Of course you can’t achieve to build a car in 5 months if it is over-complicated, you do not have at least 30 students, a good budget and several years of experience. Success in FSAE is at least a 4 to 5 years project. These days, teams are more professional and top cars are way better in performance and reliability than they were even 5 years ago so it will be extremely difficult for a new team to be competitive first year out.

6. Finishing a car 5 months before the competition is even a bigger imperative for a team which as to air or sea ship its car 3 to 10 weeks in advance….

7. This schedule does not take any glitch into account. And guess what; as with any other project in the world S**t will happen. One more reason to have plans B and to test 5 months before the competition.

8. All the reasons above support the idea of making an A team and a C car. People then machine. Not the other way around. If you try to build an A car with a C team the only thing you will get is people pointing fingers at each other.

Let’s do the math guys: if you want to compete in Michigan in May and you do not test your new car by Christmas there is very, very little chance your team will be building cars which will be successful year after year.

You could be extremely lucky once but the results and the progress won’t be consistent. I am sure there will be a lot of comments, maybe disagreements on this but that is what good principles of project management tell and this what I have seen in many FS / FSAE competitions for many years.

[Kevin Hayward]

Great post Claude, but I can't help but feel you are responding to a question that makes a big incorrect assumption. The question was:


Claude - I'm curious. If you were a team leader in charge of managing a team, how would you structure the year? Which aspect of this engineering project would you prioritize? Assuming we have 10 months to design, build, and test a vehicle what would your target timeline look like?
Ours, regrettably, looked like this last year:
Design - 5 months
Build - 4.5 months
Test - .5 months


Why must we assume only 10 months? There is no rule that says you have to design and build the car within one year.

I would suggest that any team that takes 18 months (or 24 months) to design build and test a car will have a huge advantage over a team that only uses 10 months. We did exactly that with UWA from 2002 onwards while still producing a new car every year. We figured that the further away from the winners you were the more you needed to design and test (as per Claude's statements). At the same time you need to manufacture almost two cars a year to fix problems found during tests, have spares for running, and conduct plenty of destructive testing. All of this obviously manufactured to a high quality.

Lets make an assumption that we have 12 months between one competition and the next. I would make the suggestion that you design for 12 months, test for 10 months, and manufacture for 8 months. You may need to consider different team roles to facilitate this and definitely need to have a focus on long term goals.

I find it amazing that teams lock in designs very early in a year, for cars that will be virtually identical to last years attempt. Unless they were the winners from the previous year they have no basis to believe they will improve their finishing place. I am a strong supporter of people that claim that reliability is one of the most important aspects of FSAE vehicles. By finishing the comp you almost immediately beat 50-60% of the teams. However reliability is the plinth and high performance is the trophy. There is absolutely no point in building a car that is not better than the year before (faster, more driveable, cheaper etc.).

Kev

[Claude Rouelle]

Kevin,

I was hoping you would jump in this forum because I respect your opinion and your experience.

I agree with you on the idea of a 2 years (or even more) plan.

Where I disagree with you (or misunderstand you) is the 8 months manufacturing process. Manufacturing time should be numbered in weeks not months. If manufacturing a car takes 8 months it shows significant weaknesses in the team organization (bad estimation of budget, parts not ordered on time, choice of unreliable suppliers, lack of team members discipline, communication etc..) and/or major mistakes in initial (probably incomplete) design.

I know racing is not FSAE and we have to be careful to make comparisons but if we look at percentage instead of plain duration, I do not know any serious race car manufacturer COO who would allow a manufacturing time being 2/3 of the design time, even if it is a first prototype. The manufacturing manager would be fired.

I do maintain my point; if well designed and with a well organized team, car manufacturing and assembling should be done in 6 to 8 weeks.

I understand your disagreement on "very early lock car design". I would just say that the car there is is always better than the car that could have been.

The only way out is a long term planning mix between a) Let's making it happen now even if it is not perfect and b) Let's make sure that our new ideas and what we learn form this year mistakes are taken into account on next year car. That is why you need either faculty adviser who guide students in long term planning and/or students who are involved in FSAE for at least 3 years. Some school do not allow student to get involved in FSAE for more than one year; that is a problem.

Looking at it from a larger perspective, as a design judge, I obviously do like well designed, well executed, reliable and quick car and I really also like innovations. But what I like the most is when I see that students are in control of their plan, know their strengths and weaknesses very well, measure well their limited resources and how they used these resources at their best. I prefer a team building a B car who can explain to me what an ideal A car should be, but why they reasonably decided to build the car as they present it, who come the following year with a B+ car and so on .... than a team building a A car, even made it quick but cannot explain how and why.

Recently I asked a first year team "if you had to do it again what would you do, why and how?". Not only the student had a immediate verbal answer but he opened a binder with detailed description of what next year car goals were, with already some lap time and other engineering simulations, as well as cost and manufacturing time estimations. Although I was impressed I could not give them more points because I judged this year car not next year car but I predict this team is on big raise; they are in control.

[JT A.]

Nick asked for a 10 month timeline because that's what our team operates on. As much as I wish we did work year round, our team just isn't structured that way. A few of us stay involved during summer, but it's mostly just testing, maintaining cars, & autocross practice. We technically aren't allowed to start design over the summer because at our school, the car is supposed to be completely designed by seniors as their senior design project.

Having FSAE count as a capstone design project has its advantages and disadvantages. It is very convenient for the core members of the team who would do FSAE whether they got school credit for it or not. On the other hand, we get a lot of people who sign up under the idea that it will be a fun & easy grade. They treat it like any other class and never put in the effort required to be a top team. On a volunteer-based team you could just kick these people off or demote them to less critical parts. On our team they get assigned to critical components and we (the other students) can't kick them off/reassign them because they are enrolled in the class and to kick them off the team or take away their responsibilities would basically be to block them from graduating.

One of the things we struggle with the most is the fact that we have a 90% brand new team every year with no knowledge about FSAE or sometimes even cars in general, and we have to convince them to start putting in 5x more work and dedication than they've ever done in their life.

Knowledge transfer is another huge issue, which we've tried to improve this year with some success. We've created a database of previous years' test results, SAE papers, magazine articles, a library of textbooks, etc. We also have a few grad students & underclassmen with a lot of FSAE experience who have been helping set team goals & priorities, teaching vehicle dynamics, giving CFD and FEA tutorials, basically trying to speed up the learning curve. As the saying goes, you can lead a horse to water but you can't make it drink.

I don't mean to make this sound too much like a whiny "our team has it so bad" post because in the big picture we are pretty fortunate, what I'm really getting at is this question: for a team set up the way ours is, how would that shift the 10 month timeline? I fully agree with the importance of spending several weeks on what you call "A1" stage, defining your overall concept and priorities. But to make good decisions on your concept & goals you need to have some understanding of vehicle dynamics and how to design a car tailored for FSAE events. How much time would you devote to a "FSAE 101" period to get new members caught up, without losing too much testing time at the end of the year?

[Claude Rouelle]

JT A.

No magic solutions here.

2 solutions I can think about.

1. When judging a team and a car some design judge ask "can you show me your internal engineering report?" We look for something more extended than the mandatory 8 pages design report, something which shows a substantial transfer of knowledge and experience (we made this mistake because ... do not do that because ...we choose this solution and we believe it worked because ... and here are the drawings, simulations, calculations). If you went through the pain of writing such report (more than as you wrote "just" a database of previous years' test results, SAE papers, magazine articles, a library of textbooks, etc....) it will show how committed and efficient you are to "leave a legacy" specifically applied to your team to next year(s) students.

2. If you want " to have some understanding of vehicle dynamics and how to design a car tailored for FSAE events" there are excellent books (see other part of this forum) and you may want to attend the seminar OptimumG organizes a few times a year. It is 3 intensive days. The binder each participant receives is 700 pages. I have been told by several faculty advisers that it is the equivalent of 2 semesters.

You may also want to see http://vimeo.com/41854831 for a few non-engineering advice.

[Kevin Hayward]

Claude,

I agree that building a car should only take a couple of months if well planned for. I would suggest the extra manufacturing time is for things like test pieces for mechanical testing, test equipment, alternate systems, improved parts post car testing, spares, track-side equipment, ergonomics rigs, and early mockup parts for concept validation.

Much of what you would manufacture may never get seen by a design judge and may get spread over a larger timeline than the final car build. So I would say out of the 8 months of manufacturing maybe 2-3 months of it would be hard focused build time.

The only other item that may extend manufacturing time is that University workshops are often shared facilities with time restrictions placed on the teams. Obviously this is not ideal, but needs to be planned around.

JT A.,

I am aware of other programs that run like yours does. It is definitely not ideal. Have you thought about talking to the supervisors about changing the practice and actively encourage younger students to be involved. There is a lot of education based research that extols the virtues of peer mentoring. There are both large benefits for the mentors and those that are mentored. Setting up a team where it is only final year project students reduces the education outcomes of the project and hence the value to the university.

Kev

[jlangholzj]

I've noticed that this last year we really don't stop at all. Almost immediately after competition all of the team leads are already designing and changing for the following year. At that point in the year we've been testing for however long and have been staring at the car for the last year so what needs to be changed (in handling/egress/design) is pretty fresh in our heads. Typically a month into school we're making final design decisions on the roller components and no more than a month later the rest of the car is finalized. Then money allowing we start the build process.

we follow a very similar structure to what claude has laid out.

-- previous years race
-- testing previous designs and making modification
-- Concept
-- Simulation (done almost in parallel with concept if there isn't something thats behaving like we want)
-- Modeling and documentation
-- Manufacture
-- Tuning and testing/driver training

Its a year-round thing. I've lost lots of sleep because of it but I wouldn't trade it for anything else. The fruition of parts that I've tested, run calculations on and designed just make me want to start earlier the next year.

Kind of behind the scenes we've been listening to claude's, pat's and steve's advice along with all the other design judges and i feel that it's done us well over the last few years. The timeline that's laid out before you might be a bit of a stretch (depending on team size, etc) but the harder you try to hit it, the closer you get to making good things happen!

[Claude Rouelle]

Kevin,

I will not argue with you, I respect you too much for that but I have to ask not only you but also the people interested in this topic; why would a brand new team was able to build everything from scratch in 7 weeks (and still make a very good first year car!) and another one in 8 months. Only 4 causes I can see
1. One has access to a lot of tools and the other one very little.
2. One is very organized and the other is not (I look a budget as part as the organization)
3. The team which needs 8 months wants to do things so perfectly that they can't make it happen on time and on target. Or some of their team members have been in FS / FSAE for too long in the same university (yes it does happen) and they can't think outside the box. That is the same reason why in a very competitive world like Formula One it is good (although painful) to see drivers or engineers changing team.
4. All of the above.

[Claude Rouelle]

Static event preparation will be for mornings (FSAE / FS Students usually work afternoons and nights)

http://fsae.com/groupee_common/emoticons/icon_smile.gif

[jlangholzj]

Originally posted by Claude Rouelle:
Static event preparation will be for mornings (FSAE / FS Students usually work afternoons and nights)

http://fsae.com/groupee_common/emoticons/icon_smile.gif

not true! ....well...kind of...our "nights" typically extend into the "mornings" during the month or so before competition

[Kevin Hayward]

Cole,

I wonder at the wisdom at leaving practical design projects for students until their final year when you are trying to produce engineers capable at design. The more time and practice spent in guided design makes the graduates much better. This activity doesn't have to have credit assigned to it.

The seniors will get much more from it as they will be teaching the younger students to design. And as I'm sure Claude will agree teaching someone is one of the best ways to learn.

That being said I think that Texas A&M must have close to the most successful FSAE program that is run with the approach of only seniors designing.

I am currently a faculty advisor at ECU. In 2010 the team had 4 final years (of a 3 year BTech course) and about 30 first and second year students. Fast forward to 2012 and they have a team that has most of those first and second years still in the team along with a whole lot more in training with a greater proportion being engineering students. A reasonable percentage of the team are on their 3rd car. The result is a team that needs very little faculty guidance and has processes in place to teach new members.

The most interesting design work for future cars is given to the newest team members, while the older hands take car of the essential work for the given year. Younger members are active in both vehicle and part testing, and manufacturing. They are taught skills like CAD, FEA, CFD, procurement, machining, composite manufacture and design from the older students. Each year the cars have been improved considerably.

It is not a perfect team by any stretch of the imagination and access to facilities has been a big problem, especially this year with the moving of the engineering workshop. However I would challenge any faculty advisor to provide a reasonable argument (backed by either education theory or research) why this approach is not far better than having senior only design projects. Usually if I see senior only projects it is done as a way for the faculty advisor or the school to exert more control over the project.

If the faculty advisor is the only one who is around long term, they are the only one guiding the direction of the project, including things like the switch to hybrid you mentioned. These teams are best when the culture is guided by a good advisor, but the decisions and running of the team are the job of the students.

Kev

[Kevin Hayward]

On the 8 months of manufacturing I could provide a reasonable idea of how this might be broken up:

Build of car and spares for comp: 2-3 months (Pretty close to what Claude recommends)
Manufacture of on-car testing parts for previous cars: 1 month
Manufacture of parts for destructive testing: 2 months
Manufacture of trackside equipment and test rigs: 1-2 months
Manufacture of mockups for design verification and visualisation: 1 month

This is based on the assumption of student and facility availability for about 3 days a week and the facilities available at ECU.

A mechanical workshop is an asset to a team. Most companies will not bother to build a workshop unless it is used almost continuously. I hate having assets unused. If it is free and available and you have the students able to machine then you have a chance to make something useful. That may be a part for a car or some crazy idea that may be part of a 3 year test program for a system that may never make it on the car. It may be a part to try out a new manufacturing method. It may be yet another carbon rod for destructive testing to improve the confidence of previous test results. It may be a part for younger students rebuilding one of the older cars. It might be a display stand and so on.

A design team is an asset. It should never be not designing.

A test team is an asset. It should never be not testing.

Some teams may wonder why you would ever want a team of 60 people designing and building an FSAE car. Meanwhile I see ways to keep a team of 80 or more fully occupied and actively helping the team and as an educator I would rather see 80 people well trained than half a dozen.

Kev

[Schrader]

Originally posted by Kevin Hayward:
There is a lot of education based research that extols the virtues of peer mentoring. There are both large benefits for the mentors and those that are mentored. Setting up a team where it is only final year project students reduces the education outcomes of the project and hence the value to the university.

Kev

This is taking the thread on a bit of a tangent, but it is something I have been musing over for quite some time and the importance of which I feel is under-emphasised.

The ability to comprehend something from another persons point of view is truly one of the most valuable life skills to have. It is also one of the most challenging to develop and integral to being an effective peer mentor. Galileo put it nicely..

"You cannot teach a man anything; you can only help him to find it for himself." - Galileo Galilei

The FSAE project offers the opportunity for constant "mentoring" between students in a broad context, not just limited to the technical aspects.

I think that the true value of this project is in the development of "social skills" such as these in the participating students. It is difficult to convince people of the value of such skills, without them having a context to "find it for himself". Using the allure of race cars, this project "tricks" students into learning these things in a way, usually without them being aware.

"Even in such technical lines as engineering, about 15% of one's financial success is due one's technical knowledge and about 85% is due to skill in human engineering, to personality and the ability to lead people." - Dale Carnegie

[Claude Rouelle]

Social skills.....

Matt, I vigorously applaud your comments and especially the one from Dale Carnegie (course of whom i participate in - which literally changed my life 19 5 years ago; without it I would have not created OptimumG - but that is another story)

A french poet (and also WWII jet fighter pilot) Antoine de St Exupery says in "Le Petit Prince" book that "Essential is invisible for the eyes"

As engineers we always compare visual things. FSAE students (and many judges) look at the car only and students compare FSAE team size, budget, year of experience, days and km of testing. But how do you measure and compare team spirit, team organization, students positive attitude and ability to rebound after a big car or team failure?

These invisible things are more important than any visible ones and are often not taught in engineering school or rarely used din engineering companies.

FSAE design judges can barely quantify and appreciate these teams strengths but they are so essential to team success and ... I would say .... happiness (because you can also, rarely though, achieve good results in bad mood and at an enormous cost to your mental and physical health)

I am so glad that you did bring this Matt! So glad...

[Max Trenkle]

This Summer my team was down to two people, myself and the current President. We literally had to make every single decision for the new team structure. The decisions most pertinent to this conversation were the following:
<LI>Design in the Spring, test in the Fall. This allows for better transition periods, since the wealth of knowledge of car systems will not all graduate as soon as competition is over.
<LI>Freshmen have Junior leaders, Sophomores have Senior leaders. Freshmen start out in marketing, maintenance, and fabrication, so they can learn the backbone of FSAE. Sophomores do design projects under surveillance of seniors; in this way they can get a better feel for what works and what doesn't work. Hopefully this system will promote healthy team dynamics and improve transfer of knowledge.
<LI>Fully incorporate a business structure as the team structure: in my endless efforts to entertain the interests of the business majors at my university, I have managed to acquire a few. They are invaluable. Proper 'division of labor' has completely turned this team around. Ever since we got business majors on board, the designers and I have managed to move all of our deadlines up by one month.
<LI>Move next door to Baja. They have been winning for decades, and we have not been close to them. Everyday I think of new way to combine our efforts in ares such as fabrication training, FEA training, etc. Their team models much of what this thread covers. How else could they be so consistent?

I am desperate not to repeat history: the only documentation of any design that exists is from 2001-2003. The only cars that we have are 2006-2008, one of which finished competition, I believe in the top 30. There are no documents for any of these cars. Absolutely nothing. This is why we have not competed since 2008.

For a painstaking year, the current president and I have designed the car on our own, and have documented every last bit of it. The 2013 car will be a two year car, but it was necessary. Now, we are not alone. We are now 20 members strong, and the team will continue on. I am happy that my efforts have allowed me to revive the team to what it should be. Thanks to guys like Big Bird, Claude, Z, and so many others on this forum, I didn't have to figure it all out on my own. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Big Bird]

Some great discussion going on here. I'll try to be brief because I know I am usually not.

HAPPINESS
When our team was happy, our results were good. When our team was not, our results weren't either. RMIT's earlier success can be attributed to a guy by the name of Steve Price, the regular Friday night barbeques at his place, and the strong team bonds formed accordingly.

TIMELINE
I appreciate the merits of the "design before you build" philosophy, sequential process, meeting gateways / freezes before moving on. I appreciate how it limits reworks, and how it gives checkpoints where everyone knows exactly where the project stands. I do not like the way such thinking confines your thought processes and closes your mind to alternative pathways.

I would set one goal - have a running car by end of August (our comp is in December). You might then think a little creatively about how you can use some concurrent design to meet this goal.
- If you are short on for resources, identify which components are top priority for change, and which components you can carry over their design. Hub design good enough? Lock it in, start manufacturing them in Month 1, know that you have to design your uprights around a known component. Use your workshop resources while they are still quiet*
( *At RMIT, workshop facilities are dead quiet for first semester, and booked solid in second semester as all the final year project students submit their designs for manufacture at the same time. I imagine it is the same elsewhere.)
- Think creatively about what gets your car on the ground and driving by "End of August". In June last year, Monash had an open night and I heard many an "expert" scoffing at the rather ordinary looking slap-up uprights (the condescending "ours are going to be much better than that" - consider the deeper truths hidden in that innocent sounding statement ). They were testing at end of August, getting valuable driver time and testing components and subsystems. Later in the year, some nicely done, more "perfect" uprights suddenly appeared.
- Think of how you might test stuff in isolation during the year, before the whole car is done.

I loved how Uni of Wollongong put their first car on track originally with a mule engine on carbs - about 5-6 months before comp. The race engine was concurrently tested on the dyno. By the time the race engine was ready, the chassis was well tested and drivers were trained too.

I am perfectly comfortable with concurrent design - as long as there is an understanding of the interactions and dependencies between systems.

Design time often gets blown out because designers don't make commitments. We keep all of our options open as long as possible just in case something better comes along. Sequential design ensures everyone has to make a commitment to their final design in step with each other (like some big ol' mullet-shod boot-scooting chain).

However if you are willing to make absolute commitments to some parts of your design, then you can start manufacturing those bits straight away while the workshop is quiet and the manufacturing crew is bored.

You just need to make sure that you allow an appropriate period for the fully assembled SYSTEM to be tested well before the event

(Note that I am an INFP on the Myer Briggs personality test. My recommended professions are counsellor or librarian. Most engineers are ISTJ's. Most engineers think I'm a pain in the butt to work with.)

So in answer to the first question, if I had 10 months, I'd plan my project so that I had:
10 months testing
10 months designing
10 months manufacturing

http://fsae.com/groupee_common/emoticons/icon_smile.gif

[JulianH]

It is great to read all the different approaches, different team philosophies and also different team organisations.

For the Zurich team, everything is a little bit different (again..).

We start always in September, the new team (roughly 15 mechanical engineers and 3-4 electrical engineers) come together. We keep roughly 5-8 "old" team members who -in the most cases- only advise the new students.

Therefore the new team members first have to study, check the rules and the old car to make design solutions.

Right now, four weeks into the project, we try to decide in what direction to go in the next year.

Last year, the final concept was fixed by the end ofNovember and the whole designing (CAD, FEM, CFD) was done by Christmas or around New-Year's.

The technical drawings are done by the first week of January. (We have some exams in January and February, so that's slowing us down a bit).

The monocoque was done in the first week of April, the whole car was completed (without aero) in the next two weeks. Shakedown was on May 10th or so.

We then tested (with some problems..) up to the competitions.


So basically:
September-November: Learning, Concept
November-Dezember: Designing
January-March: Manufacturing
April: Assembly
May-July: Testing

Will be interessting to see how it turns out this year...

Regards,

Julian

[Charles Kaneb]

The main difference between a 6-8 week build and a 6-8 month build for a simple car is manpower.

If you have 15 team members, including 3 or 4 experienced fabricators with the ability to commit beyond-a-full-time-job amounts of time to the project, you can build a competitive FSAE car in under 10 weeks.

If you have 5, and it's the first car build for all 5, it will take almost a year to build all the components.


Designing for a fast build has some major advantages. It forces all of the interconnections to be be designed properly - the build season's too short to allow sequential design. You can't design and build a part, then measure it to enable the next part to work with it. It also gets rid of the "we'll design it so it can accommodate either of these" syndrome that adds weight and size remarkably quickly. Overcomplicated and overweight assemblies go out the window as people realize how little time they have to make things.

On the other hand, it also eliminates any aggressive concepts that may simply not work - there's no time to redesign and rebuild in the middle of the build season. Consequently, the process of reducing weight off the "base" of a 400-lb single-cylinder/tube frame car with parts that are "block of steel or aluminum, holes here, here, and here to mount this, that, and the other" consists of a long time doing detailed FEA optimizations in the fall and a long time at the lathe and mill in the spring. It virtually guarantees you will be killed in static events. Heavy optimization of each part and using conventional processes to make them is an expensive way to end up with a simple car. You cannot convincingly lie about costs of conventional parts. The nonexistent "cool factor" and high cost fouls up your business presentation. The lack of several systems leaves you with nothing at all to discuss with several design judges.

Combine the need to minimize design risk to compress the manufacturing schedule with students that are almost all there for only one year, and you can see why Texas A&M brought a tube-frame car with a naturally aspirated engine and no aerodynamic devices to a competition in 2012. That it came within a couple points of the highest dynamic event score shows that the Bruce McLaren method works.

Moving away from this concept will take a lot of work and risk. Aero, turbo, and monocoque will each require a "Great Leap Forward" into the unknown. Each of them represents a concept that will require a high level of development to exceed the performance of existing systems - but each with the potential for significantly higher performance. 25 million people died in the last "Great Leap Forward".

[Z]

Many interesting posts here. Not surprisingly, I agree most strongly with Kevin and Geoff's views.

In particular:

Originally posted by Kevin Hayward:
Why must we assume only 10 months?
...
We figured that the further away from the winners you were the more you needed to design and test...
...
I find it amazing that teams lock in designs very early in a year, for cars that will be virtually identical to last years attempt. Unless they were the winners from the previous year they have no basis to believe they will improve their finishing place.
...
There is absolutely no point in building a car that is not better than the year before (faster, more driveable, cheaper etc.).
...
Much of what you would manufacture may never get seen by a design judge...
...
There is a lot of education based research that extols the virtues of peer mentoring. There are both large benefits for the mentors and those that are mentored. Setting up a team where it is only final year project students reduces the education outcomes of the project and hence the value to the university.
...
I wonder at the wisdom at leaving practical design projects for students until their final year when you are trying to produce engineers capable at design. The more time and practice spent in guided design makes the graduates much better. This activity doesn't have to have credit assigned to it.
Nicely summed up by...

Originally posted by Big Bird:
I appreciate the merits of the "design before you build" philosophy, sequential process, ...
I do not like the way such thinking confines your thought processes and closes your mind to alternative pathways.
...
So in answer to the first question, if I had 10 months, I'd plan my project so that I had:
10 months testing
10 months designing
10 months manufacturing
Both Kevin and Geoff have been involved with very successful FSAE teams. Both suggest that to win FSAE you should NOT adopt a one year program of sequential design->build->test, culminating in compete-once->retire.

IMO, and assuming you want to win (???), FSAE should be seen as a multi-year, non-stop, iterative, looping, convoluted, complicated exploration of the myriad possibilities inherent in the FSAE Rules, and within the Laws of Nature.

This is why I find Claude's advice so perplexing:

Originally posted by Claude Rouelle:
Here is my advice:
...
A1 Concept 7 weeks
A2 Simulation 4 weeks
...
Do not start A2 before A1 is finished...
Perhaps I'm misunderstanding you, Claude, but how are students supposed to pick the right concept for the whole year BEFORE they do any simulations at all? Toss coins? Roll dice? "I kinda like this one, 'cos it looks real cool..."? http://fsae.com/groupee_common/emoticons/icon_confused.gif
~~~o0o~~~

This "concept first" approach, which I fear is all too common in FSAE (particularly "Let's just do what we did last year, but, err..., better."), sounds like,

Team Leader of Gold-Miners, "Ok boys, our "concept" for this year is that we'll set-up camp under these trees. Yep, nice and shady... And the soil looks especially soft, so easy digging. Next, we'll..."

Newbie, interrupting, "Err, but..., excuse me..., I've done some simulations based on geological theory, and the highest probability of finding a rich seam of gold is in that igneous rock formation over there..."

TL, "What!!! Geez, you noobs are daft... That's out in the blazing sun! And those rocks are hard!
No, no no... What you noobs gotta understand is that gold-mining is all about digging holes. BIG holes!
Right, so after we set-up camp we'll spend about three months simulating, designing, and building some new hi-tech digging equipment. Now, I've heard that some of the other teams are using carbon-fibre and titanium..."

N, "Err..., but the gold...???"
~~~o0o~~~

On a more positive note http://fsae.com/groupee_common/emoticons/icon_smile.gif, my high level points simulation (= Levels 3&4 in Geoff's "Reasoning..." thread) suggests that a FSAE car should be:

1. Cheap to build. So as to win Cost, and leave plenty of cash for testing, etc.

2. Quick to build. Giving more time to practice the song and dance routines needed to score well in Presentation and Design. And, of course, for more testing.

3. Be RELIABLE!!! To score points in ALL dynamic events.

4. Have high lateral Gs. Preferably twice that of the other cars, so about 3+G. This would give ~41% faster times through the corners (= most of the lap), and thus push the opposition back towards zero points in Skid-Pad, AutoCross, and Endurance (because of the the Tmax multiplier of ~1.45).

This perfomance attribute gives by far the biggest potential points gain over the competition. Required to achieve this are either magical tyres with CoF=3+ (not likely, based on historical evidence), or good aero downforce (quite likely, given the evidence of countless aero cars well in excess of 3G on standard racing tyres).

5. Have high longitudinal Gs, mainly forwards to score well in Acceleration. However, there is only a small points gain here, so less important than lateral Gs. Also, longitudinal Gs require a high power expenditure (because power = force-vector.dot-product.velocity-vector), which implies an expensive engine and poor Fuel Efficiency.

6. The least points come from things like ridiculously complicated double-wishbone&push/pullrod&rocker suspensions, electro-pneumatic paddle-shifters on multi-speed gearboxes, steering with bevel-gearbox AND rack-and-pinion, extensive "sexy" sidepods that take ages to build but have negligible aero effect, etc., etc. Yet these are the concepts many teams seem to decide on at the beginning of the year. http://fsae.com/groupee_common/emoticons/icon_confused.gif

Methinks that if there were more "heads-on-chopping-blocks", then more FSAEers would be out there in the hot sun, chipping away at that igneous rock formation.

Ahh..., when the 3G-lateral FSAE car...??? http://fsae.com/groupee_common/emoticons/icon_smile.gif

Z

[Claude Rouelle]

I understand the confusion that some of you can have so here are some additional explanation

Of course every step has some overlap; No FEA without drawing or vice versa... Similarly, concept and simulation are interlaced.

However first things first.....

1. By concept I mean decisions to be made in car main characteristics
- Basic car definition: Engine (1, 2, 3, 4 ... cylinders, turbo or not)? Or EV? Wings or not? Tire choice etc.. A lap time simulation software could help you to decide where are your priorities are in just a few days
- Other concept issues are definition of target such weight distribution, CG height, some essential parts weight and stiffness or acceptable compliance target (chassis, upright for example)
- The most important part of the concept is the reading and full understanding of the rules. In terms of car concept that is often the # 1 weakness of FSAE teams
- Also understand what make a good team, what makes winners win and it is not only about the car itself. That research takes time
- The concept part will also involve some education, reading and learning process. I will also involve objective evaluation and comparison on one hand of goals and on the other hand team effective abilities to reach these goals (budget, education, experience, team sizes, manufacturing capabilities etc...). The car concept is inextricably connected with project and risk management.
- In fact by concept I understand not only basic car concept but also team concept as well as collection of information.

2. Once the main car concept has been decided then detailed simulation can be used. For example
- Using tire model to decide target of camber variation in roll, heave, and steering and from there your chassis and upright pickup points coordinates
- Chassis torsional stiffness and weight optimization
- Choice and calculation of spring, ARB, dampers etc....
- Inlet manifold and exhaust flow simulation and optimization...

I hopes that helps

[Kevin Hayward]

Claude & Z,

I think there is a little bit of confusion due to terminology about concepts and simulation. When Claude mentioned Simulation before I took it as detailled design work. At no point did I think he meant not doign conceptual lap time simulations early.

When I was at OptimumG I saw a number of teams (professional) that just didn't do any calculations or have any indepth understanding when running cars. It is what kept (and still keeps) OptimumG a very valuable company to these teams.

Furthermore I would challenge people reading this thread to see that Claude, Z, Big Bird, and myself agree far more than we disagree on these ideas. One of the common themes is that a year of FSAE is not solely about building the car. Also without careful planning and management this is exactly what a team will do for 10 months. Building these cars is very time consuming, and without proper planning it can chew up all of your time.

As a brief summary we all hold that design and testing should take as much or more time than the manufacturing process. If this means that you have to make design decisions that reduce how much you manufacture in order to have the necessary time for design and testing then you need to do it. Also a car is not finished when it is running. It is mostly finished once it has been reliability tested and tuned.

Claudes orginal percentage of the three activities was roughly:

40% design
15% manufacturing
35% testing

While I think these activities can each fill the year and be concurrent I generally agree with this breakdown as how much resources (in time) should be allocated to each activity. If this was a team of 60 people and each was allocated to a single activity it would end up being:

24 on design
9 on manufacturing
21 in testing

I think this is about right.

Kev

[Max Trenkle]

One of the greatest things a team can do is gain a full understanding of the points system for FSAE. Z mentioned very many optimizations that are common, such as maximizing Lateral G's, deciding how much sidepods are worth, etc. In order to win, you have to optimize your points through your available skills.

A good team discussion about next year's car is useless. A good team discussion about next year's goal to achieve highest number of points possible is key to becoming a competitive team.

If you're a new team, or revived team like mine, it's easy to see through some historical analysis that just finishing all dynamics events will net you in around the top 1/3 of teams. Now getting top 1/3 seems a lot simpler than getting first place in skidpad.

Once you decide where your time will net the most points, then I think you can sit down and discuss stuff like longitudinal vs. lateral acceleration, aero vs. no aero, tire data vs. magic camber settings... http://fsae.com/groupee_common/emoticons/icon_wink.gif