+ Reply to Thread
Page 10 of 12 FirstFirst ... 8 9 10 11 12 LastLast
Results 91 to 100 of 116

Thread: Ideas to pursue for innovation points

  1. #91
    Originally posted by Z:
    <BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by james17:
    In 2001/2002 our team set about building a VERY innovative car. The vehicle was very close to being an aluminum and carbon fiber uni-body. It had 5 machined aluminum bulkheads with large diameter square wall aluminum tubing running between them with structural carbon outside that.
    It's funny how people who constantly remind you that "I've been around racing my entire life, real racing..." (James' earlier post), always seem to think that innovation = complication. For example James' idea of an innovative FSAE engine is a "CF and titanium 20,000 rpm V-12". On the other hand, I (who claim no experience) have been advocating a much simpler 600cc single-cylinder "1/8 of smallblock V8" because I think it would work (ie. package very well), could be done cheaply, and would be very reliable.

    Similarly, James thinks that an innovative chassis must have "5 machined aluminium bulkheads" and lots of carbonfibre. Again, I have most often suggested a sheet-steel tub that could be cut, folded up, and welded together in half a day (by one person!), and would cost ~$100 in materials. And as well as being lightweight, it would be very stiff, strong and reliable.

    James, read Big Bird's post on the previous page.

    Z </div></BLOCKQUOTE>

    Maybe the reason people who have been around racing there entire life all have the same argument is because they are all right, maybe?

    , I give up, atleast when freshman tell me they can build a car in half a day they can see me roll my eyes...

  2. #92
    You can come up with all sorts of crafy ideas on paper. Absolutely brilliant, very innovative stuff. But when you start going hog wild with it, you wind up not getting your car done on time. Or something fails. And forget any sort of time estimate if you haven't been involved in FSAE or any motorsport. If you haven't done something hands-on, take whatever estimate you have and multiply it by 5 or 7, and that is how long it will take.

    There are no amenities on a FSAE car. Every subsystem is a critical path item - without it, the car doesn't function. Reliability is key, and simplicity usually yields reliability.

    Combine this with the fact that you have a team of undergraduates, for many of which this is their first design project. You don't have a bunch of guys working for Ford SVT.

    I would leave innovation to independent studies, of which there is on average 1 a year relating to FSAE at CU. Or leave really crafty stuff to the top 1-2 design engineers on the team.

    I'm not knocking innovation. Its good to blank-page redesign major things here and there if there will be a worthwhile performance increase. But don't just do wild things for the hell of it. That time and money can be spent better elsewhere. Your goal is to build a fast, reliable, winning car. Determine what is restraining you from doing better, and address those issues one at a time.
    Colorado FSAE | '05 - '07
    Goodyear Tire & Rubber | '07 - '11
    NASCAR Engineer | '11 - ??

  3. #93
    Senior Member
    Join Date
    Mar 2005
    Location
    Australia
    Posts
    1,690
    Originally posted by Big Bird:
    Firstly, speaking to Dr Sano from Tokyo Denki University, ...the designer of Honda's first F1 grand prix winning car... he thinks FSAE is a wonderful opportunity for us, but we tend to make our designs far more complicated than they need to be.

    Secondly,... Ross Brawn (Ferrari F1)... thinks that... we tend to make our designs far more complicated than they need to be...
    Anybody care to comment on what they think these excessive "complications" are, and how they might be deleted from the standard car?

    Would this require any "innovative" thinking?

    Z

    "An engineer is someone who can do with one dollar, what any fool can do with ten."

  4. #94
    i suspect that that sano and brawn, come from a type of racing where everything is custom made, could see a 600cc single, air-cooled, possibly forced induction, with a NASCAR style restrictor plate as the simplest way to do fsae. hence, adapting a water-cooled, naturally aspirated, multi-cylinder sportbike engine is not "ideal"...however, i think given the development timeline, budget, and personnel avaliable to most teams, and the results some people are able to achieve, the sportbike engine is a much more feasible option.

    i'd love to see our team, for example, go to a 2-year design cycle, which would give us 4 months more testing/tuning time while also extending the design timeline from 9 months to one year. in this way, we'd probably be open to a bit more innovation. but it means either not going to competition one year (unacceptable, as we'd lose most of our multi-year sponsor deals), or designing 2 cars in one year (when one is hard enough!)
    Mike Miles
    Carnegie Mellon SAE/Carnegie Mellon Racing -- Formula SAE 2003, 2004, 2005, 2006

  5. #95
    hey guys
    i've been following this discussion/slug fest a feel it's about time to chip in.
    Z, in response to your question about Dr. Sano's view on complication in FSAE, i would first point out that i completely agree with what he said. I think we try to make our cars too complicated. Case in point: the embarassingly low number of cars that finish endurance. Its only a 20 minute race. I admit that our team hasn't finished endurance since 2002 because of parts failures. Last year we were pulled out because or problems on our turboed single cylinder. And the thing i would like to point out is that directly after we were pulled out, Michigan Dearborn with that briggs twin that everyone seemed to like to dismiss away as a cometitor drove right by us and finished about 40 places overall better than us. So what was the better product, our bored out, turbocharged, intercooled, single cylinder sitting between a carbon fibre monocoque, carbon fibre rear differential box, and all carbon suspension tubes or a briggs twin with a cvt and a steel frame that did twice as well as we did in the end? There's your elimination of complication to get a better result right there.
    that, of course is a pretty wild simplification example though. there are other, less extreme examples, such as sticking with a manually actuated shifter instead of going pneumatic or electric and for some reason losing downshifts halfway through enduro.
    every day i'm in our shop and have to listen to our baja team (who we share our shop with) give me grief about spending too much time on the computer and not building. it doesn't help that they won the the baja west 4 hour endurance race in arizona either.
    i, and the rest of the fsae'ers on this forum who live and breathe formula want to build the best car we can. we all lose sleep thinking about placement for the steering rack, or draw new uprights on our notes in class. we all fall into the trap of over designing the car on paper, but then having it roll out in the few weeks (if even that early) before comp. sometimes we can't draw the line. using my shifter example from above, there has been interest every year to design paddle activated pneumatic shifting, but how much do we really need it? i don't know what the average system can achieve in terms of shift times (i'm not including auburn in this generalization because their shifting system just blows my mind, just listen to the 1st to 2nd shift on their video on their website! did it shift? i dont know! its that fast) but over the summer the TPS data logs were showing our good drivers were hitting .1 seconds from full throttle to full throttle between gears. how much time do you want to invest on a system that is arguably more "complicated" than an indy-style push pull lever with a mountain bike brake clutch compared to the gains from it?

    the other thing i would like to comment on is the "ideal" engine discussion. i think a project such as machining your own engine is a little out of scope in this competition. the only team to come close to that is the washington v-8. i agree that was one beautiful car, with amazing execution. i'm not sure how many years it took to implement that, but they also took the heads (and pistons...? not sure on that one) from two kawi 250s and made their own bottom end. i'll be very impressed if i see another team raise the bar further in terms of scope of a project like that. so for in terms of a fully machined 600cc single, probably won't see one of those anytime soon. and don't even tell me how much simpler singles are to run than f4s. saying that around our team is opening pandora's box. it would be a multi year project just to design the starter motor to get that huge thing turned over. our 470 starter had 5 bolts shear off due to kickback during excessive starter motor use (running starting and idle enrichments) this past summer. also they're alot louder and much less fuel efficient. for a more technical and indepth discussion of the merits of singles i can refer you to our engine guy who practically jumped for joy when the decision was made to go back to an f4 this year after 3 years of singles.
    so intead of falling into the trap of saying you're wrong and not offering my opinion, i'll give my ideal formula engine. we need to pick motors which make the same amount of power before and after we restrict and put custom ECUs on. the cbr's are great, they're pretty much bullet proof, and you can make some wicked power with them, but the fact that their built for give or take 110 hp, and we put on average 75-80 horse (until we get choked) through them shows that the motors are more than we need. their overbuilt (for formula) and too heavy for our application. on the other hand we ran a suzuki ltz-400 bored out to 470ccs and turbocharged last year. the motor makes 32 hp stock, and we ended up pulling 65 hp out of her with the turbo. once this happend, all the drivetrain components started failing. we sheared a total of 5 teeth from two sets of transmissions from our 5th gears while tuning, we smoked one race clutch, and one stock clutch with the race springs shimmed with washers. we also sheared a weld on our right driveshaft. so that was definately way too much power to put through that motor, and it showed. i think RMIT has the right idea, they use a motor that makes 59 hp stock, and ended up with a bit more power after tuning (perhaps someone from RMIT can give the specific numbers), that coupled with a centrifugal clutch, manually actuation paddle shifters and possibly the best driver in all of fsae allowed them to do very well.
    they chose their engine well, yet i still believe there are better engines out there. i still think singles are better left in go karts, and f4s left in race bikes, so what's in between. there aren't many teams running twins out there. i would favour an inline instead of V configuration for drivetrain packaging reasons. i could see a twin that makes somewhere in the vicinity of 80 hp being the perfect engine for us. so instead of spending three months of 2 am nights on [insert preferered solid modelling software here] and mastercam to try and design your own engine which by some magical means will be better than one designed by a company with teams of people being paid to design just spend some serious time investigating the "innovative" choices out there.
    John Valerio
    Queen's FSAE
    http://engsoc.queensu.ca/formulacar/

  6. #96
    Senior Member
    Join Date
    Sep 2002
    Location
    Perth, Western Australia
    Posts
    717
    John,

    Just a small point. Melbourne University made an engine of there own for a few years. It ran in comp. Finished enduros. Was a single cylinder I think.

    Probably not brought up as much as the Western Washington v8 because of either the "cool" factor or that it wasn't done in Australia.

    The thing did not use very much fuel at all and hopefully someone from Melbourne might post some details of it.

    Cheers,

    Kev

  7. #97
    Originally posted by CMURacing - Prometheus:

    i'd love to see our team, for example, go to a 2-year design cycle, which would give us 4 months more testing/tuning time while also extending the design timeline from 9 months to one year. in this way, we'd probably be open to a bit more innovation. but it means either not going to competition one year (unacceptable, as we'd lose most of our multi-year sponsor deals), or designing 2 cars in one year (when one is hard enough!)
    When I speak to student groups, I usually recommend that they have a look at how things are done in industry. All the good ideas aren't necessarily dropped into the product at once. There is no "best" mousetrap.

    One can establish an R&D function, where ideas are tried and tested first, before being integrated into the entry in the future. The ideas don't have to work right away, but what is learned (even if the idea fails) is worth a lot. Once a concept is proven and developed it can be integrated into the car for a given year.

    Now, that takes an organizational structure and culture in which that type of activity can work. A lot of student grouops have trouble getting a single entry done, but so many are "mature" nowadays (as opposed to when I was a student and these competitions were new) that enough could do so. Of course the problem with students is that everyone wants their great idea incorporated, and they want it done NOW - not after they've graduated. Again, it takes a culture that encourages an R&D approach to innovation.

  8. #98
    Senior Member
    Join Date
    Sep 2002
    Location
    Perth, Western Australia
    Posts
    717
    One of the things that we did in UWA was move to an 18month cycle a few years ago. This meant that in the second half of the year the preliminary design and studies would be started for the next car.

    This shifted to a 24month cycle for the International campaigns.

    The end result of the 18month cycle is that there is always design occuring. The first 6 months do not tend to be mind-blowing in terms of achievements. However it definitely accelerates the team coming together and minimises the work lost at the start of the year.

    Simple fact is that innovative design takes time to develop.

    Cheers,

    Kev

  9. #99
    kevin,
    sorry for the mistake on my part, i had not been aware of melbourne's achievements. after checking out the history section of their webpage, it seems that their engine did take 2 years to develop and maufacture, was a parallel twin cylinder, and had an overall displacement less than 600ccs. and then in 2004 they turbocharged it, all with good success. if any of this is wrong, i welcome corrections as the website is a bit less than 100% specific. i think that what they did was an amazing feat, and whoever conceived that plan definately knew the score. its a shame they went to the RR for their '05 car but they must have had their reasons and that is a sweet motor regardless.
    it also seems they (or the engine team/guy, at least) agreed with my theory that inline twins are the "ideal" (ideal in that if you were going to build a motor specific for the competition, what would you build) way to go for formula.
    it also makes sense they got better fuel economy with a twin, as the singles are really thirsty engines. we're cutting our fuel tank size down this year due to the shift to the f4.
    John Valerio
    Queen's FSAE
    http://engsoc.queensu.ca/formulacar/

  10. #100
    Basically my idea is to have a planetary gear set, drive the sun gear with an input shaft from the IC engine, and connect the planet-cage to the output (ie. it drives the diff). The ring gear (which can rotate in the housing) is connected to a variable disp. swash-plate pump/motor, which in turn is hydraulically connected (ie. 2 short pipes) to a fixed disp. roller-vane pump/motor on the input shaft (sun gear). This could all fit in the space of a typical bike engine clutch/gearbox.

    With the swash-plate set for zero displacement there is no oil flow, the ring gear is locked solid, and the input just drives through to the output like a normal planetary gearset. This would happen in "top gear" (ie. top speed), and since no oil flow, minimal viscous frictional losses. With the swash plate set for more displacement, it sends oil to the vane pump on the inlet shaft giving it more torque. That, combined with the ring gear "slipping backwards", gives a higher torque output at lower speed - ie. low gear. There is more oil flow now, so more viscous losses, but the car typically has more than enough power at low speeds (ie. traction limited, not power limited)
    I like the idea. Since we never talked about
    the planetary gears in circuits class, I had to look them up. They're awesome little devices....although I've pulled them in and out of an automatic tranny, I was in a hurry to get the car working, so I didn't study them like I normally would... rather fascinating really and quite simple.

    Your concept is similar to the drives that John Deer came up with...

    Anyway thanks for the reply.. I've got to think about it some more before I comment on it further.

    As a side note, I'm starting to see where Z is coming from.....and I must apologize insulting him when I first got agitated with him for stirring the pot...I had the same feeling many others have voiced. But in the end, I've found that he's quite open for discusssion, teaching, and learning. Sometimes it's good to set the ego aside and think outside the box.
    But I admit, it's hard to not take shit personally sometimes... At least Z's critisism and comments are followed by alternatives, not just bitching. I hate it when somebody says no, without offering suggestions or alternatives.
    UNM FSAE 2003 to 2005

+ Reply to Thread
Page 10 of 12 FirstFirst ... 8 9 10 11 12 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts