+ Reply to Thread
Page 1 of 18 1 2 3 11 ... LastLast
Results 1 to 10 of 180

Thread: Vehicle Dynamics starting points and design process

  1. #1

    Aston University Design/Build Thread 2015

    Hi everyone,

    A little introduction about me and my team first, I am currently a third year Mechanical Engineering student at Aston University and a member of the Formula Student team. I certainly don't know everything but I'm willing to learn even if it might be a painful process at times, similarly, I am not a maths genius, I am competent with most maths but I'm sure there will be times where I will be asking things which seem trivial to most of you! The team structure consists of 16 members and is built around the idea of FSAE/FS being a year project for third years (of which there are 9), as such the 'main' areas (if you like) are split between the 9 third year students and have been defined as:

    1. Transmission and Engine
    2. Uprights and Brakes
    3. Suspension
    4. Pedal Box
    5. Bodywork
    6. Chassis
    7. Driver Environment
    8. Intake/Exhaust
    9. Suspension Arms

    Whilst you may not agree with the splits and grouping of the items that is unfortunately how it is and has been defined/fixed by the project supervisor.

    My area is 'Suspension' however, I'd prefer to call it general Vehicle Dynamics since I have control of all the inboard and outboard attachment points within reason.

    So what is the purpose of this thread?

    The reason I've created this thread is not to get everyone to answer my questions for me, that's not what I want to do and I don't learn anything from someone telling me "you want xx amount of camber compensation in roll because we did it last year and it worked pretty well". I don't wish to be spoon fed what to do but in some instances I've found myself hunting around in the dark with no proper direction and it's these times where I think "well I could really do with someone who has experience looking over what I've done and putting me back on the right track or finding my errors". What I'm trying to say is that I would like opinions on my train of thought, direction, method (analysis) and how it could be improved to ensure that I haven't overlooked something critical.

    As such, what follows is my basic interpretation of how I see the problem I've been given and the solution path I believe I should be going down.

    My thoughts

    In the simplest form my job is to design a system which maximises the tyre performance and hence car performance in order to extract the best possible laptime and in turn gain us the most points possible in the dynamic events. As the tyres are the only part of the car in contact with the ground then it seems clear that building a system around what the tyre 'likes' is the way it should be done, similarly, it is important that the car be responsive to the driver inputs (whilst have predictable handling characteristics) so that the driver has confidence in the car and is not frightened to drive it at its limit.

    Thus, it appears the system should be designed 'outside in' in the sense of design driven from the tyre, not by building a chassis and then bolting suspension to it. To me it appears that the next step is to study the tyre.

    What do I already know?

    I accept that I actually know very little and that this will be a long journey in furthering my understanding of these systems but as a starting point I have a copy of RCVD of which I have read both sections in regards to tyres and tyre data (I understand that these are only introductions and are not the full story) hence I know the definitions of the basic parameters: slip angle, tyre aligning torque, scrub radius, KPI, pneumatic and mechanical trail, tyre load sensitivity, cornering stiffness, slip ratio, camber, basic pressure effects, etc.

    I have some experience of the competition having helped out last years team at the FSUK competition and I've looked over various threads on here for the past few years and picked up a few things, likewise I have read the Pat's Corner articles on FSG.

    I know that bump and roll steer are to be avoided and can be done so by pointing the tie rods to the instant centers.

    I know that I know not nearly enough

    What is fixed?

    Wheels -13x6 w/ 0.5inch positive offset
    Tyres - 6.2/20-13 Avon A92 compound

    What do I have available to work with/from?

    RCVD
    OptimumG folder
    Avon tyre data - http://www.avonmotorsport.com/resource-centre/downloads
    Optimum K,T&D (Subject to purchase by supervisor)
    Excel
    Pen + paper
    Last years (broken) car

    What have I done so far?

    In order to determine a rough wheelbase and track widths I used the information available from Racecar Engineering along with the times for Skidpan, Acceleration and Autocross at FSUK last year to produce graphs and lines of best fit in excel, whilst the data had considerable variance due to the variety of powertrains used and design philosophy (wings or not, etc) the suggested outcome was that the minimum allowable wheelbase (1525mm) and a track width only sufficient for the car to not roll at its max lateral acceleration event was sufficient. I also noticed that most cars appear to run narrower rear tracks than fronts but plotting the track ratios against times for autocross suggested that this actually wasn't optimum and that a equal front and rear track was optimum. What I'd like to note about this last point is that I fully understand that the data for this comparison may be unreliable and that these initial conclusions may be proven incorrect at a later date.

    I have also looked into the Avon Tyre data and the testing method. It appears to me that the idea of running the tyre on the outside of a drum is not really a fair comparison to running a tyre on a flat surface as it would do actually, surely the length of the contact patch is affected by the fact the road surface isn't actually flat and this in turn will affect the aligning torque readings? I'm certain that there are more discrepancies between this testing method and a flat road including the actual surface differences. Whilst I will look further into this I'd like to move onto my next point, As it is my desire to design the car around the tyre it seems that creating a tyre model should be the first step before deciding on anything else.

    I understand that the idea is to normalise the loadings in order to 'fit' the data to a curve, the difficulty I am having is that whilst Avon provide co-efficients for a Pacejka '96 model I'd like to understand where these co-efficients actually come from, the actual 'fitting process' and how it works, if anyone could point me in the right direction (or work through it with me) on this I'd be very grateful. My intention is to add progress to this thread as I go along, as such there's not much here currently but if the thread lasts the lifespan of the project it should be decently comprehensive by the end!

    Thanks,

    Christian
    Last edited by ChristianChalliner; 11-02-2014 at 04:55 PM.

  2. #2
    ChristianChalliner,

    Designing the whole car depends on your understanding the competition and setting your targets. Setting your targets will help you to define ( Optimum Suspension Design ).
    What is the response you are expecting your car to do ?
    with the help of your tire data and your understanding of dynamics you can set initial suspension sketch and expect how this suspension will act when you simulate it. and again edit your design and repeat the process until you find your optimum.
    it's a good idea to analysis other teams data like static weight dist wheel base track width etc . but try to think about them what are the effects of these things on your target performance? So again it depends on your targets set them and try to reach them, finish your car early and test it for more than couple of months
    Last edited by Ahmad Rezq; 10-24-2014 at 08:10 AM.

  3. #3
    Thanks Ahmad I appreciate your input and it's given me something to think about (whilst jogging my memory on a few points),

    The main issue of last years car was a general lack of response to driver inputs, having looked over the car since we have found considerable rear toe compliance due to: slop in rose joints; uprights of poor stiffness; unnecessarily small steer arms (<50mm); undersized bearings; poor chassis load path for the tie rod and a massive scrub radius (100mm+) all quite visible from applying very little force to the wheel. We also found considerable bump steer at the front along with: very short steer arms (55mm); limited steering angle due to the linkage inverting because of poor layout; heavy steering and again a large scrub radius (60mm).

    A crude analysis of the forces acting about the steer axis at both ends of the cars shows that the tyre has considerably larger advantage over the tie rods than the tie rods do the wheel (Assuming that the mechanical trail is all the advantage the tyre has (I am fully aware that the pneumatic trail will be added to this and as such the tyre has even more advantage)). Thus I am not surprised that the car was unresponsive to driver inputs.

    My initial aims are to solve these problems before going further in optimising the camber gain curves to suit the tyre, decide on roll center height and it's affects on how the weight is transferred (through the arms and the springs), decide on a rocker design (to give linear or rising rate springing) and then investigate anti-dive and anti-squat properties.

    With regards to looking at others, as I see it a shorter wheelbase gives a lower moment of inertia so under acceleration there should be more weight transfer to the rear, more force on the tyre = more traction thus short wheelbase cars should perform well in acceleration. This also works both ways though and a shorter car will snap faster than a longer one making it susceptible to rapid handling changes (that the driver may struggle to deal with) however, for the slaloms the ability to change direction quickly must surely be a good thing. On a similar note a narrower track will mean your effective 'corner radii' for the slaloms will be larger so you will be able to pass through the gets utilising less steering angle and at a higher speed before the tyre lets go at the expense of the car being less stable (possibly a problem in braking zones?). I will throw out there that having a narrower track at the rear will move the balance towards oversteer and should also reduce the chance of your rear wheels clipping a gate but that is pure speculation on my part and I don't have the simulations to prove it.

    I perhaps have failed to make my initial issues clear, the one which is causing me the biggest issue is that I don't fully understand the tyre data and what it means, for example: the Avon curves show data for various camber angles but I cannot make the connection as to how that will allow me to decide on what a suitable FVSAL length is and if the FVSAL length should shorten in compression or not, the only thing I see is that excessive camber gain with travel will mean that the contact patch is compromised severely in bump reducing the cars tractive and braking abilities but may be beneficial in roll as the camber angles will remain closer to the perceived 'ideal'.

    All I am seeing is these lines on a page with the fairly obvious that as Fz increases Fx does and that the tyres peak at a given slip angle before the Fx value begins to drop off, likewise, the Fx values go up as camber is increased but does that mean I should endeavour to run with as much camber angle as possible without incurring extreme inside tyre temperatures and wear problems? I'm not sure it does. I feel like I'm really missing something vital to connect the dots here but I keep trying to read everything I can find on the subject but my understanding is not progressing

    Our plan is to have the car ready at least two months prior to the competition so we can carry out significant testing to validate the design, optimise the setup and get the drivers comfortable in the car. The problem is I cant afford to wait until then to find out that I've missed something vital.

  4. #4

    Smile

    Christian,

    1. At least you are stating your assets, your goals and even your doubts in a objective way. That is impressive.

    2. Bump steer and roll steer is not necessarily a bad thing unless your tracks are very bumpy which is not really the case in FSAE/ FS. In any case, even without any pure kinematics bump or roll steer, you will have bump and roll steer by compliance to a point that most FSAE/FS teams have no idea of, do not calculate, or measure or integrate in their simulations and their car setup.

    3. Yes it is about maximizing tire performance and car drive-ability. But its not only about maximizing each and all tires performances (that is grip) it is about also about balance (that means having the exact amount of yaw moment when you want / need it) and it is about control and stability. "THE" BEST Method to understand grip, balance, stability and control and tune your car using a decently relevant tire model understanding is to create your own Yaw Moment Vs Lateral Acceleration with Stability and Control simulation tool which is well described in RCVD and in our OptimumG seminars. It will help you to evaluate each car design and tuning parameter influence on those 4 criteria. That is THE method that makes our customers winning races and championship.

    4. In your tools you should add OptimumLap. It is free of charge and it will help you to decide the priority on performance of weight, grip, aero and power. Yes it is "only" a mass point (that is also why it is free of charge) but we used it on a LMP2 (on pole) on the Le Mans circuit and our prediction was wrong by just 0.6 " on a 3'36" lap time.

    In OptimumLap try different tire traction ellipse shapes and you will see which of lateral or longitudinal grip influences the most the lap time. When you will switch to your 4 wheel model that will be a good beginner to decide if you want to control more your camber variation in heave and pitch (braking and acceleration) or in roll (corner)

    5. I have serious doubt about the relevance and the usefulness of the Avon tire data for FS: look at the graphs: you can't even see the peak slip angle for Fy..... You may want to consider using TTC data (if you do do not forget the Calspan sticker on your car)

    6. Be careful about the track width size just to avoid tipping over under lateral acceleration; that doesn't take any transient effect, tire Mx, delta altitude between non suspended mass CG and instant center etc...

    7. Equal front and rear track COULD be optimum ( I like that word ) except that when car do not have enough yaw angle (that means cars not driven at the limit which is the case of 90 % of FSAE 90% of the time) the driver most often will hit the cones with the inside rear wheel

    8. I do not think the smallest wheelbase and smallest inertia in a FS/FSAE will make the car "too" snappy and un-drive-able; loot at competitive go-karts in tight corner; less inertia and still control-able.

    Here is my advice for car design

    1. Start by a simple mass point simulation
    2. Then a steady state 2 wheel model
    3. Then a steady state 4 wheel model with linear kinematics; roll and pitch centers do not move, spring and ARB stiffness and their motion ratio are constant, ratio camber variation per degree of roll or mm of heave constant
    4. Non linear kinematics 4 wheel model, no compliance yet, still steady state but now you input your tire model to use the Yaw Moment Vs lateral acceleration simulation

    At that time and only at that time you go to CAD. Not before.


    There will be another OptimumG seminar in the UK around the Autosport show early January (dates to be decided soon) you might want considering it
    Last edited by Claude Rouelle; 10-24-2014 at 08:25 PM.
    Claude Rouelle
    OptimumG president
    Vehicle Dynamics & Race Car Engineering
    Training / Consulting / Simulation Software
    FS & FSAE design judge USA / Canada / UK / Germany / Spain / Italy / China / Brazil / Australia
    [url]www.optimumg.com[/u

  5. #5
    Senior Member
    Join Date
    Mar 2005
    Location
    Australia
    Posts
    1,690
    Our plan is to have the car ready at least two months prior to the competition so we can carry out significant testing to validate the design, optimise the setup and get the drivers comfortable in the car.
    Christian,

    I can see that although you have the right intentions, you are already HEADING WAY OFF COURSE. Sorry to be blunt, but because you are clearly putting a lot of thought into these posts I do want to help you.

    First little tip, if you "plan to have the car ready two months prior to comp", then you more likely will still be bolting it together two hours before Scrutineering closes.

    New plan - SIMPLIFY EVERYTHING, and get the car ready much, much earlier.
    ~~~o0o~~~

    But biggest-picture stuff now.

    Is your Team's Overall Goal to:
    1. WIN OUTRIGHT!!! (or at least within the next few years)?
    2. Learn to play video games?
    3. Tread water, then enjoy the trip to the circus at the end of the year?

    If 2 or 3, then I cannot help you.

    If 1, then please read on. (BTW, well done Ahmad for also recognising the importance of setting clear targets.)

    VERY IMPORTANT. This competition is there for the taking. By anyone. Your Team can win it!
    * It DOES NOT require huge resources. Simple and cheap cars win regularly.
    * It does not require huge experience. The Rules limit all of you to "students".

    What IS required, and ALL THAT IS REQUIRED, is the WILL TO WIN.
    ~~~o0o~~~

    Still big-picture stuff, but a bit more focussed.

    By now I hope you have read Geoff's (Big Bird's) "Reasoning..." thread, stickied at top of thread list. Print it out and get all the Team members to read it. Several times, so as not to miss anything.

    I also suggest the "Any way to objectively choose engine?" thread, and for your own area, the "Suspension Design" thread. (And, if you have time, any of the other threads that I have successfully hijacked. )

    The important point here is that at this stage you are FOCUSSING TOO MUCH ON LITTLE DETAILS, yet you have given us NO IDEA OF THE OVERALL CAR CONCEPT!

    * Will your car be a low-power, non-aero, "super-light and agile", Tokyo Denki style car? (See also the quite neat Chinese Xiamen (splng?) car at a recent comp.)

    * Or will it be a mega-horsepower, super-wide-tyred, Stuttgart-Panzer-tank style car?

    * Or will it be a "brown go-kart with aero-undertray" Z-car?

    It is important that your whole Team gets answers to these big questions early on, and STICKS WITH THEM. Your current Team structure probably works against this. (Frankly, I think it is a recipe for chaos.) However, you seem to be responsible for many of these big issues, so I hope you can get the rest of the Team to, err..., "sing from the same song sheet".
    ~~~o0o~~~

    Focussing down on one particular medium-level issue.

    As an example of what I was shouting about above, I strongly suggest that you DO NOT WASTE TIME on sophisticated tyre models. You already seem to have a pretty good grasp of the most important tyre issues.

    Importantly, you have already found out that COMPLIANCE ISSUES, stemming from sloppy detail design, make any "tyre curves" that you come up with theoretically, utterly worthless in practice.

    To repeat my mantra, at this stage your Team's goal is to build a four-wheeled-chair that can propel itself and the driver at an average speed of 50+ kph, over a distance of 30 km, WITHOUT FALLING APART. My good friends the toothless-hillbillies (lovely people) can do this easily with four wheelbarrow wheels and tyres, and absolutely no knowledge of "tyre curves", whatsoever.

    To restress this yet again.
    "... my job is to design a system which maximises the tyre performance and hence car performance..." <- WRONG!!!!!!!!!

    Your job, and the rest of the Team's, is simply TO WIN!

    A simple, reliable, "brown go-kart with a GOOD aero-undertray" could do this using only 80% of the maximum available tyre performance. Maybe with only 70 %, or less...

    Anyway, tyres are consumables. They are easily changed and easily adjusted. The "static settings" of tyre brand, size, compound, rim-width, inflation-pressure, toe, and camber angles will get you 98% of the way to "maximum performance". Note that none of these depend on the suspension or any other part of the car (except, of course, for those compliance issues!).

    On your billiard-table smooth FS tracks, things like suspension kinematics and springs & dampers make very little difference to the BIG-picture.
    ~~~o0o~~~

    Probably too much for you to think about above, but a few more quick comments on your posts so far.

    * Sell your 13" wheels and tyres. In case you already have a sponsorship agreement in place, Avon make good 10" tyres...

    * DO NOT aim for the minimum "theoretical" track width. Your Team's inexperience may cost you dearly (ie. rollover!). Add at least ~10cm.

    * DO NOT use "push/pullrods&rockers". These are a complete wank, and will just slow you down. Plain old "direct acting" SDs are enough.

    * "I will throw out there that having a narrower track at the rear will move the balance towards oversteer..." NOPE! Think about it, but more importantly, DO NOT confuse yourself with such subtle details. KISS, and keep F&R tracks about the same.

    * "I cannot ... decide on what a suitable FVSA length is ..." There is quite a bit on this on the Suspension Design thread (admittedly, it is only the rantings of some stupid old farts...). Also a very simple suspension with good FVSALs is shown at the top of page 25 (I think on this new forum).

    * And many, many other details, which can be covered later.

    Anyway, please keep "blogging the build". At the very least, your next year's Team should benefit from it...

    And, like Jonny from Tasmania's thread, I really enjoy watching these things develop. Thanks!

    Z

    (PS. Just saw that Claude posted while I was writing this. No time now, but may comment more later...)
    Last edited by Z; 10-24-2014 at 10:02 PM.

  6. #6
    Thank you both for your thoughts, both of you have made me think deeper about certain issues and refocused me on the big picture, I'll answer both in order so firstly:

    ------------------------------------------

    Claude, Thank you for the compliment

    With regards to bump and roll steer not necessarily being problems I can see the value in having designed and controlled suspension driven steer characteristics however, I don't believe I am at the stage where I could currently implement a system that would improve the cars performance so for now I have decided I will focus on minimising the situation until my understanding of the system as a whole is better. Similarly, I was pondering the thought that all systems will have compliance to some extent so maybe analysing and 'tuning' the compliance may be more beneficial to car performance than just saying "well if I make it as stiff as possible it will be negligible" because in reality I'm pretty sure that the compliance can't/wont be negligible.

    Your next point I think is where my understanding is most lacking, through past experience I have found that cars with better balance are both more fun to drive and turn better laptimes than ones with super sticky tyres and a poor balance so I can see the difference balance makes but I have a hard time quantifying the result. I'm away from my books currently but on Monday I will go through the OptimumG folder and start putting together a Yaw Moment Vs Lateral Acceleration calculator, I'll come back with the calculations and post them here once I've gotten somewhere for you all to see/comment on/etc.

    I have looked into OptimumLap before and I did do some very basic experiments before the start of the year but they were mostly limited to mass vs laptime simulations, what I did find is that before I inputted some dummy values for load sensitivity the simulation was suggesting heavier cars were faster It did not occur to me to alter the effective friction ellipses in order to simulate setups favouring roll moments or bump and pitch moments but that does make sense to me in my head so I will revisit the program and perform those tests. I have a feeling that I will find lateral grip to be much more important though due to the nature of FSAE tracks.

    Your comments on the Avon tyre data has somewhat validated my thoughts that I was potentially wasting my time trying to make something from it. I was pushing for the faculty advisor to purchase the TTC data over summer but I did not get very far, I will cover the reasons why at the end of this post in the 'big picture' summary I will do.

    Regarding track widths, you have raised good points to suggest I was somewhat too keen to come to an early conclusion so I think I'll take a step back and look into it again once I've created a simple calculation model, it makes sense to me that a transient loading would (and indeed does it seems in the case of SUV's) be a greater factor in rolling the vehicle than just a steady state lateral acceleration scenario. Similarly, the chance of the car being driven at its maximum with students rather than professional drivers does seem very unlikely.

    Indeed go-karts are still controllable despite their short wheelbases, I think I maybe overstated un-drive-able, maybe reduced stability would have been a more fitting statement, either way, I think I should be tying in steering ratio and geometry considerations to match the wheelbase since a 'faster' ratio will be needed to catch the slides as the car will rotate faster and also a longer wheelbase = more steering lock necessary to turn a given radius.

    I will take your advice and update this thread as I go through the stages you have listed (I have a feeling I'll be getting stuck at stage 4! (probably earlier than that actually)).

    I did check the seminar dates, the closest one to me (Oxford) is listed as November this year, I will look into it.

    Once again, thank you for taking the time to construct such a detailed response.

    ------------------------------------------

    Z, firstly, thanks for taking the time to help out (not for the first time either I might add!)

    There is no harm in being blunt, I much prefer it to dancing around the point and it moves everything along at a faster pace.

    You are correct in the time management issue, this is exactly what happened last year, the plan was to have a car finished a month in advance, they finished the car at the competition! We have put in place steps to ensure that wont be repeated this year.

    Likewise, I agree with the simple car concept but the point where I disagree is on the aims. We all want to win, always, over the past 2 years I've had 4 projects which have been effectively mini competitions so of course I would like to win outright but in those competitions the field was equalized, that is not the case in FSAE for various reasons which are not the fault of anyone but it is important to remember that is indeed the case. I will elaborate further on this at the end.

    I have read both the Reasoning thread and the Suspension Design thread, I haven't checked the Engine thread but I doubt it will make much difference to my situation since the engine choice is fixed by the supervisor.

    I don't disagree that the team structure works against this, I admit and know that it does but again, it's important to know the situations which have caused this and why they are non-negotiable and cannot be changed.

    I will definitely keep blogging the build providing I feel that there is something to be gained from doing it, if this thread turns into just another argument (like most on here do) then I just wont bother and everyone loses so as long as it stays civil and there is interest I will update it

    Thank you for taking the time to reply and throw your thoughts into the ring, you like Claude have made me go back to considering 'the big picture'.

    ------------------------------------------

    As has been pointed out I have so far failed to lay out the main goals I and the team have so hopefully this next section will resolve that matter.

    The big picture

    Last year we placed 77th at FSUK, we missed acceleration and skidpan because our diff carrier repeatedly ripped out of the car whilst trying to pass the brake test, we failed to complete either the sprint or the endurance within the time limits and handed paperwork in late. All of this because people did not fully understand the competition or the initial goals fully. The goal was to build a car which would place in every event, the team failed in that objective. Why? Organisation and poor car performance.

    Deadlines were missed because people failed to understand the implications of missing them and the cars performance was poor because the analysis and understanding was poor, every time the analysis was incomplete the standard response was to beef up the components to a stage where they thought they would not fail. As such the car used M8 bolts, rod ends and spherical bearings everywhere, the tubing sizes were too conservative and the car ended up weighing 290kg (once all the diff reinforcement had been carried out) WITHOUT DRIVER! Yet the car still failed.

    Our main aims are: slash the car weight to at least 220kg; improve the cars responsiveness; meet all our deadlines; get testing as soon as we can; finish inside the top 20 and to make the best use of the equipment we have available whilst furthering our own understanding. Due to various reasons it is not possible for us to sell parts in order to purchase new ones, I wont go into these here but lets just say it is not up to me or anyone in the team. Likewise, for the purchase of new parts, they have to go through a long and complex process before receiving approval and the order paid.

    As such, our engine is fixed, it is the Honda CBR600RR 03 model, we are told the chassis must be a spaceframe, the wheels are 13 x 6 and our tyres will be the Avon's, our diff is the Drexler FSAE model and our brakes are 4 pot AP Racing Calipers with motorcycle discs. The engine is a definite weak link as it weighs roughly 60kg without considering the diff, driveshafts, chain,etc and we don't make good use of the power it has.

    These are the constraints we have to work within (there are more but non spring to mind right now) I believe we can still fundamentally build a good car around these parts provided that they are integrated properly into the design and we make use of their strong points whilst recognising the weaknesses.

    But in reality the main objective I and most of the team members have is to actually learn something! If we get to the competition, finish inside the top 20 and I feel like I know no more than I do now then I will still feel like we have failed, yes of course we want to win or at the very least place in a position we can be proud of but in my eyes it must come from having learnt what it takes to build a good car and not from gut feeling of 'what we think will work'.
    Last edited by ChristianChalliner; 10-25-2014 at 05:28 PM.

  7. #7
    600cc motorbike engine, tube frame, 13" wheels, Drexler are the starting ingredients for the Missouri S&T cars, winners of the autocross at Lincoln 2012 and 2013. Lots of fast cars use 'em. The weight of the engine makes it easier to get the weight distribution right. Don't second-guess it too much. If you're racing in July you need to be running at racing speeds in May which means a complete and running car in April.
    Charles Kaneb
    Magna International
    FSAE Lincoln Design Judge - Frame/Body/Link judging area. Not a professional vehicle dynamicist.

  8. #8
    Charles,

    Adding weight to get to the ideal weight distribution? Patch on patch? You can get a light car with the ideal weight distribution. Many teams have prove it before.
    Claude Rouelle
    OptimumG president
    Vehicle Dynamics & Race Car Engineering
    Training / Consulting / Simulation Software
    FS & FSAE design judge USA / Canada / UK / Germany / Spain / Italy / China / Brazil / Australia
    [url]www.optimumg.com[/u

  9. #9
    Senior Member
    Join Date
    Mar 2005
    Location
    Australia
    Posts
    1,690
    Christian,

    More BLUNTNESS follows...

    But in reality the main objective I and most of the team members have is to actually learn something!...
    ... yes of course we want to win ... but ... it must come from having learnt what it takes to build a good car and not from gut feeling of 'what we think will work'.
    Very sadly, it is the above attitude that will see you repeat last year's performance this year.

    I know that this comment doesn't seem to make sense to you, but it is true.
    ~o0o~

    Last year we placed 77th at FSUK, we missed acceleration and skidpan because our diff carrier repeatedly ripped out of the car whilst trying to pass the brake test, we failed to complete either the sprint or the endurance within the time limits and handed paperwork in late. All of this because people did not fully understand the competition or the initial goals fully. The goal was to build a car which would place in every event, the team failed in that objective. Why? Organisation and poor car performance.

    Deadlines were missed because people failed to understand the implications of missing them and the cars performance was poor because the analysis and understanding was poor, every time the analysis was incomplete the standard response was to beef up the components to a stage where they thought they would not fail. As such the car used M8 bolts, rod ends and spherical bearings everywhere, the tubing sizes were too conservative and the car ended up weighing 290kg (once all the diff reinforcement had been carried out) WITHOUT DRIVER! Yet the car still failed.
    I am sure that last year's students also "wanted to learn". But, clearly (!!!), NO ONE WAS TEACHING THEM!
    ~o0o~

    I had pulled out quite a few of your quotes that I was going to go over in detail (eg. all the theoretical details that you brought up). But I think it better to just sum this up as simply as possible.

    * LITTLE LESSON - The main lesson you should have learnt by now is that planning to finish "a month in advance" results in complete and utter failure. Your current plan to finish two months in advance will result in a mere utter failure.

    If you really want to learn anything about VD, etc., then you MUST have the car driving at least FOUR MONTHS before comp. There will still be a mountain of work to do after that.
    ~o0o~

    * BIG LESSON - Many long-time ex-FSers will tell you that their deepest understanding of VD, or engine, or whatever in FS, only came to them many years AFTER they finished their last competition. You should realise that this whole education process is not a linear one of,
    Learn-some-theory -> use-theory-to-design-a-car -> build-car -> race-it -> collect-trophies.

    Real education DOES NOT WORK LIKE THAT.

    This is probably contrary to everything that you have ever been told by your modern teachers, but in real-world, old-school education you;
    1. First learnt HOW to do it, by DOING it...,
    2. Then, much, much later, you discovered WHY you did it that way.

    You can only "understand" WHY, by first "knowing" HOW. And you can only know HOW, by actually DOING it.

    (As I said, contrary to everything you have ever been told... Criticisms welcome... )
    ~o0o~

    So, get a working car built ASAP.

    You might start this process by posting as many concept sketches as you can come up with here, and then consider the reasonableness of the criticisms that the sketches receive.

    The very first of the concept sketches should only show the locations of the major components. The most major component is the driver. So start with the "ergonomic concept sketches".

    Err..., you do have your "ergo-rig" built already, don't you???

    Yes, of course, you must have this built by now. So, perhaps start with photos of the driver in their preferred position in the ergo-rig, and with the engine and four wheels in their proposed positions.

    Z

  10. #10
    Senior Member
    Join Date
    Nov 2010
    Location
    NSW, Australia
    Posts
    352
    Christian,

    Excellent thread so far, thanks for that.

    Just wanted to re-enforce the 'build it early' sentiment. 2 months is NOT enough. Something like 4 months is more applicable and instead of saying "we're going to build it as early as possible" you set a date and remove all obstacles to hitting that date. This may sound impossible but if you approach each part with "is it actually required and will it cause us to miss our drive date?" then it is achievable. I'm talking here about items that people will typically want to play engineer with (turbos, push/pullrods, wings, fancy electronics, etc.).

    You have been given some items to build around. This is not a bad thing and plenty of teams have been successful with your ingredients (including UoW, though we've had many different diff types over the years). It should also help to speed up the process because you are wasting less time choosing things that your team might not fully understand. If you feel like you don't have a solid enough understanding of tyres and how to build around them, build something in the ballpark and include plenty of adjustment (different mounting holes, adjustable rod-ends, swappable A-arms/geometry, different springs, etc.). The season will be gone before you have a chance to fully develop an "optimum" package (and inevitably the rest of the team will have also "optimised" and when you bolt it all together you end up with <<<< sum of the parts).

    Sorry if I'm preaching to the converted, just figured I'd add my 2c
    Jay

    UoW FSAE '07-'09

+ Reply to Thread
Page 1 of 18 1 2 3 11 ... 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