UPDATE: Added rule proposal thing so we can work bit better, chime in your thoughts Well, I woke up at the kinda late for a morning and got this silly idea, why not make Formula SAE inspired car, then I kinda was thinking that BeamNG is actually really incredible platform for virtual FSAE competition. Here is what I have so far: Total time spend maybe around 6 hours, that is efficient working hours. I kinda would like to see more of Formula SAE inspired cars, every FSAE vehicle is unique sample of engineering, so as universities compete in car making and driving, why modders could not do the same? Teams currently enlisted (I don't have your team names yet, you need to send them via pm or posting to thread): Modders that are participating: atv_123, maybe, if finds time ben_arkless CrimsonCrocodile Missonsystem Miura Mr. Twiglesworth, might? SebastianJDM General advisor: Blijo (also looking for a team) 3D Modelers looking for a team: Nicelittle Sitthy? Skinmakers looking for a team: Welshtoast? Keeping things 1 step away from chaos (that is general management, planning, running things, receiving complanits, etc.): Fufsgfen For judging I'm thinking system where everyone can cast a scoring vote, but votes directly can't work because of varying skill level etc. Very much I would like to have judge board from highly experienced modders, so if you are interested from such task, be in contact! Formula Student germany has really nice rule books with illustrations, pages 26 and 53 has aerodynamic device illustrations that I think works well for us: https://www.formulastudent.de/fsg/rules/ Proposed rules (to be changed, schedule etc to be determined, Test run of the competition runs to end of 2018) : -Pigeon 600cc engine, max RPM limit changed to 10500, engine will be provided, permitted changes are positions of nodes (to place engine) naming of engine and slottype (to avoid conflicts) -6 spd-sequential gearbox (ratios will be provided with the gearbox) -CVT gearbox is possible, modder needs to provide link to information about gear ratios used in similar vehicle -Differential type and ratios are free (ratios between 2.36-5.36 are considered realistic) -Number of wheels free -Number of driven wheels free -1550mm minimum wheelbase (in case of more than 2 axles wheelbase will be distance between two axles closest to center point of vehicle) -350kg minimum weight, with an exception if one can make 200kg vehicle that he/she can prove to be realistic there is 100 bonus points -Minimum weight has to include 75kg driver weight set to node that is 30cm above the seat cushion/vehicle floor. -Safety rollbar has to be 116cm above seat cushion / vehicle floor and if vehicle is rolled over, driver should have enough safety marginal to ground, so design enough wide rollbar. -Aero elements should use triangle values from SBR4 Hillclimb (wings, splitters). -No indestructible parts allowed, deformation will be scored, unnaturally stiff structures will score lower etc. Winner is who gets highest score if two gets same score then for example skidpad and endurance result will determine winner Scoring is done by several categories (proposal, scoring to be determined) Presentation(Repo page, model, textures) Acceleration(0-100kmh) Novelty of the concept (Unique features, weird stuff etc) Skidpad Autocross Fuel Economy Endurance race (Automation test track, 2 laps?) Autocross type test course UPDATED 16.11.2018: https://www.beamng.com/threads/form...petition-for-modders.59114/page-3#post-959423 (it is very simple and crude, any track maker is welcomed to make it stand alone track, beautified, cone positions in it are pretty accurate though). This Blog might get you inspired: https://www.formulastudent.de/pr/news/details/article/pats-corner/ old stuff: Spoiler So this is kinda my thought, to get other modders attempt to make one of their own, well maybe some day, I need to get mine running and figure out how to loose lot of weight, about twice the weight that I want it to be, but jbeam is not tuned too much yet. I'm quite certain that I will not win, but maybe we see few better ones before year ends. Rules: https://www.fsaeonline.com/page.aspx?pageid=e179e647-cb8c-4ab0-860c-ec69aae080a3
Usually when I build something, something unexpected happens, like this time, all I was trying to do was to use nodegrabber to suspend engine so I can observe node names better, instead all tires bursted to flames... It might be that after succesfully placing 3 suspend to air points, 4th one I managed to couple two engine nodes toghether, ones that form back of the cylinder head, engine flies off and tires burst to flames, now what would you except really, it is normal for me. I 'found' this kind of engine, it was not attached anywhere (much) so decided to grab it to my FSAE inspired car: Weight balance is not that bad even with added engine, there just is too much of it, but I worry it later, like everything it will sort itself out eventually Wheels alone weight 60kg, so mine will probably be bit on heavy side no matter what I do to it. Next to eat and then to difficult stuff, to get suspension to work, steering to operate and some brakes, kinda would like to have engine running already, but that has to wait until things stops and turns I have only 9 hours or so today to make it drive, might be bit tight, but let's see.
Hmm, interesting Maybe it is an idea to make this a challenge with our own set of rules? Proposal for rules for the BeamNG challenge: Height(driver's helmet) Lenght and width Amount of aero and values for the aero(or none at all?) Use universal wheels Power to weight ratio Minimum weight Suspension travel/ground clearance Amount of driven wheels Skins? And maybe with teams of modders+drivers(3 people per team for instance)
There could also be electric and ICE classes. In FSAE under 200kg are using 1 cylinder engine and above 200kg they can use bigger engines for example, competitions is different then at heavier class, but even in heavy class they tend to be under 250kg, mine is just way too heavy, not sure what would work best for our modding, but kinda would be interesting to try to meet weights of those real things, but maybe it makes things too hard? I'm thinking that aero is free, but it should follow some sort of limitation, like aero element nodes should add certain weight per added downforce and I think it should be possible to define maximum downforce per area of wing for example. Also I think that it would be good to run short time so that there are bit of challenges that need compromises, IRL you can't have everything, I think same would go for this, it encourages to make more creative choices I think. We could try to run with single engine, like Pigeon 600cc which kinda would fit to this kind of vehicle, no modifications to engine would be allowed, so it is your skills in designing chassis and suspension etc. that makes thing to handle. Very core I would think should be handling, skidpad tests, slalom tests, etc. Oh no, I need to reduce track width a bit, I'm aiming that thing should fit to bed of D15, not sure if that would be sensible for the size? Might need to check actual measurements of some of FSAE vehicles, but I'm not that excellent They seem to run wings, there might need to be wing and wingless class or season, there are no exactly hundreds of competitors in our modding 'race'. Another bit I was thinking was that has to be scratch made, wheels, engine, seat, steering wheel can be sourced from vanilla parts, but chassis, suspension etc. should be scratch made and especially jbeam. Perhaps we need to do pre-run to form a rules first? Again it would handy to have a group, but Blijo you have experience from league stuff, do you want to handle rule thing? I know you would be lot better at that, but how is your available time? For me, I just thought this kind could be something fun and interesting which might be useful for learning more, would love to see what different kind of solutions each modder / team finds.
I think that we should use 1 engine(pigeon sounds good) Then the weight is 200kg minimum with 0L fuel Then to simplify stuff I would say that we use the triangle values from the splitters of the SBR4 Hillclimb stuff Track width: To be determined Lenght of the car: Same, not too long I reckon The jbeam, I don't really care what people use but chassis should be scratch made, maybe set a minimum hub weight? And time-wise, I am waiting for someone for the Fintray and other stuff is very low priority , the only thing is university and scouts Also tracks, maybe someone can make nice autocross tracks? EDIT: It should also have a driver , that meatsack in the middle of the car makes it more realistic
I am good with skins but cr*p with anything else which makes me basically useless. Also. what about driver aids? [ABS, TCS etc.]
For tracks, any enough large flat parking lot / stretch of street in existing track could be used, it is quite easy to spawn cones or install them permanently which would improve performance, but they eat cars when installed permanently. Kinda like config for existing maps would work just fine, but of course dedicated autocross arena would be really sweet. Spawning cones is something I can do, not sure about whole map, spawning cones can be game crashing experience when I'm doing it Driver is of course quite crucial, not sure how Stiggy would hold up? TBH currently I don't know how to add Stiggy to my creation and of course would need permissions for such. Minimum weight sounds good to me, also that is great idea about using SBR4 Hillclimb splitter values! I wonder if official FSAE rules have something about dimensions, those might work, I did just eyeball this experiment of mine, it can be far off actually. ABS and TCS, I'm not sure if those are allowed in official competition, Electric car variants are allowed to use torque vectoring though, so there is that. Somehow I managed to slip few bugs into my creation, but I'm happy it did not explode. Kinda forgot to add brakes though. Needs many iterations, but it is drivable car done in a weekend, which I'm quite surprised I managed to get it to work: Well, bugs exists so that you learn something when fixing them
Hmm, I started something tonight. I found out that the wheelbase of those cars should be more than 1550mm and started on a chassis. I currently have a rolling chassis with pushrod suspension in the front and in the rear, just with the wrong wheels and no drivetrain. And it's too heavy. I might make a set of wheels for this competition and make a set of rules with @fufsgfen if more people want to join.
i would love to attempt to join. not sure if i'd have the time to complete it, but this would be pretty good practice for me. a spec rim/tire setup (for different weight classes or whatever) would be pretty smart too. http://www.imeche.org/news/news-article/hoosier-release-new-formula-fsae-tyre-size also, i recommend not requiring the stiggy, since it sounds more intimidating and difficult, which would sort of scare people away.
Here is pdf on this page that has few illustrations about aero bits and driver protection, also some stuff about driverless vehicles etc. Those might be interesting for future at least if not for now. As BeamNG research project is about autonomous vehicles, this all might be quite interesting area for some in bit further ahead in future: https://www.formulastudent.de/fsg/rules/ I'm not sure if fsae and student germany will have differences in rules, but I would think they follow international rules in Germany too. I did read about wheelbase yesterday too, checked mine and I have 1.9m wheelbase, D'oh! Anyway it is first ever car I have made from scratch, so I'm surprised it even did work, so redoing the thing I think Even more D'oh is that most of that extra is behind the seat, so fitting of engine in a way I hoped might not be possible, it can only be transverse setup I think. Single engine and single tires sounds really good as it would give all similar opportunities and challenges. Spotted a map also from here http://www.fsae.com/forums/showthread.php?8800-Course-Design-Feedback-Discussion-FSAE-Lincoln/page10 Some clips from Lincoln 2015 I spotted from YT:
in the official rules, it states that any engine/transmission combination can be used. i think we should use the 600cc pigeon motor with a 6-speed sequential gearbox (specified ratios, i can help with that if you'd like!) it also states that all four wheels must be able to lock up during the braking test (so no ABS), and no drive/brake by wire (so no ESC either).
I would think that as most use 600cc sport bike engines, research would be needed to see if those have different kind of ratios and use what those have as it is likely what real teams would need to use, unless they choose to manufacture something, but from such I have not seen information and as ratio reduction is done after box it is easier for teams to set ratios to best suitable range there, so I would think they would run stock sport bike transmissions. Also for a diff there are several options teams use, spool is quite common, but other kinds are also allowed, there are no restrictions on such really. EV are allowed bit more electronic control what I have seen in rules, but maybe those could be something for later as then we need to worry about realistic energy amount, voltage and power limits, appropriate battery weights etc. After all designs should be something replicating realism and using as close realistic values as possible, which would be one challenge as it is one area of skill to be tested. Similarly to real world competition where they have design segment that is evaluated and given a score. Scoring, from https://en.wikipedia.org/wiki/Formula_SAE Obviously not all of those apply to mod making, we have other aspects, which could be like unique naming / conflicting probability, repo page and car selector info + screenshot could be presentation, design could be jbeam and how realistic values are, uniqueness and novel solution could add to score etc. For example node weight and beam spring/damping can present different kind of materials, for steel there is certain range of values that can be accepted, which would be interesting aspect to be scored, but that brings then questions, who could then be qualified to define ranges of values. For example I think it is possible to work out some sort of table of values that would match these (from pdf that is in https://www.formulastudent.de/fsg/rules/), but I suck a bit and have not learned quite yet relation of node weight / beam spring etc. and this kind of project is wonderful in learning more of those: Code: The steel properties used for the calculations in the SES must be: Non-welded strength for continuous material calculations: • Young’s Modulus (E) = 200 GPa • Yield Strength (Sy) = 305MPa • Ultimate Strength (Su) = 365MPa Welded strength for discontinuous material such as joint calculations: • Yield Strength (Sy) = 180MPa Formula Student Rules 2019 Version: 1.1 29 of 133 T3 General Chassis Design • Ultimate Strength (Su) = 300MPa So using super light and strong NB structure would result bit less of score in one of the two top category than using realistic values, but then again modder could try to go that route if it would allow him to get away with it by scoring higher in endurance even for example. Also damage and deformation should be part of scoring, so one doing excessive FLT_MAX (me?) would loose points. Challenge is to make scoring such that one would feel it is fair, not too impossible, but still encourages good practices. One great thing in FSAE is that there are many possible ways to win, while endurance event gets lot of points, one can focus on other areas more and win despite not being best in endurance event. Visual aspects are then of course one important area, but I feel as a race car it should not be most defining aspect, but certainly important one. Also I think use of LUA could give extra points, for example instrument cluster etc. would definitely need recognization (is that a word) in form of extra points. Anyway these are just thoughts that came up sneaking on me, just something for discussion, would love to hear if some of them are bad or not working.
I actually really like this idea... I have summer holidays coming up and I am determined to learn how to mod. Mine will likely completely fail but I'm really willing to give this a shot. I hope people get involved and try and challenge themselves, as with enough people this could be an awesome event... maybe even an annual thing?? I wonder if the devs would be willing to judge the cars and make this an official BeamNG event? or is that too much of a stretch? Looking forward to seeing how this turns out EDIT: Also I think either team or individual submissions should be allowed, and depending on numbers there will be an overall winner or an individual winner and team winner... Just my opinion
I think it is something that could motivate to go bit further, there are also very good modders that might find new wind by trying this, then new modders that might find this to be a goal to get things rolling, great thing is that there is ton of information available about formula student vehicles. Also as we know BeamNG is top of the line simulation, we might be able show what it really can do. Speaking of simulation, I stumbled upon this: http://www.fsaesim.com/documentation.html They claim accuracy within 5% I wonder if we can do better? My only concern is lack of tire temps, which are kinda important as with them it is not possible to make setups that abuse tires excessively, but maybe some day that too is in. There are all kind of huge potential in this, much more than I originally thought, so this might be actually really good even for BeamNG when more modders join in. At some point it might be official event as potential in marketing and benefiting long term goals is there, but I think that might be dreaming at this point, but when enough momentum is there, anything is possible, so more people joining in and giving their input is certainly helpful. I might not have best of abilities, certainly I am a dreamer that dreams big, but I do have time to put in for many kind of things, as working on things that benefit BeamNG works well on me, that is what I do.
