I have done some search around here and did not manage to find an answer. How correct/wrong I am to assume that body base drag value is drag equation at: air density at 1.2kg/m^3 (air pressure 101 kpa, temperature 20C) vehicle speed 1m/s So in case of for example NSX. If my short research is correct: A=1.78m^2, cd=0.32 nsxD=1.2/2 [kg/m3] * 1.78 [m^2] * 0.32 * 1^2 [m/s] = 0.34 In case of cobra: A=1.54m^2, cd=0.5 cobraD=1.2/2 [kg/m3] * 1.54 [m^2] * 0.5 * 1^2 [m/s] = 0.46 However rF2 cars has base drag balues like that: NSX has 0.31, and cobra 0.502 plus has some extra bit of drag applied to rear and front. I also saw that post 2000 NSX might have been bit more aerodynamic, down to 0.3 CD. Whats confusing is that for NSX I get larger value, and for Cobra I get smaller value than is set in rF2 physics .hdv files. And I used largest cd coeficient I find for cobra, while some other sources had 0.42-0.45 values, but they might have been for a car with windshield removed. I just make an assumption that cobra drag base value of rF2 might have been inflated for rough underbody, which I suppose CD coeficients might not include. NSX underbody is much more streamed, so I guess it matches acctual cd better:
ANSYS CFD Webinars is a good place to start. We used it at our oven business to design closed loop flow systems and it was pretty accurate. I sold out my share when I retired a few years ago, but here is a link to the webinars. https://www.ansys.com/products/fluids/mini-meshing-webinar-series That said, keep in mind it took about 10 minutes to run a flow/velocity/density analysis, and 40-50 minutes to run thermal analysis. Running a simplified version of aero drag in realtime is all we can hope for and leaves room for error particularly when there are adjustable wings, and thats found in a handful of otherwise very accurately modeled mods. About a dozen of the mods I run most are customized as I don't run online or even share them out of respect to the modders. Nearly all of them have had adjustments to BodyAeroDrag variables. For a few of these revised mods there was data available for power, top speed, and aero device settings and for those using some trial & error to match actual performance was my solution.
Yes, it would be awesome to plug use of CFD tools into aero research. But the thing I am wondering is what is needed for the basedragvalue in the first place. It is not just Cd, it is not CdA, because then values would be greater than CD as almost all of the vehicles has frontal area greater than 1 square meter.
Hard to say, but I suspect the overall frontal profile is calculated from the graphical model and the BodyBaseDrag is applied to that area. Maybe try matching the physics of a smaller carbody to a larger carbody and see if the top speed drops, that would be a good indicator.
From the physics spreadsheet, drag (N) is calculated with (rF2 drag coefficient) * v^2. The usual drag formula is drag = 0.5 * air density * v^2 * Area * drag coefficient So the (rF2 drag coefficient) = 0.5 * air density * Area * drag coefficient... no?
Indeed it does. I was a bit stymied by where the Area was coming from. Body area input variables point back to C184-E187 from what I could trace. HDV comments like "// aero forces from moving left" probably should have been a sufficient description but apparently I had some brain fade on the interpretation....maybe it's time for a haircut Going through the aero worksheet was an overdue exercise, but it's quite thorough in replicating accurate conditions for closed wheel cars. Open wheelers can have gaps between body/wing/sidepods where aero forces that may add a bit of chaos to the mix. From the FISI example in V035 they added Fender Parms and were rather light on wing drag to downforce as compared to a couple other mods I've hacked for personal use. If only I had access to pitot tube data..... So it all sort of came together for me, not what I had intended on spending a few hours doing today.....but time well spent.
>>>>>Whats confusing is that for NSX I get larger value, and for Cobra I get smaller value than is set in rF2 physics .hdv files. Well, BodyDragBase is only one of many parameters to set an rfactor vehicle's drag. There is also contribution by BodyDragHeightAvg, BodyDragHeightDiff which can be assigned a respectable percentage of the total drag, and 'bodyrake' parameters, too. Furthermore, if there is a lot of lift it will add a bit of drag, depending on rake angle. I mean, the same total drag can be obtained by combination of different parameters. It depends on how much height/rake sensivity you want to achieve and other features.. That's why i wouldn't be obsessed with bodydragbase value. Normally you want to achieve realistic acceleration and top speed. Once you are sure the engine is producing the right power at the wheel (which is already difficult to know, because information sources are often misleading) it is forgivable to tweak a bit the drag coefficient until the result you obtain is the same as the real counterpart. If you want to try the opposite way, a maximum theorical aero drag coefficient can be obtained with a simple formula: Cx= Power / top_speed^3 (power in watts and speed in m/s) However, some of the drag should correspond to tyre roll resistance, spindle friction,... so that aero drag should be slightly smaller than the maximum calculated. Rfactor 2 doesn't take in account any frontal area data for calculations. Perhaps it is on the spreadsheets to help obtain the drag coefficients for the hdv file. Hdv drag coefficients can be considered the usual 'Cx' coefficients published by car industry. (Edited)
