CFD - 2022 F1 Car

[Vanja #66]

Where is throat located in 2022 cars, I expect here lowiest pressure?
Is there any real picture of undertray of these cars?

You'll have to look at car threads for that, haven't seen good photos so far.

[Likes_race_cars]

What a fantastic post, not sure i understand even half of it. This is good stuff.

[Fluido]

Very similar setup to the Perrinn model; straight ahead solve here, no crossflow or curved flow, steady-RANS w/ SST turbulence inc. wall functions, 31.5million elements, etc etc...

How number of cells affect accuracy, if you use high resolution grid(more cells), results will be closer to real one?

[Stu]

Very similar setup to the Perrinn model; straight ahead solve here, no crossflow or curved flow, steady-RANS w/ SST turbulence inc. wall functions, 31.5million elements, etc etc...

How number of cells affect accuracy, if you use high resolution grid(more cells), results will be closer to real one?

More cells do create greater accuracy, but require a longer calculation time. Even the teams rationalise cell count; before increasing resolution where it is required.

[Fluido]

More cells do create greater accuracy, but require a longer calculation time. Even the teams rationalise cell count; before increasing resolution where it is required.

What does it mean, only super computers can do real results?
Can our "home computers" gives correct results or calculation would take a couple of years to have correct resluts?

[CAEdevice]

More cells do create greater accuracy, but require a longer calculation time. Even the teams rationalise cell count; before increasing resolution where it is required.

What does it mean, only super computers can do real results?
Can our "home computers" gives correct results or calculation would take a couple of years to have correct resluts?

I depends on the data you are looking for.

[Fluido]

I depends on the data you are looking for.

Downforce at front/back wheels(within 20N accuracy), static pressure distribution at all parts of car body, position of center of pressure,overall drag, separation point at undertray, cofficient of pressure change in relation to speed(Re number),accurately predict vortex paths at undertray, find thickness of boundary layer at every point of body.. at F1 car

?

[CAEdevice]

I depends on the data you are looking for.

Downforce at front/back wheels(within 20N accuracy), static pressure distribution at all parts of car body, position of center of pressure,overall drag, separation point at undertray, cofficient of pressure change in relation to speed(Re number),accurately predict vortex paths at undertray, find thickness of boundary layer at every point of body.. at F1 car

?

All together? You would need K*10^7 cells. At this point, I would not neglect a transient simulations. Yes, you need something more than a desktop. Even with a multi xeon machine tens of hours would be required.

[johnny comelately]

What do you aero chappies think of bullet point 2?
"Doug McLean (Boeing)| Common Misconceptions in Aerodynamics"

[johnny comelately]

[Fluido]

How number of cells affect accuracy, if you use high resolution grid(more cells), results will be closer to real one?

More cells do create greater accuracy, but require a longer calculation time. Even the teams rationalise cell count; before increasing resolution where it is required.

Where is your autosave, backup folder from CFD and CAD, on SSD or HDD?

Isnt bed to write so much on SSD and I have small SSD, but if I put on HDD will program run slower?

[jjn9128]

How number of cells affect accuracy, if you use high resolution grid(more cells), results will be closer to real one?

More cells do create greater accuracy, but require a longer calculation time. Even the teams rationalise cell count; before increasing resolution where it is required.

Where is your autosave, backup folder from CFD and CAD, on SSD or HDD?

Isnt bed to write so much on SSD and I have small SSD, but if I put on HDD will program run slower?

Without the quote I've no idea what you're asking.

[Fluido]

How number of cells affect accuracy, if you use high resolution grid(more cells), results will be closer to real one?

More cells do create greater accuracy, but require a longer calculation time. Even the teams rationalise cell count; before increasing resolution where it is required.

Where is your autosave, backup folder from CFD and CAD, on SSD or HDD?

Isnt bed to write so much on SSD and I have small SSD, but if I put on HDD will program run slower?

Without the quote I've no idea what you're asking.

My question is not related to any quote..
If I just write @username you will not be notified...

[Fluido]

G'day! It's been a while... hasn't it?

With the 2019 season now concluded, and jjn and I being in stage 4 withdrawals right now, we figured that we needed "something" to keep our mi... ooooooo!! Is that the 2021 regulations???

So, without further ado, I would like to present to you all our first, and legal (to the best of our knowledge), 2021 spec. F1 car!! And the first car that is wholly of F1T design!!

-sigh-... who would have thought that Aerosaruman and Aerogandalf would ever work together again, huh?

https://imgur.com/PYnaTVH.png
https://imgur.com/guOyb4N.png
https://imgur.com/IeQhAHn.png


So.......

What forum post from Vyssion would be complete, without some CFD: COLOUR FOR DIRECTORS!!

Very similar setup to the Perrinn model; straight ahead solve here, no crossflow or curved flow, steady-RANS w/ SST turbulence inc. wall functions, 31.5million elements, etc etc...


https://imgur.com/P6ED0ny.png
https://imgur.com/5R58Ahv.png
https://imgur.com/ThzSwAg.png


Here is a list of bullet points that, so far, we have twigged, and are gonna work on next. Now I am gonna shamelessly make excuses for any "iffy" performance or flow features you see, because... IT WAS OUR FIRST GO AT IT, OKAYYYY?? Overall, for a first go at it, the car performs very, very well; hats off to jjn.

• CzS = -4.95
• CxS = 1.80
• Front Wheel wake is pretty crazy now... we will work on this
• Going to reduce pitch setting of FW flaps
• Inboard floor fence vortex isnt under control (just spreads out)
• 2021 Force Breakdown = Floor (50%), FW (37%), RW (25%)
• Compared to Perrinn = Floor (55%), FW (30%), RW (40%)
• Higher than 100% is due to Lift generated on the chassis and wheels (6% lift on 2021 wheels, 9% lift on Perrinn)
• Balance of 2021 is ~49% which is quite far forward, so we will correct this (starting with FW flaps as mentioned above)
• Surprisingly large amount of low pressure below the car
• Front suspension is stalled
• Top of floor inlet is stalled
• Rear of Lower sidepod expansion has stalled
• RW flap stalled in curved EP sections (and slightly along TE) and RW as a whole is very draggy

-edit- replaced old gifs with larger ones - I made a mistake when I was creating them... oopsies
https://imgur.com/s0WIUwj.gif
https://imgur.com/rUTxoyq.gif
https://imgur.com/KYFQvuG.gif
https://imgur.com/2faOvGm.png
https://imgur.com/1wXjL5V.png
https://imgur.com/zTHRlNz.png
https://imgur.com/zueL8hT.png
https://imgur.com/bBaIjl3.png
https://imgur.com/9fPR6mf.png
https://imgur.com/kLcnClK.png

How much is project size of CFD and CAD ?

[Zynerji]

I have an odd question for the CFD folks.

Does any team actually model the chassis vibrations and energy lost to noise during these runs? Or is even the most extreme cases still modified approximations?

I'm sure it's a ton more complex, and probably very minimal in its helpfulness, but I would expect it could become necessary as the aero gets more knife-edge.