2017 front wings downforce compared to 2010

[jjn9128]

The simulation is half wing so I have to double the force value it get the value for the entire wing. I’ve only got a wing in setup so I don’t think I’m including the forces from anything else.

Forces on the ground plane (should be opposite so probably not) or symmetry plane, could have included a fluid integration box as well as the surface pressure integral... you'd be surprised some of the mistakes that can be made.

Are your reference values correct?

[Dipesh1995]

The simulation is half wing so I have to double the force value it get the value for the entire wing. I’ve only got a wing in setup so I don’t think I’m including the forces from anything else.

Forces on the ground plane (should be opposite so probably not) or symmetry plane, could have included a fluid integration box as well as the surface pressure integral... you'd be surprised some of the mistakes that can be made.

Are your reference values correct?

I think I have found the source of my force value errors. So I did a few hand calculations integrating the pressure coefficients for the plan area for each wing element to get the coefficient of lift which came to -1.2931 +- 3% margin of error and then using the standard equation to get a force and I got 1902 N which is far more sensible than the value I was getting. Having investigated further, I found that STAR was giving the coefficient of lift based on plan area as 2.297 which is absurd. If I put that value into my hand calculations I get 3531 N which is pretty much the same value as STAR is giving me so my hand calc method is correct. The question now is why is it doing that?

[Dipesh1995]

If the ratio of error between the coefficients of lift is the same for the complex wing as well, it would generate 4055 N which seems correct although most likely it won't be though I suspect it won't be too far off.

[jjn9128]

I think I have found the source of my force value errors. So I did a few hand calculations integrating the pressure coefficients for the plan area for each wing element to get the coefficient of lift which came to -1.2931 +- 3% margin of error and then using the standard equation to get a force and I got 1902 N which is far more sensible than the value I was getting. Having investigated further, I found that STAR was giving the coefficient of lift based on plan area as 2.297 which is absurd. If I put that value into my hand calculations I get 3531 N which is pretty much the same value as STAR is giving me so my hand calc method is correct. The question now is why is it doing that?

Sounds to me like there's a reference value incorrectly input somewhere... are you asking star to output a force or a coefficient? I don't think the reference area would affect the force calculations but may well affect your coefficient. Would -2.3 be the ClA, i.e. Cl based on a reference area of 1m*m?? So if you divide that by your reference area you get -1.3?

But that doesn't explain why the force is wrong, unless it's working out a coefficient then calculating the force from that???

I'd advise checking all the reference values in each drop down.

[Dipesh1995]

I think I have found the source of my force value errors. So I did a few hand calculations integrating the pressure coefficients for the plan area for each wing element to get the coefficient of lift which came to -1.2931 +- 3% margin of error and then using the standard equation to get a force and I got 1902 N which is far more sensible than the value I was getting. Having investigated further, I found that STAR was giving the coefficient of lift based on plan area as 2.297 which is absurd. If I put that value into my hand calculations I get 3531 N which is pretty much the same value as STAR is giving me so my hand calc method is correct. The question now is why is it doing that?

Sounds to me like there's a reference value incorrectly input somewhere... are you asking star to output a force or a coefficient? I don't think the reference area would affect the force calculations but may well affect your coefficient. Would -2.3 be the ClA, i.e. Cl based on a reference area of 1m*m?? So if you divide that by your reference area you get -1.3?

But that doesn't explain why the force is wrong, unless it's working out a coefficient then calculating the force from that???

I'd advise checking all the reference values in each drop down.

I think you’re on to something because if I multiply 2.297 with the planform area which is 0.569, I get 1.31 which is very close to what I calculated, well within rounding and reading error range. So 2.297 may well be Cl/A based on a planform area of 1 m^2 although I did put the planform area to get it.

I’m asking STAR to output a force value and there is no option as far as I’m aware to input a value for area. It only asks for the direction and ref pressure which is 0 and makes very little difference if I put anything else in anyway. I can only input at an area value for the coefficients but it still gave me 2.297. I don’t know how STAR calculates the force but I would have thought it integrates the each bit of area with its corresponding pressure coefficient to calculate Cl which it then uses it to get the force. This is all a bit odd to be honest. I’ll check again to make sure I’m not being an idiot but I doubt it.

[Dipesh1995]

After reading the notes, STAR uses pressure*area to calculate the force where area is a face area vector.

[jjn9128]

After reading the notes, STAR uses pressure*area to calculate the force where area is a face area vector.

Yessss... but the pressure over the area of a given surface cell, i.e. a cell on the bottom surface of a wing with negative pressure inclined at 30degrees has a downforce and a drag vector. The surface cell vectors are then summed for the body and added to some wall law and turbulence modelling to add viscous drag effects. This doesn't explain why your absolute forces are wrong... and by a factor of your planform area. Is there a possibility that's a coincidence???? I think not, all I can think of is to start the simulation from scratch and take car inputting all the values (not copying from the current case) see if you get the same result.

[Dipesh1995]

After reading the notes, STAR uses pressure*area to calculate the force where area is a face area vector.

Yessss... but the pressure over the area of a given surface cell, i.e. a cell on the bottom surface of a wing with negative pressure inclined at 30degrees has a downforce and a drag vector. The surface cell vectors are then summed for the body and added to some wall law and turbulence modelling to add viscous drag effects. This doesn't explain why your absolute forces are wrong... and by a factor of your planform area. Is there a possibility that's a coincidence???? I think not, all I can think of is to start the simulation from scratch and take car inputting all the values (not copying from the current case) see if you get the same result.

Ok so I’ve made a bit of a blunder, 1.29 is the ClA so if I divide that by A to get Cl, I get pretty much the same value as the sim so if there is an error, its somewhere else.

Anyway, I ran the sim again but this time with the full wing with small reduction in AoA of the flap to get rid of trailing edge separation. The planform Cl value I got was -2.15 and the planform Cd I got was 0.14. The planform area of the wing is 1.214 m^2. Having looked at some papers from the University of Southampton (https://www.southampton.ac.uk/~mb4/xin_zhang/29.pdf & https://www.southampton.ac.uk/~mb4/xin_zhang/48) , comparing their figures with mine, I can see that my pressure coefficients are in the ball park as are my Cl numbers assuming that they’re also based on planform area. My force values are 6958 N (downforce) and 460 N (drag).

I'm going to see a CFD Prof tomorrow, hopefully he'll shed some light on any errors with my setup.

[Dipesh1995]

After reading the notes, STAR uses pressure*area to calculate the force where area is a face area vector.

Yessss... but the pressure over the area of a given surface cell, i.e. a cell on the bottom surface of a wing with negative pressure inclined at 30degrees has a downforce and a drag vector. The surface cell vectors are then summed for the body and added to some wall law and turbulence modelling to add viscous drag effects. This doesn't explain why your absolute forces are wrong... and by a factor of your planform area. Is there a possibility that's a coincidence???? I think not, all I can think of is to start the simulation from scratch and take car inputting all the values (not copying from the current case) see if you get the same result.

Ok so I’ve made a bit of a blunder, 1.29 is the ClA so if I divide that by A to get Cl, I get pretty much the same value as the sim so if there is an error, its somewhere else.

Anyway, I ran the sim again but this time with the full wing with small reduction in AoA of the flap to get rid of trailing edge separation. The planform Cl value I got was -2.15 and the planform Cd I got was 0.14. The planform area of the wing is 1.214 m^2. Having looked at some papers from the University of Southampton (https://www.southampton.ac.uk/~mb4/xin_zhang/29.pdf & https://www.southampton.ac.uk/~mb4/xin_zhang/48) , comparing their figures with mine, I can see that my pressure coefficients are in the ball park as are my Cl numbers assuming that they’re also based on planform area. My force values are 6958 N (downforce) and 460 N (drag).

I'm going to see a CFD Prof tomorrow, hopefully he'll shed some light on any errors with my setup.

I've had my setup checked and there is no issue with it. He also checked the coefficient of lift with experimental results of another two-element wing and my Cl was in the ballpark so I'm sure the results are not massively inaccurate.

So for the complex wing, the Cl is ~ -2.78 and the planform area is ~ 1.02 m^2. The air density STAR uses is 1.18415 kg/m^3 and the inlet velocity is 67.056 m/s. So that gives approximately the same value as the simulation for the entire wing.

There are a few reasons that I have come up with that may explain why I have such a high downforce value.

1. Realizable K-Epsilon, as seen for wing simulations, tends to overestimate lift coefficients by roughly 9-10% and sometimes even more compared to experimental results.
2. The wing is much larger than an F1 wing. The span is the same at 1.8m but it is 75mm taller and has a 95mm longer depth.
3. The setup does not incorporate all the downstream components seen on an F1 car which may affect downforce generated by the front wing.

Anyway, I'm now going to start to wind this down as I need to turn my attention to my FYP. Once again, I really do appreciate the help you and Vyssion have given me throughout these last couple of months.

Also: https://photos.app.goo.gl/rWK6EQg88fxs81Mq2

I've fixed my Y250 . Still a bit weaker than I hoped but it'll probably work better if there were some bargeboards and brake ducts.

[Dipesh1995]

For those who are interested

Whilst searching for articles on F1 front wings, I came across an article written by Willem Toet.
https://www.linkedin.com/pulse/secrets- ... illem-toet

In the comments, Willem Toet says that an F1 car generates about 1 tonne of downforce at 160 KPH. F1 front wings typically contribute between 25-30% of that so if what Toet says is true, by taking that into account and scaling the downforce to what it would be at 240 KPH, F1 front wings generate between 560-675kg of downforce.

[Vyssion]

Apologies for the lack of communication recently.

I made some minor changes to the wing around the tunnel vortex serrations and the leading two elements of the wing but none of those parts don't really influence the Y250. The vortex leaving the inboard tips is the same.

Given that the main premise of the edge vortex is to blast the front surface of the rotating tyre to try and minimize the 6 vortices being shed by it, if youre playing around with anything in that area, you may have upset the system from what you had before and its now no longer doing that.

I would also strongly recommend you ALWAYS begin your simulations from timestep zero with a full initialization, rather than continuing from current data; especially if you are changing geometry and thus your mesh!!

Anyway, I ran the sim again but this time with the full wing with small reduction in AoA of the flap to get rid of trailing edge separation. The planform Cl value I got was -2.15 and the planform Cd I got was 0.14. The planform area of the wing is 1.214 m^2.

You mentioned planform area? In my experience, the way that Cd and Cl are used in motorsport is based on the frontal area rather than the planform area. Aviation use planform area, however, racing tends to use the same area figure (frontal usually) for both lift and drag. Also, teams often wont even use a specific frontal area due to the fact it changes with all the little tweaks that they make to the wing pitches, rake angles, etc etc... They will often just use a "standard constant" value for all calculations to keep things homogenous.

1. Realizable K-Epsilon, as seen for wing simulations, tends to overestimate lift coefficients by roughly 9-10% and sometimes even more compared to experimental results.

I would agree with the over-estimation with regards to the standard k-eps model as it has been shown experimentally to break down when measured next to complex gemetries with strong adverse pressure gradients (i.e. wings in ground effect). The realizable k-eps model though is slightly different in that it uses a changeable variable for the C(mu) value rather than a constant. It also has a new transport equation for the dissipation rate which comes from an exact differential equation derviation; these two things means that it is substantially better at modelling complex geometries, boundary layer shear/flows, strong adverse pressure gradients, etc, than its standard version. There are obviously better turbulence models out there for this specific type of thing, but it isn't all that bad.

Anyway, I'm now going to start to wind this down as I need to turn my attention to my FYP. Once again, I really do appreciate the help you and Vyssion have given me throughout these last couple of months.

You're welcome mate. Always welcome back with your new questions etc!