GreenPower Racecar Aerodynamics - azza46

[azza46]

Hi, this is my first post on this forum although I often scroll through as I have a keen interest in aero and have always found the posts to be very informative and interesting. Now though, i have a project of my own and could do with a bit of guidance as I am far from experienced in this field.

For my project I am designing a bodyshell for a car to enter in the Greenpower series. The main priority for the shell is low drag. I have started by taking some measurements of a fictional driver and made a basic shape using two aerofoils, one cambered and one symmetrical. I have made a model in solidworks and run an initial flow simulation but my main problem is i do not really know how to translate the results i can see on the screen into a potential next step of development. I cant really identify potential problem areas or what needs to be changed.

I study motorsport engineering and we do learn a small amount about aero but only the very basics. So i would like to take it to the next level and start developing my knowledge.

So, after my waffling, i was hoping to put up a picture of my flow results and see if anyone could offer any advice as to what i am looking at and what my next steps should be.

Right, i hope this works.







Anything i can learn would be useful, even if its what information i should be displaying. I did put a floor in but i have hidden it. The car is running a ride height of around 60mm so that what i set it at.

Obviously i havent got the driver in yet but i was just hoping to get an initial reaction before i start taking it further.


Thanks.

[hollus]

I think this thread might be of interest to you.
viewtopic.php?t=6677

[kiev the great]

I just completed my masters project with 2 others on a greenpower corporate challenge car in association with company. We designed the bodyshell of the car. We went the slim as possible bullet route (ended up looking like the Silesian entrant from last year), but lots of things were considered. With regards to your analysis, the cockpit opening is a very big deal. We spent lots of time looking at the shape and angle of the openings to reduce drag. The drivers head is also a big deal as the regs force you to have the drivers eye line above bodywork in front and to the sides.

[wesley123]

I think this thread might be of interest to you.
viewtopic.php?t=6677

Seems like he already visited that topic.

But more OT, having rounded edges on the model would already help a little in reducing drag. As the topic hollus already showed, the shape you're using already seems to be getting pretty close to the solution with the least drag

[Greg Locock]

The basic shape of the fuselage should have a Cd (frontal era basis) of about 0.07 to 0.15, depending on the distance for which you can maintain laminar flow (60% giving the 0.07). BUT the devil is absolutely in the details, wheels, cockpit opening, suspension and so on will easily double the Cd if care is not taken.

[Tommy Cookers]

Going back to stuff done about 50 years ago, I remember Cd of 0.19 for Bluebird and 0.116 for Goldenrod

I don't think the laminar flow would have extended more than about 15-20 %

Is this apparent discrepancy relative to your Cd values due to scale/Re number effects ?

Goldenrod certainly did the business

[Greg Locock]

One of my favorite cars http://calteches.library.caltech.edu/528/2/Bettes.pdf

I'm not a real aerodynamacicicst so take the following with a huge pinch of salt.

For sensibly designed cars there are roughly 3 parts to the drag equation. The first is L/D. A car is a lousy shape for a wing, and at best L/D is around 10, 5 is more typical. So minimising lift or downforce is a high priority

Second, skin friction. That's where laminar flow comes in and for 'streamlined' bodies it is the dominant source of drag. The simple summary I gave more or less covers that. In the case of Goldenrod they would have maintained laminar over the nose cone only.

Thirdly, inertia recovery. With a nicely tapered tail section they'd be fine on that.

So on skin friction I'd guess they would be around where a non laminar flow solar car is, 0.12-0.15

If you want to argue the difference between 0.12-0.15 and 0.119, let me introduce you to the real world of windtunnels. Same car in two different tunnels can have 30% errors in pitch moments, and 0.03 on Cd (ie 0.3 and 0.33). Did the caltech tunnel have moving floor?

[N12ck]

One of my favorite cars http://calteches.library.caltech.edu/528/2/Bettes.pdf

I'm not a real aerodynamacicicst so take the following with a huge pinch of salt.

For sensibly designed cars there are roughly 3 parts to the drag equation. The first is L/D. A car is a lousy shape for a wing, and at best L/D is around 10, 5 is more typical. So minimising lift or downforce is a high priority

Second, skin friction. That's where laminar flow comes in and for 'streamlined' bodies it is the dominant source of drag. The simple summary I gave more or less covers that. In the case of Goldenrod they would have maintained laminar over the nose cone only.

Thirdly, inertia recovery. With a nicely tapered tail section they'd be fine on that.

So on skin friction I'd guess they would be around where a non laminar flow solar car is, 0.12-0.15

If you want to argue the difference between 0.12-0.15 and 0.119, let me introduce you to the real world of windtunnels. Same car in two different tunnels can have 30% errors in pitch moments, and 0.03 on Cd (ie 0.3 and 0.33). Did the caltech tunnel have moving floor?

one way to reduce drag is vortex generators across the body to reattatch the flow

[Greg Locock]

no, they reattach separated flow, it is still, or even more, turbulent. Separated flow increases the inertia recovery drag, in the simple model I gave.

[N12ck]

no, they reattach separated flow, it is still, or even more, turbulent. Separated flow increases the inertia recovery drag, in the simple model I gave.

they reattach separated flow as I described, a side effect of this can be reduced drag, depending on the design of the VG's

for example, this shows the principal perfectly:

http://www.mitsubishi-motors.com/corpor ... 16E_03.pdf

[Greg Locock]

Sorry, yes, you didn't say they affected the laminar flow. Yes, energising the boundary layer can keep the flow attached where it would otherwise have separated, and it can reduce, or increase, the drag.It's a tradeoff between increased skin friction due to the higher velocities, against the reduction in inertia recovery drag.

[N12ck]

Sorry, yes, you didn't say they affected the laminar flow. Yes, energising the boundary layer can keep the flow attached where it would otherwise have separated, and it can reduce, or increase, the drag.It's a tradeoff between increased skin friction due to the higher velocities, against the reduction in inertia recovery drag.

yes I possibly should have been clearer, the placement of them is a major factor, and it is a trade off as you say,

if you put them very far upstream (depending on the design) it can create extra drag,

if you put them further downstream with the correct design it can reduce drag by reattatching flow and yes you get more skin friction, but if it designed correctly you can overcome this negative,