Over and Under or around the sides

[Carlos]

It's very impressive, really well done; I especially admire the ingenuity of the underpan's tunnel.

[greenpower dude reloaded]

I've always felt it's vitally important to cover the wheels as much as possible and despite several attempts at that it was always giving odd results and funny pressures, the latter we seemed to be getting regardless of whether we included wheel covers or not and so I thought i would try this. That way there should be very little if any pressure change once it's in the tunnel.

The big area of low pressure just on the top of the roll hoop is down to poor meshing, the leading edge is elliptical. But it meshes it square and jaggedy so that's an area that has already been improved really.

Credit is also due to MarcusH who mentioned channels previously, but i think this is a fair bit bigger than he meant.

Other developments are we have been offered a discount on some already very cheap carbon fibre and the loan of a Vacuum pump. So I may be looking to give that a go early next year.

I'm just running a model in the software atm of this years champion.... once you know tyre sizes of the top car it then becomes very easy to scale a picture and make a model.. I'll let you know FA and Cd once the run is done and how it compares to our car.

[marcush.]

luke,very cool looks of the wonderwoosh ,I feel this converges towards a very good 2010 car...

did you actually try a more droplike shape in plane view ? lets say with the biggest width of the bodywork at the front wheels -of course sacrificing some frontal area ,but maybe giving better overall flow ?

very exciting the whole project..... =D>

[BreezyRacer]

Very nice indeed. Looks pretty slippery and still easy to build. Do you have any flow line images?

[greenpower dude reloaded]

yeah, I could bore you all with literally 100's of images etc. but instead I'll just post up the bigger developments.

The first big jump was including these long wheel covers. The second came just last night when I took away the extra width i had included (between updates) to make it a more teardrop kind of shape.

This caused a lot of low pressure and it seems that with this nose and tail config flatter sides are better. It really is all trial and error.

I'll get some flow images up 2mo eve I'm on Linux atm and i'm using paraview on vista, it didn't want to work on linux

[BreezyRacer]

yeah, I could bore you all with literally 100's of images etc. but instead I'll just post up the bigger developments.

The first big jump was including these long wheel covers. The second came just last night when I took away the extra width i had included (between updates) to make it a more teardrop kind of shape.

This caused a lot of low pressure and it seems that with this nose and tail config flatter sides are better. It really is all trial and error.

I'll get some flow images up 2mo eve I'm on Linux atm and i'm using paraview on vista, it didn't want to work on linux

No need to post them, just as long as you're happy that you've been able to chase out much of the flow separation at the tail area. You've really learned a lot during this project! Given your drive and ambition you will be victorious. Start planning on how to leverage all that success into your next project.

[tok-tokkie]

OK!! sorry (again) been busy running what seems like hundreds of models this is the best one i've managed so far.



I haven't had (until now) figures as low as the models I had somebody run for me before. Even though I ran the same models. SO although this figures sound high I don't think that the Cd is all that believable and that it's better to look at the relative improvements between designs.

The first car we tried and had someone else ran output a Cd of 0.28 on their program and 0.42 on mine.

After a lot of minor changes I have got that down to 0.26 on my software so that's a drop of 39%. The same drop on the other software (which is much closer to reality) would give us 0.17 which i suspect is about right. We also have a much smaller FA now as well.

So @ 40mph now we would require 0.3hp to over come the aero drag.

Before we started this we would have required... 0.4hp so thats roughly 70 watts saved! thats very nearly 3amps!! and when we only draw 20 that's a big difference!!

This continues to be a fascinating & informative thread.
Why is the rear view model (old?) not the same as the front view (current?)?

[greenpower dude reloaded]

This continues to be a fascinating & informative thread.
Why is the rear view model (old?) not the same as the front view (current?)?

thank you and erm i dunno It should be the same they were taken during the same session


Breezy,

I've tried to get rid of as much seperation as possible but have still found to the tail to cause a little bit however, a run using a Kamm tail produced a higher Cd, so I decided to stick with a little separation rather than cause a lot. I suspect that to be effective you have to design a Kamm back very carefully

[marcush.]

I read about the shaping of the backside-Kamm heck some time ago and they stated not to end in a blunt surface but actually recess the rear surface a few inches into the body so you get a real trailing edge all around the rearend surface.this should help ovoiding the flow trying to not cleanly separating from the surface but trying to lick the surface of the blunt back.
I will look if I can retrieve that piece ...

[BreezyRacer]

When your goal is an ultimately low drag number I think what you've got is about as good as it's going to get. A Kamm tail has lots of space saving advantages for it's design but this appears to be one area of your body that you have room to get close to optimum.

If you wanted to try something I would raise the back "diffuser" area and raise the whole rear tail section.This might give you lower drag numbers if your roll hoop and drivers fairing are pretty clean. If they cause turbulence though, it will just still be turbulent at the tail, thus no improvement.

Of course you can also play with some vgs around the driver's helmet area across the top and see if that doesn't reduce turbulence at the tail. You have to mesh really tight see the vgs in CFD.

It's a lot of effort to "tune" your CFD technique, getting numbers that are repeatable and useful. It seems like you've done well there. I knew something was up a few weeks ago when you proclaimed a vast difference in drag numbers between your previous two designs. Glad to see you've got a reliable baseline now.

[greenpower dude reloaded]

Thanks Breezy, although (being a bit of a perfectionist) I have to admit my CFD is still a long way off but it is getting there. As you can see the pressures are slightly to one side on the front of the car.

The run you mentioned was one of those very perculiar fluke ones. A problem with the mesh which seemingly cannot be helped. I still get those from time to time, but have realised it best not to mention those so you all think I've improved haha

[flynfrog]

Just wait until you move on to cross winds

At some point you have to say ok thats good enough and move on.

Btw looking good. do you know your Cp vs Cg at your speed its not a huge deal I supose.

[greenpower dude reloaded]

Haha, I've been dreading putting it in the tunnel at at 45degs, but It's got to be done at some point though, I think that's when we will really see a need for more rounded edges.

I was loathed to include the long wheel covers at one point because I thought it would make too much of a difference in a cross wind but they improved things so much that really I don't think a 35mm strip is all that bad..

It's all looking encouraging though...

trying to Get CG as low as possible and it's not that bad in all honesty, there is no weight worth mentioning above say 200mm above the ground and not entirely sure where the CP is.

[flynfrog]

Haha, I've been dreading putting it in the tunnel at at 45degs, but It's got to be done at some point though, I think that's when we will really see a need for more rounded edges.

I was loathed to include the long wheel covers at one point because I thought it would make too much of a difference in a cross wind but they improved things so much that really I don't think a 35mm strip is all that bad..

It's all looking encouraging though...

trying to Get CG as low as possible and it's not that bad in all honesty, there is no weight worth mentioning above say 200mm above the ground and not entirely sure where the CP is.

45 is over kill most likey but 15 to 25 might be useful The CP vs CG is more of a front to back problem you want your CP on the correct side of your cg or you will aero unstable.

[greenpower dude reloaded]

Oooops, of course, Unfortunately it's very tail heavy.. I've looked in to changing that problem but there is not a vast amount we can do we are really pushed for any useful space in this car. I haven't put it on the scales yet but at an estimate I would say that fully loaded it'll be about 120 KG 36kg batteries 44kg driver (you've got to love little sisters!) + 40kg car. All educated geusses. There is one battery directly infront of the rear axle and one directly behind there is also about 10KG of drive train not far behind that.

FL 20 FR 20

1100mm Wheelbase

RL 42 RR 38