The Underbody (flat belly pan is suboptimal - explanation & solutions offered)

[Gealii]

Originally Posted by AndrzejM

I'm just thinking... What about flat bellypan with small bumps instead of dimples? Any thoughts? Because in my opinion it's much simpler to glue some bumps to the flat surface of the belly pan than to make dimples in coroplast for example.

Edit: In addition, I have those bumps on the inside wall of Berta's rearview mirrors. Don't know the exact meanning of those, for now. I've heard that cut the noise of the air passing the mirror, but I'm not sure that's the only reason for that.

Its easier to glue bumps onto coroplast instead of taking a ball-pin hammer and hitting the coroplast to relieve anger and making the dimples?

[MetroMPG]

Hard to tell for sure from a single photo, but I was curious to see what GM had done on its remarkably slippery EV1.

Food for thought, looking from back to front:



Not shown is the portion behind the rear wheels, which was surely upswept to aid pressure recovery at the rear.

[aerohead]

Originally Posted by MTrenk

freebeard:
--> This is why almost all racing series have height restrictions on exhaust exits, and have banned skirts that touch the ground.
--> Nowadays, I am studying the effects of placing exhaust above the diffuser on my team's car, trying to pull air under the car to get the diffuser to 'up to speed' faster, since underbodies can become effective at as low as 20 mph under normal conditions (our average speed in the Formula SAE competition is around 30-40 mph, so it's vital we make as much use of these things as possible).
--> It wouldn't be negative pressure, just less pressure (probably).

Frank Lee:
--> The underbody is very different from the rest of the car. Yes, the further you go along any surface, the thicker the boundary layer (because of air molecules getting 'stuck' on the surface molecules, resulting in a zero velocity at the surface), but if you look at a flat panel boundary layer vs. an airfoil boundary layer, I think you'll get an idea of why it matters so much how things are curved.

aerohead:
--> Dimpling a surface changes up Reynold's numbers. Size matters, but so does everything else.

If we're limiting our discussion to belly pans,then,for a conventional automobile,it will have a fully turbulent boundary layer and constant coefficient of drag without dimples.Adding dimples would be superfluous.It would be like throwing a lit match into a roaring fire.Dimpling might serve an accoustic function,limiting frequencies which can be generated.
-------------------------------------------------------------------------
As to automobile-size spheres,both smooth,and dimpled spheres would have turbulent boundary layers.It's simply a supercritical Reynolds number effect.
-------------------------------------------------------------------------
As to everything else,sure.But bear in mind that streamlining is dirt simple.We've had perfectly streamlined forms for 500,000,000 years.It's working around hard styling which is difficult.I'm sure Hucho tore his hair out at VW trying to clean up Georgio's designs while maintaining the 'look.'

[2000mc]

Originally Posted by MetroMPG

Not shown is the portion behind the rear wheels, which was surely upswept to aid pressure recovery at the rear.

[AndrzejM]

Originally Posted by Gealii

Its easier to glue bumps onto coroplast instead of taking a ball-pin hammer and hitting the coroplast to relieve anger and making the dimples?

Yeah you are right that will be much easier, but I think coroplast will be weaken in those "anger relieve" spots, and if you glue something to the coroplast's surace it can be even stronger than just a sheet of coroplast.

But that wasn't my question anyway. I just wanted to know your opinion on dimples vs bumps according to aerodynamics

[AndrzejM]

I just wanted to show you those bumps I've mentioned before.

[Smokeduv]

I was thinking on putting a complete flat underbody because my car already has it but it had those dimples, but I never thought that they are better than a fully flat surface, so maybe I'll just leave it as is and then flatten the parts that can cause turbulence.

This is the original underbody panel in my car (sorry for the very bad picture):



Can i improve it?

[MTrenk]

I don't believe that bumps would work. Dimples create low pressure zones that keep the boundary layer attached for as long as possible. Bumps won't really create that effect.

aerohead:
I know about the Reynold's numbers and blah blah blah.
--> A flat panel will have a short laminar attached boundary layer, then an attached turbulent layer, then lose any attached flow.
--> A dimpled (flat) panel will keep the attached turbulent flow, reducing drag compared to the flat panel.
Sure things are different based on how big your object is vs. how large your dimples are and stuff, but in no way can you say that a flat panel will be the same as a dimpled panel.

Still the best way is to not have a flat panel; an airfoil shaped bottom is ideal.

[freebeard]

I've come up with a solution that may be the optimum for ecomodding.
--> Front air dam to move the air around the car
--> Underbody that slopes upward for the airfoil shape in the back
--> Kammback extension to match the airfoil shape from the underbody

Front air dam plus side air dams? Front dam ahead of the bumper vs between the front wheels? Curved vs straight?

How would you match the K-back to the diffuser?

--> Nowadays, I am studying the effects of placing exhaust above the diffuser on my team's car, trying to pull air under the car to get the diffuser to 'up to speed' faster, since underbodies can become effective at as low as 20 mph under normal conditions

I'm still trying to picture this.

[Otto]

Questions:

Flat panel belly pan: Why not slice up several old tennis balls into crescent shaped domes of ~1/2" height, then glue or velcro them to a flat belly pan and test for difference?

Corrugated belly pan: The Ford Trimotor and I think a Junkers transport plane of that era had corrugated skin, with the ridges/valleys oriented in the direction of flow. This reportedly was an elegant way to stiffen the skins while reducing aerodynamic drag, as the flow wedges (wakes) generated by surface imperfections were contained and prevented from growing laterally downstream. Corrugated fiberglass material is available very cheaply from every Home Depot or Lowes. Anybody tried using it as belly pan material, or made aerodynamic tests of it?

The Toyota video of their new SUV: I'm underwhelmed. The guys in that video crow about their design, but frankly it seems to me they left LOTS of money on the table. Maybe they studied aero at the same school whose grads did the Chevy Volt. I'd bet dollars to doughnuts that various people on this website could make significant improvements to that vehicle in a weekend, working in their driveway, for less than $100.

MTrenk, are you a seasoned, practicing aero engineer or a beginning student with years to go before that? Read Hucho yet?