Why do many velos have flat bottoms?

As an inspiration on a DIY HPV build, I looked at the most popular velomobiles, like the Quest and the Milan. They all appear to have flattish bottoms, i.e. the cross sections are not round/oval, they are more liked 'rounded triangles', with the 3 wheels sticking out of the mostly flat base.

In designing my moulds, I am thinking that having a rotational symmetry would make making the moulds the easier from a DYIer standpoint (e.g. a cylinder, or an ellipsoid could be turned on a large lathe). I do not want to scpulture by hand too much, and it's OK for me to lose 10% in drag coefficient for the gain of the ease of manufactuting.

Is the flat bottom important? The Aptera, the most aerodynamic car that has never been, had a mostly round-ish bottom. Vehicles that set speed records for HPVs look oval-ish in cross-section, e.g. http://www.popularmechanics.com/technology/a22946/human-powered-speed-record-aerovelo/

In side wind, one the HPV starts to get lifted from the side, wouldn't the flat bottom pick up the wind and make the problem even worse?

Geoff -- assuming this can be done, is there any obvious *disadvantage* of having a rounded bottom? Would a round bottom be more, or less resistant to lift off of the front wheels at high speeds, esp when hitting a bump? Ditto about the sidewinds -- thumbs up/thumbs down? I agree, cars are wide and heavy (no lift-off), so it's the most obvious to have a flat bottom.

I will sure post my progress if I make any. E.g. about intermediate crash tests of separate small modules (don't want to crash test the whole thing). I tend to have crazy ideas about making something 'novel', spending months researching, then realising there is an obvious obstacle that it won't work, and it never gets anywhere.

RoyMacdonald wrote: My own opinion for what is worth is that air directed under the bike will be compressed

Roy, if this is the case, then a rounded bottom would save energy...

You might imagine there would be room for big tunnels between the wheels and the 'cockpit' - that's what I thought until I tried it, but, as I say above, there isn't room. Here is a front view of a practical trike to illustrate. The remaining tunnels are very shallow and could potentially create more drag rather than less, especially if the air tries to flow across them rather than along them:

In design, many things become clear when you actually try them rather than just talking about them. That's why I hate it at work when people try to make design decisions in meetings! Meetings are good for brainstorming and evaluating alternative designs, but not for actual designing.

Edited by GeoffBird - 30 May 2017 at 11:08pm