Lower Cd, but larger wake

[aerohead]

The wake pressure is determined by the pressure at the point of separation.
The 'wagon' style roof is undergoing slight convergence both in elevation and plan as the flow moves rearward with the air slowing and building pressure continuously until it breaks away finally at the back.The wake formed is at a higher static pressure,closer to the forward stagnation pressure so smaller delta-P,and lower drag.
----------------------------------------------------------------------------
The sedans experience separation at the backlight header and C-pillars which are 'upstream' of the wagon style roof and at a lower static pressure,so the wake,necessarily must be also at a lower pressure,with a higher delta-P compared to the nose,and higher drag.
----------------------------------------------------------------------------
The sedans can also experience attached vortices which typically do not form on wagon style bodies.The vorticity robs a remarkable amount of power from the engine/gas tank.

[jakobnev]

I'm thinking there are two kinds of wake at the extremes: One where a bubble is formed behind the car that has the same air in it all the time, the other where new air that has been accelerated flows into the wake constantly. (In reality you would always be somewhere between these extremes)

The first kind would of course be better since air would flow around the car and then around this bubble and be left behind not much disturbed and the second kind would leave behind a trail of fast moving-energy sucking air.

This effect could be measured, for example, in a wind tunnel by turning a smoke generator on and off and observing how long the area behind the car stays smoky.

[max_frontal_area]

the arguments made here make sense... but my 2 vehicles dont.
1980 rabbit and jetta identical except for trunk on jetta and slightly "cleaner"
front. both have NA diesels, and manuals with very similar final drive.
the rabbit does 60 mpg, the jetta does 70. what gives???

[aerohead]

Originally Posted by jakobnev

I'm thinking there are two kinds of wake at the extremes: One where a bubble is formed behind the car that has the same air in it all the time, the other where new air that has been accelerated flows into the wake constantly. (In reality you would always be somewhere between these extremes)

The first kind would of course be better since air would flow around the car and then around this bubble and be left behind not much disturbed and the second kind would leave behind a trail of fast moving-energy sucking air.

This effect could be measured, for example, in a wind tunnel by turning a smoke generator on and off and observing how long the area behind the car stays smoky.

I believe that all separated wake flow is described as random eddy turbulence of which its kinetic energy cannot be converted to useful static pressure and is therefore lost to friction heating of viscous attrition.Typically,there is very violent mixing into and out of this region.
When you describe a 'bubble' I believe you are referring to a special case which can occur at very small aft-body truncations in which the wake behaves like an attached vortex in 3-dimensions,embodying the form of a 'phantom' boat-tail of which the surrounding flow field will pass over as if it was a solid boundary.
This can only occur in about the last 20 % of the tail.

[aerohead]

Originally Posted by max_frontal_area

the arguments made here make sense... but my 2 vehicles dont.
1980 rabbit and jetta identical except for trunk on jetta and slightly "cleaner"
front. both have NA diesels, and manuals with very similar final drive.
the rabbit does 60 mpg, the jetta does 70. what gives???

max,would you happen to have photos of the two cars taken from the side?
The difference in mpg if attributed strictly to a difference in Cd would require a difference of over 33 %.

[jakobnev]

aerohead: What's your take on rounded rear ends(think beetle and TT) being worse for Cd that straight and sharp edges? (basically any modern hatch exept the new scirocco (which has a Cd of 0.34)

To me it certainly looks like the round edges direct the air in behind the car and the straight sharp edges let the air flow past the volume or air that moves with the car.

[SvdM]

I also remember that the Mk2 and Mk3 Jettas having a lower Cd than the Golfs, but could this then be entirely due to the difference in the grille design between the two?

[euromodder]

Originally Posted by jakobnev

To me it certainly looks like the round edges direct the air in behind the car and the straight sharp edges let the air flow past the volume or air that moves with the car.

Too much rounding and you get the Coanda-effect pulling back on the car.
Drag.

[aerohead]

Originally Posted by jakobnev

aerohead: What's your take on rounded rear ends(think beetle and TT) being worse for Cd that straight and sharp edges? (basically any modern hatch exept the new scirocco (which has a Cd of 0.34)

To me it certainly looks like the round edges direct the air in behind the car and the straight sharp edges let the air flow past the volume or air that moves with the car.

jakobnev,your's is a good question and also one of the most difficult to answer,so I'm going to let Dr.Hucho explain it as he did in his SAE Paper from 1976 which involved his work at Volkswagen AG.And the caveat is that you should not make generalities about this kind of complex flow.
-------------------------------------------------------------------------- The short answer is that fastbacks and square-backs do well with rounding,notchbacks do not.
--------------------------------------------------------------------------
A good fastback will have the top and side flow re-joining at similar angles,velocities,and pressures.No longitudinal vortices will form,the flow will separate at the slowest velocity,highest pressure,into the smallest wake of greatest base pressure and lowest drag.Curvature of the C-Pillar buttress and body sides was demonstrated to be beneficial. The separation point CAN move around a bit,and from Kamm's/Fachsenfeld's research,it's best just to make a sharp chop.
--------------------------------------------------------------------------
A square-back typically has top and side flow of similar velocity and pressure and as these flows separate at the back of the body they form a wake of homogenous turbulence,free of longitudinal vorticity,there is no lift.
Curvature along the roof is good.It's okay at the back,although it's ther same situation with separation point,and a chop is preferred

--------------------------------------------------------------------------
The notch-back coupe or sedan,if the C-pillar is rounded improperly,can direct the side flow into the highly cambered top flow at the backlite centerline which will create counter-rotating longitudinal vortices of high drag.
-------------------------------------------------------------------------- The original basic Jetta form had over Cd 0.40.
Pushing the notcback up to create a wagon type roof drop the drag to Cd 0.38.
Raising the basic form trunklid 150mm got to Cd 0.37.
Changing to a 10-degree fastback got the same car to Cd 0.34.
At 28-degrees fastback angle,the flow changed back to squareback flow with full wake.

[aerohead]

Originally Posted by max_frontal_area

the arguments made here make sense... but my 2 vehicles dont.
1980 rabbit and jetta identical except for trunk on jetta and slightly "cleaner"
front. both have NA diesels, and manuals with very similar final drive.
the rabbit does 60 mpg, the jetta does 70. what gives???

max,I found some data at home for the VWs.
The Jetta is created by ADDING length to the Golf/Rabbit,about a foot.Frontal areas and Cds are virtually identical for the two cars with each generation.
Popular Science Magazine did some comparison tests in Jan/Feb 1983.At a steady 35 mph both cars got within 0.5 mpg,at 55 mph within 2.4 mpg.
The Jetta was turbo'd,the Rabbit NA.