Thoughts on gain from boat tail on 94 eclipse

[kach22i]

Originally Posted by Xist

From: http://www.nasa.gov/centers/dryden/p...ain_163113.pdf

Nice post.

Originally Posted by aerohead

*My boat tail was also 20-degrees in 1980,with a length equal to 75% of body height,bending gently into the slope......

If we take a car of approximately 4.5 feet in height, the boat-tail would be 3 Ft 4-1/2" long, right?

The 75% of body height, that's a scaling factor, right?

In effect, we can scale down the template 70-75% (range listed) and get 97% (31/32x100) of it's benefits, right?

Why don't we just scale down the template then?

Is there a chart showing the losses when scaling down?

Say for instance:

X.100/100% (1:1)

X.75/97%

X.50/39%? (just a guess)

It's that last ratio I'm interested in most. (see edited comments far below)

If I had at least three reference points I could start to chart this scaling relationship of L to H and percent of boattail affect.

EDIT-1:
Automobile 2 - Odds And Ends Photos by kach22i | Photobucket




EDIT COMMENTS:
Scaling down the template for "boattailing purposes" by a factor of 75% to get it to 1/4th of it's original size results in getting 31 mpg in lieu of 32 mpg on a Dryden/NASA style van, right?

[ennored]

I'm not on the scaling bus. Just want to state that.

But, there is something that I don't quite have a grasp on. I posted in another thread, but it was a bit of tangent there. What I don't think I have ever seen is a "template" that describes what the minimum rear "tail" needs to look like to achieve attached flow. This shape is NOT ideal relative to drag, but should have much less drag than say a flat rear like a semi trailer, or even a hemisphere. We see this over and over, less than ideal shapes making good aero numbers. From the Dryden van, to Trailer Tails, to Joe Schmoe's kammback, to the latest hybrid car, etc. We hear a lot about attached flow and tuft testing here as being a reasonable metric. Just what shape do we need to get that? (And just how good is it?)

The Dryden van is a good example. It has a boattail. Does it match the template? No. We can conclude from that it could have been better. OK, got that. But there were tuft tests done that showed it achieved what looked like attached flow. And was A LOT better than no tail, or rounded rear edges.

(And on the 25% - 75% "fits" above. Doing the same for the plan view would find a closer fit. Everyone seems so stuck on the side view....)

Tim

[kach22i]

Originally Posted by ennored

What I don't think I have ever seen is a "template" that describes what the minimum rear "tail" needs to look like to achieve attached flow.

Tim, this is a great idea.

I've promoted the idea of one template for econoboxes and SUV's, and another for notchback coupes.

There is too much focus on the side view, but as most of the air is going up and over the top of the car/truck, perhaps this is warranted.

[Xist]

Originally Posted by kach22i

Tim, this is a great idea.

I've promoted the idea of one template for econoboxes and SUV's, and another for notchback coupes.

There is too much focus on the side view, but as most of the air is going up and over the top of the car/truck, perhaps this is warranted.

Aerohead!

How, exactly, would I go about making an Aeroheadsignal?

[aerohead]

Originally Posted by kach22i

Nice post.



If we take a car of approximately 4.5 feet in height, the boat-tail would be 3 Ft 4-1/2" long, right?

The 75% of body height, that's a scaling factor, right?

In effect, we can scale down the template 70-75% (range listed) and get 97% (31/32x100) of it's benefits, right?

Why don't we just scale down the template then?

Is there a chart showing the losses when scaling down?

Say for instance:

X.100/100% (1:1)

X.75/97%

X.50/39%? (just a guess)

It's that last ratio I'm interested in most. (see edited comments far below)

If I had at least three reference points I could start to chart this scaling relationship of L to H and percent of boattail affect.

EDIT-1:
Automobile 2 - Odds And Ends Photos by kach22i | Photobucket




EDIT COMMENTS:
Scaling down the template for "boattailing purposes" by a factor of 75% to get it to 1/4th of it's original size results in getting 31 mpg in lieu of 32 mpg on a Dryden/NASA style van, right?

If we go back to the fundamental premises which form the methodology of the 'Template' construction we run into scaling snags:
*We're going after pressure drag which is a symptom of flow separation.
*The contour of the 'Template' is based on the mirror-image,ground reflection of the Cd 0.04,2.5:1 fineness ratio streamline body of revolution which is separation-free.
*This body is separation-free only because of its exact contour,based on its distance above the ground plane.
*Any 'steepening' of the contour (which would come about from scaling) triggers flow separation immediately,due to the adverse pressure gradient created by the rapid pressure rise generated by the steeper slope.
*Once separated,the flow will never re-attach and all the kinetic energy which might have energized the wake is lost to viscous attrition.
-------------------------------------------------------------------------
If you're going to do a 'composite'(combination form) body,with body,wheel area,and greenhouse agglomerated together you'll need to rely on Hoerner's work.It is a 'BIBLE' of aerodynamics from which Hucho leans heavily.
There are 'scaling' relationships there,and it runs in the opposite direction from what you might think.
-------------------------------------------------------------------------
For 'typical' passenger cars and pickup trucks the 'Template' should be used only as illustrated.You can go 'longer' as THIS will not introduce flow separation,but bear in mind that the structure becomes weaker as you move away from a spherical/ovoid form,requiring internal reinforcement to counteract aeroelastic effects.
-------------------------------------------------------------------------
For buses,motor homes,tractor-trailers you might consider wing section/struts.
If you intend to do a limited-length tail,with no intention of future streamlining,then consider the GM Optimum boat tail.It will have hard angles of :
19-degrees @ top
10-degrees @ sides
10-degrees @ bottom
panel intersections chamferred into a radius which grows with length.
-------------------------------------------------------------------------
Fachsenfelds 1936 Onmibus passenger bus with inflated boat tail followed the 'Template' fairly close and came in at around Cd 0.257 if memory serves me.
It's the best I've ever seen.

[Xist]

Aerohead, I am doomed!

[aerohead]

Originally Posted by Xist

Aerohead, I am doomed!

No you're not!

[ennored]

Thanks Phil. I have already started to consider my RV a vertical wing with respect to the tail. That is, it's 100" wide, and about 130" tall. The top 50" can get a tail designed for a 50" height, and the lower 80" gets it's tail solely based on the sides.

I'm still a bit lost with less than optimal tails though. How come the Dryden van tail looks to have attached flow? It was 80" long (and truncated to 40") on an 80" wide vehicle. The optimum tail would have been 142" long based on half the width (and much longer in side view). Does it really only have attached flow for part of it's length? Did the tufts lie? We KNOW it's better than a flat rear end. And we KNOW it's less than optimum. Is it just the filling of the turbulent wake that is improving the drag? Makes me think about the straight edge tail, like the trailer tail, or the flat 10° stuff you're talking about. Would they would better with a rounded shape?

Really makes me want to build something. Been investigating removing my rear cap. a 36" tail would fit in my garage. Another 36" or so could be folded out a la trailer tail. I only need 50" to replicate the half tail from the Dryden van. Suggests I could get as good as it or better than it!

[kach22i]

Originally Posted by ennored

I'm still a bit lost with less than optimal tails though. How come the Dryden van tail looks to have attached flow?

I don't want to speak for Aerohead, but as I understand it the Dryden van with all treatments and modification combined achieved a 15% improvement over that of the stock Econoline van. There is a PDF paper or two which might define which part of that is the boat-tail alone.

This 15% improvement I assume is a far cry from an optimum 3D teardrop shape.

The flow attachment is one thing, but once it lets go of the body, there could be aerodynamic retribution (back lash) over that of a more gentle slope preceding termination.

[ennored]

The Dryden van went from a best Cd of about .35 without a boattail to .24 with. Over a 30% improvement for the tail. I accept that with a longer tail it could have been better. But the tuft testing they did looked pretty good. And a 30% improvement is pretty good for a non-attached flow improper tail, right?