View Single Post
Old 01-25-2013, 08:22 PM   #439 (permalink)
Tesla
Master EcoModder
 
Join Date: Feb 2012
Location: Australia
Posts: 355

The Truck - '02 Nissan GU Patrol ST
Thanks: 5
Thanked 76 Times in 50 Posts
Quote:
Originally Posted by aerohead View Post
In the near future I'll have a better answer,but for today,let me say the following:
*research on boat-tailing was conducted on bodies of revolution of which all body camber originated from a common location at the point of maximum cross-section.
*if you build a mirror-image half-body based on the body of revolution,the plan taper would also originate at the point of maximum roof camber.
*so in a 'literal' translation of a half-body car,the plan taper would be identical to the roofline taper.
--------------------------------------------------------------------------
Then there's the real world!
*All contemporary automobiles will have some side body camber.
*Some,like EV1 and Insight Gen-1 will also have measurable plan taper.
*Every car should be taken on a case-specific basis,utilizing whatever the manufacturer provided,and building upon that.
*So a 'CUBE' won't be like a 'MG EX 181' when streamlining.
--------------------------------------------------------------------------
*For roofline use of the 'Template',the full body height is the only parameter necessary,pin-pointed at the bodies location of max. camber.(It's good to ballast the car with at least 300-pounds to get it to settle first,if it'd going to do that).
*For body sides plan tapering, we first need to determine the location of max body width.
*And it could be different for the lower body and greenhouse.
*From these locations we've got to analyze what we're working with,and see how to morph the sides as best we can to mimic what the roofline would be doing.
*If the 'interference is so great as to be able to closely match top and sides (which will be the case more than not),then as Paul Jaray did in 1921,we streamline the main body separately from the greenhouse,creating the 'combination form'.[I had to do this with the T-100]
--------------------------------------------------------------------------
Technically,it would probably require a wind tunnel and hundreds of pressure taps to tune the body.
Short of that,perhaps these 'proven' separation-free bodies suggest contours which will maintain attached flow.
--------------------------------------------------------------------------
I'm working on a plan-taper thread but always seem to be months away from posting.
--------------------------------------------------------------------------
*Examples of 'Template' cars which include plan taper have already yielded Cd 0.12 or lower.
*If the plan taper is excluded,drag can be as high as Cd 0.21 (Kamm)
*Pseudo-Jaray 'fast', 'fast-back' bodies of low effective fineness ratios yielded Cd 0.20 - 0.186 (Heald- Schl'o'r)
*Bochum University's solar car has a 'fast' roofline,with Cd 0.14.
*GM's Sunraycer showed Cd 0.089 with wheel fairings in model form,Cd 0.12 without the fairings @ full-scale as raced.
*HONDA's Dream 2 solar car is Cd 0.10
*Without the PV arrays,these cars would have lower drag and easier ingress/egress.
--------------------------------------------------------------------------
These vehicles are looooooooooooooooooooooooooooooong! And it's why Kamm and Fachsenfeld advocated the tail truncation at 50% frontal area wake area to ease driving in traffic and parking issues.
Fachsenfeld also envisioned inflatable boat tails to add length for highway travel.
--------------------------------------------------------------------------
I'll keep chipping away at future threads.
Thank you so much,
Confirming some things I suspected and adding more.

Please don't dangle the carrot for a plan taper thread,
As you've probably picked up I am particularly interested in the "Plan taper" and how this is applied in a real sense, don't want to pre empt too much on the upcoming thread,

but just two more things and I'll stop bothering you.
1 / Please confirm your views on how to apply template to plan taper, as I understand it, it is applied as a symetrical wing view centred on vehicle centre line, resulting in a much shorter tail than the roof taper side view a lot of the time, there seems to be some conflicting views on this.

2 / I have read some comments regarding possible issues with crosswinds by using the full template plan taper and that less angle is better on the sides. Is this actual fact or does it relate more so to poor merging between roof and sides resulting in angles >22?

Going back to read through again.
  Reply With Quote