Biggest gains are at the back???

[aerohead]

Originally Posted by Ken Fry

Thanks for your comments, here, and elsewhere. I put the template over an SM image, and it is not until the very aft edge of the backlight that the angle goes to 24 degrees. At the top of the backlight frame, it is about 11 degrees. I could be off in both cases by a couple degrees. The SM actually rode a little nose-low, I think -- at least the picture I used seemed about like my old one. But clearly the backlight begins to drop away from the template at its forward edge.

I agree it would be interesting to see the whole back end tufted.

It would also be fun to reposition the self-levelling linkage, to raise the rear and lower the front. Makes me wish I still had the car. Somewhere there is probably a fully meshed SM model sitting in someone's CFD machine.

Regards, Ken

Oh, and no insult taken at all re your Nuna post. I enjoy all your posts.

In my photo,I can't tell if the hydro-pneumatic suspension is not pumped up,or if the car has been lowered,as would be likely for a dry lakes racer.The front wheel is up in the well pretty far.
The 'rake' seems like in Chaz's photo.It's hard to know if any of the dimensions are in true-length or not.
In 1970 Citroen published an ad for the GS,with Cd 0.38.This photo is in true-length and the rear slope was 25-degrees.You can see the smoke immediately detach just as the roof-line falls away and follow a path no steeper than 17-degrees at the tail of the car.If you take a french-curve and extrapolate Citroen's body it creates a tail with maximum 27-degrees slope,violating Mair's limit by 5-degrees.
This car has a length-to height ratio of 2.913.
--------------------------------------------------------------------------
The racing SM has a L/H ratio of 3.704,really good!
--------------------------------------------------------------------------
The CX that Setright tested achieved an indicated 124 mph on 82 bhp.Very impressive!
--------------------------------------------------------------------------
In 1980 Motor Trend did a small article on the GSA with Cd 0.345.The X3 version had a rear deck spoiler which cut drag to Cd 0.318.( an RX-7 of the day was Cd 0.340) With 1.275-liter engine,the GSA would return 28 mpg at 75 mph.Pretty sweet for a 4-dr sedan!
--------------------------------------------------------------------------
Perhaps a new member will surface,who owns a Citroen.It WOULD be fun to tape yarn all over it!

[aerohead]

Originally Posted by KamperBob

AND the template applies only down centerline (vertical plane of symmetry lengthwise). Away from centerline sides play a role. It's a 3D problem. If the sides taper more quickly then the top, air flows over top and down the sides to fill. If top drops more quickly than the sides slope, air flows up the sides and over top to fill. Flow will establish its equilibrium. Equations will balance. Even without boundary layers, simpler pressure/flow CFD should help here. Tufts help and are more approachable technology.

Yes,so far that's all we use it for.She's based on a complete body of revolution,but if done would be prohibitively wide,so that's a non-starter!
I'd started a plan-view drawing,then got distracted digging for other plan views to homogenize into some sort of generic form.
Morphing the sides into the roof is tricky 'cause the velocities and dynamic pressures are different.Not a problem in the original body of revolution.
So far,my gut feeling is that,after locating the position of max camber on the body side,mimic the roof curvature starting from the side zero point .
Progressing to the rear,begin to roll the sides into progressively steeper tumblehome ( as we see in the HONDA Fit B,C,and D-pillars),blending into the roof-line with slowly increasing radii.(extremely difficult to fabricate!).
I've been completely absorbed reducing the Trailer trip data since early October and have basically let everything else slide.

[jime57]

Originally Posted by aerohead

Morphing the sides into the roof is tricky 'cause the velocities and dynamic pressures are different.Not a problem in the original body of revolution.
So far,my gut feeling is that,after locating the position of max camber on the body side,mimic the roof curvature starting from the side zero point .
Progressing to the rear,begin to roll the sides into progressively steeper tumblehome ( as we see in the HONDA Fit B,C,and D-pillars),blending into the roof-line with slowly increasing radii.(extremely difficult to fabricate!).

Phil, I'm a little confused by this description, mostly because I don't know what the word "tumblehome" means. I don't find the term in Hucho, so I am at a loss, that being my only substantial reference.

Also, thinking the "side" problem through a bit, I think the sides of most street cars may already be badly compromised in terms of the ultimate aero shape. For example, the "max camber" of the sides is not a well defined point and may extend over several feet. The problems seems to manifest itself quickly when one considers a home built boat tail. The max camber point of the roof and the sides is going to be at significantly different points, keeping one well away from anything resembling a shape of rotation. 3-Wheeler resolved this problem on his boat tail by pretty much maintaining the initial rear convergence angles on his Insight. We haven't seen any tuft tests, so it is a bit difficult to know how that approach worked out. Your comments?

JimE

[KamperBob]

Originally Posted by aerohead

Yes,so far that's all we use it for.She's based on a complete body of revolution,but if done would be prohibitively wide,so that's a non-starter!
I'd started a plan-view drawing,then got distracted digging for other plan views to homogenize into some sort of generic form.

One size fits all sounds to me like a tall order.

Originally Posted by aerohead

Morphing the sides into the roof is tricky 'cause the velocities and dynamic pressures are different. Not a problem in the original body of revolution.

That's where I'm hoping good-enough CFD can help. Even without turbulent boundary perfection my hope is that pressure and flow balance as a function of roof/side taper aggression can be compared. In the case of an aerocap for a pickup, for example, the cab and box fixed geometries no doubt limit how much streamlining can be done.

Originally Posted by aerohead

So far,my gut feeling is that,after locating the position of max camber on the body side,mimic the roof curvature starting from the side zero point .
Progressing to the rear,begin to roll the sides into progressively steeper tumblehome ( as we see in the HONDA Fit B,C,and D-pillars),blending into the roof-line with slowly increasing radii.(extremely difficult to fabricate!).

That was a design goal for me when I drafted a concept cap.

Originally Posted by aerohead

I've been completely absorbed reducing the Trailer trip data since early October and have basically let everything else slide.

Focusing on your own priorities is completely understandable, Phil. It's too bad the weather was sh*t when I visited last month. For future reference next time I'd be happy to lend a hand wherever I can. Tuck the offer in your back pocket for consideration.

[aerohead]

Originally Posted by jimepting

Phil, I'm a little confused by this description, mostly because I don't know what the word "tumblehome" means. I don't find the term in Hucho, so I am at a loss, that being my only substantial reference.

Also, thinking the "side" problem through a bit, I think the sides of most street cars may already be badly compromised in terms of the ultimate aero shape. For example, the "max camber" of the sides is not a well defined point and may extend over several feet. The problems seems to manifest itself quickly when one considers a home built boat tail. The max camber point of the roof and the sides is going to be at significantly different points, keeping one well away from anything resembling a shape of rotation. 3-Wheeler resolved this problem on his boat tail by pretty much maintaining the initial rear convergence angles on his Insight. We haven't seen any tuft tests, so it is a bit difficult to know how that approach worked out. Your comments?

JimE

Hi Jim.Tumblehome is used by automakers to describe inward leaning of the cars greenhouse sides,beginning at the beltline (bottom of greenhouse).
If you stand behind today's HONDA Fit and look at the A,B,C,and D-pillars,you notice that each consecutive pillar is leaning in towards the center of the car a bit steeper.This would be the 'morphing',as the sides of the car are integrated into more of a body of revolution type of architecture.
If done correctly,the velocity of airstreams from top and sides will meet at similar velocity,preventing the formation of high-drag longitudinal vortices.
-------------------------------------------------------------------------
And yes,up until the EV-1,passenger cars were aerodynamically compromised for the sake of interior volume.AeroVironment knew they would have to pull out all the stops to meet battery range issues,so the Impact side max camber point was at the center of the door,with generous plan taper,necessitating the narrower rear wheel track.
HONDA went part way with Insight in 2000,tapering the upper 2/3rds more than all as in Impact.Impact had a virtually perfect windshield and almost completely flat belly,besting Insight in the 'tunnel.
-------------------------------------------------------------------------
When I did the VW bus,I just followed VWs lead,mimicking the sides and top,wrapping them around to a 20-degree slope with compound curve edge integration.
-------------------------------------------------------------------------
On the CRX,I did allow the sides of the boat-tail to morph from HONDA's faceted sheetmetal body side into a more organic highly radiussed integration as it rolled into the top.
-------------------------------------------------------------------------
On the DODGE pickup I dispensed with the original bed altogether which eliminates the upright end rails at the tailgate,This has allowed me to keep increasing the tumblehome such that at the tailgate there is an elliptical body cross-section now,very much like a body of revolution.The wake is reduced significantly do to this alone and there's little chance for any vorticity.I maintain 'room' for a 48" load and narrow tailgate.The spare tire has its own 'closet' at the rear,of which its floor creates the diffuser of the bellypan.Eventually,I'll ditch the T-100 bed and do likewise,allowing a more 'ideal' tail,morphing right into the boat hull.
Of course,this is easier with a truck compared to uni-body.We've got to work with what we have.
With a long straight-edge,with ends equidistant from the center of the vehicle,the side max camber point can be found easily as well as a 'look' at existing side camber.From there we can see what we can do,and it will of course be different for each vehicle.
Had Alfred P.Sloan Jr.and Harley Earl never existed,we might have rational vehicle design today,making it much easier to to reach 'ideal' flow.But things are what they are so we'll have to act accordingly.

[aerohead]

Originally Posted by KamperBob

One size fits all sounds to me like a tall order.



That's where I'm hoping good-enough CFD can help. Even without turbulent boundary perfection my hope is that pressure and flow balance as a function of roof/side taper aggression can be compared. In the case of an aerocap for a pickup, for example, the cab and box fixed geometries no doubt limit how much streamlining can be done.



That was a design goal for me when I drafted a concept cap.




Focusing on your own priorities is completely understandable, Phil. It's too bad the weather was sh*t when I visited last month. For future reference next time I'd be happy to lend a hand wherever I can. Tuck the offer in your back pocket for consideration.

Hey Bob! Yeah,with everything that's been published I think I'll just put everything up and let folks digest it on their own.With so many vehicle 'designs',we're not gonna get the 'one-size.'
And yes,let's lean on everything that's available.Especially with respect to universities with CFD that we as individuals can't afford.Last time I went to the IHPV competition at Battle Mountain,the College teams told me that since their school did not have a wind tunnel,they were relying on CFD to predict their bikes performance (by the way,they were all basically Cd 0.11,Varna included according to them).
I'm pretty confident that your cap would be rewarded at the pump.
Over time I think the pickup bed will evolve into something like I showed you on the Dodge.Those square corners are really tough on airflow potential.
I appreciate the help offer.Things happen kind o' spontaneously so it's hard to forecast project time.Hope you found dry desert for riding.And maybe next time things will be drier around here.You're a good trooper! Really enjoyed the visit!

[jime57]

Originally Posted by aerohead

-------------------------------------------------------------------------
On the CRX,I did allow the sides of the boat-tail to morph from HONDA's faceted sheetmetal body side into a more organic highly radiussed integration as it rolled into the top.
-------------------------------------------------------------------------

Is there a picture around somewhere of this project? I'd really love to see it since the shape is somewhat similar to the Insight.

[aerohead]

Originally Posted by jimepting

Is there a picture around somewhere of this project? I'd really love to see it since the shape is somewhat similar to the Insight.

Jim,I've never done a thread on the CRX.And I think that there's only one or two photos of it in the 'Archive.'
I will do one sometime.Most of the photos I took were Kodachrome slide and will have to be transferred to an SD card so Al can work with them in PhotoShop.
I can say that the boat-tail achieved the lowest drag reduction and best top speed enhancement.
With the short nose,front gap-fillers,piece-meal belly pan,rocker panel extension,diffuser,rear wheel skirts,and 12" boat-tail extension she came in at Cd 0.235 according to Glenn Sharf at GM's Aero Lab, based on highway delta-mpg .
Top speed was redline limited in 3rd gear so I didn't learn all I wanted there at Bonneville.The 'kit' did add 9-mph at 4,200' elevation compared to the CRX completely stripped.
The 1st-gen Insight is very similar to the World of Speed CRX.I think basjoos' AEROCIVIC will be the poster child for the 2025 CAFE standards.And AEROCIVIC would really scream at the Salt Flats just as it does at the gas pump!

[jime57]

Originally Posted by aerohead

Hi Jim.Tumblehome is used by automakers to describe inward leaning of the cars greenhouse sides,beginning at the beltline (bottom of greenhouse).
If you stand behind today's HONDA Fit and look at the A,B,C,and D-pillars,you notice that each consecutive pillar is leaning in towards the center of the car a bit steeper.This would be the 'morphing',as the sides of the car are integrated into more of a body of revolution type of architecture.
If done correctly,the velocity of airstreams from top and sides will meet at similar velocity,preventing the formation of high-drag longitudinal vortices.

Found a Fit today over in the Lowes lot while looking for grill block parts NOW I see what you are talking about. Stand behind the Fit and the morphing into a better transom shape is very apparent. When I got home I stood behind my Gen 1 Insight and decided that the Insight had done a much better job, but of course they didn't make nearly as many compromises. Now you are making me want to put some tell tails on the corners of the rear to see which way the flow is going.

Do you think the Insight is cursed with the longitudial vortices at the upper corners of the rear (i.e. the corners of the tail lights)?

[jime57]

Originally Posted by aerohead

I can say that the boat-tail achieved the lowest drag reduction and best top speed enhancement.
With the short nose,front gap-fillers,piece-meal belly pan,rocker panel extension,diffuser,rear wheel skirts,and 12" boat-tail extension she came in at Cd 0.235 according to Glenn Sharf at GM's Aero Lab, based on highway delta-mpg .
Top speed was redline limited in 3rd gear so I didn't learn all I wanted there at Bonneville.The 'kit' did add 9-mph at 4,200' elevation compared to the CRX completely stripped.

I've been interested for a while with this Bonneville fascination, but I must say I don't fully understand. It is true I suppose that at"level" top speeds of well over 100 MPH, ALL the power is going to overcome aero resistance. So one can make some calculations of Cd with reasonable accuracy. Is that the point?