comment on 1st-gen Porsche 911/912 lift

[aerohead]

Originally Posted by JulianEdgar

Raining and a miserable day here, so out of interest I just went back to my aero textbooks and looked at what they had to say on the subject of lift.

Aerohead's theory is completely wrong, and even his favourite reference (Hucho, 1987) doesn't support his theory.

If anyone is interested, I can quote specifics from Katz, Barnard and Hucho. Barnard is the clearest.

I'd really prefer that you address lift in your own voice. So far,it appears that you have only a rudimentary understanding of lift issues. And I believe that you're unwilling to introspect and self-examine your 'belief' system.
My experience from the mechanical engineering curriculum exposed me to an understanding of fluid dynamics fundamentals which runs counter to some of your thesis. And I freely admit that I'm prejudiced towards, and defer to the expertise of career, PhD level aerodynamics investigators.
If you can scientifically argue against the body of empirical evidence,collected for going on a hundred years, I look forward to it.

[JulianEdgar]

Originally Posted by aerohead

I'd really prefer that you address lift in your own voice. So far,it appears that you have only a rudimentary understanding of lift issues. And I believe that you're unwilling to introspect and self-examine your 'belief' system.
My experience from the mechanical engineering curriculum exposed me to an understanding of fluid dynamics fundamentals which runs counter to some of your thesis. And I freely admit that I'm prejudiced towards, and defer to the expertise of career, PhD level aerodynamics investigators.
If you can scientifically argue against the body of empirical evidence,collected for going on a hundred years, I look forward to it.

Your arrogance is just breathtaking. Not only, apparently, am I wrong - but so are Hucho, Barnard, Katz and Skibor-Rylski.

You have developed a theory of lift that is completely wrong as it applies to any car of the last 30-odd years, but rather than admit that, you continue to defend the indefensible.

No wonder so much of the advice that you give here is wrong.

I don't care if your beliefs include conspiracy theories and incorrect understandings of how lift occurs in cars, and that you haven't kept up with any aerodynamic research of the last 25 or 30 years.

But I do care that you are misleading others. In fact, I think it is disgusting - wasting other people's time and energy through, fundamentally, your arrogant belief that you are right and everyone else is wrong.

[JulianEdgar]

Originally Posted by aerohead

*The tuft flow images cannot be trusted with respect to attached flow. Hucho explains why.

Can I have a page reference please?

In fact, as someone else has asked, can we have page references each time you invoke "Hucho says"?

After all, the last page reference you gave from Hucho (on purportedly supporting the template) did not do so.

[California98Civic]

Hi Julian,
I posted a reply to you with page numbers above in post #9 but you seem to have missed it. Had some questions for your reply, if you want to make one:
https://ecomodder.com/forum/showthre...tml#post625144

[JulianEdgar]

Yes I did miss this.

Originally Posted by California98Civic

When I read this post the first time I appreciated the references, but I got stuck on a few things. Your example from Hucho (2cd Edition, p.281) seems to be an image of three differently streamlined bodies. Yet you seem to regard only the last as streamlined. Why? And isn't the difuser angle also relevant, not just the sweep of the top of the body?

No, of course they're all streamlined. Not sure how I suggested otherwise? Aerohead has argued that streamlined bodies do not create lift. Here is an example, where the more streamlined the body (lower Cd), the higher the lift. And yes, the diffuser angle is relevant, so it's interesting how the more heavily cambered body has more lift, even with the diffuser.

Originally Posted by California98Civic

A second point you made is contradicted by the evidence you show from Katz’s Race Car Aerodynamics. Katz’s images on p.48 show increases in drag and lift between body shapes 4, 5, and 6. The “low drag body of revolution” has zero lift and the lowest drag. The half-body has the second lowest lift and drag. And the generic contemporary production shape has higher lift and drag than either 4 or 5. Yet you claim that “low drag shapes that have upper curves coming down to a small wake area have higher lift.” But body shape #6 has the higher lift.

There is no point in looking at a streamlined body of revolution (Cl = 0) and then applying that to a road car. When such a streamlined body is cut in half and placed near a road, it develops lift.

I am not saying that in every case as drag goes down, lift goes up. You can have low drag bodies with low lift, or high drag bodies with high lift - as in #6 (and that's fairly common). You can have almost any combination you want, but in a road car chasing low drag with a small wake, you need to be very careful that you don't develop a body with lots of lift due to the high degree of camber. Many of the shapes suggested on this site will have very high lift. (In fact, just refer to the thread on a wing in ground effect, a profile that was optimised to produce maximum lift. A car shaped like that, with the suggested ground clearance and angle of attack, would probably fly).

There are solutions, and the easiest is a proper undertray. But to ignore lift/downforce is to ignore literally half the aerodynamic forces acting on a car.

Originally Posted by California98Civic

Lastly, you gloss aerohead's text this way: “Aerohead's theory is that lift is caused by separation, rather than attached flow.” And you conclude: “The primary cause of lift on road cars is the attached airflow on upper curved surfaces. To argue otherwise is to show a grave misunderstanding of the mechanism of lift on road cars.” But aerohead actually argued almost the same thing as you: “Lift is a function of low pressure acting 'over' a portion of the body.All separation should be at the back of the car. If it's not,it's in a lower pressure regime,as it's at a higher velocity.The closer you get to the windshield the lower the pressure and the higher the lift.”

Yes lift is a function of the pressures acting on the body. The debate is not about that, it is about the causes of that low pressure. Aerohead has again and again argued that low pressure is caused by separation. In fact, as the references I have given show, it is caused by attached flow over the cambered surface of the car - just like an aircraft wing. The greater the curve, the lower the pressures over it. In fact, premature separation will reduce lift - the opposite of what Aerohead argues. That this is the case can be clearly seen on the tuft pictures of the Porsche, where earlier separation was produced in order to reduce lift. (But according to Aerohead, those tufts don't show real flows!)

Originally Posted by California98Civic

I'll grant that aerohead sometimes writes haikus, or else with a lot of jargon, but I think his point about lift and pressure was clear enough.

Unfortunately, I think that around half of what Aerohead writes is misleading or simply wrong. I do agree with him on some things: coastdown testing on the road is pointless, and testing of small models in wind tunnels is very likely to be misleading. But of all I have seen him write, they're about the only things I can think of that are helpful guidance (with good evidence underpinning them) for people modifying their cars.

[aerohead]

Originally Posted by JulianEdgar

Your arrogance is just breathtaking. Not only, apparently, am I wrong - but so are Hucho, Barnard, Katz and Skibor-Rylski.

You have developed a theory of lift that is completely wrong as it applies to any car of the last 30-odd years, but rather than admit that, you continue to defend the indefensible.

No wonder so much of the advice that you give here is wrong.

I don't care if your beliefs include conspiracy theories and incorrect understandings of how lift occurs in cars, and that you haven't kept up with any aerodynamic research of the last 25 or 30 years.

But I do care that you are misleading others. In fact, I think it is disgusting - wasting other people's time and energy through, fundamentally, your arrogant belief that you are right and everyone else is wrong.

* I went back through your book and believe I located the source of this controversy:
Here's an evolution of the relevant details:
- page 79- ' wind tunnel photos... smoke trails show flow separation from the rear end of the roof... explains the low recorded pressures across the rear.'
( this is in perfect agreement with Hucho's comments on the relationship between separation,pressure,lift,and drag )
- page 80,Figure 4-13 '... The base pressure of the newer car is twice that of the older cars!' ( in perfect agreement of Hucho's relationship between drag, pressure drag, an separation ).
- page 80, Figure 4-14 ' ... flow separation at the end of the roof of the older car is responsible for the low measured pressures on the rear on the rear of the car.' ( when I comment that separation is responsible for the low pressure, then this statement somehow becomes invalid )
- page 80, ' higher pressures on the rear of the ( 2003 M-B E 500 ) help explain the lower Cd.' ( when I imply the same conclusion, my reasoning is flawed and am attacked for it ).
- page 85, ' [T]he very gentle radius at the top of the windscreen... result(s) in a higher than expected pressure...' ( although an even more gentle radius of the 'template' somehow produces a higher suction peak ).
- page 85, '... implying this area was in a separation bubble.' ( which would not exist with the 'template' contour,as it is incapable of producing flow separation,by definition,implying that the 'template' would lack the energy loss of the E 500's locked-vortex,greater efficiency,and lower drag ).
- page 85, ' ... on the boot lid - a positive pressure..., the base pressure in the wake is much higher...' ( with zero separation, and full pressure recovery ( minus surface friction losses, somehow the 'template', by the very same metric, would not represent an even higher pressure, absolute highest base pressure, and lowest possible pressure drag ).
- page 86, CFD image of the Jaguar XE: the yellow/orange region of the C-Pillar area registers the existence of attached,longitudinal, counter-rotating vortices, agreed upon on page 91: '... with the two rotating vortices that are coming off the C-pillars... likely to leave this area of the car in separated flow' ( somehow this is lost on the Porsche 911 Carrera, with it's even steeper roof contour and no body structure on which to re-attach to [ and remarkably lost on the 'template' which is incapable of generating C-Pillar vorticity,by definition ).
--------------------------------------------------------------------------------------
- Finally, on page 173 we find the epicenter of the lift argument :
( The more pronounced rear lift in the case of the Porsche reflects the airflow generating low pressures over the rear half of the car as the air wraps over the long curve.' )
All aerodynamicists would agree with this comment, however not with the implied source of the low pressure.
------------------------------------------------------------------------------------
* The 911 body is 43.8% ft-body.
* The 911 roof contour descends 'early' and too 'steeply' to support attached flow.
* Flow separation on the 911 occurs ahead of the backlight header.
* The rapid pressure rise induced by the steep contour precipitates strong,attached longitudinal vortices, so strong ,that they create the intense fastback downwash, responsible for holding the tufts along the body, giving the observer the illusion of attached flow.
* Should the 911 had had attached flow, it's lift would be at around 79% of the forebody, just ahead of the windshield header, where the 'minimum pressure' resides.'
* Should the 911 body have had fully attached flow, by the time the aft-body flow reached the rear bumper, all the velocity of accelerated flow would have been replaced by pressure recovery,as per the Bernoulli Theorem, leaving only surface friction losses, a fraction of drag compared to pressure drag.
* The only solution to the 911 situation is, to loft the entire body contour,up to the top of the separated flow, which lies within a 22-23-degree slope, off the rear of the car. ( see Hucho, 2nd Edition, pages 281,169,175,141,142,144,124,114,153,154,155,61,15, 16,104,209.
* Solving the 911 issue would violate Ferdinand 'Butzi' Porsche's 'silhouette', and so first the 'duckbill',then the 'whale-tail' and 'tea-tray' spoilers were added as palliatives, protecting the major silhouette, and thereby lessening risk to sales.
-------------------------------------------------------------------------------------
The spoilers do not 'spoil' lift-producing airfoil-shaped bodies.Spoilers raise the aft-body upwards out of the turbulence they're already submerged in,which was caused by flow separation of an improper contour in the first place,attempting to reach up into the inviscid flow above, where kinetic energy can attack the structure and facilitate re-attachment, capture a locked-vortex in the case of an non-aspirated spoiler,killing this lift- do -to -separation.
-------------------------------------------------------------------------------------
As to the hated 'template,' as a 'streamline body' without any flow separation, positive pressure fore and aft, cancels any lift at the location of minimum pressure, delivering 'neutral' lift,as Spirit of EcoModder registered at DARKO. Just as predicted in the 1930s. Static wheel loading via gravity is all that's necessary for stability at posted speed limits ( in the United States of America,anyway).

[JulianEdgar]

It's very hard to argue when everything I write is twisted to suit your particular model of reality.

I've said it all before but I'll try again.

1. Yes, premature separation causes low pressures in the separated area. That's why I can write: "flow separation at the end of the roof of the older car is responsible for the low measured pressures on the rear of the car." It happened a lot on older cars.


but....

2. Flow separation seldom occurs on cars of the last 30 odd years. Therefore, the low pressures do not come from separation. They come from, as I said in the book, the airflow "airflow generating low pressures over the rear half of the car as the air wraps over the long curve*".

You then write "All aerodynamicists would agree with this comment, however not with the implied source of the low pressure" completely ignoring that in fact none of the four professional aerodynamicists who read the draft raised any issue with that statement.

Basically, like with many of your statements with regard to car aero, they reflect:

1) You haven't kept up with the literature...


2) ...so you apply models and understandings that were relevant 30-40 years ago but not to most cars now.

If you were to get out on the road and measure some pressures on recent cars, you would not have to rely on your theories about what you think is happening, and instead could actually see what is really happening.

(*I could have said instead "the lift comes from the camber of the body", but I try to write as simply as possible.)

[aerohead]

Originally Posted by JulianEdgar

It's very hard to argue when everything I write is twisted to suit your particular model of reality.

I've said it all before but I'll try again.

1. Yes, premature separation causes low pressures in the separated area. That's why I can write: "flow separation at the end of the roof of the older car is responsible for the low measured pressures on the rear of the car." It happened a lot on older cars.


but....

2. Flow separation seldom occurs on cars of the last 30 odd years. Therefore, the low pressures do not come from separation. They come from, as I said in the book, the airflow "airflow generating low pressures over the rear half of the car as the air wraps over the long curve*".

You then write "All aerodynamicists would agree with this comment, however not with the implied source of the low pressure" completely ignoring that in fact none of the four professional aerodynamicists who read the draft raised any issue with that statement.

Basically, like with many of your statements with regard to car aero, they reflect:

1) You haven't kept up with the literature...


2) ...so you apply models and understandings that were relevant 30-40 years ago but not to most cars now.

If you were to get out on the road and measure some pressures on recent cars, you would not have to rely on your theories about what you think is happening, and instead could actually see what is really happening.

(*I could have said instead "the lift comes from the camber of the body", but I try to write as simply as possible.)

1) Flow separation occurs on all cars, except streamlined cars. It's the reason they don't have as low as Cd 0.12. Hucho clearly explains this. And the rules haven't changed in the last 20-30-years.
2) The low base pressure of modern cars is a function of the local pressure at the separation line. It has absolutely nothing to do with 'airflow wrapping over the body.' That doesn't even happen on wings.
3) The only way your four professional aerodynamicists could be factually correct, with respect to low pressures over the aft-body is, that the intense,fastback, long-lived, attached, counter-rotating longitudinal vortices are inducing a downwash, which impells the 911's tufts to 'adhere' to the rear slope, which in no way demonstrates 'attached flow'. The low pressure originates from the low local pressure at the separation line at the base of the backlight, where the body contour exceeds 23-degrees off the horizon, which would show up in smoke flow visualization. Your premise is incorrect.

3) Until you study and master the Bernoulli Theorem you're never going to understand the ramification of flow separation as it relates to pressure at the separation point(s), how that impacts base pressure, pressure drag,and overall drag.
4) Current literature has not overturned fluid dynamic principles.
5) The models and understandings I present are founded in empirical research leading back to the 18th-century. Ram-tubes, static tubes, and manometers aren't a recent technology. Nor are moving-floor wind tunnels.
6) As to measuring pressures, the protocol would include 42-pressure taps and manometers, photo-recorded, simultaneously, in a laboratory, where atmospheric conditions could be closely monitored, and all results corrected for standard SAE atmosphere. If you're in an income bracket which can absorb the cost of 42-Dwyer Magnehelics, good for you!
7) Looking at your measurements for the 1979, Mercedes-Benz 230 sedan, and averaging the six pressure points listed, it looks like the top of the car produces 0.00625 atmospheres difference at 100-miles per hour. And it's quite odd that Jaguar placed the top of the XE's spoiler right on the 'template' line.
8) In 1969, one of the car magazines found that they could eliminate any front or rear lift with a couple dollars worth of sheet metal or plastic. Hucho later basically claimed the issue wasn't worth fretting over, unless people drove like you do. Which is completely illegal in the US, and considered 'homicide' in some states.
9) As to the 'template,' the streamline body of revolution (torpedo) from which it is derived, produces zero-lift. (Goro Tamai,MIT, The Leading Edge).
10) The streamline half-body from which the 'template' is further derived, 'could' generate lift, however this depends upon its inclination (angle -of-attack ) which,so far, you haven't researched.[ At 135-miles per hour, the Spirit of Ecomoder.com developed 30-pounds front downforce, and 22-pounds rear lift ( with a compromised belly and diffuser), on a 4,280-pound vehicle. Using the velocity-square law, you can see that driving even at 85-mph (the highest legal speed in the United States) would produce 'meaningless' 'lift.' Essentially, the 'template' produces a zero-lift body (measured in a laboratory under controlled conditions).
11) It is the roof 'camber' which helps decelerate and increase the pressure of the flow. The further the flow is from the suction peak ahead of the windscreen header, the higher the local pressure, and higher the base pressure once the flow does separate. On the 'template'(s) body side camber and boat-tailing, this is taken to the extreme, providing the entire deceleration ramp (twice the length of the acceleration ramp), which experiences zero separation, full pressure recovery (in light of the surface friction losses), producing the lowest possible drag.
None of this will mean anything to you until you get Bernoulli behind you. Ask your 'experts.'

[JulianEdgar]

Originally Posted by aerohead

1) Flow separation occurs on all cars, except streamlined cars. It's the reason they don't have as low as Cd 0.12. Hucho clearly explains this. And the rules haven't changed in the last 20-30-years.
2) The low base pressure of modern cars is a function of the local pressure at the separation line. It has absolutely nothing to do with 'airflow wrapping over the body.' That doesn't even happen on wings.
3) The only way your four professional aerodynamicists could be factually correct, with respect to low pressures over the aft-body is, that the intense,fastback, long-lived, attached, counter-rotating longitudinal vortices are inducing a downwash, which impells the 911's tufts to 'adhere' to the rear slope, which in no way demonstrates 'attached flow'. The low pressure originates from the low local pressure at the separation line at the base of the backlight, where the body contour exceeds 23-degrees off the horizon, which would show up in smoke flow visualization. Your premise is incorrect.

3) Until you study and master the Bernoulli Theorem you're never going to understand the ramification of flow separation as it relates to pressure at the separation point(s), how that impacts base pressure, pressure drag,and overall drag.
4) Current literature has not overturned fluid dynamic principles.
5) The models and understandings I present are founded in empirical research leading back to the 18th-century. Ram-tubes, static tubes, and manometers aren't a recent technology. Nor are moving-floor wind tunnels.
6) As to measuring pressures, the protocol would include 42-pressure taps and manometers, photo-recorded, simultaneously, in a laboratory, where atmospheric conditions could be closely monitored, and all results corrected for standard SAE atmosphere. If you're in an income bracket which can absorb the cost of 42-Dwyer Magnehelics, good for you!
7) Looking at your measurements for the 1979, Mercedes-Benz 230 sedan, and averaging the six pressure points listed, it looks like the top of the car produces 0.00625 atmospheres difference at 100-miles per hour. And it's quite odd that Jaguar placed the top of the XE's spoiler right on the 'template' line.
8) In 1969, one of the car magazines found that they could eliminate any front or rear lift with a couple dollars worth of sheet metal or plastic. Hucho later basically claimed the issue wasn't worth fretting over, unless people drove like you do. Which is completely illegal in the US, and considered 'homicide' in some states.
9) As to the 'template,' the streamline body of revolution (torpedo) from which it is derived, produces zero-lift. (Goro Tamai,MIT, The Leading Edge).
10) The streamline half-body from which the 'template' is further derived, 'could' generate lift, however this depends upon its inclination (angle -of-attack ) which,so far, you haven't researched.[ At 135-miles per hour, the Spirit of Ecomoder.com developed 30-pounds front downforce, and 22-pounds rear lift ( with a compromised belly and diffuser), on a 4,280-pound vehicle. Using the velocity-square law, you can see that driving even at 85-mph (the highest legal speed in the United States) would produce 'meaningless' 'lift.' Essentially, the 'template' produces a zero-lift body (measured in a laboratory under controlled conditions).
11) It is the roof 'camber' which helps decelerate and increase the pressure of the flow. The further the flow is from the suction peak ahead of the windscreen header, the higher the local pressure, and higher the base pressure once the flow does separate. On the 'template'(s) body side camber and boat-tailing, this is taken to the extreme, providing the entire deceleration ramp (twice the length of the acceleration ramp), which experiences zero separation, full pressure recovery (in light of the surface friction losses), producing the lowest possible drag.
None of this will mean anything to you until you get Bernoulli behind you. Ask your 'experts.'

It's an odd world you live in when tufts showing attached flow aren't really showing attached flow, and where measured on-road aero pressures can't really be measured pressures! Much better to sit at a keyboard and make stuff up about what, according to you, should be happening.

I think I'll go on what I can measure, and on what some of the top aerodynamacists in the world tell me. Both have vastly more credibility than 'Aerohead' on a discussion forum.

[freebeard]

It's very hard to argue when everything I write is twisted to suit your particular model of reality.

I've said it all before but I'll try again.

This is a general problem in society in general. Scott Adams describes it as two movies on one screen. Trying again will be futile.

Dueling walls o' text. My eyes glazed over. Do you guys use speech-to-text?

The only reason I got to the bottom because of the first line:

When this topic came up it triggered something we talked about back in 5/16/2009, when Ernie Rogers shared his thread,'Describing my Beetle wing ( VW New Beetle)',here at EcoModder.

Good practice would be to provide a link: https://ecomodder.com/forum/showthre...etle-8412.html

Beetle drag reducer picks up 5 to 8 %

I'll just point out another Ecomodder thread: Compact camper alternative: New Beetle TDI with hard mounted pop up tent Permalink #15

Beetle w/o any attachments/slick top= ave 48mpg :{)
Beetle with factory VW Roof racks and CVT tent= 38mpg :/
Beetle with new areo-hitch mounted rack/CVT tent= 49.3MPG WhoooRah!
That is a gain of 11.3 mpg boys and girls!

Unfortunately, the pix were hosted on tinypic.