Fossati's Aero-Hydrodyamics

Just a quick reply before I go sailing . I will respond fully tomorrow to your significant contribution:"The manometer tube reading will be a direct (true) pressure measurement and can be laid out as the normal vector to the foil profile without resolving anything."

It's interesting that even the youTuber academic (Richard Stadelman) adds to the confusion by, at 3:28, mis-labelling the region below the fluid levels as the negative region!

 

For pressure transducers which do not move then mounting orientation does not matter. Any offset due to gravity will be very small and is compensated for by zeroing the transducers before making measurements. For pressure transducers which are subject to high accelerations then other measures may be needed.

 

While it is quite possible for a wind tunnel to be open with no walls,I have yet to see such a beast used for more than a demonstration of the principles of using a tunnel.All of the tunnels I have seen collecting useful data had a closed test section so that a uniform speed affected the entirety of the region.We could get bogged down with a diversion about blockage ratios,but that is getting a bit abstruse for this thread.I will point out that the more useful tunnels for vehicle work have a rolling road that is synchronised to the wind speed and this takes quite a bit of engineering. If you are looking for a good primer on measurement techniques,this video covers it well without being too long:

 

A few open test section (aka open jet) wind tunnels which are used for serious work:

Technical University Munich Wind Tunnel A and Wind Tunnel B Wind Tunnels https://www.epc.ed.tum.de/en/aer/wind-tunnels/ A Numerical and Experimental Evaluation of Open Jet Wind Tunnel Interferences using the DrivAer Reference Model https://www.sae.org/publications/technical-papers/content/2016-01-1597/
DNW Large Low Speed Facility Large Low-Speed Facility » DNW https://www.dnw.aero/wind-tunnels/llf/
BMW Group Wind Tunnel BMW Group Wind tunnel - Bob Aerodynamic Tests of the Bob- und Schlittenverband für Deutschland (BSD) https://www.press.bmwgroup.com/global/photo/detail/P90282230/bmw-group-wind-tunnel-bob-aerodynamic-tests-of-the-bob-und-schlittenverband-f%20%20r-deutschland-bsd
Pininfarina Wind Tunnel BMW Group Wind tunnel - Bob Aerodynamic Tests of the Bob- und Schlittenverband für Deutschland (BSD) https://www.press.bmwgroup.com/global/photo/detail/P90282230/bmw-group-wind-tunnel-bob-aerodynamic-tests-of-the-bob-und-schlittenverband-f%20%20r-deutschland-bsd
NASA Full Size Wind Tunnel Full-Scale Wind Tunnel - Wikipedia https://en.wikipedia.org/wiki/Full-Scale_Wind_Tunnel#:~:text=The%20Full%2DScale%20Tunnel%20(abbreviated,was%20a%20National%20Historic%20Landmark
Mississipi State University Wind Tunnel​

For thos interested the blockage effects with an open test section are oppposite of those with a closed test section. Blockage effects are worst at transonic speeds and slotted walls or similar are frequently used to minimize blockage effects.
There has been recent publicatons on blockage effects and correction with open section wind tunnels.

https://elib.dlr.de/66914/1/A_CFD_P...ections_Ciobaca_KatNetII_Bremen_22_120509.pdf
https://arc.aiaa.org/doi/10.2514/1.15656
Analysis of open test section flow in a closed return wind tunnel https://scholarsjunction.msstate.edu/td/651/​

 

Thanks both @wet feet and @DCockey for your replies. The 'Air Shaper' YouTube channel looks like a great resource!

@Sailor Al - The demonstration here uses pressure transducers instead of manometers on a foil with test taps:
Pressure Profile Over a NACA 0012 Symmetric Airfoil
The explanation and presentation is incredibly clear and thorough, and well worth watching. And @wetfeet, he mentions using Scanivalves - 2'nd time I've come across that term!

 

By p. 316 of Appendix 1, experimental evidence has been presented for the existence of over- and under- pressures on the foil in an airstream. Using data digitised from the images (attached in my second post), the geometry of the airfoil retrieved from Airfoil Tools and some python code, the forces generated by the pressures can be represented in the following vector diagram :

which, when rotated fairly faithfully corresponds with image 2.26 from p. 31

(I included the this part about resolving the pressure into force to test the validity of my code. I attach a high-level overview of my method.)
Using much of the same logic, the pressure values can be presented with over-pressure values indicated in red and under-pressure in blue.


In the middle of the same page 316, we read the following:
"These pressure variations are attributable to a variation in dynamic pressure induced by a flow speed variation it is in fact evident that where the section where the flow passes narrows because of the conservation of fluid flow there must be an acceleration of the flow which must correspond with a reduction in pressure"
This seems to say that variations in pressure within the air flow are caused by changes in flow speed caused by the air passing through a narrower section.
The above illustration of the pressure data from the text, it can be seen that the low pressure (in blue) occurs around the first four points from the LE on the upper surface, where it could be argued that the presence of the foil somehow narrows the section, there is also low pressure over the rear section of the upper surface where, using the same argument, it could be said that the section widens.
Also along the lower section, the airstream could be said to be narrowing but that's associated with an increase in pressure, not a reduction.

So this is another of the challenges I have encountered in Fossati's Appendix 1. I don't see how the explanation is supported by the experimental data presented in the text.

 

I think I get your process from the PDF.
Can you confirm that what looks at first glance like incorrectly ordered vectors in the black and white diagrams actually represent incremental force from the local pressure reading x the interval length between taps? Basically, it's a crude numerical integration across both the upper and lower surface. Total force vector is a graphic addition of these individual vectors.

This is where you need to look at velocity or pressure distributions across the entire airfoil, and get a better visualization of the flow field around the test section. Fossati doesn't do this.
Go here and read: Velocity Distributions https://www.mh-aerotools.de/airfoils/velocitydistributions.htm
Go here and watch:


After that, your results will make much more sense.

If you are still interested after that, install a copy of the open-source airfoil analysis program XFLR5 and Martin Hepperle's JavaFoil applet and mess about with them. I've attached a couple of screenshots. I'll leave it to you find the software!

 

That's correct, it's a standard representation of the effect of adding vectors.

But what about my question? How does Fossati's explanation account for a) the low pressure along the rear of the upper surface (which contributes a significant amount of the aerodynamic force FA), and b) the high pressure along the forward part of the lower surface?
Remember this thread is about Fossati's modern, highly regarded, technical textbook, published in association with a recognised academic institution.

 

I get it. And in his defense, the book covers the the entire sailboat in the mixed environments of air and water. An explanation of fluid flow around an object in an appendix is bound to be summary.

Perhaps it would make more sense to think about it in terms of the velocity gradient, which is the inverse of the pressure gradient. The velocity of the flow at streamlines near the wing surface is higher than the freestream velocity above the wing (just as the local pressure all along the upper wing surface is lower than ambient pressure). Sure, the local velocity is highest near the leading edge, but the local velocity is still higher than the freestream velocity near the trailing edge.

Analogously for the bottom surface of the airfoil: The velocity of the flow at streamlines near the airfoil surface is lower than the freestream velocity below the wing (just as the local pressure all along the lower wing surface is higher than ambient pressure). This all assumes a symmetric foil at a positive angle of attack. Same principle for a cambered foil, except the cambered foil can generate lift at 0 angle of attack.

Here is a screenshot from JavaFoil for a generic cambered section (I don't recall which) set at 10 degrees AOA, with a pressure field map and streamlines turned on. Net effect is lower than freestream pressure across almost the entire top surface and higher than freestream pressure along most of the bottom surface. Hard to tell from this pic, but there is downwash at the trailing edge as the mass of air following the wings surface gets directed downward. So it's both Bernoulli (or Euler) and Newton. Different ways to look at it.

 

Defense of what? Are you suggesting that he "misspoke" in that single-sentence paragraph?

 

Not at all. I meant that in a book that comprehensive, he shouldn't be expected to cover everything in a summary appendix on airflow. That work is up to you. I hope the explanation I wrote and screenshot I attached after that helps answer your previous question.

 

I'm reading Fossati to understand the science behind the origin of the aerodynamic force.
Watching popular science videos and experimenting with 3rd party software won't further that objective. Fossati appears to be a highly qualified academic who has published a highly respected textbook on the subject.
I am suggesting that his technical statement about pressures, flow speed, acceleration, conservation and passage width does not appear to be supported by the experimental data. I think your observation "I get it" acknowledges that.

 

I have a feeling that some of the misunderstanding comes from the translation-which isn't wonderful and you may have found examples.I have no doubt that somebody as experienced as Fossati gave a correct explanation but finding a translator with a detailed knowledge of the topic can't have been easy.One of the examples of this is the definition of Prismatic Coefficient that I referred to earlier,if you look at page 20 you will see it.

 

If you believe the explanation fails due to a poor translation, then aren't you implying it's not supported by the experimental data?

 

I am suggesting that the translation of the entire book may not be perfect.Bernoulli had the general principles in place a long time ago,but he was Swiss and not Italian.