Single degree of freedom flutter may occur due to low structural damping, lateral dynamic response associated with Vortex shedding from the structure produces loads in excess of the design loads and the Strouhal number (fd/v) which depends on fundamental bending frequency and diameter of structure.

Perhaps this can help?

https://rules.dnvgl.com/docs/pdf/dnv/codes/docs/2010-10/rp-c205.pdf

/C

flutter is a phenomenon related to aero foils, wings etc. Buildings donot have flutter however, they may undergo different vibrations depending upon structure and excitation mechanism.

Generally structures with circular cross sections do not suffer galloping - that excludes most chimneys and risers. Such structures typically suffer from along-wind and across-wind vibrations due to vortex shedding (plus along-wind buffeting) but not galloping. As a practical matter it is possible to compute response in both directions independently and combine them (in a statistical sense because the wind is generally turbulent). Of course if there are any asymmetries in cross section other mechanisms, including galloping, can result. The best book I know that describes these mechanisms without all the detailed mathematics is

Flow-Induced Vibrations: An Engineering Guide, Eduard Naudascher and Donald Rockwell (Dover)

Dear Roger

I have never worked with galopping modes. I happened to note on Prof. Rienstras homepage that such occur for transmission lines, i.e. horisontally oriented structures, usually, with a circular corss section.

See the animations on his hope page.

http://www.win.tue.nl/~sjoerdr/

Sincerely

Claes Fredö

Dear Claes:

While transmission line conductors are generally circular in shape they have asymmetries caused by helically wound wires which lead to some of the behavior referred to. Often additional asymmetries are caused by ice accretion which can lead to shapes that can lead to galloping in the right conditions. The same phenomena has been observed on tower guy cables.

Regards,

Roger

Hi Roger

This is what is great about RG. Always a possibility to pickup new learnings. As stated, galloping modes is nothing I have worked with, so please bear with me.

A flexible cylinder exposed to flow, e.g. a tube bundle or a valve stem exposed to cross wise outlet flow would be exposed to a force that rotates around its circumference. Said force would cause it to move in a circular pattern.

Changing the cross section from cylindrical to a more jagged shape would change its body wake width, vary its bending stiffness and so on but not greatly change its core feature of being exposed to this rotating force.

Of course, making the shape uneven decorrelates excitation along its length - which is tractable with respect to reduced excitation.

For wires, I suppose that the extra factor lies in gravity affecting the mode as well as lines being long spans, i.e. I guess that we have a case of propagating wave motion rather than that of standing waves.

These look fairly symmetric to me, though video resoltion is of course on the course side

http://www.win.tue.nl/~sjoerdr/plem_clip.wmv

http://www.tdee.ulg.ac.be/player.php?filename=gal_croa512k.flv

(more videos here)

http://www.tdee.ulg.ac.be/doc-27.html

I assume that a 'mode' is a 'mode', i.e. that the shape and behaviour is not very much different between a circular and a nearly circular cross section (in particular if cross section variation is random with the system length).

Out of ignorance & curiosity,- what is the game changing physics that makes a circular structure not being exposed to flow excitation & do you believe it could be used on something like a tube bundle in a heat stack?

Thx.

/Claes

Hi again Roger

Sorry to pester you more. I just happen to find the topic interesting.

Found this one, slides 20 & 39

http://web.mit.edu/13.42/www/handouts/VIV-2.pdf

I see that towed cables move in a figure eight pattern due to the tension. If this is the main thing, then I guess cross section symmetry would help decouple excitation and cross section assymetry increase excitation?

Then again, it seems on slide 39 the cross section shape affects the stability for gallopping. The PPT does not show circular cross section, but I suppose that what you are driving at would be that a perfectly circular cross section would have a beta valcue close to zero?

Sincerely

Claes

Dear Claes -

You bring up a lot of very interesting questions. Some I can comment on and others are outside my area of expertise. For the time being here are my comments:

"A flexible cylinder exposed to flow, e.g. a tube bundle or a valve stem exposed to cross wise outlet flow would be exposed to a force that rotates around its circumference. Said force would cause it to move in a circular pattern."

Not sure what is meant by "a force that rotates around its circumference" This is very complex flow. I don't know much about this subject but I would guess one of the main effects on a cylinder that is part of an array would be fluctuating flow from vortex shedding from a cylinder causing transverse fluctuating lift forces on downstream cylinders.

"Changing the cross section from cylindrical to a more jagged shape would change its body wake width, vary its bending stiffness and so on but not greatly change its core feature of being exposed to this rotating force.

Of course, making the shape uneven decorrelates excitation along its length - which is tractable with respect to reduced excitation. "

Another very complex topic. My feeling is if the body is prismatic the shape should not have much influence on correlation. I know that vortex shedding correlates less spanwise if the diameter on a circular cylinder varies. Then there is a tendency for the vortex shedding frequency to vary along the length more or less according to the Strouhal relationship. If the cylinder vibrates correlation increases with amplitude. Another factor that complicates the picture is the influence of turbulence.

Circular cylinders cannot gallop because they do not experience steady lift at any angle for flow transverse to the axis of the cylinder. This amounts to beta = 0. They do, however, experience fluctuating lift due to vortex shedding.

Hi Roger

Sloppy of me. A cylinder exposed to a flow is excited by a force that can be seen as a rotating vector from the cylinder cross section centre.

OK, it is the steady lift drag factor that is the driving excitation mechanism in galloping, rather than the oscillating excitation force that I have been babbling about. This explains the matter and was the missing piece of my knowledge.

Thanks

Claes

Hi Feroz

Then I believe I am able to help you a little with your question.

As pointed out by Dr Naseer, buildings and chimneys do not act the same way as is the case for flutter which interacts with lift, i.e. the static lift force. As pointed out by Dr Basu, galloping, again, is a different situation as it to some degree involves interaction with rigid body motion, i.e. motion of mass, and lift.

Vibration on the other hand is a different beast as it does not move mass - it merely oscillates it without net motion.

This conondrum is a bit the same as the three kinds of pressure (Line/Ambient Pressure, kinetic pressure (wind) and pulsation/sound) we are faced with when working with aeroacoustics (http://qringtech.com/2010/09/15/wave-steepening-increase-peak-pressure-piping-pumps/).

As buildings and chimneys are anchored to ground, these do not encompass net mass transport and therein lies the crux of the matter in distinguishing between these situations.

I believe this movie is enlightening

https://www.youtube.com/watch?v=Eh_vOcXazaU

Please note that the cylinder in the movie exhibits only a vertical direction motion as it is elastically sprung in the vertical direction. A cylinder anchoed at its base or from both ends would rotate around its cross section center in a circular motion pattern.

Also, the autor mentions telegraph wire's singing which is explained by aeolean tones (https://www.amazon.com/Notes-Acoustics-Uno-Ingard/dp/9380298323/ref=sr_1_3?s=books&ie=UTF8&qid=1528492057&sr=1-3&refinements=p_27%3AUno+Ingard) rather than galloping.

Just to clarify, I believe that flow might also interact with torsion modes if such motion entails a time varing fluctuation of the object's drag factor.

A very good book on the topic of aeroacoustic that is available for download is found here

http://www.win.tue.nl/%7Esjoerdr/papers/boek.pdf

As an introduction to the subject, Uno Ingaard's book "" is a very good starting point.

https://www.amazon.com/Notes-Acoustics-Uno-Ingard/dp/9380298323/ref=sr_1_3?s=books&ie=UTF8&qid=1528492057&sr=1-3&refinements=p_27%3AUno+Ingard

Sincerely

Claes

Please share me the best answer might you get...