Strake Design & Analysis

[ss770640]

hello fellow like minded engineers!

I am seeking to transfer strake technology from aerospace to a subsea environment as a method of mitigating Vortex induced vibrations. A problem i am sure the aerospace industry must be familiar with. The effect of fluid flow induced vibration has on a wing is very similar to that of a subsea pipeline subjected to inflow and cross flow currents.

But my question is, what's the best method/program available on the market to simulate the design/pattern/sequence etc of proposed strakes? (Abaqus?)

Also, as an example, given the dimples on a golf ball are pretty much designed to minimise turbulent flow/vortexing. Why aren't airplane wings covered in them to assist inflight stability, ViV problems and thus efficiency? It is the dimple design i am seeking to verify as a method of achieving this.

thanks in advance

ss770640
CEng (UK)

 

[rb1957]

"given the dimples on a golf ball are pretty much designed to minimise turbulent flow/vortexing. Why aren't airplane wings covered in them to assist inflight stability," ...

i'd hazard a gusee that the flow field around a wing is a little different to that of a golf ball; consider that a golf ball is spinning.

 

[GBor]

So, rb, I should stop my "spinning wing" design?

I'm a former marine guy...let me just say that before you try and transition something from one industry to the next, you have to understand both. I'm not convinced from your many posts in the marine forums that you understand either sufficiently.

 

[rb1957]

maybe we can get rid of airports ... just have tees, and Really big golf clubs ??

 

[ss770640]

I understand mine perfectly. But the strake technology is more applicable to aerospace. Hence why I am asking.

The only strakes we get subsea are clamp on. I'm looking to put it into the pipeline coating but have never gone into depth analysing strake design. Hence the question and EXAMPLE golf ball.

Ps. helicopters have rotating wing design...

BEng(Hons) CEng (UK) MIMECHE

 

[rb1957]

you posed a question "why don't airplane wings have dimples ?" ... you got appropriate responses (IMHO).

PS, helicopters have rotating wings, not spinning (like a golf ball).

PPS congratulations for understanding your field perfectly ... me, i can only manage an imperfect understanding, and keep learning things all the time.

 

[ss770640]

Still waiting for an engineering answer.....

Don't think I will get one here

BEng(Hons) CEng (UK) MIMECHE

 

[rb1957]

ok, seriously, aren't you already in a very low Re ? i don't think boundary layer "tricks" will be very effective.

 

[GBor]

OK, I'll quit the fun at your expense, too, since rb has clearly taken the "high road".

Trusting that you understand the modeling requirements, do you know what the amplitude and frequency of the driving force is? Or are you looking to include the fluid flow calculation into a multi-physical assessment?

If the former, then several software packages come to mind so long as they can do some type of forced vibration calculation. If the latter, You may want to look at a software package called AMPS (

http://www.ampstech.com)

for such a complex problem if you are trying to use FEA, although Abaqus may work as well.

Are you building a subsea vessel? Or does this have to do with your pipeline analogy?

Sorry for the earlier ribbing...I haven't had enough to eat today.

 

[ss770640]

we spent a fortune in offshore vessel days installing helical strakes onto a pipeline riser. These were literally helically wound to disrupt the flow of seawater around it at a approx speed of 1-10m/s, against a flowline that could be anything from 6"-12" OD. To make it more complex the current speed varied with depth. (Seawater Density 1025kg/m3).

So i started looking at ways of disrupting laminar flow, golf balls and airplane wings (and ultimately here)

Helical Strakes:

http://www.mecaenterprises.com/helical_strakes.htm

In order for it to be cost effective, the method of disruption would have to go into the riser and not extrude from it.

How does Aerospace manage ViV? Do you just throw it all into CFD/wind tunnels, and watch for turbulent flow?

BEng(Hons) CEng (UK) MIMECHE

 

[btrueblood]

So, V = 1-10 m/s, D = .15 to .3 m, rho = 1025 kg/m3, viscosity (nu) ~= 1.5x10^-5 m2/s.

I get Re = VD/nu = 10,000 to 200,000

From Roberson & Crowe's Engineering Fluid Dynamics, the Strouhal no. for a cylinder in crossflow is

S = fd/V ~=.2 for those Reynold's numbers. I.e. the frequency of vortex shedding, n, will be

f = 0.2 V/d = from 0.6 to 13 Hz. The upper bound could go as high as 16 Hz given the spread in the referenced data. I.e., if your pipeline is tied down such that the lowest resonant frequency is above 20 Hz or so, then no problem.

If you can't tie the pipes down, then certainly helical strakes can work, by providing disruption of the boundary layer. These are typically used on vertical towers and smokestacks, where the wind direction is unpredictable. They work because they disrupt enough of the b.l. to eliminate large eddy pockets along the entire span, and a benefit is gained from axial flow generated by the helices as well. In other words, they will reduce the strength and span of vortices, but not completely eliminate them. If the direction of the tidal/current flows is predictable and regular, you might have more benefit from axial strakes (trips) at the approx. 90-110 degree positions relative to the flow direction. There are textbooks and articles that discuss the optimal location for trips on cylinders in such conditions (it varies with speed/Reynolds no.)

The Re range you are in IS also the right zone for turbulence generators (e.g. golf ball dimples) to disrupt the b.l. and generate a smaller eddy pocket. Don't know, and have not seen, articles that discuss reductions in Strouhal no., or shedding vortex strength, due to turbulence generators on cylinders.

If you really want to go to town, put a light streamline fairing (tail wedge) around the pipe, and allow it to swivel (weather-vane) to match the flow direction.

How does the aero world handle vortex-induced vibrations? Most typically by similarity to past designs, some CFD analysis. Once a problem shows itself (hopefully by the prior methods, more troubling in wind tunnel work, and worst case during flight test) the most typical "fixes" are turbulence generators, and stiffening of affected (resonant) structures. Occasionally, the resonances are attributable to 'off-design' operation that can be limited (e.g. don't open the landing bay doors above a certain speed).

 

[flash3780]

I always thought that dimples in a golf ball produced additional lift due to the spin of the golf ball (the top of the ball moving in the direction of the flow around the ball, and the bottom of the ball opposing it).

The drag created by the dimples on the spinning ball decreases the speed of the air beneath the ball (higher pressure) and increase the speed of the air above the ball (lower pressure). I think that the pressure differential is what produces extra lift.

If you look at your standard Moody chart, greater surface roughness = greater resistance to flow (i.e. more drag).

If the above assertion is true, it wouldn't be of any benefit to add dimples to an airplane wing (unless your airplane is tumbling).

 

[KENAT]

So, 10 years since I looked at this kind of thing at university but fundamentally the type of thing you're talking about was as I recall all about 'energising' the boundary layer. While causing the flow to be more turbulent (which intuitively you'd think was a bad thing) it actually delays separation and so done properly the net effect is a good thing. As I recal they use tricks like this on engine pylons etc.

I think my Prof was actively involved in trying to find approaches to model these situations.

I think what Btrue says about their use on smokestacks etc. might be a good place for you to start looking.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[btrueblood]

The effect is only valid for spheres and cylinders in a 'narrow' Reynold's no. range, i.e. 10^4 to 5x10^5, where the flow is in transition from laminar to turbulent. If the flow seperates before the b.l. goes turbulent, the result is a larger seperation bubble (wake), high drag, and stronger vibrations due to vortex shedding. If the b.l. is forced to be turbulent upstream of the seperation (via trip rings, strakes, surface roughness, etc.) the seperated zone is much smaller. Most fluid dynamics texts demonstrate this with plots of the drag coefficient for cylinders across Reynold's no. ranges spanning that gap. My uni's aero department would have us undergraduates conduct wind tunnel tests on smooth and "tripped" (wire ring at the 108 deg. location) spheres in the 10^5 Re range in air, showing the effect quite clearly. Also, see recent Mythbusters episode, where they put a thick layer of modelling clay on a car, and cut "scaled" dimples into it, and showed appreciable reduction in aero drag via fuel consumption at speed.

The idea of spinning golf balls generating lift is not incorrect (but smooth balls do this too - try it with a beach ball sometime), but even un-spun golf balls (cannon-launched, or just squarely hit) show drag reductions and/or range improvements. It's why the number and configuration of dimples is very closely controlled for regulation balls. Apparently, a 5-20 yard improvement can be gained if the circular dimples are instead made hexagon-shaped. But they'll kick you out of the country club for using them (ok, maybe not that, but they'd certainly take away your green jacket if you had one).

 

[btrueblood]

Sorry to be such a spotter, but re-reading what I wrote on ViV in the aero world, the overall subject is termed "aero-elasticity" or "aero-structural interaction", two more good search terms for past ASME and AIAA papers on the topics. The first term is the earlier coining. Aeroelasticity has its classical example in what is called flutter - where an airplane wing has a maximum speed at which it can be safely operated. Above this speed, the torsional stiffness of the wing allows the lift moment to twist the wing enough to cause the angle of attack to increase, causing higher lift moment...etc. A classic non-linear forced vibration problem. If the airspeed is increased near the flutter speed, the wing will begin to vibrate/oscillate in a torsional/bending mode. Beyond that speed, the airfoil will make one oscillation, and then be torn right off the plane. Before it was a known and calculable effect, it caused no end of consternation to pilots making high speed dives. Early testing of the phenomena was done with very carefully scaled wind tunnel models, and flight testing. Imagine being the pilot on those flights, with the geeky engineers telling you to keep cranking the speed up just a few more knots - until the wing is vibrating like a jackhammer, just before "divergence"...hoping the stress guy got his number correct, and the chute you have is packed right.

Another example are the vibrations of aero fairings (the outer "skin) of launch vehicles (rockets), especially as they pass "max Q", the speed where the dynamic pressure is a maximum, usually right around Mach 1.

Look carefully at the upper surface of the airplane wing you ride, next time you fly. Good aerodynamicists give you nice, smooth upper wing surfaces. Bad ones have lots of little stubby wing-shaped things poking up, to solve all kinds of downstream aero problems that weren't found until after the plane was in flight testing. More pointedly, you will find such "fixes" on planes with a lot of longevity, as changes to the planes (engine upgrades, size/weight capacity changes, etc.) change the design enough to warrant tweaking things.

 

[KENAT]

Just look at the big ass fences on many 50's thru 70's era Soviet designs if you want examples of what might be termed "aerodynami cluges'. Though as I recall they were for a slightly different reason to what's being discussed above.

Or the amount of aircraft that have a strake blending the vertical stabilizer to the top of the fuselage.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[WKTaylor]

The USN has seen a lot of these type problems... suggest You go to DTIC on-line and search for undersea cable [morings, towed-arrays, etc] and undersea pipeline stability issues

http://www.dtic.mil/dtic/

Regards, Wil Taylor

 

[BigInch]

Oh cool. Something to do with my golf ball skins. Glue them on my pipelines.

**********************
"The problem isn't working out the equation,
its finding the answer to the real question." BigInch

http://virtualpipeline.spaces.live.com/