Vortex Shedding on Lamp Poles 1

[blmcghee]

I am in the process of vibration testing 30 ft lamp poles showing fatigue failure at the base. I'm wondering if I drive the pole with a electric eccentric vibrator will I acheive the desired resonant frequency? I feel that perhaps this will only give me the 1st modal frequencies and I believe the 2nd modal frequencies(mid point of pole movement) caused by vortex shedding is the real cause of the failures. Any one have any past experience with this problem they care to share?
Thanks in advance.

 

[tomirvine]

If the purpose of your test is to identify the natural frequencies, then you can strike the top of the pole with an impulse hammer. You call also use the shaker which you mentioned.

Sincerely, Tom Irvine

http://www.vibrationdata.com

 

[blmcghee]

Thanks tom,
I am worried that the failure is being caused my the 2nd mode vibrations and I am worried that will only show up with a wind tunnel test to get vortex shedding. On a square pole ,8" square and 30' long anchored to mount, will this shedding cause that much oscillating action? Do you think with a simple vibration as described I can acheive all frequencies needed?
Thanks again

 

[tomirvine]

Wind tunnel tests are very expensive and may be unnecessary for your problem.

Here is a recommendation.

1. Perform a simple modal test on an installed pole during a windless condition. Strike the top of the pole with an impluse hammer. Record the pole response with an accelerometer. The result data will give the natural frequencies and damping ratios of the vibration modes.

2. Measure the response of an installed pole under a variety of natural wind conditions. Use accelerometers. The resulting data will show which modes are being excited.

3. The resulting data from steps 1 and 2 can be compared against analytical calculations to determine whether vortex shedding, or some other mechanism, is the cause of the problem.

4. If vortex shedding is the problem, there are a variety of possible design modifications. Some of these changes could be performed as a retro-fit on existing pole installations.

I am available to help with this project on a consulting basis.

Sincerely, Tom Irvine
Phone: 480-820-6862

http://www.vibrationdata.com

 

[blmcghee]

thanks Tom. I'll see if the company would like to enlist your expertise.

 

[sms]

As was mentioned, you'll probably get the best data by catching it in the act, in other words get data when the posts are being excited by the wind. That will tell you what mode is being excited. I suspect it is one of the first couple modes any way. The higher modes typically are less likely to be excited. I would be curious to know where the fatigue failure is occuring. Is it in the bolts that hold down the lamp post? If so were the bolts properly sized and torqued? Are you really sure this is a dynamics problem and not a case of high static loading that is being pushed over the edge to failure with a small dynamic load?

 

[blmcghee]

The failure is occuring in the weld where the mounts are welded to the poles. This is a rather common problem with all forms of light poles and the modes causing the excitation of the center of the poles seems to be the culprit. Several universities and larger companies have done research but the results are always in debate. Seems like a rather simple problem which is not so easy to nail down and the corrective measures are particular to each case(as expected when dealing with so many free variables I suppose)
Thanks for the discussion

 

The phenomenom you are using is KARMAN VORTEX (see Den Hartog - Mechanical Vibrations p305). The formula for calculating this is

fD/V=0.22

where
f=frequency (Hz)
D=diameter (inches)
V=wind speed (inches/second)

ron frend

 

[kverspoor]

Sorry for this late post but it seems to me that, while vortex shedding may excite lateral natural frequencies and the wind would provide a horizontal load, there may be some other action going on here. What about the torsional natural frequency(s) and/or acting in concert with residual stresses in the weld?

I have no experience in this area and I don't know the effect the wind direction would have on whatever the pole is supporting. I just pictured the old STOP sign problem we dealt with in school and didn't know if it were at all applicable.

Regards... kv

 

[temel2]

blmcghee

hi,

1. First of all, I think vortex shedding is not the only reason for the vibrations of the traffic poles. Assuming laminar flow (low wind speed) to form a vortex shedding, the vibrations would only be in the in-plane of the pole. However the vibrations for a traffic pole are both in-plane and out-of-plane. Also, a good source of understanding the vibrations of the traffic lights is, NR412 Fatigue-Resistant Design of Cantilevered Signal, Sign and Light Supports Report. It is stated here that vortex shedding is not even a factor affecting or initiating the vibrations. The large amplitude vibrations should be self-exciting, so galloping phenomenon would be the main reason. Also,trucks and cars passing through has an effect on the vibrations. The effect of turbulence is not clear yet but there are some recent papers that may be helpful. The research is mostly on wind-induced vibrations of flexible-/tapered cylinders.
2. The natural frequencies of the structures are pretty low around 1Hz. However the forcing is random, so the post is being loaded with a variety of frequencies. Therefore, unless you have a parametric or an autoparametric damper, you may not need to force it at the natural frequency as it would not simulate the natural conditions.
3. to kverspoor
what did you mean by torsional natural frequency? The tip of the light pole is oscillating in an elliptic or circular shapes (looking from through the cylinder). However, the researchers are mostly concentrating on an uncoupled problem, meaning the in-plane and out-of-plane vibrations are independent of each other. This is an assumption to simplify the problem actually.
4. Also a square pole again may not be feasible because of the flow pattern difference between a cylinder and a square. Also, the natural frequencies may not be same, so for the determination of the frequencies it would be better to test it on the real structure itself.
5. Finally, actually a different manufacturing process may help to increase the fatigue life (compressive residual stress formation), retrofitting the structure (such as assuming that the vortex shedding is the problem, changing the roughness or adding bumps to get rid of laminar flow) , adding a damper or even drilling some holes through the cylinder and putting tubes so that there will not be any tension or compression on the structure (galloping).
I hope I could give some ideas.
Thanks..