Resonance in structures, how to avoid and calculate it?

[JoeH78]

Hi all,

I would like to know what would be your design sequence for structures which are vulnerable to resonance. This resonance issue occurs especially when you have light weight structure where cycling loads (wind, earthquake, etcc...) has great effect on natural vibration frequency of structure. Theoretically, each structure has natural vibaration modes and if any loads comes with in range of that modes( same vibration frequencies) this will amplify the effect of applied load and inevitably will result in collapse.

But how we are supposed to calculate how much time that structure will sustain to the applied load when it has the resonance?

Your comments will be appreciated?

 

[frv]

I don't think collapse is inevitable; but if you design something with a natural frequency that is similar to the frequency of whatever the use is, then you will have usability issues.

How to avoid it? Determine what some of the frequencies will be depending on the use of the structure (people walking, people running, equipment vibration, etc.) and then determine whether the natural frequency of the structure is similar. If so, you can usually do one of two things: increase the mass or reduce the span.

There are many publications on this. Have you tried searching these fora? Google? The particulars, if you want to get very precise, can be fairly involved, but there are several simplified methods to check your structures.

 

[JedClampett]

Wind and seismic resonance are pretty uncommon, unless you're talking about very low frequency structures. Individual light poles and suspension bridges are the only examples I've heard of of wind induced dynamic loadings.
Where it's more common is machinery supports. If you have a high frequency (1800 rpm) amchine, it's hard to avoid some vibration and you might get into trouble (although failure is rare), if you don't consider it at all.
Rather than try to calculate how much resonance you can take, try to avoid it.

 

[GregLocock]

The main tricks we use in the automotive game, in order of preference are

1) move the resonance so that it does not align with the excitation frequency

2) Attempt to attenuate the excitation at source - probably not practical with wind and earthquakes

3) Add some sort of isolation to the foundation of the system

4) add damping so that the response at the peak is reduced

5) add a tuned harmonic damper

6) increase the strength of the structure so it can run at resonance (ha, it is done but that is so ugly)

7) Add a subframe to isolate the system from the excitation

8) add an active cancellation system

Not all of those are appropriate or economical.







Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

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

 

[JoeH78]

Please don't take me offensive here, the question arised when I tried to implement a dynamic analysis program. I googled throughly for that ( one day) resonance issue but results are not quite satisfying so decided to post here.

I don't think collapse is inevitable; but if you design something with a natural frequency that is similar to the frequency of whatever the use is, then you will have usability issues.

I think that's been answered by engineering intuition and experience. I'm not convinced yet. Because theoritically it is possible, we usually refer to well known dynamic equation when we trying to solve those kind of problems M*U''(t)+C*U'(t)+K*U(t)=F(t), where we absolutely have no control over in the RHS of equation ( F(t) ) . So I can generate or improvise any arbitrary function F(t) which will match the exact frequencies of structure, in real life this is very rare ,probably chance in a million but still possible.

Rather than try to calculate how much resonance you can take, try to avoid it.

Do we simply rely on that? Probably my best approach would be to conduct some time-history and push over analysis, with simple load functions like sinusoidal, ramp func. etcc.


The items you summarized basically matches the already mentioned methods , try to avoid resonance increase mass or stiffness, use isolators or sub-structures for that. What you mean by active cancellation system ? Can you give an example?
One of the fields that I was wondering how they perform dynamic anlysis was automative field. How they really model a car chasis induced by engine harmonic motions when it works, probably the crashing of car chasis is out of question here but should be done to some extent one way or another.

I also would like to hear how aircraft wing is designed from someone in aerospace field.

Thanks for helping.

Regards,

 

[JedClampett]

Quote (JedClampett)
Rather than try to calculate how much resonance you can take, try to avoid it.

You avoid it by tuning the structure out of the range of resonance. If the forcing function has a frequency of 1800 rpm, make sure the structure has a natural frequency of 2400 rpm. Alternatively, you could tune your structure to 1200 rpm (low tune), but that's a lot riskier due to higher modes and equipment start up.
As far as your automotive analogy, I doubt that every piece of a car is dynamically analyzed. The engine is on motor mounts which are there to isolate the vibrations. But in many cars (including my piece of junk), you can feel vibration at certain rpm.
If you had an unlimited amount of money, you (you, not me) could run time histories and push over analysis. But if you had to do this for every structure, you'd see that for normal designs, it's not a huge consideration. For a 550 mw turbine generator at a power plant, it is a huge issue.

 

[PicoStruc]

First : Take a structural dynamic AND earthquake engineering course, it is now as simple as it seem !
Second : Structure are MDOF system that will respond with multiple mode which can be differently damped depending of your structure.
Third : Light structure such as tower are generally sensitive to wind due to their frequency range.
Heavier structure such as multi-story building (> 5 story) are a lot sensitive to earthquake (T=0.1 to 5 sec).
Fourth : Wind and earthquake force function frenquencies content is wide and not specific such as mechanical dynamic load.

For these reasons, we design structure to resist wind and earthquake load. For the earthquake load, you can reduce force by using a energy dissipation mechanisism such as plastic hinges... It all well prescribed by codes.

 

[PicoStruc]

You can use too period shifting technique such as adding mass or using damping devices

 

[GregLocock]

We model, fairly successfully, both vibrations from the engine, and crash. The effort required is enormous, for one vehicle line you are probably talking about 70 CAE analysts for 2 years, two supercomputers, and a lot of test work. Crash is the more difficult of the two problems.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

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

 

[GregLocock]

And

http://en.wikipedia.org/wiki/Active_vibration_control

for a brief overview on active

and

http://en.wikipedia.org/wiki/Tuned_mass_damper

for passive dampers

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

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