Structural Vibrations 6

[Euler]

I am designing a structure that will support a vibrating screen. The screen will operate at about 21 hz and must be mounted on a platform about 28 feet in the air, supported by 4 columns. The platform must straddle some existing tanks with a rigid frame in the direction perpendicular to the vibration motion and will be fairly heavily braced with tube steel x-braces in the direction parrallel to the vibration motion. Usually, we try to design the structure with a natual frequency of at least 1.5 times the machine frequency. In this case, that would be very expensive and impractical, if not impossible. We plan to design the platform with a natural frequency of .5 or less times the machine frequency. Will the structure respond as though a shock load was applied, as the machine frequecy matches the structure nf(10 hz), even though it is momentary? If the machine "spins up" in about 3 to 4 seconds, is the force developed, when the machine hits the structure nf, so momentary that the it is not a realistic force? Is there any reasonably priced software out there that could be used to simulate this process? Thanks

 

[PXC]

You will have to check that higher frequencies are also distant from the operating frequency. It may be quite possible to "tune" the structure to achieve this. You must be careful with frequencies that are not "in the direction of the excitation force", as they might still be excited.

I would not expect the screen to excite any lower modes on startup as it is so quick. They can take a long time to slow down though, and so they could excite the structure then. This is exactly what happens to the screen itself, as the springs it is mounted on have a low natural frequency, say 4 Hz. I do not know if you can get a screen with brakes on the drive to stop it quicker.

I use Strand FEA for dynamic analysis. The first analysis is to calculate natural frequencies. The next analysis calculates the amplitude of vibration as a funtion of forcing frequency. The big variable here is damping. This is typically so small that the amplitudes of vibration will be excessive when you are too close to resonance eg within +/-15%. Some screen manufacturers give forces in only 1 direction, some in 2 or even 3. The forces in the 3rd and possibly the 2nd direction would be say 10% of the main direction. I would always put a small force in these other directions eg 10%, even if the manufacturer had not given one. If you dont, the harmonic analysis will not pick up on any possible resonance in that direction. What might cause such a force? Uneven loading on the screen deck perhaps, or maybe a mechanical misalignment (I know of that having occurred).

And Strand is a very good price. Ref

http://www.strand.aust.com/

 

[aggman]

Euler,
Startup will not be a problem, as the screen passes through the natural frequency very quickly. What will be a prolem is shut down. The screen will shut down slowly over a period of time. I design screen towers every day. Most of the towers have 2 to 4 screens in them. The company that I work for has about 30 years experience in the industry also, so we have a lot of standard type bracing arrangements that we use. You are right in that it is not economical to force the natural frequency of your structure to be greater than the screen. What we do is force the manufacturer to give us the resonant loadings of the screen and their frequency. The loads that the screen gives off at this point will be 5 times higher or so, maybe more, than any live loading that they give you. Make sure that the natural frequency of your structure and the resonant frequency of the screen are not the same. For the most part I design for these values. Experience shows that as the screen slows, it will pass the natural frequency of the structure, which wil cause a moment of slight shaking, and then settle out. As the screen reaches the resonant frequency, again the structure will have two or three cycles of shaking and then will settle out. The mangitude of the shaking will depend on the closeness of the screen and structure natural frequencies. It is company policy to be conservative to the tune of unity numbers in the 60%-70% ranges, and to allow a 1.5 multiplier on all loads supplied by the screen manufacturer, per AISC. You can do a more involved analysis, but it probably is not worth it, and you will get a satisfactory structure. Also, you will have better results if you can support the screen with columns directly under the spring pads, or at least close to it. If you can't, and have to put the screen on a beam, spanning out to columns, then I highly recommend placing in plane bracing for the screen base to transfer lateral loads to the column line. This will help reduce vibration in handrails, etc. Also, one of the most important things to remember is that a screen is constantly changing in mass and rigidity. All manufactures put a nice disclamer that all their loadings do not take into account rock, etc. The screen media decks will change on a daily basis, and their will be wet aggregate and dry aggregate, some of which will stick to the screen and some will not. What I am getting at is that we can all do as much dynamic analyis as we want, but in this industry their will be things that change all the time and you can't control it. The method of design that we use now reflects a lot of experience designing these structures, some good and some bad. We have just proven to ourselves over time that designing these types of structures has just as much to do with your gut, as with the theory behind it. Hope my long winded response has helped.

aggman

 

[Euler]

Thanks for the great suggestions! We will design the platform to be very durable by using extra strength in the bracing connections. We do a lot of industrial building design where screens and other vibrating equipment are used. The convention in the past has been to design for extra stiffness without really knowing how much and then just tuning vibrations out during the debug phase. Not very economical, and sometimes impossible after all the structure and equipment are in place. Lately though, we have taken the approach of designing the beams based on the stiffness needed to avoid resonance in the members and providing enough bracing and columns to keep the structural vibrations to a minimum. This project was different in that there is so much congestion of existing equipment that we cannot locate the columns or brace it the way we would prefer. We have done some calculations for the force developed at resonance and it will be significant, even if only momentary, and the structure will be "beefed up" to account for the abuse it will see over the years. Thanks for your help.