Mass Participation factors 2

[mechwitch]

Dear all,

I am reletivly new to vibrational analysis and am trying to work through some of the basics.

I have been doing some basic modal analysis so far successfully and am now about to embark on dynamic random analysis. I have been told by my collegues to look out for mass participation factors in my results and that they have to be above 80%, but so far nobody has been able to tell me what they are or what they signify - not in a way that I can understand anyway!!

So can someone please tell me what is and what the significance of a mass participation factor is.

I find it difficult to 'go do' without understanding.

Regards

 

[cloche]

As far as I know, a mass participation factor gives a representation of how much of the part is contributing to the eigenmode, or in other terms, how much vibrational energy is dissipated through the eigenmode. The infinite sum of the m.p.factors is equal to 1 or to another normalization quantity (e.g. total mass of the vibrating system). The significant eigenmodes are those for which the sum of m.p.factors is at least a certain percentage of the norm (e.g. 0.9 if the norm is unity, 90% of total mass if the norm is the total mass,...)

 

[jcfoley]

Mechwitch,

cloche is correct from what I understand. I would suggest that you post this in the "Finite Element Analysis Engineering" forum. I know someone there can advise you. The mass participation factors are used to distinguish between "local" and "global" eigenvalues. I work with a lot of structural steel skids and we get a lot of local modal frequencies for stiffener plates. The only way to sort these out is to use mass participation. The real trick is to figure out what % = Global...

Hope this helps.

Chris Foley
Midland, TX

 

[cloche]

The threshold % depends on the type of analysis you are performing, but principally on the structure type: if it's a single bulk component, or a multi-linked structure, or... As Chris says, it's a bit a problem to estimate the correct % where to set the threshold.

The complete explanation of m.p.f. and related matters is somewhat complex, but for sure you can find plenty of good books (e.g. "The Vibration Analysis Handbook", "Theory of Vibration with Applications", "Machinery Vibrations - Measurement and Analysis"...) that will clarify this.
I myself will buy some of these to better understand all implications of this (and other...) matter(s)...

Claudio

 

[Drej]

In simple terms the (mass) participation factor squared gives you the effective modal mass for that mode for your structure. Notice when you see these factors that they are associated with a "direction" (X, Y, Z, etc..). Think of it like this: there is a simple point mass on a (massless) spring, and 1 DOF in the system only. Dynamically loaded, the point mass is allowed to bounce happily up and down on the spring in one direction, hence the whole of the mass (in theory) is participating in the bouncing in that direction. The square root - in this case - of the point mass is the participation factor, since the whole of the mass is participating in the mode.

Depending on what you're looking at as well (the type of structure, the code you're using, etc.), I'd always go with 90% effective system mass and above to make sure you've captured the response of the system.

Cheers,

-- drej --


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[cloche]

I agree with drej as regards the % threshold of mpf.
In fact, I can figure out only one exception to this: when you do analytical interpolations based on eigengrequencies in order to calculate "best-fit" Rayleigh damping coefficients Alpha and BEta that would respect an imposed structure response damping factor: in this case it is usualy wanted to take into account all eigenfrequencies whose sum of mpf gives at least 99%.

 

[tomirvine]

As an aside, I have posted a paper on modal mass and participation factors at:

http://www.vibrationdata.com/tutorials2/ModalMass.pdf

Tom Irvine

http://www.vibrationdata.com

 

[cloche]

Well done, Tom Irvine!
I remember now having already read (perhaps in another life, I don't remember... ;-) ) something very similar: probably it was your Revision A of the same document.