Shell mode or beam mode ?

[GMarsh]

Hi,

I have a question regarding identification of particular mode as shell or beam mode. I am attaching the picture of mode shape here.

Till recently I was thinking it is i=1, j=2 shell mode (i-circumferential, j-axial). But when reading some other literature I am reminded that beam mode possibility also is there. So I am not able to differentiate it.

Dimensions of shell: 350mm diameter, 3mm wall thickness, 100mm height. This mode shape is at 4700 Hz where as the fundamental mode is around 1200 Hz. There are nearly 20-30 other shell modes before this 4700Hz mode.

Also is there any formula to identify at what frequency beam mode (of a shell) onsets and if so whether it will be dominant ? My harmonic response analysis shows that this is one of the modes with significant amplitude. Any hints as to why this should be dominant in terms of amplitude?

Thank you.

Geoff

 

[hacksaw]

the dominant modes will have less stiffness than the others,
will help narrow the field,

you can also look at the modal participation relative to you excitation

 

[GregLocock]

Mode shape descriptions are rather arbitrary. But I agree, that does look very like second mode of a cantilever.

Cheers

Greg Locock


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

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

 

[MikeyP]

Looks like it is only deforming in shear but not bending which makes it a little odd.

Ta

M

--
Dr Michael F Platten

 

[GMarsh]

Thanks to everyone who responded here. I was looking at this problem last few days and concluded (at least as of now) that it may not be a beam mode. Two ways to confirm this:
- when I draw a mode order (x-axis) vs frequency (y-axis) graph, as given in a book by Leissa, the mode in question is nicely fitting with remaining data to give a elongated U-shape curve
- when I calculated 1st natural frequency of beam mode for this structure, it is way above this value of 4700 Hz.

With this two observations I feel it may be shell mode only.

If it proves otherwise I will update here.


hacksaw - can you elaborate how to identify this low stiffness ? I calculated stiffness using m*w^2, m is generalised mass calculated in natural frequency extraction step. But ordering the modes in that way is not helping much. On the contrary ordering the modes using just on generalised mass basis shows 4700Hz mode as dominant. But this concept is not working on other structures where also there is a dominant mode behaviour. Am I missing something ?

Regards
Geoff

 

[hacksaw]

was suggesting that if you have the modal mass and the natural frequency then a stiffness may be calculated; the least stiff modes that are also compatable with the excitation are the ones most likely excited

okay that is a bit different from the question asked, but the approach has been used with thin shells in identifying the dominant modes

 

[hacksaw]

actually the modified bending stiffness after K. Saijyou (2002) is required not the stiffness originally described; am in the process of working through Leissa, what a classic