This content has been marked as final.
Show 6 replies

Underwater pressure modelling
Peter Gillespie Jan 4, 2007 3:29 PM (in response to Cedric Floquet)I'm not experienced in working with submersed objects, but I'll share what I do know.
Your model needs to be fully constrained without inertial relief, even though it is not actually attached to anything. if it were a submarine, I would put a restraint on the propeller axially, and the rudders in the other two directions. I don't know your specific application, but you should be able to figure out something that is actually 'restraining' or constraining it in real life. The only thing I can think of would be a full sphere that is too heavy to float and too light to sink to the bottom. With that I would probably constrain something internal, since you should have small forces anyway if your pressure is equal.
I think with the inertial relief on that your displacement may not be accurate, so I won't comment further on that until the restraints are in place correctly.
Underwater pressure modelling
Cedric Floquet Jan 5, 2007 11:43 AM (in response to Peter Gillespie)Dear Peter,
Thanks for the answer. I thought quite a lot about the modellingissue and found a partial answer in the CosmosWorks Online tutorial'Symmetry restraint for solid and shell model'. That should pointme towards the right direction.
Obviously it will work for the casing but I will struggle as soonas the model will not present an axial symmetry.
The device I am working on is a kind of probe dangling at the endof a 4km long cable into seawater. It could be assimilated to apendulum in equilibrium if you want.
As the instrument is under hydrostatic pressure, the onlyparameters which is important will be the bulk modulus.Unfortunately I could not find any mention to it in the SW or CWhelp. The change of volume in the material induced by the highpressure will create some misalignments for some of my opticalcomponents and as I use different types of material, the bulkmodulus is not the same everywhere hence inducing stress at somecritical points.
One thing I am really interested in will be to know if CW is takinginto account the bulk modulus into its calculations. Does anybodyhere knows anything about it?
I know it can be calculated from the poisson and Young moduluseasily for metals, but I am not convinced the formula can beapplied to plastics.
Any input on that matter is welcome.
Best regards,
Cedric
Underwater pressure modelling
Peter Gillespie Jan 5, 2007 1:20 PM (in response to Cedric Floquet)I would think that you should be able to assume that the cable restrains it in the vertical direction, and you probably can include both up and down because of the weight and cable, respectively. That is assuming the weight at that depth overcomes the buoyancy.
And assuming you don't have other forces acting on it beyond the pressure, the cable will resist twisting, which takes care of the rest of the restraints. I realize that is not going to give you perfect results, but it would be worth a try at least and see what reaction forces at the restraints are.
The symmetry restraint may provide the restraints other than vertically, I am not sure exactly how that works, but it sounds promising.
And I couldn't tell you anything worthwhile about CW and bulk modulus, sorry!
Underwater pressure modelling
Richard Larder Jan 9, 2007 8:03 PM (in response to Peter Gillespie)In the theory of Elasticity for Isotropic materials there are threedependent Moduli plus poisson's ratio. The two elasticconstants are usually expressed as the <a href="elastic_constants_E_nu.cfm">Young's modulus</a> <i>E</i> and the<a href="elastic_constants_E_nu.cfm#poisson">Poisson'sratio</a><span style=" fontfamily: 'Times New Roman'; fontsize: 12.0pt;"><img height="24" width="14" src="/jvqjvfup2872.gif"></span>. However, thealternative elastic constants <i>K</i> (<a href="elastic_constants_G_K.cfm#bulkmod">bulk modulus</a>) and/or<i>G</i> (<a href="elastic_constants_G_K.cfm">shear modulus</a>)can also be used. For isotropic materials, <i>G</i> and <i>K</i>can be found from <i>E</i> and<span style=" fontfamily: 'Times New Roman'; fontsize: 12.0pt;"><img height="24" width="14" src="/cfvzcrfc2873.gif"></span> by a set of <ahref="calc_elastic_constants.cfm#Table">equations</a>, andviceversa.<br><br><p>In response to the hydrostatic load, the specimen will changeits volume. Its resistance to do so is quantified as the <b><spanclass="green">bulk modulus <i>K</i></span></b>, also known as themodulus of compression. Technically, <i>K</i> is defined as theratio of hydrostatic pressure to the <a href="#volchange">relativevolume change</a> (which is related to the direct strains),</p><p> </p><div align="center"><span style=" fontfamily: Verdana; fontsize: 6.5pt;"><img height="96" border="0" src="/bkazhtii2870.gif" width="202"></span><br><br>Cosmos requires E, and <span style=" fontfamily: 'Times New Roman'; fontsize: 12.0pt;"><img height="24" src="/zrmhyebr2874.gif" width="14"></span> or G. Ifyou have only the bulk modulus you can assume a poisson's ratio of 0.3 for most metals and solve for E. Poisson'sratio cannot be larger than 0.5. If you need morerelations just Google elastic constants and you can find moreequations.<br><br>Dick Larder<br><br><br><br><br align="left"><br></div>




Underwater pressure modelling
Richard Larder Jan 9, 2007 8:51 PM (in response to Cedric Floquet)Sorry the equation did not come thru in my last post. If Nu =Poisson's ratio then it is;
E = 3*K*(12*Nu)
Underwater pressure modelling
SolidWorks Corp Feb 1, 2007 1:56 PM (in response to Richard Larder)The only thing I have to add is regarding Inertial Relief. Ths effect is meant to counter dynamic loading that keeps a moving body in equilibrium and is not meant for a static body. Try your best to restrain where possible and realistic (take advantage of symmetry if you can.) Then use the Soft Springs option to satisfy equilibrium instead of Inertial Relief. Last caution.. don't rely on Soft Springs to cover for restraint mistakes. Make sure you are using it when your system truly is unrestrained and force balanced.
Good luck!
