## Forum

Forum breadcrumbs – You are here:
Please or Register to create posts and topics.

# Design on substructure elements

Hi David,

I’m a student trying to learn this software to my research on the substructure. I’m focus on the research about the impact of different subelements size. As the screenshot 1 shows below, when I change the diameter of the “subelementrigid – elementID2 – diameter form 12 to 16”,  almost the whole system was floating over the water. You can see it as the screenshot2, the green line is the original OC4 system, the blue is the one that I changed its subelement’s diameter.

I notice that it may because I didn’t change its mass value. Here is what I confused: If I simply change the value of diameter on the location screenshot1 shows, will the mass of the subelement automatically adjust? Or I need to change the mass value by myself? If I need to change the mass, would you please give me some guidance about how can I adjust the mass of the subelements when I change its diameter?

Best Regards,

Easey

Hi Easey,

the member mass is specified in the 2nd column of the SUBELEMENTSRIGID table (BMASSD in your screenshot) in units of kg/m. This value represents the mass per unit length of the element. Since you didn’t modify this value, but only changed the diameter of the member definition, the mass of the members remains unchanged. To calculate the mass density, you can multiply the cross-sectional area of the members by the density of the material they are made of.

If you’ve enabled the IsBuoy option for the members in the SUBMEMBERS table, then the buoyancy force is automatically computed based on the volume of water displaced, which is determined by the diameter and length of a member. This is why the modified floater is floating upward.

In the provided example that you have modified, the mass density for the members is set to 1 kg/m, which is effectively a very low value and can be considered close to zero. In this example, the floater’s mass is defined as a lumped mass using the 6×6 SUB_MASS matrix, rather than being represented as a collection of distributed member masses. This is just one of the various approaches for modeling a floater in QBlade. You can find an overview in the Modeling Options for an Offshore Substructure section of the QBlade documentation.

BR,

David

Hello David,

Thanks for your answer! Now I’m facing some new problems in the matrix you mentioned. Following your previous guidance I found the SUB_MASS matrix and got the way to change the value of mass. But after I changed the mass, I noticed that the value of inertia should also be modified, they are respectively in the last 3 columns of the matrix. Then I went to check the definition of OC4, and I found that I can’t get the value of roll inertia by using the fomula I_roll=Mass multiply distance^2 (1.1373E+7 × 13.46), as you can see in graph 2 I uploaded. Did I select the wrong central point of mass? Could you give me some guidance about how to calculate the Platform roll/pitch/yaw inertia the graph 2 shows?

Best Regards,

Easey

Hello Easey,

the 6×6 SUB_MASS matrix contains the total mass properties of the OC5 floater.

While the formula you mentioned for obtaining the inertia is correct, it’s important to integrate the mass distribution multiplied by the square of the distance from the reference point where the lumped mass matrix is generated. In this case, the reference point is REF_COG_POS, as this is where the SUB_MASS matrix is located.

However, the table 3-3 you provided lacks information about the internal mass distribution within the floater. To obtain this data, you would need a detailed model of the floater that includes such information.

An alternative approach to calculating the lumped mass matrix directly is to utilize an explicitly distributed mass. This is achieved by directly assigning mass within the elements table(s) (SUBELEMENTSSUBELEMENTSRIGID, etc.). This way, the mass is automatically included in the members, and no lumped mass matrix is needed. The mass per length of the members should be much easier to obtain than a floater model of distributed masses.

BR,

David

Hi David,

As your answer said, is that meant if I want to calculate the inertia of the floater, I need to sum up every subelement’s “mass  × square of the distance” one by one?

And when I’m using the alternative approach you provided, I also get some confuse.

As you can see in figure 1, the original BMASSD is 1 for each element. Then I tried to modified the diameter of element 2 from 12 to 16, so from the area ratio, the BMASSD should be changed to 1.778 (1.2²/1.6²=1/1.778), may I comfirm that this fomula is correct?

After doing so I ran the procedure, and the result of heave-time curve seems the same as BMASSD=1, Diameter=16 when I just change the diameter to 16 without modifying the value of BMASSD. As graph 2 shows after changing the BMASSD the result is still the same.

Would you please give me some guidance about whether my approach is correct and how can I modify the value of BMASSD while changing the diameter?

Thanks a lot!

Best Regards,

Easey

Hi Easey,

in the model you are using, the total mass is assigned using a 6×6 SUB_MASS matrix. The value of 1 for BMASSD is a very small mass (only 1kg/m length) used only for numerical stability, and can be considered negligible. Thereby, changing this value by a small percentage has no effect on the model.

To obtain representative mass values for a member, simply multiply the cross-sectional area of the member (that can be obtained from the member diameter and wall thickness) by the density of the member material.

BR,

David

Easey has reacted to this post.
Easey

Hello David,

So can I assume that the 6×6 SUB_MASS matrix is the original total mass, and if I wanna add the mass of subelements, I just need to add the Excess mass over the original model?

For example, the original cross-section area of element2 (d=12) is 2.25 m², and the modifed area (d=16) is 3 m², the density of materal is 7850 kg/m³

And as you can see in graph 1, should I change the BMASSD to (3-2.25)*7850=5918?

or I should modify the BMASSD to 3*7850=23550, as you said “multiply the cross-sectional area of the member by the density of the member material”?

Thanks a lot!

Warm Regards,

Easey

Hello Easey,

based on your description, it seems that you are adding only the ‘additional’ mass, due to the change of its size, to each element, which should yield the correct result. However, your proposedmethod combines lumped mass contributions (SUB_MASS) with distributed mass contributions, which, while generally correct, may not be the most consistent approach.

As for your last question, you only need to include the ‘additional’ mass once in the SUBELEMENTSRIGID definition. This way, each member generated from this definition will automatically be assigned this new mass.

BR,

David

Easey has reacted to this post.
Easey

Hi David，

I’m really greatful for your help. I’ll try this approach on my research first to see if I can get effective results.

Thanks again!

Best Regards,

Easey 