## Modeling of elevated water tank

02-Getting started with the modelling
zthaer
Posts: 57
Joined: 12 Apr 2014, 15:07

### Modeling of elevated water tank

Hello ,
Im just wondering if there are any modeling considerations to account the sloshing pressure as a second lumped mass as per Hausner 63
This is the first question , the second is that the lumped mass assigned in the three orthogonal directions , insted of applying it as a gravity load is some models of structures that i saw before .

Thank you ,

huffte
Posts: 832
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

### Re: Modeling of elevated water tank

Hi zthaer. The referenced document proposed an essentially 2DOF system with part of the mass being the sloshing mass with associated spring. So I see no reason to prevent one from modeling such a system in SeismoStruct.

As for the second issue, it is completely up to you how masses are generated in SeismoStruct. That is actually a nice feature to have. You may:

1. Create your own lumped and distributed masses
2. Generate masses from frame element area and density properties
3. Generate masses from specified loads
4. Specify exactly which of the 6 global degrees of freedom for which you wish to consider masses.
5. Generate loads from element masses in any direction you wish.

Best of luck and happy modeling zthaer.
Tim Huff

zthaer
Posts: 57
Joined: 12 Apr 2014, 15:07

### Re: Modeling of elevated water tank

Hhelo huffte,
Thank you to response and to be specific my question is the paper proposed a formula to calculate equivalent k to the slushing mass
But how to model this spring stiffness , is it correct to consider it as a concrete column with that k ?

Best regards ,

huffte
Posts: 832
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

### Re: Modeling of elevated water tank

zthaer, there are many, many ways in which the stiffness could be modeled. As you suggest, one way would be to provide a column with the appropriate stiffness and the appropriate connectivity. Another way would be to use link elements with the explicit stiffness. So this is where you get to be a creative engineer and decide which of the many ways is best for your particular situation.
Tim Huff

zthaer
Posts: 57
Joined: 12 Apr 2014, 15:07

### Re: Modeling of elevated water tank

thank you huftte , your suggestions would be always appreciated , and to be frank im trying now to model alink between the two masses but seems seismostruct define the link as just nodal element , i have to model another element such as column to connect them together , the question is do you have an example ? [:D]

Thank you again ,

huffte
Posts: 832
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

### Re: Modeling of elevated water tank

zthaer, I have e-mailed a very simple and basic example which includes a link element with no frame member between any of the link nodes.

As always, I encourage you to scrutinize the model and try to understand why it is working as intended. Do not simply use the model and assume it is appropriate for your particular case. Use it to gain an understanding of how links are defined and how they work in SeismoStruct.

This particular example has masses at each of the first 2 link nodes. So the masses are originally coincident. Once the acceleration loading progresses you will eventually see separation between the two masses.

Best of luck zthaer.
Tim Huff

ss.rouintan
Posts: 1
Joined: 31 May 2018, 12:05

### Re: Modeling of elevated water tank

Hello
may I have this example email too?I ve got to model tendon spring elements with particular lenght for rocking system but seismostruct define the link as just nodal element can you please help me.?

huffte
Posts: 832
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

### Re: Modeling of elevated water tank

Can I appeal to zthaer to recollect the name of the model. As this post was 4 years ago, I cannot seem to recall which of my models depicted such an element. Or, if you like ss.rouintan, describe your request in more detail and I'll see what models I have which might help.
Tim Huff