Topic Options
#801 - 11/18/02 02:09 PM Pump nozzle design as per API 610
skumar Offline
Member

Registered: 11/18/02
Posts: 1
Loc: kuwait
Hi friends, I am a new user to CAESAR so if my question is not in place, please forgive me. I have a project involving centrifugal pumps. I am not able to understand exactly how to model pump discharge and suction nozzles such as C nodes etc.
Also which load case to be considered for analysis?
Could any one of you help me by giving step by step procedure or please mail me any typical problem.
Thank you very much.
_________________________
regards

Top
#802 - 11/18/02 03:54 PM Re: Pump nozzle design as per API 610
John Breen Offline
Member

Registered: 03/09/00
Posts: 482
Loc: Pittsburgh, PA (& Texas)
Hello Skumar,

May I suggest that you first use the search function on this board. This is a topic that has been discussed here before with some interesting input from the community. Use "API 610" as the search key words and you will find previous discussions on this topic.

Also you might want to get your company to obtain a copy of theis document:

"Guidelines for the Design and Installation of Pump Piping Systems", by Vincent A. Carucci and James C. Payne – February 2000 (48 pp) (ISBN: 1-58145-456-2) - $85.00

You can find it here:

http://www.forengineers.org/pvrc/wrcpvrcbulletin.htm

Regards, John.
_________________________
John Breen

Top
#803 - 11/19/02 08:34 AM Re: Pump nozzle design as per API 610
Richard Ay Offline
Member

Registered: 12/13/99
Posts: 6226
Loc: Houston, Texas, USA
There are several ways to model this situation.

<font color="#0000ff">Without the Pump</font>
In this case, just model to the pump flange, and call it an anchor. In this situation, you wouldn't need CNODEs.

<font color="#0000ff">With the Pump</font>
This case requires that you model the pump as a series of rigid elements, down to the base, which is then anchored. At the pump flange, you simply transition from <em>pipe</em> to <em>rigid</em>. You really don't need CNODEs in this instance either.

Including the pump allows you to connect the suction and discharge lines, and obtain an overall graphical representation of the system. However, assuming the pump is rigid, and acts as a point of fixed support (i.e. an anchor), the lines will not affect one another, so running both together isn't necessary.

<font color="#0000ff">With CNODEs</font>
A CNODE can be thought of as <em>an association of Degrees of Freedom</em>. In essence what you are telling the software is that the "X" DOF for node 123 is the same as the "X" DOF for node 456. You typically use this technique to connect pipe to nozzles or structural steel, and expansion joint tie-bars.

A CNODE is treated as an <em>internal restraint</em>. Because it is a <em>restraint</em>, the loads on the CNODE appear in the restraint report.

One <em>trick</em> some users employ is to connect pipe to pump flanges via a CNODE. This then shows the loads on the flange in the Restraint Report. This isn't strictly necessary, because you can always get the loads from the <em>element force report</em>.

I hope this helps.
_________________________
Regards,
Richard Ay - Consultant

Top
#804 - 11/20/02 12:03 PM Re: Pump nozzle design as per API 610
Anindya Offline
Member

Registered: 09/14/02
Posts: 58
Loc: India
Kumar,

To put things in a very simple manner, let us consider you have a end suction-top discharge nozzle.

Give a global anchor at the pump centre, this is well defined in API 610, else you can see it on the online help screen in Caesar II, under the API 610 analysis module.Let us say this is node 10.

From node 10, you model a rigid upto the pump inlet nozzle of the same dia as the inlet pipe . Say this is node 20.Next at node 20 you model the inlet pipe flange say this is node 20-30.However number the element not as 20-30, rather number it as 21-30 and at node 21 use an anchor , with connecting node at node 20. You follow an identical pattern for the discharge side of the pump.

If you model the node and C-node in this way , you will get the correct sign of the forces and moments on the nozzle. If you switch node and C- node between nodes 20 and 21, you will get the same magnitude of forces and moments, only sign will be reversed and you can infer that physically from the behaviour of the system.

By definition, a Node and the connecting node will have same displacement in a degree of freedom if they are connected by a rigid stiffness in that degree of freedom, else if you connect them by a non rigid stiffness they wiill have relative displacement in that Dof. Sign of forces and moments on nodes and C- nodes follow the action-reaction force signs as per Newton's third law.So there is really nothing to worry about which node is C-node and which is the restraint node, so long you physically interpret the sign of the results properly.

Hope this answers your question.

A. Bhattacharya

Bechtel Corporation
_________________________
Anindya Bhattacharya

Top



Moderator:  Denny_Thomas, uribejl 
Who's Online
0 registered (), 31 Guests and 3 Spiders online.
Key: Admin, Global Mod, Mod
April
Su M Tu W Th F Sa
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30
Forum Stats
12065 Members
14 Forums
16973 Topics
75151 Posts

Max Online: 303 @ 01/28/20 11:58 PM
Top Posters (30 Days)