Modelling Pump Nozzles

[nikolastrojman]

Hi there piping stress analysts

I'm new to this field of mechanical engineering(I've started working in this field last year) so there are a lot of things I still need to learn and understand, and I'm hoping that you guys can help me along this path. Currently I'm trying to model piping connected to a centrifugal pump which isn't fixed to a ground(the pump is connected to the skid and the skid is supported by legs which are not fixed to the ground so the whole thing can move due to thermal growth of piping connected to a pump).
Can anybody give me an idea or better yet show me through an example the best way to approach this problem.

I would be very grateful if someone could help me with this matter.

Thanks in advance.

 

[C2it]

How are you intending to ensure where the pump might be in a few years time ?

Modelling is simple, but it seems like you have a rubbish piping design.

What are the boundary conditions ?

Are there any fixed points ?

Can you be sure that thermal expansion will not cause localised gross deformation due to lack of position control ?

 

[nikolastrojman]

Hi C2it,

I'm sending you a photo of the actual piping arrangement of the above mentioned pump. You will see from the photo what I'm talking about. This kind of piping arrangement, or should I say piping installation procedure for centrifugal pumps is often preformed by my client. They have this installation practice where they let the pump move freely (pump will adapt to the thermal growth of the piping by moving from the installation position to the operating position) thus decreasing the stress, and forces and moments on the pump nozzles. Also, as you can see from the photo, they have an anchor on the suction side of the pump (usually located as near as it can possibly be to the suction nozzle) so that they ensure that the pipe doesn't move to "much".

I hope this will give you a clearer picture of the situation I have.

 

[DSB123]

Hi I have come across this before with an existing system. You need to model the pump and both suction and discharge pipework together. You need to include the weight of the pump and skid and apply a conservative friction factor to the supports in order to get the effects of the friction back onto the pump nozzles.

 

[nikolastrojman]

Hi DSB123

Do you have maybe an example of how to do this in Caesar II correctly?

 

[ZippyDDoodah]

Model the pump as a rigid element using the pump weight just as DSB suggested. Support it with +Y supports with a friction factor for concrete (.6 or .7 ??). Model discharge and suction as DSB suggests, because discharge piping will add to the weight for increased friction resistance.

 

[DSB123]

ZippyDDoodah,
my oh my what a wonderful day. What additional information have you given above and beyond my response apart from friction factors for concrete which are questionable due to the type of adjustment supports on the pump skid.

 

[curtis2004]

Hi there,

If a pump assembly is not anchored into concrete pad, does it vibrate all the time?
It will destroy the pump!