sole plate natural frequency 2

[mehere11]

I am asked to verify a sole plate thickness supporting discharge column of a submersible pump. (please see the attachment for configuration).

One of the criterions for the thickness is, static deflection of fully loaded sole plate to produce a natural frequency note more than 0.707times that of driver.

Any hints or references on how to calculate natural frequency of sole plate in this case?
Is it possible to calculate it without using FEA analysis softwares like Ansys etc..?

 

[GregLocock]

There are two modes of interest, vertical bounce, and a pendulum mode.

I'm guessing you want the first.

Roughly spekaing you have the mass of the tube and the pump and any water bouncing on the bending stiffness of the plate.

So, can you work out the deflection of the plate at the inner edge of the sole plate, for a unit load on that circle?

Looks tricky to me but not impossible. FEA would be quicker, and would also allow you to investigate other modes.





Cheers

Greg Locock

I rarely exceed 1.79 x 10^12 furlongs per fortnight

 

[Tmoose]

What is the "driver frequency" that you must be 30% lower than? Floor spans possess flexibility that are poison if ignored.

The gussets don't appear to get particularly close to the anchor bolts. That invites increased flexibility, which tends to produce lower resonant frequencies.
The installation, in regards to shimming and grouting, can pretty drastically change the flexibility and finally the as-built resonant frequencies. It seems pretty popular to expect grout to "stick" to steel. Even with very particular prep of the sole plate's underside and epoxy grout I think that is expecting a lot, so any inclination for upward bow of the plate must be resisted by plate stiffness, or, deep ribs to transfer the load to the anchors. And anchors must tighten against metallic spacers or shims.

 

[mehere11]

I do not have enough information about driver frequency. what I interpret by 'driver frequency' in this case is the rpm of pump converted in the frequency unit. I might be wrong though. Some one having better idea please correct me if I am wrong.


Thanks

 

[GregLocock]

You need to get back to whoever set the spec. i'd also be concerned by rpm*number of blades, tho I think pump rpm is a good start.

A Tmoose implied, the damping and support offered by the surrounding soil/grout renders this calculation a bit odd.





Cheers

Greg Locock

I rarely exceed 1.79 x 10^12 furlongs per fortnight