Reasons for the failure of the impeller in centrifugal pumps
Thread Content
Dear teachers: Recently, the impeller of a 520 m3/h centrifugal pump in our unit broke off at its root; after it was repaired, it broke again after being in use for a month. I’m not sure what the cause is Does anyone here know anything about it?
In general, this situation is related to stress concentration that occurs here and the heat treatment of the material.
Could you send a cross-section of the fracture? Is the medium density high? Generally speaking, fractures are nothing but quality issues: either due to stress concentration or because the drive wheel itself has defects ； Fixed the issue of no stress-relief treatment, which led to fatigue fracture once stress concentration occurred
That shaft isn’t thick to begin with, so it’s barely possible to tighten the impeller nut. If the structure isn’t modified and only an inducer wheel is added at the location of the impeller nut, problems will arise once stress concentration occurs on the shaft, especially since heat treatment makes it too hard. It is recommended that the inducer wheel and shaft be connected using keys to transmit torque, with a sturdy impeller nut attached to the shaft end, rather than combining the two into one unit.
As I thought, the impeller is simply screwed onto the threads of the shaft head, and the hollow grooves in those threads represent a weak point – any slight flaw in the design or manufacturing can lead to fracture. Given the large weight of such an impeller along with the resulting unbalanced forces, and considering the fact that the cantilever arm is quite long, it’s clear that the threads of the shaft head alone are not sufficient to meet the strength requirements. It is recommended to lengthen the shaft, ensure that the induction wheel is connected to the shaft via a proper shaft hole key, use a key for this connection, and then secure it with a nut at the shaft end; this design is safe and reliable. :)
The analysis is good; thanks for the guidance!
The fracture site is in the heat-affected zone of the weld, I guess