Abstract

Parallel operation of hunchback curve pumps leads to stable or unsteady operating conditions according to the system they feed and its opening variations. Stable operating conditions are obtained when the two pumps operate in the stable part of their characteristics (flow condition A) and in certain cases, when one operates in the stable part and the other in reverse flow (flow condition B). When the two pumps have slightly different curves, the latter stable flow condition divides into B1and B2 depending on which pump is operating in reverse flow. There is a series of unsteady flow conditions between these conditions and in the neighbourhood of low or no flow they finally reduce to two pumping cycles, P1and P2 ; P1 is between stable flow conditions A and B1 in which the pump with the lowest maximum head operates in reverse flow ; P2 is between flow condition B2 and conditions in which both pumps operate in reverse flow. When changing over from one open circuit to another, including up to the point at which valves are closed, operating conditions can change from stability to another mode, or from one pumping cycle to another or from stable conditions to a pumping cycle or conversely. In some cases, it can be observed that at some times, operating conditions of pumps change very rapidly, the faster the smaller the difference in characteristics. The initial visual examinations were carried out at rated operating conditions, i.e. the ideal state (VI/2 II. RN= 0.55). They confimed the regularity of flow outside the limit layers on the hub and keel (intrinsic oscillations of lines of current near the foot of blades, milling of the limit layer of the casing), and outside a stall extended by an alternating helicoidal track in the neighbourhood of mobile vannes. This track which moves with the vanes of course causes disturbance on each passage to the turbulent flow around the fixed rectifier blades, when there is a rectifier. Operating in partial flow (V1/2TL RN ~ = 0,275) reveals the rotating stall : the various visual examination methods followed allows the diagram of this stall to be drawn, which in the given testing conditons (R = 104) reveals itself as one or several pockets, always on the keel side and upstream from the mobile runner. These tests also show the trend of a phenomenon in the course of a test (time of occurrence, speed and number of pockets) or as a function of parameters (volume of pockets as a function of flow, etc). A few tests with a rotating keel, i.e. without no play at the extreme points have shown that although this formula which is quite difficult to implement avoids milling of the limit layer of the keel, it does not prevent the occurrence of the rotating stall while operating at partial flow.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call