To explore the influence of the diffuser clocking on the performance of the multistage centrifugal pump, 8 test schemes were designed through orthogonal test. What’s more, the influence of diffuser clocking on the energy performance and pressure pulsation of the pump was studied through experimental test.
The multistage pump synchronous test experimental device is shown in Fig. 2. The test rig consisting of a water tank, an inlet valve, an outlet valve, a flowmeter, an inlet and outlet pressure transmitter, a vibration signal collector and sensor, a pressure pulsation signal collector and sensor, a multistage pump, a frequency conversion cabinet and the other parts are shown in Fig. 3.
The data acquisition system including electromagnetic flowmeter, meterexplicit pressure transmitter, pressure pulsation sensor, threephase electrical parameter acquisition instrument and rotational speed measuring instrument realized the synchronous acquisition of performance parameters. The performance parameters of the instrument required for system testing are summarized in Table 2.
3.3 Test Scheme
Eight orthogonal experimental schemes were designed for the 5stage centrifugal pump. Two adjacent pump stages were taken as a test factor, and the clockwise staggered angle of the secondary diffusers relative to the primary diffusers was passed for the level value of this test factor. Since the number of diffuser blades was 6, two horizontal numbers were set, which were 0°and 30°. Therefore, there are four horizontal factors included in the pump, namely, factor A, B, C and D. The orthogonal test table is summarized in Table 3.
Table 3
Test serial number

CL1

CL2

CL3

CL4

CL5

CL6

CL7

CL8

A(12)

0°

0°

0°

0°

30°

30°

30°

30°

B(23)

0°

0°

30°

30°

0°

0°

30°

30°

C(34)

0°

0°

30°

30°

30°

30°

0°

0°

D(45)

0°

30°

0°

30°

0°

30°

0°

30°

The pressure pulsation monitoring point at the outlet of multistage pump arranged at twice the pipe diameter at the outlet of multistage pump, as shown in Fig. 4.
3.4 Test Results
A comparison chart of the head and efficiency of multistage pump under design flow rate is shown in Fig. 5. The dimensionless comparison method was adopted which dimensionless parameters H/HCL1 and η/ηCL1 were used to characterize the change rate of the head and efficiency of each diffuser clocking scheme relative to the diffuser clocking scheme CL1.
As can be observed from Fig. 5, the scheme CL6 is the best one to improve the multistage pump external characteristics. The pump head and efficiency reach the maximum of the eight schemes, which are about 100.4% and 100.5% respectively of the scheme CL1. It can be observed that the diffuser clocking has little influence on the pump head and efficiency.
Figure 6 shows a time domain diagram of the pump outlet pressure coefficient in one cycle under different diffuser clocking schemes.
The outlet pressure pulsation curve under each diffuser clocking scheme fluctuates within the same range, and the periodicity of this is not obvious. The outlet pressure coefficient of the clocking scheme CL1 reaches the maximum value of about 0.032 at 0.4T. In addition, the maximum values of outlet pressure coefficient under the scheme CL2 to CL8 are lower than that under the clocking scheme CL1.
Further, the arithmetic root mean square of pressure coefficient is taken as the effective evaluation standard of the outlet pressure pulsation, and the calculation formula is as follows:
The mean square root, the instantaneous value, the average value and the number of values of the pressure coefficient are Up, Cp(i), and, respectively.
The arithmetic root mean squares of pressure coefficients of each scheme are compared, as shown in Fig. 7.
Based on Fig. 7, the diffuser clocking effect has a significant influence on the pressure pulsation at the pump outlet. Compared with the CL1, the effective values of pressure coefficients under each diffuser clocking position are decreased to a certain extent, with ranges from 3.5–27.2%. Besides, the root mean square decreases of pressure pulsation in the CL3 to CL8 are 10% more than the CL1, which have a significant impact on improving the stability of multistage pump outflow. Moreover, it can be presented that the clocking effect of the CL4 is the most obvious than others, and the effective value of outlet pressure pulsation is decreased by 27.2%.
Figure 8 presents a frequency domain distribution figure of pressure coefficient. As can be observed from the Fig. 8, the distribution of outlet pressure coefficients is basically the same in each diffuser clocking scheme. Furthermore, the main frequency is the axis frequency (1IRF). This explains the outlet pressure pulsation frequency domain distributions are not changed owing to different diffuser clocking schemes. It suggests that the outlet pressure pulsation is still mainly affected by the rotation of the rotating shaft, but the main frequency amplitudes of pressure pulsation are varied under different diffuser clocking schemes.
Under different schemes, the amplitudes of the outlet pressure coefficient at the shaft frequency are shown in Fig. 9. In all schemes, the axial frequency amplitude under the CL1 is the largest, because all diffuser circumferential positions are the same under this scheme. Therefore, the axial frequency amplitude generated by each pump stage will be conducted to the next pump stage and it will produce cumulative effects. However, the other seven schemes all have different circumferential positions between one or more diffusers of previous pump stages, which leads to certain losses in the fluctuation phenomenon that the transmission shaft frequency is the main frequency. This explains that the amplitude of the shaft frequency and the effective value of the pressure coefficient are reduced. It makes the outlet flow more stable.
Corresponding to the greatest decrease on the effective value of the pressure coefficient of the CL4, the amplitude of the diffuser clocking scheme CL4 has the biggest change. Meanwhile, the amplitude of the CL4 is 35.8% less than the CL1, indicating that the scheme CL4 is the better one.
Since the amplitude of the main vibration frequency at the outlet of the multistage pump is more than that at the inlet, the clocking scheme CL4 is the most significant to improve the operation stability of the multistage pump, and the amplitude at the main frequency is reduced by 35.8%, which shows that the vibration at the outlet of the multistage pump caused by the rotation of the rotating shaft under this scheme is greatly suppressed. Therefore, the CL4 is the best one to improve the operation stability of the multistage pump among the eight schemes.