Dynamic Instability Analysis of a Double-Blade Centrifugal Pump

Round 1

Reviewer 1 Report

2.1. Pump model

• To clarify the figure 1 (b), please provide an isometric view. (It is hard to visualize the suction and volute fluid domain)
• Why did you choose this size pump with defines operation condition listed at table1?
• Mesh
  • What it the wall y+? It would be better to zoom in the mesh at impeller edge, also.
• CFD model
  • One of the main usage of this pump is liquid (with solid particle) transportation. Why, you did not model the multiphase(fluid+solid) flow for this study?
  • Did you use a commercial CFD software?
    • Yes : please provide the name of the software and the version. Moreover, more details of the setup is required.
    • No: if it is an in-house, please provide the main equation sets you used through the numerical analysis
  • Previously, you showed the cavity fluid domain, did you use these domains in your simulation? If yes, how did you handle the mesh of the small area?

2.3. Experimental setup

• A table next to fig 4 that describes the number will be a representation
• More details about the accuracies of the measurement devices
• Error of the measuring devices is defined, what about the repetitively of the experiment ?

2.4. Layout of monitoring points

Again, CFD, what tool did you use to get/analyze the signal?

 

2.5. Simulation validation

• Please be more specific with ref [25], what chapter did you get the disc friction loss from?

3. Results and Discussion

• Line 181 : please refer the plot to show the time of the pressure pulsation peak are not synchronous
• Fig 8: there is no straight correlation between flow rate changes versus pressure pulsation. For instance, point 4 is out of the trend, why?
• Fig 11: Can you compare the velocity stream line at different flow rates as well? It would be interesting.
• What are the Reynolds numbers at inspected data points?
• Fig 14: define the starting point.

Author Response

Please see the attachment.

 

Author Response File: Author Response.docx

Reviewer 2 Report

The article deals with the analysis of pressure pulsations in a radial pump by means of numerical calculations. For this purpose, CFD is used to calculate the flow in the pump under different load conditions. Subsequently, pressure pulsations are evaluated at selected points in the pump housing and presented partly in the time domain and in the frequency domain. 

This procedure corresponds to the state of the art. A numerical calculation method is used. It is not clear what is meant by "finite element volume method". Presumably a finite volume method is meant.  Is a commercial code used, or an opensource or in-house code. Which turbulence models are used. 

What is meant by "domain frequency" (line 244).

The authors come to the conclusion that oscillations at the blade passing frequency are induced by the flow. This is actually self-evident. 

The work is largely comprehensible and described in a compact form. The results of the calculations are partly supported by experimental data. The paper should be linguistically revised before publication.  Some (not all) errors are marked in the enclosed PDF.

Comments for author File: Comments.pdf

Author Response

The article deals with the analysis of pressure pulsations in a radial pump by means of numerical calculations. For this purpose, CFD is used to calculate the flow in the pump under different load conditions. Subsequently, pressure pulsations are evaluated at selected points in the pump housing and presented partly in the time domain and in the frequency domain.

Q1: This procedure corresponds to the state of the art. A numerical calculation method is used. It is not clear what is meant by "finite element volume method". Presumably a finite volume method is meant.  Is a commercial code used, or an opensource or in-house code. Which turbulence models are used.

Answer: Yes, the “finite element volume method” is corrected to “finite volume method”. In this study, we use the commercial CFD code, ANSYS CFX 19.0. The SST k-ω turbulence model is used for simulation. Please find this information in page 5 of revised paper.

Q2: What is meant by "domain frequency" (line 244).

Answer: The domain frequencies mean the main excitation frequencies. To avoid this confusion, we use the main excitation frequencies instead.

Q3: The authors come to the conclusion that oscillations at the blade passing frequency are induced by the flow. This is actually self-evident.

Answer: Thanks for your kind suggestion, it is a self-evident conclusion, we delete this conclusion in the revised paper.

Q4: The work is largely comprehensible and described in a compact form. The results of the calculations are partly supported by experimental data. The paper should be linguistically revised before publication.  Some (not all) errors are marked in the enclosed PDF.

Answer: Thanks for your kind suggestions. We revised the paper based on your inputs.

Round 2

Reviewer 1 Report

-Fig. 2 shows the seal gap, It is not clear that you simulate the seal or not. In case you are simulating the seal section of the pump, what is the leakage flow rate at various conditions.

- More details of the CFX setup is required. (advection scheme, transient scheme, etc. )

Author Response

Please see the attachment.

Author Response File: Author Response.docx