Effects of the Impeller Blade with a Slot Structure on the Centrifugal Pump Performance

Round 1

Reviewer 1 Report

Very informative and valuable paper with good structure and methodology. However I miss following points:

• Height of the slot is very small (quarter to three quarters of slot width). Why such small slot width was investigated? Explanation is required.
• Authors draw several conclusions about the effect of slot, but his information is not sufficiently supported by provided information. Contours in Figs.10-11 are very small and do not offer detailed analysis. I suggest plotting graph of the blade loading (for Case 1 and original blade) and also plotting streamlines in the blade channel to provide more evidence for the presented conclusions (to see how local eddy is influenced by the slot) and also to show the real effect of the slot on the flow field.
• Authors confuse the effect of splitter blades and slot in the introduction. While splitter blade could have similar effect if placed in the front part of the blade (close to the leading edge), usually (probably always) splitters are placed between mid of the blade channel and trailing edge of the main blade. Splitter also extends over whole height of the meridional section.
• Authors chose Case 1 to show impact of slotted blade compared to impeller without slots. It would be nice to see also head and efficiency comparison between Case 1 and original impeller in one graph (H-Q and eta-Q curves).
• I did not get figure 12. Should it show more uniform distribution of relative velocities across whole blade channel? Would it not be better to show relative velocity profiles on 3-5 stations from leading to trailing edge?
• English used throughout the paper is very good (in my humble opinion) and only marginal errors and typos are present: …Navier-Stokes… (p.1), … effects of blades … (p.4, down).

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

My general appraisal is the paper has the potential to be of large interest, because it is a systematic analysis of the effects of blade slots on the performance of a centrifugal pump by a numerical simulation of a complete pump.

But, the formulations in the paper should be much clearer on the possible benefit of a small slot in the blades of a centrifugal pump.

Therefore:

1. The original geometry without blade slot should also be one of the cases in Table 3. If wished to limit the number of cases to 16, case 3 may be left out, because this is clearly always the worst (large width and large depth of the slot).
2. The case without slot should then be added in the tables 5, 6, 7, such that it becomes visible if there is gain or loss with respect to the case without slot.
3. The performance (head or efficiency) of the case without slot should be made visible in figures 7, 8, 9, so that, once more, it becomes visible if there is gain or loss by a slot in a blade with respect to the blade without slot.

The authors conclude that there may be gain in head and efficiency by a slot of limited width and depth for low flow rate. The explanation of this possible gain in Fig. 10 is not clear. There is obviously incidence at the blade passage entrance with formation of a zone of low pressure at the suction side of the leading edge. It is however not clear by the figure by which mechanism the pressure distribution is improved by the presence of the slot. Therefore:

4. Fig.10 should make clear by which mechanism the pressure distribution is improved, and how this leads to a gain in efficiency and head. The figure should be much larger. The slots should be visible. It should be specified what the distance is of the shown surface to the shroud. Very likely, flow paths through the slot will be necessary for explanation. The most relevant is to make the comparison between the blade without slot and the blade with slot that reaches the largest efficiency.

The authors conclude that there may be gain in head and efficiency by a slot of limited width and depth for high flow rate. The explanation of this possible gain in Fig. 11 is not clear. There seems to be lower pressure at the blade passage without slot. The explanation by the authors is changed angle at the blade passage inlet. It is however not clear by the figure by which mechanism the pressure distribution is improved by the presence of the slot and how this may lead to better performance. Therefore:

5. Fig.11 should make clear by which mechanism the pressure distribution is improved, and how this leads to a gain in efficiency and head. The figure should be much larger. The slots should be visible. It should be specified what the distance is of the shown surface to the shroud. Very likely, flow paths through the slot will be necessary for explanation. The most relevant is to make the comparison between the blade without slot and the blade with slot that reaches the largest efficiency.

Small remarks

6. The paper should be carefully verified on typing errors. E.g.: uncompressible; Naivier-Stokes; slit; extant…. And many more.

7. The paper should be carefully verified on sentences or statements with an absurd meaning. I see:

Page2: The analysis shows that the impeller with the slotted structure has a circumferential speed and pressure. (I have no idea what this may mean)

Page 6: 0.5 accuracy

Page 6: according to similar theoretical formula for pumps. I understand: according to the similarity laws for pumps.

Page 7. The figure should be Fig 6. Further, these results suggest that the orthogonal design case accurately reflects slotting effects numerically. I have no idea on how this may be concluded from the comparison  between experimental and numerical results of the pump without slots

8. Please specify the specific speed in nondimensional form. I have no idea how the value of specific speed = 85 is obtained. I calculate a non-dimensional value of 0.44. This is then indeed medium.

9. Please specify the blade angle at entrance of the rotor.

10. Page 5 one but last line. Use "axial distance" instead of "vertical distance" 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I recommend the paper for publishing. It presents very interesting results. Authors improved presentation of their research and also discussion of the results by adding appropriate figures and clarifying text.

Author Response

Thank you very much for your good suggestion.

Reviewer 2 Report

General remarks.

The language has much improved. The paper is now quite well readable. There are still a limited number of typing errors.

I am reviewer #2 and I do not have access to the remarks of the other reviewers and to the answers to these remarks. Moreover, I cannot see text fragments that may have been changed as a consequence of remarks by other reviewers. So, I would like to know the reaction of the other reviewers.

Remarks on details.

1. Section 2.1. Write the formula of the specific speed. The used specific speed is dimensional, contrary to what the authors say about it.
2. Table 2. Include the blade inlet angle = 37.5 ° in the table in the paper.
3. Section 3.1. I do not understand the statement that the instrument accuracy is 0.5. Is meant the uncertainty is 0.5 %? If so, correct.
4. Just above Table 4: write: “ … according to the rules of similarity theory.”  
5. Beginning of section 3.3. I think that it cannot be said that the head increases at the large flow condition for some cases. The increases above the reference value are within the uncertainty band. The increase of the efficiency above the reference value is clear for some cases. So, just say that the efficiency is improved for the large flow rate.
6. In Fig 10, performance curves are shown (new figure), not internal flow distributions, as said in the text.
7. Page 13, beginning of the new red text: the pressure distribution is shown on the blade surface at the mean circumferential flow surface (middle line is confusing).
8. Next sentence. I do not understand what means: “The flow direction ranges from 0.1 to 1.0 forming an edge from the blade inlet to the outlet.” But, for me, this sentence may be omitted.
9. Page 13, last line. “The velocity amplitude distribution in case 1…”
10. Fig 16. I assume that the surface integral of the distributions has to be unity, thus equal for the two distributions.
11. Fig 13. Right: original model 1.4 Qdes should be in black.

Remarks on interpretation of the results; important for understanding

12. The new figure 13 helps in understanding how the slot jets influence the main flow at the inlet of the blade passages. The pressure is increased in the suction zones of the blades at the inlet due to the slot flow. This is already visible on the figures 11 and 12. But the new figure 13 is useful for further understanding. Figs. 11 and 12 are surfaces at span 90 %. Fig. 13 is at span 50 %. So, the global shift towards higher pressure in the inlet zone of the rotor is illustrated by Fig. 13. This is an effect that improves the cavitation margin, as is known from the literature. But Fig. 13 cannot explain increased head for 0.6 Q and increased efficiency for 1.4 Q.
13. The figures 14 and 15 are not useful for me. I cannot see the difference in the pressure contours. And the pressure distribution plots of figures 14 and 15 do not give more information than Fig. 13. Much more relevant for understanding is the flow distribution at the inlet. I see now that I was not explicit in my first review on what I think may help for understanding. I asked for flow paths through the slots, but I did not ask explicitly for the flow distribution at the impeller inlet. I let it to the authors to decide what precisely may be shown, but they should illustrate the better flow uniformity that they claim in the presence of slots. Thus better uniformity of flow velocity magnitude or better uniformity of flow vector directions; not just pressure. Could a plot of relative velocity vectors at a radius somewhat inside the rotor, just downstream of the slot exit, in a circumferential surface at 90 % span (thus such that the flow through the slot can be seen and thus can be proved that the grid resolution in the slot is sufficient for a reliable representation of the slot flow) demonstrate that it is correct that the flow becomes more uniform, what the authors claim is the benefit of the slot flows. Or contour plots of the magnitude of the relative flow velocity on a cut with a grid surface somewhat inside the rotor? Whatever can be invented for illustration of the distribution of the relative velocity on a surface that is approximately perpendicular the main relative flow, somewhere a the rotor inlet, but inside the rotor. Comparison of flow without slot and flow with slot.
14. Say that can be illustrated that the flow at the inlet of the rotor is more uniform for 0.6 Q and 1.4 Q in the presence of slots; it is still a mystery why this increases head, but decreases efficiency, for 0.6 Q and improves efficiency, and leaves head almost the same, for 1.4 Q. Clearly for the design flow (Q), there is always decrease of head and decrease of efficiency, which seems evident due to the energy consumption of the slot flow. I hope that the authors can formulate an explanation of the different reaction concerning head and efficiency for 0.6 Q and 1.4Q.   

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report

The paper is now Ok.