An Electric Load Simulator for Engine Camless Valvetrains
Round 1
Reviewer 1 Report
1. There are some typos. 1. The name of first author in reference [15] is not correct. 2. p2 L77 “an electric load simulator. for linear rudder.” the period before “for” should be removed. 3. p11 L262 voltage v, not u.
2. There is no spring force Fs in Fig. 1.
3. Fig.2 shows the prototype of the electric load simulator. But the cross-section diagram, cores and coils of the simulator should be shown to understand it. Also please show the parameter values of the simulator.
4. In a magnetic device, there is few magnetic flux in the areas with large gap. But in Fig.2, some part with large gap still has high flux. Why?
5. It is very hard to understand the magnetic flux in Fig. 2 without a 2D cross-section diagram of the proposed actuator. Please show a detailed structure diagram to show the permanent magnets and electro magnets with coils.
6. In Table 3, for the initial engine condition, please also give the crank angle.
7. p6 L186 and P7 L197, authors said “Since only the exhaust stroke of the engine is considered, the intake valve in the 3D engine model is always closed”. However, engine has an overlap period for IVC and EVO. If this is counted in this simulation, then the intake valve is not always closed in engine simulation. Then the simulation results in the manuscript might not correct.
8. p7 L205 “The exhaust stroke… is simulated... The piston moves from the bottom dead center to the top dead center.” But this is not in the real situation. For any SI engine, for volumetric efficiency, intake valve will open early (EIVC) and exhaust valve will close late (LEVC). So the exhaust stroke simulated for piston moves from BDC to TDC is not consistent with the real engine condition.
9. In Fig. 6 The exhaust valve lift is simplified as a linear profile, but there are some problems. 1. The gas flow depends on the effective open area of valve, the simplified one has incorrect open area at each crank angle and the exhaust gas flow. 2. the mover has mass and inertia, it cannot change its displacement abruptly at starting and at ending.
10. p11 L255 In the manuscript “Assuming that eddy currents and magnetic saturation are ignored”. But if the applied current is too high, magnetic saturation often occurred. Is this a reasonable assumption? Can design a large actuator to avoid magnetic saturation? Please verify the magnetic saturation of this actuator by Fig.2.
11. Also, in conclusion part, In Fig.12, there is large time delay between the experimental and simulated forces. If this error is caused by the magnetic saturation as authors said, can the magnetic saturation be simulated before, and change the design of actuator? Can change the assumption and put magnetic saturation into models.
12. p11 L279, If the transfer function only has constants Ks, Ts, and 1, there is no time-lag function. Is an S missed in the denominator of TF?
13. p11 L269 Please provide the values of the gain Kp, KI, and Kd of the designed PID.
14. In Fig. 10, the simulated displacement is not a linear line as in Fig. 6. How to get this simulated displacement curve?
15. Fig. 9 shows the simulation models. Can show the transfer functions of all blocks?
Can provide other cases for small cylinder pressure under partial-load condition?
Author Response
Response to Reviewer 1 Comments
Point 1: There are some typos. 1. The name of first author in reference [15] is not correct. 2. p2 L77 “an electric load simulator. for linear rudder.” the period before “for” should be removed. 3. p11 L262 voltage v, not u.
Response 1: (1). Reference [15] has been revised as shown in p14 L415. (2) The period before “for” has been removed in p2 L77. (3) p11 L262 voltage v has been revised.
Point 2: There is no spring force Fs in Fig. 1.
Response 2: There are not springs in the device we designed, so there is no spring force Fs in Fig. 1
Point 3: Fig.2 shows the prototype of the electric load simulator. But the cross-section diagram, cores and coils of the simulator should be shown to understand it. Also please show the parameter values of the simulator.
Response 3: The cross-section diagram of the electric load simulator has been added in Figure 2. The parameter values of the simulator are shown in Table 2.
Point 4: In a magnetic device, there is few magnetic flux in the areas with large gap. But in Fig.2, some part with large gap still has high flux. Why?
Response 4: The permanent magnet of the electric load simulator is designed as Halbach structure, which can enhance the flux density in the air gap and reduce the magnetic saturation.
Point 5: It is very hard to understand the magnetic flux in Fig. 2 without a 2D cross-section diagram of the proposed actuator. Please show a detailed structure diagram to show the permanent magnets and electro magnets with coils.
Response 5: The detailed structure diagram of the electric load simulator has been shown in Figure 2 to show the permanent magnets and electro magnets with coils.
Point 6: In Table 3, for the initial engine condition, please also give the crank angle.
Response 6: The crank angle has been added in Table 3 for the initial engine condition.
Point 7: p6 L186 and P7 L197, authors said “Since only the exhaust stroke of the engine is considered, the intake valve in the 3D engine model is always closed”. However, engine has an overlap period for IVC and EVO. If this is counted in this simulation, then the intake valve is not always closed in engine simulation. Then the simulation results in the manuscript might not correct.
Response 7: Since we only considered a typical operating condition, there is no valve overlap period of the engine.
Point 8: p7 L205 “The exhaust stroke… is simulated... The piston moves from the bottom dead center to the top dead center.” But this is not in the real situation. For any SI engine, for volumetric efficiency, intake valve will open early (EIVC) and exhaust valve will close late (LEVC). So the exhaust stroke simulated for piston moves from BDC to TDC is not consistent with the real engine condition.
Response 8: Since the simulation is solved in a typical ideal condition, the actual displacement of the piston in the real engine condition has not been added into consideration.
Point 9: In Fig. 6 The exhaust valve lift is simplified as a linear profile, but there are some problems. 1. The gas flow depends on the effective open area of valve, the simplified one has incorrect open area at each crank angle and the exhaust gas flow. 2. the mover has mass and inertia, it cannot change its displacement abruptly at starting and at ending.
Response 9: The actual area value, transition time, open time and closing time of the valve in the simulation are consistent with the actual situation. As shown in Figure 11, the displacement of the mover doesn’t change abruptly at starting and at ending.
Point 10: p11 L255 in the manuscript “Assuming that eddy currents and magnetic saturation are ignored”. But if the applied current is too high, magnetic saturation often occurred. Is this a reasonable assumption? Can design a large actuator to avoid magnetic saturation? Please verify the magnetic saturation of this actuator by Fig.2.
Response 10: Through the finite element simulation analysis, it can be found that the current and force are within the reasonable range, so the phenomenon of magnetic saturation will not occur.
Point 11: Also, in conclusion part, In Fig.12, there is large time delay between the experimental and simulated forces. If this error is caused by the magnetic saturation as authors said, can the magnetic saturation be simulated before, and change the design of actuator? Can change the assumption and put magnetic saturation into models.
Response 11: The delay time lies in the fact that there are many non-linear factors in the actual system, for example the current and displacement cause great disturbance to the output force of the simulator, while the simulation model is more ideal.
Point 12: p11 L279, If the transfer function only has constants Ks, Ts, and 1, there is no time-lag function. Is an S missed in the denominator of TF?
Response 12: The transfer function has been revised in p11 L284.
Point 13: p11 L269 please provide the values of the gain Kp, KI, and Kd of the designed PID.
Response 13: The values of the gain Kp, KI, and Kd of the designed PID have been provided in p13 L320.
Point 14: In Fig. 10, the simulated displacement is not a linear line as in Fig. 6. How to get this simulated displacement curve?
Response 14: The displacement value obtained from the test is input into the simulation model, so that it is consistent with the displacement curve in the test system. The values of current in the simulation model are compared with values in the test system, under the same voltage and displacement input.
Point 15: Fig. 9 shows the simulation models. Can show the transfer functions of all blocks?
Can provide other cases for small cylinder pressure under partial-load condition?
Response 15: The transfer functions of the simulator model are shown in p11 L263, L278 an L284. Other simulation conditions are higher than 1200r/min.
Reviewer 2 Report
About this manuscript, I have following commands:
1. In figure 1, the valve should be turn 180o to show the correct mechanism.
2. Figure 2, can be eliminated, due to a magnetic fluxes of the actuator cannot give any information about the capability of the actuator. A force vs displacement diagram can give more information about the actuator’s properties and prove that the actuator is good to this study.
3. What is the software used to simulate the 3D engine model? If the software is written by authors please show the theory used.
4. The time period of the piston is about 25ms, why the exhaust valve only open 5.76 ms?
5. In Figure 7 and 8, how to get the experiment results? Authors should explain the experiment process and so on. If the experiment results are come from sec.5.2, please move that section to the position before Fig.7 and 8.
6. The structure of this manuscript is not so good. The authors should reconstruct the manuscript to let the reader can understand the manuscript easier.
Author Response
Response to Reviewer 2 Comments
Point 1: In figure 1, the valve should be turn 180o to show the correct mechanism.
Response 1: In figure 1, the electromagnetic linear actuator on the left is a gas-physical simulation load device, which simulates the gas load of the engine valve in the exhaust stroke and exerts the gas-physical force on the right valve. The valve on the right side should be placed as shown in figure 1 to truly simulate the force of the valve on the left side.
Point 2: Figure 2, can be eliminated, due to a magnetic fluxes of the actuator cannot give any information about the capability of the actuator. A force vs displacement diagram can give more information about the actuator’s properties and prove that the actuator is good to this study.
Response 2: Since other reviewers have raised questions about figure 2, please do not delete figure 2 now. In order to fully reflect the characteristics of the electromagnetic linear actuator, the characteristic curve of the electromagnetic force, namely the force and displacement curve, has been placed in the paper, as shown in figure 3.
Point 3: What is the software used to simulate the 3D engine model? If the software is written by authors please show the theory used.
Response 3: Because we do not purchase software copyright, so it is not convenient to reveal the name of the software.
Point 4: The time period of the piston is about 25ms, why the exhaust valve only open 5.76 ms?
Response 4: The exhaust stroke piston moves for 25ms. The process from closing to full opening of the exhaust valve is 5.76ms. We mainly study the gas physical force during the process from opening to full opening of the valve, during which the gas physical force will constantly change due to the change of displacement, which is our research content.
Point 5: In Figure 7 and 8, how to get the experiment results? Authors should explain the experiment process and so on. If the experiment results are come from sec.5.2, please move that section to the position before Fig.7 and 8.
Response 5: As a result of the addition of an electromagnetic force and displacement curve, that is, figure 3 in line 166. The figure number has been changed, which is explained as the new figure number below. Fig. 9 is a complete simulation with no experimental values. The test curve in figure 8 is the measured data of engine test. With the constant change of crankshaft Angle, cylinder pressure is measured by cylinder pressure sensor. The results in figures 8 and 9 are not derived from Sec.5.2.
Point 6: The structure of this manuscript is not so good. The authors should reconstruct the manuscript to let the reader can understand the manuscript easier.
Response 6: I will fill in the details for the reader to understand easily.
Reviewer 3 Report
The manuscript has considered the electric load simulator design and development for engine camless valve-trains under semi-physical conditions by designing an electromagnetic actuator according to the system scheme and performance. The authors confirmed the effectiveness of their proposed method by experimental data as well.
The paper is well organized and the results analysis are clear and detailed. The paper could be useful for readers in the field as it covers an important practical industrial issue.
I believe that the paper could be considered for publication in Applied Sciences after addressing the following minor room by the authors:
1- In the introduction chapter, a gap analysis is required to justify the necessity of doing this piece of research more clearly.
Author Response
Response to Reviewer 3 Comments
Point 1: In the introduction chapter, a gap analysis is required to justify the necessity of doing this piece of research more clearly.
Response 1: In the introduction part of this paper, the importance of this study has been written in p2 L51-L62. For example, the study of gas simulated load capacity introduced in this study can replace the real engine test. This method can save research and development cost and improve test reliable. It turns destructive experiments into predictive studies under laboratory conditions, and helps researchers foresee and detect the problems that occur in the mechanism and control system before they operate actually ease provide.
Reviewer 4 Report
The paper is a very good one and deserves publication after some small improvements
The paper validates also the results using real measuremets.
I would like to suggest just a few improvements:
Major:
please clarify the sentence written a page 13:
"The control parameters of the PID controller are continuously adjusted so that the output force".
Please clarify the adaptetion technique which you used. From the paper it is not clear.
Minor:
The control parameters of the PID controller are continuously adjusted so that the output forcePlease clarify equation (2) using physical arguments and/or references.
The following paper can be considered to improve the cited literature.
The following paper considers an adaptive PID Controller to be applied in Motion controli for valve control in Camless Engine Systems.
Mercorelli, P., Werner, N. An Adaptive Resonance Regulator Design for Motion Control of Intake Valves in Camless Engine Systems (2017) IEEE Transactions on Industrial Electronics, 64 (4), pp. 3413-3422.
The following paper taken into account different Problem related to identification in the context of electromagnetic drives also with applications in camless valvetrains.
Bascetta, L., Rocco, P., Magnani, G. Force ripple compensation in linear motors based on closed-loop position-dependent identification (2010) IEEE/ASME Transactions on Mechatronics, 15 (3), art. no. 5169985, pp. 349-359.
Author Response
Response to Reviewer 4 Comments
Point 1: please clarify the sentence written a page 13:
"The control parameters of the PID controller are continuously adjusted so that the output force".
Please clarify the adaptation technique which you used. From the paper it is not clear.
Response 1: About the PID regulation method, I have added in page 14 L319-L331, and write clear PID parameters of the three links.
Point 2: The control parameters of the PID controller are continuously adjusted so that the output force Please clarify equation (2) using physical arguments and/or references.
The following paper can be considered to improve the cited literature.
The following paper considers an adaptive PID Controller to be applied in Motion control for valve control in Camless Engine Systems.
Mercorelli, P., Werner, N. An Adaptive Resonance Regulator Design for Motion Control of Intake Valves in Camless Engine Systems (2017) IEEE Transactions on Industrial Electronics, 64 (4), pp. 3413-3422.
The following paper taken into account different Problem related to identification in the context of electromagnetic drives also with applications in camless valvetrains.
Bascetta, L., Rocco, P., Magnani, G. Force ripple compensation in linear motors based on closed-loop position-dependent identification (2010) IEEE/ASME Transactions on Mechatronics, 15 (3), art. no. 5169985, pp. 349-359.
Response 2: The two articles you recommended have been read and papers have been added to enrich the references in [26] [27].
Round 2
Reviewer 1 Report
1. The response 7 and 8 are not quite correct. A normal engine has overlap period. It is not a variable-timing engine, so all of the operating conditions have overlap period. And even the authors consider typical ideal condition, but the ideal condition must be reasonable.
2. For the response 10, even the current and force are within a reasonable range, it doesn't mean that the magnetic saturation does not occur.
Author Response
Response to Reviewer 1 Comments
Point 1: The response 7 and 8 are not quite correct. A normal engine has overlap period. It is not a variable-timing engine, so all of the operating conditions have overlap period. And even the authors consider typical ideal condition, but the ideal condition must be reasonable.
Response 1: Considering that there is no valve overlap period, the cylinder pressure is higher when the exhaust valve is opened, which is equivalent to a limit condition.
Point 2: For the response 10, even the current and force are within a reasonable range, it doesn't mean that the magnetic saturation does not occur.
Response 2: In order to comprehensively consider the power density and other factors, magnetic saturation does exist in the case of large current. But the impact on performance is small, so it can be considered permissible.
Reviewer 2 Report
Please put response 4 into the manuscript.
Line 240, the experimental results first appeared, but no experiment equipment setup was mentioned. the first experiment set up were appeared between line 295 to 299. Is this experiment set up is used to obtain the experiment results in line 240. If it is true please move 295 to 299 to the place before 240. If not, please give more explanation about experiment.
Author Response
Response to Reviewer 2 Comments
Point 1: Please put response 4 into the manuscript.
Line 240, the experimental results first appeared, but no experiment equipment setup was mentioned. the first experiment set up were appeared between line 295 to 299. Is this experiment set up is used to obtain the experiment results in line 240. If it is true please move 295 to 299 to the place before 240. If not, please give more explanation about experiment.
Response 1: Response 4 has been added in line 220 to 223 in blue. Engine cylinder pressure measurement test is to place a cylinder pressure sensor in the engine to measure the cylinder pressure with the engine running. Lines 245 to 247 have been added and illustrated, in red. The trial described in lines 299 to 303(previously 295 to 299), instead of 245(previously 240), is two trials.
Reviewer 3 Report
The authors have addressed my minor comments and the current version of the manuscript could be considered for publication is applied sciences.
Author Response
Response to Reviewer 3 Comments
Point 1: The authors have addressed my minor comments and the current version of the manuscript could be considered for publication is applied sciences.
Response 1: Thank you for your time and valve comments.
Reviewer 4 Report
The paer is a good shape now. It can be published in this Format.
Author Response
Response to Reviewer 4 Comments
Point 1: The paper is a good shape now. It can be published in this Format.
Response 1: Thank you very much for your repeated reading of this article and your Suggestions for revision .
