An Extended Approach for Validation and Optimization of Position Sensor Signal Processing in Electric Drive Trains

Round  1

Reviewer 1 Report

The topic of the paper is interesting, however the contents could be made somewhat more informative.

Below some more specific comments.

Title: "A new approach..."

Avoid using general expressions telling actually nothing:

Here 'the new approach' remains obscure even after having read the draft.

Please tell the busy reader what is your main contribution.

Abstract:

'The new approach' is not clarified in abstract:

Rewrite the abstract and clearly tell what is your contribution.

Keywords:

Think over about keywords: e.g. is this really 'signal processing' article.

To my mind the approach is more on system level (architecture).

1. Introduction:

Line 32 (typo) "targets in particular" not "targets in a particular"?

Figure 1: This simple (ancient Egyptian style pose perspective) figure could be given in a more conventional (classical) engineering style.

A good figure caption tells the notations used (colors, arrows, boxes, etc).

BTW, quite many figures are from other sources and the introductory text is more or less basic tech. Could the authors try to compact the material that is already available elsewhere. Use citations...

Line 56-57:  'a new approach' again here...

'highly accurate': avoid using this kind of qualitative expressions; give exact numbers, what is the precision. What was the target and how did you manage to reach it.

Line 65: 'extended sensor characterization' what does that mean in exact figures?

2. Material and methods:

Line 91: EMC: the authors could briefly tell about the EMC shield (in figures).

Figure 2: Notations? What is the figure '2' with backslash line between DUT and SSC boxes?

BTW, what is the horizontal dual line: axis/tube/cable?

Figure 3: is the grey box called DUT actually the climate chamber for the DUT? What is the big shining grey metal cabin called 'climate chamber'?

How it is used here (via the tubes)? The authors could explain in the figure caption.

Figure 4: from outer source: is it necessary here?

Figure 5: Figure 6: Basic stuff: is these figures necessary here?

Figure 7: Is that the resolver you used here? If not why here?

Lines 128-130: The authors seem not to be strictly aware of the numerical problem here: division of sine by cosine or vice versa lead to division by zero!

That is why the division is not done but atan2() function with two parameters (here sine and cosine) used in stead.

(you can actually find atan2() function in practically any programming language having trigonometric function for this division by zero and quadrant reason)

Noise: are you sure that the noise is equal on both windings?

Line 157: 'absolute position': what is the precision in figures?

2.4 Synchronization approach

Is this the 'new approach', if so the authors should state what is the 'new' by comparing their 'approach' to that already found in literature.

Figure 11: If here is the main contribution, the authors could take some more effort to clarify this figure, like what are the signals etc.

LInes 250: the referee does not quite understand the 'approach' here.

Could the authors try to explain e.g. what is the basic problem to be solved and how it has been solved in literature?

Figure 12 and Figure 13: how these are related to each other. Are the signal showed the ones shown in red?

Here the authors could use the same colors for clarity?

Figure 14:

In stead of giving a flow chart, the authors should tell how the errors are calculated and checked. Matlab is just for computing. The authors should know the precision of the related quantities and errors deduced on that.

So estimate the measuring errors and their influence on results.

Formulas (1)-(15): try of give the essential difference between the sets.

And do not forget error analysis.

Line 308: PP: is that relevant for the results?

Engineering is understanding error tolerances not slide shows.

3. Results

Figure 15: Resolver inside the climate chamber?

Line 324: are the .m files available?

Figure 17: avoid longer than one page images.

Figute 17c) this is actually a table, and quite simple one.

Could you replace it with a more informative statistical overview of the test data?

4. Discussion:

Here should tell the most important 'new' features of the proposed 'approach'.

What was the goal and how it was reached (error statistics)?

Line 374: To my mind the precision is not much dependent on the clocks.

Could the authors tell why it would be?

(Always when claiming something, be prepared to tell why in detail...).

Excel and Matlab are nice tolls, no doubt, but the beef should be in the code.

References:

Quite few solid journal articles. The authors should carefully stude literature to see what is the 'new approach'.

Author Response

Dear Reviewer, 

thank you very much for your high quality review inputs. Enclosed you find our response to your comments.

Yours faithfully, 

Christoph Datlinger and Mario Hirz

Author Response File: Author Response.pdf

Reviewer 2 Report

The electric and hybrid cars request precise sensors for exact electric drive train control. The paper presents a new approach for enhanced rotor position sensor evaluation, enabling to merge real-time data processing with test bench measurement (a calibration procedure in fact).

In my opinion, the experimental implemented procedure and the obtained technical results are correct.

Some aspects presented in the paper can be improved, and I have some suggestions for the authors.

‍‍Major revision‍‍

1. The authors already presented and published a lot of papers on a similar subject. This references are indicate at the bottom of this review. The improvements brought by the current paper compared to previously published ones must be emphasized in Introduction.

2. Most of the cited paper are internet pages and conference papers, and in my opinion this in not acceptable for publication in a prestigious journal. Please cite at least 5 more I.S.I. papers, published on similar or kindred subject.

3. The actually obtained results contains errors produced by the use of multiple clock generators. I suggest to the authors to use the same quartz clock frequency if possible, as they hinted for future research. As the authors presented and published more papers on the same subject, the results of the present article must be enriched in order to be published.

Minor revision

1. Line 53: "system represents includes multi-objective optimization" is "system includes multi-objective optimization".

2. Line 58: "The exemplary examined sensor" should be "The examined sensor".

3. Line 60: "high precise reference system" is "high precision reference system".

References on similar subject presented and published by the same authors:

https://www.researchgate.net/profile/Mario_Hirz/publication/327000506_Investigations_of_Rotor_Shaft_Position_Sensor_Signal_Processing_in_Electric_Drive_Train_Systems/links/5b827a5f92851c1e12335559/Investigations-of-Rotor-Shaft-Position-Sensor-Signal-Processing-in-Electric-Drive-Train-Systems.pdf

https://graz.pure.elsevier.com/en/publications/evaluation-of-signal-processing-effects-on-rotor-position-sensor-

https://www.researchgate.net/profile/Jens_Gaechter/publication/278353832_Evaluation_of_Angular_Sensor_Systems_for_Rotor_Position_Sensing_of_Automotive_Electric_Drive/links/59dddac3458515f6ef0888e7/Evaluation-of-Angular-Sensor-Systems-for-Rotor-Position-Sensing-of-Automotive-Electric-Drive.pdf

https://www.researchgate.net/publication/319025138_The_Effect_of_Rotor_Position_Errors_on_the_Dynamic_Behavior_of_Field-Orientated_Controlled_PMSM

https://www.researchgate.net/profile/Jens_Gaechter/publication/301243079_Evaluation_and_Modeling_of_Rotor_Position_Sensor_Characteristics_for_Electric_Traction_Motors/links/59dddaf90f7e9b53c1b0625d/Evaluation-and-Modeling-of-Rotor-Position-Sensor-Characteristics-for-Electric-Traction-Motors.pdf

 https://www.sae.org/publications/technical-papers/content/2016-01-1065/

 https://ieee-peds-2017.com/wp-content/uploads/2017/11/130.pdf

Author Response

Dear Reviewer, 

thank you very much for your high quality review inputs. Enclosed you find our response to your comments.

Yours faithfully, 

Christoph Datlinger and Mario Hirz

Author Response File: Author Response.pdf

Round  2

Reviewer 1 Report

Thank you for the authors for taking so much effort to revise the draft.

Quite many figures has been revised.

There are still some questions related to the figures:

Figure 1: replace magnified resolver by test "resolver". Now seems like a steering wheel.

Figure 2: Better than before, but usually photos are not very good: like a CSI photo!

BTW, you could use ellipses to enclose objects in image instead of arroes so that the size of the object could be clearly seen. (The arrow does not give any hint of the size of the object pointed to)

Figures 3 and 4: these are nice: BTW, have you got permission to use them?

Figure 5: atan and atan2 should give the same slope, but the scales in the panels are different.

Figure 6: Could align boxes/lines. AC-machine used three phase currents. What is the standard notation?

Figure 7: small font especially on right panel.

Figure 8: What is the direction of information flow? Too small font. Permission to use this image?

Please use image caption to explain what is actually seen.

Figure 9: Use space more economically. Why lines change color/width?

Figure 10. Quite nice: fonts could be a bit larger.

Figure 11: Difficult to see text: use larger font. Explain colors.

Figure 12: Use ellipses (with short marking) instead of arrows and text boxes. ECU seems fuzzy.

Figure 13: Much space wasted: use two separate images: one for online and the other for offline.

The processes are separated so that the images can be also separated.

The image within the magnifying glasses readable? Magnifying glass: use thinner outlines.

Figure 14: Scale markings in extremely small font: Just increase the font size to be readable; and also the lines could be wider. Notations: 'Electrical' wires as rainbow: replace by blue/green/yellow.

Figure 15: Quite good and clear! But the caption is quite long.

Figure 16: Difficult to see the idea of this graph.

Figure 17: why two separate panels, while scales are equal and curves already of different colors. The comparison(?) is easier when both curves are in same panel, right?

Figure 18: As above.

Figure 19: Quite small fonts and large area used.

Figure 20: Figure parts are usually called panels (not plots) and marked by a) b) etc.

Figure 21: Again too small fonts and line widths.

Table 1: Give only significant digits.

References: DOIs and URLs make the list less readable. Hopefully replaced by links in the final publihed version?

Author Response

Dear reviewer, 

please find attached the point-by-point response regarding your valuable comments.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors took into account all my suggestions and answered to all my questions. The paper was substantially improved and the references were updated. As I previously mentioned, my single reservation is now the fact that most of the procedures and the results from the paper were prior presented and published (see my first review report). 

Author Response

Dear reviewer, 

please find attached the point-by-point response regarding your valuable comments.

Author Response File: Author Response.pdf