Design and Measurements of an Electrothermal Filter Using CMOS Technology^†

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper presents the design, fabrication, and testing of an electrothermal filter (ETF) implemented in 3 μm CMOS technology. The work is positioned as both a sensor and a potential building block for temperature-to-frequency conversion. I appreciate that the authors have implemented their design and included experimental results to validate it.

However, the paper requires significant revisions before it can be considered for publication.

1. The abstract is too brief and does not conform to MDPI guidelines. Moreover, it is poorly written, with grammatical errors and awkward phrasing throughout. For instance, the first sentence should use “also serve” instead of “serve also.” Additionally, the phrase “in an ASIC” in the second sentence (lines 12–13) is unnecessary. I strongly recommend that the authors rewrite the entire abstract to improve clarity, structure, and grammar.
2. For a journal article, the introduction is insufficient in depth and quality. Almost every sentence contains grammatical issues. For example, in the sentence spanning lines 32–33, the word “also” should precede “should.” Likewise, the sentence between lines 22–23 could be better phrased as: “However, in numerous applications, power and performance are not paramount.” I encourage the authors to carefully review the MDPI author guidelines and rewrite the introduction accordingly, with attention to both language and structure.
3. This section presents only a single equation to explain the design concept, which is inadequate for a journal submission. The authors should include a schematic of the circuit and, if possible, provide simulation results to give readers better context for Equation (1).
4. I appreciate that the authors mention the color of the measurement plot lines in the text, which is helpful since the plot labels themselves are hard to read. This is a commendable effort to improve clarity. However, the results section, while including experimental data, lacks the rigor expected in a journal article. The authors should compare their design with existing published work to highlight its novelty and performance.
5. The authors mention that the input signal frequency affects measurement accuracy. It would be helpful to include a summary table that shows the effect of frequency on accuracy and gain. This would enhance the reader’s understanding and improve the clarity of the presented findings.

Comments on the Quality of English Language

The manuscript contains numerous grammatical errors, awkward sentence constructions, and inconsistencies that affect readability. I strongly recommend that the authors consider using MDPI’s English editing service or seek assistance from a professional language editor to improve the clarity and quality of the manuscript.

Author Response

Dear Reviewer,

Thank You for the valuable comments concerning our manuscript. We have tried to address all Your remarks and to introduce the suggested changes. The altered parts of the manuscript were highlighted in the current version with the yellow background. Below we describe in detail the actions, which have been taken regarding each individual comment.

Regarding the English language question raised by the Reviewers, the manuscript has been read by a Canadian native speaker from the Ontario province, an IEEE Fellow, and the corrections suggested by him have been introduced in the revised version of the manuscript.

Best Regards

Mariusz Jankowski, Michał Szermer and Marcin Janicki

Reviewer 1

Preliminary part:

This paper presents the design, fabrication, and testing of an electrothermal filter (ETF) implemented in 3 μm CMOS technology. The work is positioned as both a sensor and a potential building block for temperature-to-frequency conversion. I appreciate that the authors have implemented their design and included experimental results to validate it.

Reply:

Thank You for the overall assessment of our manuscript.

Particular questions and remarks:

1. The abstract is too brief and does not conform to MDPI guidelines. Moreover, it is poorly written, with grammatical errors and awkward phrasing throughout. For instance, the first sentence should use “also serve” instead of “serve also.” Additionally, the phrase “in an ASIC” in the second sentence (lines 12–13) is unnecessary. I strongly recommend that the authors rewrite the entire abstract to improve clarity, structure, and grammar.

For a journal article, the introduction is insufficient in depth and quality. Almost every sentence contains grammatical issues. For example, in the sentence spanning lines 32–33, the word “also” should precede “should.” Likewise, the sentence between lines 22–23 could be better phrased as: “However, in numerous applications, power and performance are not paramount.” I encourage the authors to carefully review the MDPI author guidelines and rewrite the introduction accordingly, with attention to both language and structure.

Reply:

We have tried to adhere to the MDPI guidelines and the template provided on the journal webpage. Both the abstract and the introduction have been extended providing deeper theoretical background. The entire manuscript has been carefully revised and read by a native speaker.

2. This section presents only a single equation to explain the design concept, which is inadequate for a journal submission. The authors should include a schematic of the circuit and, if possible, provide simulation results to give readers better context for Equation (1).

However, the results section, while including experimental data, lacks the rigor expected in a journal article. The authors should compare their design with existing published work to highlight its novelty and performance.

Reply:

Section 2 in the original manuscript was intended to introduce briefly the ETF principles of operation with the proper references to the publications of Makinwa et al. Now, except for the additional information provided in the introduction, we have decided to include it as a part of the previous Section 3. The newly created Section 2, including the required schematics, describes in detail both the theory and the practical realization of the ETF. Unfortunately, it was not possible to simulate the ETF performance before its manufacturing, because resistor models provided by the manufacturer do not take into account the variations of resistor values with temperature and the only way to verify it was to carry out measurements.

Additionally, in the summary of the experimental part our design is compared with those presented in other published works. However, we still believe that the main novelty consists not in the improved performance of the ETF, but in the analysis of its operation at different input signal frequency values.

3. I appreciate that the authors mention the color of the measurement plot lines in the text, which is helpful since the plot labels themselves are hard to read. This is a commendable effort to improve clarity.

Reply:

We have introduced in the text the descriptions of the colours used for plot lines in the figures. Moreover, when possible, the font size has been increased, so that the labels should be legible when printed on A4 or letter size sheets.

4. The authors mention that the input signal frequency affects measurement accuracy. It would be helpful to include a summary table that shows the effect of frequency on accuracy and gain. This would enhance the reader’s understanding and improve the clarity of the presented findings.

Reply:

Indeed, the problem of the influence of signal frequency on measurement accuracy has not been properly formulated in the original manuscript. We would like to point out that the phase shift is more sensitive to temperature at frequency values for which the amplitude of the thermopile output signal is significantly attenuated, what in turn might affect the measurement accuracy because of the input range of the ADC used in the measurement equipment. The part of the text concerning the measurement accuracy has been rephrased and an appropriate table has been added in the revised manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

This paper investigates the design and measurements of the electrothermal filter. 

1. What is the difference between the electrothermal filter proposed in the paper and the existing electrothermal filter?
2.Whether the calibration results in Fig. 4a is the measurement results when the silicon die eoposesed directly to the enviroment? If no, please supplement the comparative experiment.
3.Why 500mV/div and 10mV/div were selected? If authors wanted to compare with electrothermal filters, please show this. Otherwise, authors should mention the reason of this channel sensitivity.
4.Some typos. and format, font size mismatching (ex. customer needs. [4-6] at intro. etc.)

Author Response

Dear Reviewer,

Thank You for the valuable comments concerning our manuscript. We have tried to address all Your remarks and to introduce the suggested changes. The altered parts of the manuscript were highlighted in the current version with the yellow background. Below we describe in detail the actions, which have been taken regarding each individual comment.

Regarding the English language question raised by the reviewers, the manuscript has been read by a Canadian native speaker from the Ontario province, an IEEE Fellow, and the corrections suggested by him have been introduced in the revised version of the manuscript.

Best Regards

Mariusz Jankowski, Michał Szermer and Marcin Janicki

Reviewer 2

Basic questions and remarks:

1. What is the difference between the electrothermal filter proposed in the paper and the existing electrothermal filter?

Reply:

We are not quite sure what the Reviewer means. The electrothermal filter considered in this manuscript is the same as in the original conference paper related to this special issue of the journal. Compared to the electrothermal filters presented in the cited literature, the manufacturing technology and the geometrical parameters are different.

2. Whether the calibration results in Fig. 4a is the measurement results when the silicon die exposed directly to the environment? If no, please supplement the comparative experiment.

Reply:

The chart presented in Figure 4a shows the measurement results obtained with the package opened, but this does not affect the measured steady state resistance values. On the other hand, the removing of the lid reduced the time of IC temperature stabilisation after each change of a preset value in the chamber and significantly accelerated measurements. An appropriate explanation has been added in the text.

3. Why 500mV/div and 10mV/div were selected? If authors wanted to compare with electrothermal filters, please show this. Otherwise, authors should mention the reason of this channel sensitivity.

Reply:

The amplitude of the sinusoidal input signal applied to the heating resistor is approximately 5 V_p-p, whereas the amplitude of the output signal at the thermopile is significantly attenuated (see Figure 6a), even lower than 5 mV_p-p. Thus, when using the same vertical scale for both channels, the output signal would look like a flat line. This is why we are using different scales. Another, even more important reason for using different scales, is the measurement accuracy. Namely, when a signal on the screen extends vertically through more divisions, the measurement results recorded in the file are more accurate, because the input range of the ADC is better used. Nevertheless, we have changed Figure 5b presenting the screenshot in which both sinusoidal curves have comparable height, thou still the vertical scales in the measurement channels are different, and the phase difference can be better observed. An explanation concerning the use of different scales also has been added.

4. Some typos. and format, font size mismatching (ex. customer needs. [4-6] at intro. etc.)

Reply:

We have tried to adhere to the MDPI template provided on the journal webpage. The entire manuscript has been carefully revised and read by a native speaker.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I appreciate the authors for thoughtfully incorporating my suggested changes. I am satisfied with all the revisions made.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

1. Whatis the reason why the total resistance of the designed ETF in this paper expected to be 960Ω? Is it a performance requirement for ETF? Or is it mentioned in the existing literature?
2. What are the conditions for obtaining 9 different samples? Or what are the differences between 9 different samples?

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf