the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Quad-Mag Board for CubeSat Applications
- 1Climate and Space Sciences and Engineering, College of Engineering, University of Michigan, Ann Arbor, MI, USA
- 2The Johns Hopkins University Applied Physics Laboratory, Laurel, MD, USA
- 3Mechanical Engineering, College of Engineering, University of Michigan, Ann Arbor, MI, USA
- 4Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA
- 5Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA
- 6Made In Space Incorporated, Moffett Field, California, USA
- 7General Dynamics Land Systems, Sterling Heights, MI, USA
- 1Climate and Space Sciences and Engineering, College of Engineering, University of Michigan, Ann Arbor, MI, USA
- 2The Johns Hopkins University Applied Physics Laboratory, Laurel, MD, USA
- 3Mechanical Engineering, College of Engineering, University of Michigan, Ann Arbor, MI, USA
- 4Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA
- 5Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA
- 6Made In Space Incorporated, Moffett Field, California, USA
- 7General Dynamics Land Systems, Sterling Heights, MI, USA
Abstract. The design, characteristics, and performance of a CubeSat magnetometer board (Quad-Mag) equipped with four PNI RM3100 magnetometers is presented. The low size, weight, power, and cost of the RM3100 enables the inclusion of four sensors on a single board, allowing a potential factor of two reduction in the noise floor established for an individual sensor via oversampling with multiple sensors. The instrument experimentally achieved a noise floor of 5.345 nT (individual axis), averaging across each axis of the four magnetometers, at a 65 Hz sampling rate. This approaches the previously theoretically established limit for the system of 4.37 nT at 40 Hz. A single on-board, Texas Instrument MSP430 microcontroller handles synchronization of the magnetometers and facilitates data collection through a simple UART-based command interface to a host system. The Quad-Mag system has a mass of 59.05 g and total power consumption of 23 mW while sampling and 14 mW while idle. The Quad-Mag enables 1 nT magnetic field measurements at 1 Hz using commercial-off-the-shelf sensors for space applications.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(3913 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3913 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Brady P. Strabel et al.
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-293', Boris Ginzburg, 26 May 2022
Manuscript describes characteristic and performance of a magnetometer board (Quad-Mag) equipped with four PNI RM3100 magnetometers.
Description is detailed and clear.
To my mind, there are two technical characteristics that should be given consideration:
a) Mutual influence of magnetic sensors placed on the same board (could be evaluated, for instance, with the help of another magnetometer board)
b) Magnetic influence (interference) of board electronics on magnetic measurements being taken (probably similar to a))
Technical: page 5, line 4 - probably, term "thermal drift" instead of "thermal gain" is more customary? (and further in the text);
page 5, line 12 - should be Figure 3a
page 13, line 1 - should be "configured to gain sample"
-
AC1: 'Reply on RC1', Brady Strabel, 04 Aug 2022
First, we want to thank you for taking the time to read the paper and provide useful feedback. Please find our replies to the comments:
a/b) These are valid concerns. We have added a section “Interference” detailing the influence of multiple magnetometers and of board electronics on measurements. Specifically, we quantify the effects on the resolution of a single magnetometer as it samples in different scenarios (i.e. with the quad-mag board absent entirely, on the quad-mag board with no other magnetometers present, and on the quad-mag board with all magnetometers present). We show that the differences in resolution between the cases are negligible and thus we can say the board electronics and multiple sensors do not influence our results.
Technical:
*page 5, line 12 - changed to figure 3a
*page 13, line 1 - changed to “the system was again configured to sample”
-
AC1: 'Reply on RC1', Brady Strabel, 04 Aug 2022
-
RC2: 'Comment on egusphere-2022-293', Anonymous Referee #2, 29 Sep 2022
General comments: An inexpensive and lightweight magnetometer concept is presented. Losses in quality compared to conventional instruments are explained in the preprint. The quality of the instrument is examined with comprehensible tests.
A very similar article was presented in 2018 as "Investigation of a low-cost magneto-inductive agnetometer for space science applications" at https://doi.org/10.5194/gi-7-129-2018. What is new here is that four sensors are operated simultaneously.
Stacking the four simultaneous measurements results in a halving of the expected data errors for statistical reasons. The four measurements at different locations on the circuit board would make it possible to identify interference fields generated by the device itself. (As differences in the measurements). The preprint does not go into this direction. But it is mentioned as an outlook ("undetermined blind source separation").
All in all his preprint is relevant and of interest for the community.
Specific comments:
In chapter2 principles of magneto inductive sensing is explained. I found it hard to understand. The terms „driving the circuit with a positive (forward) or negative (reverse) voltage ...“ pussled me first. Reversing the supply voltage of a Schmitt trigger is in practice certainly not possible. Looking also at the PNI release notes of the PNI-11096 circuit I understand: Coil, resistor and Schmidt trigger form an oscillator, as properly explained in the preview. During this oscillation, the magnetic core material is driven into saturation. This reduces inductance and accordingly influences the oscillation period. The effect is advanced if the surrounding field is parallel to the field produced by the current in the coil. It is reduced if thefield is antiparallel. The coil is reversed by means of electronic switches in the ASIC included in the RM 3100 and both periods are compared to produce readings of the field.
Chapter 4.2 Is it clear that the offset you measure between different sensors, are due to board-borne constant fields? Later on you state large sensor offsets. Can the sensors be swopped on the board to see if it are the sensors itself?
Chapter 4.4 Stability: You state in chapter 4.2: „The values of these offsets generally range from a few hundred nT to a few thousand nT, depending on the axis. In practice, the internal offset changes slightly after every power cycle of the sensor due to its digital components. As a result, the Quad-Mag board requires careful calibration if used for absolute field measurements.“ I think this is a stability item an should be mentioned there.
Technical corrections:
Introduction P1L15. The SWARM satellite mission uses three spacecraft to overcome just this problem.
P9L23: please better use the term „stacked“ than „overlapped“
In the Abstract the last sentence says: The Quad-Mag enables 1 nT magnetic field measurements at 1 Hz using commercial-off-the-shelf sensors for space applications. Don‘t you think this is a bold statement seeing offsets of tenth to hundreds of nT at a single sensor and a resolution above 1nT in a zero Gauß chamber? How about „ The Quad.Mag allows for almost 1nT resolution under optimal conditions.“
-
AC2: 'Reply on RC2', Brady Strabel, 24 Oct 2022
Thank you for taking the time to review and comment on this manuscript. Please find our replies to the comments:
Chapter2: As outlined by Luezinger and Taylor (2010), the described circuit acts as a comparator-based L/R relaxation oscillator that alternates positive and negative current through the non-grounded side of the sensing coil. Indeed, the reverse and forward biasing is done via electronic switches in the RM3100 control ASIC that effectively change the polarity of the coil. Measurements are taken in both directions and compared to generate a zero-centered, positive/negative field value.
Chapter 4.2: An interference section has been added to address related concerns. In short, the offsets present are in fact a combination of board electronics and mutual sensors present on the board.
Chapter 4.4: Added a paragraph to the Stability section reacknowledging this.
Introduction P1L15: Added an additional reference to the SWARM mission in following paragraph covering recent multi-spacecraft missions.
P9L23: Changed to “stack”
Abstract: Changed to “The Quad-Mag enables nearly 1 nT magnetic field measurements at 1 Hz using commercial-off-the-shelf sensors for space applications under optimal conditions.
-
AC2: 'Reply on RC2', Brady Strabel, 24 Oct 2022
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-293', Boris Ginzburg, 26 May 2022
Manuscript describes characteristic and performance of a magnetometer board (Quad-Mag) equipped with four PNI RM3100 magnetometers.
Description is detailed and clear.
To my mind, there are two technical characteristics that should be given consideration:
a) Mutual influence of magnetic sensors placed on the same board (could be evaluated, for instance, with the help of another magnetometer board)
b) Magnetic influence (interference) of board electronics on magnetic measurements being taken (probably similar to a))
Technical: page 5, line 4 - probably, term "thermal drift" instead of "thermal gain" is more customary? (and further in the text);
page 5, line 12 - should be Figure 3a
page 13, line 1 - should be "configured to gain sample"
-
AC1: 'Reply on RC1', Brady Strabel, 04 Aug 2022
First, we want to thank you for taking the time to read the paper and provide useful feedback. Please find our replies to the comments:
a/b) These are valid concerns. We have added a section “Interference” detailing the influence of multiple magnetometers and of board electronics on measurements. Specifically, we quantify the effects on the resolution of a single magnetometer as it samples in different scenarios (i.e. with the quad-mag board absent entirely, on the quad-mag board with no other magnetometers present, and on the quad-mag board with all magnetometers present). We show that the differences in resolution between the cases are negligible and thus we can say the board electronics and multiple sensors do not influence our results.
Technical:
*page 5, line 12 - changed to figure 3a
*page 13, line 1 - changed to “the system was again configured to sample”
-
AC1: 'Reply on RC1', Brady Strabel, 04 Aug 2022
-
RC2: 'Comment on egusphere-2022-293', Anonymous Referee #2, 29 Sep 2022
General comments: An inexpensive and lightweight magnetometer concept is presented. Losses in quality compared to conventional instruments are explained in the preprint. The quality of the instrument is examined with comprehensible tests.
A very similar article was presented in 2018 as "Investigation of a low-cost magneto-inductive agnetometer for space science applications" at https://doi.org/10.5194/gi-7-129-2018. What is new here is that four sensors are operated simultaneously.
Stacking the four simultaneous measurements results in a halving of the expected data errors for statistical reasons. The four measurements at different locations on the circuit board would make it possible to identify interference fields generated by the device itself. (As differences in the measurements). The preprint does not go into this direction. But it is mentioned as an outlook ("undetermined blind source separation").
All in all his preprint is relevant and of interest for the community.
Specific comments:
In chapter2 principles of magneto inductive sensing is explained. I found it hard to understand. The terms „driving the circuit with a positive (forward) or negative (reverse) voltage ...“ pussled me first. Reversing the supply voltage of a Schmitt trigger is in practice certainly not possible. Looking also at the PNI release notes of the PNI-11096 circuit I understand: Coil, resistor and Schmidt trigger form an oscillator, as properly explained in the preview. During this oscillation, the magnetic core material is driven into saturation. This reduces inductance and accordingly influences the oscillation period. The effect is advanced if the surrounding field is parallel to the field produced by the current in the coil. It is reduced if thefield is antiparallel. The coil is reversed by means of electronic switches in the ASIC included in the RM 3100 and both periods are compared to produce readings of the field.
Chapter 4.2 Is it clear that the offset you measure between different sensors, are due to board-borne constant fields? Later on you state large sensor offsets. Can the sensors be swopped on the board to see if it are the sensors itself?
Chapter 4.4 Stability: You state in chapter 4.2: „The values of these offsets generally range from a few hundred nT to a few thousand nT, depending on the axis. In practice, the internal offset changes slightly after every power cycle of the sensor due to its digital components. As a result, the Quad-Mag board requires careful calibration if used for absolute field measurements.“ I think this is a stability item an should be mentioned there.
Technical corrections:
Introduction P1L15. The SWARM satellite mission uses three spacecraft to overcome just this problem.
P9L23: please better use the term „stacked“ than „overlapped“
In the Abstract the last sentence says: The Quad-Mag enables 1 nT magnetic field measurements at 1 Hz using commercial-off-the-shelf sensors for space applications. Don‘t you think this is a bold statement seeing offsets of tenth to hundreds of nT at a single sensor and a resolution above 1nT in a zero Gauß chamber? How about „ The Quad.Mag allows for almost 1nT resolution under optimal conditions.“
-
AC2: 'Reply on RC2', Brady Strabel, 24 Oct 2022
Thank you for taking the time to review and comment on this manuscript. Please find our replies to the comments:
Chapter2: As outlined by Luezinger and Taylor (2010), the described circuit acts as a comparator-based L/R relaxation oscillator that alternates positive and negative current through the non-grounded side of the sensing coil. Indeed, the reverse and forward biasing is done via electronic switches in the RM3100 control ASIC that effectively change the polarity of the coil. Measurements are taken in both directions and compared to generate a zero-centered, positive/negative field value.
Chapter 4.2: An interference section has been added to address related concerns. In short, the offsets present are in fact a combination of board electronics and mutual sensors present on the board.
Chapter 4.4: Added a paragraph to the Stability section reacknowledging this.
Introduction P1L15: Added an additional reference to the SWARM mission in following paragraph covering recent multi-spacecraft missions.
P9L23: Changed to “stack”
Abstract: Changed to “The Quad-Mag enables nearly 1 nT magnetic field measurements at 1 Hz using commercial-off-the-shelf sensors for space applications under optimal conditions.
-
AC2: 'Reply on RC2', Brady Strabel, 24 Oct 2022
Peer review completion
Journal article(s) based on this preprint
Brady P. Strabel et al.
Model code and software
Quad-Mag Data Analysis Brady P. Strabel https://doi.org/10.5281/zenodo.6515198
Brady P. Strabel et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 262 | 85 | 13 | 360 | 7 | 6 |
- HTML: 262
- PDF: 85
- XML: 13
- Total: 360
- BibTeX: 7
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3913 KB) - Metadata XML