0201 Evaluation of a Commercial-Off-The-Shelf Fluxgate Magnetometer for CubeSat Space Magnetometry

8/11/2019 0201 Evaluation of a Commercial-Off-The-Shelf Fluxgate Magnetometer for CubeSat Space Magnetometry

1/14

Copyright A. Deepak Publishing. All rights reserved.

www.JoSSonline.comwww.DeepakPublishing.com

JoSS, Vol. 2, No. 1, p. 133

Abstract

A low-cost, commercial-off-the-shel (COS) fluxgate magnetometer suitable or space magnetometry ap-

plications on a CubeSat mission is proposed as a cost effective science grade magnetometer. Tree commercial-off-the-shel fluxgate magnetometers were identified and evaluated against the ollowing criteria: power efficiency,weight, noise, linearity and adaptability. Te evaluation was concluded with environmental testing o the selectedmagnetometer, which included thermal cycling, unctional temperature and vibration tests as a measure o spacequalification o the instrument. Separation o the sensor and electronics was one o the major modifications toenable the sensor to be deployed on a boom, which is important to minimize magnetic intererence rom the satel-lite, while the electronics board is placed within the satellite body. Te LEMI-011B fluxgate magnetometer passedthe environmental tests and was proposed as suitable or CubeSat deployment. Te perormance characteristics othe magnetometer afer modification and environmental tests are as ollows: 0.7 n (rms) noise at 12.83 Hz, 2 nover 60000 n non-linearity and 0.03 W power consumption.

Evaluation o a Commercial-Off-the-ShelFluxgate Magnetometer or CubeSat Space

Magnetometry

Electdom Matandirotya1, Robert R. Van Zyl1,

Daniel J. Gouws2

, and Elda F. Saunderson2

1Cape Peninsula University o echnology, Faculty o Engineering, Department o Electrical Engineering, Bellville,Western Cape, South Arica

2South Arican National Space Agency: Space Science, Hermanus, Western Cape, South Arica

1. Introduction

CubeSat developments at Universities offer a plat-orm or students to realize space-based research goalswithin a short time rame and at relatively low cost. A

CubeSat is a nano-satellite with standard dimensions o100 x 100 x 100 mm and mass less than 1.33 kg. Withthis orm actor the CubeSat is generally reerred toas a 1U (Munakata, 2009). Te use o commercial-off-the-shel (COS) components in CubeSat designs

Matandirotya E., et al. (2013): JoSS, Vol. 2, No. 1, pp. 133-146(Peer-reviewed Article available at www.jossonline.com)


2/14


lowers its cost and development time. Consideringthe mass and volume o a CubeSat, associated pay-loads must be compatible with the satellite bus to avoidstrain, especially on the satellite power budget, which istypically limited to 2- 3 W or a 1U. A candidate mag-

netometer is proposed or CubeSat space magnetom-etry. Te work presented in this paper was conductedin parallel with the development o two CubeSats, a 1Uand a 3U, by the students at the Cape Peninsula Uni-versity o echnology under the FSAI nano-satellitedevelopment program.

Te use o magnetometers in space dates back to1958, when a fluxgate magnetometer was used or de-termining satellite orientation on Sputnik-3 (Diaz-Mi-chelena, 2009). Magnetometers in satellite applicationsare used or either orientation purposes or as science-

grade payloads. Orientation magnetometers are incor-porated in the attitude determination and control sys-tem (ADCS) or the stabilisation o the satellite in orbit.Science-grade magnetometers are designed or definedscientific experiments such as measurements o thegeomagnetic field in the vicinity o the satellite (Acuna,2002). Magnetic measurements are a significant tool inthe derivation o undamental characteristics and be-haviour o high energy particles and plasma in space,as well as mapping o the Earths magnetic field (Langelet al., 1982; Olsen et al., 2003). During adverse spaceweather, i.e., during magnetic storms, the high ener-gy particles have an influence on the Earths magneticfield pattern and, hence, magnetic measurements aid inthe understanding o the effects o such space weatherevents. (Space weather is defined as conditions on theSun and in the solar wind, magnetosphere, ionosphere,and thermosphere that can influence the perormanceand the reliability o space-borne and ground basedtechnological systems, and can endanger human lie(Moldwin, 2008).)

Both vector (e.g., fluxgate; vector helium) and sca-lar (e.g., proton precession; optically pumped) magne-tometers can be used or magnetic measurements inspace (see also Acuna, 2002 and reerences therein),depending on the mission specification. Progressivetechnological advancement led to a shif rom ana-

logue to digital signal processing, which comes withthe benefit o lower power consumption and reduc-tion in overall size o the magnetometer unit (Ciudadet al., 2010; Forslund et al., 2007). A suitable deploy-ment boom and a space-hardened enclosure are neces-

sary to minimize magnetic intererence rom the mainsatellite body, as well as shield the sensor rom radia-tion (Diaz-Michelena, 2009; OBrien et al., 2007) re-spectively. Te desired operational characteristics o aspace-grade nano-satellite magnetometer are mechani-cal and thermal robustness, compact orm actor, lowpower consumption, and low mass. Low noise levelsand stable offsets enhance measurement accuracy. Temeasurement range o the magnetometers should bewide enough (above 50000 n) to accommodate theEarths field, as well as any fluctuations resulting rom

space weather anomalies (Ciudad, 2010). Te recom-mended attributes o a space magnetometer are listedin able 1 (Balogh, 1999).

able 1. Characteristics o Space Magnetometers (adopted romBalogh, 1999)

Measurement Range, n 60000

Band width, Hz 0 - 30

Resolution, n 1 - 5

Noise, n < 5

Te evaluation procedure, environmental tests

and results are described. In this paper, we discuss thereasons which lead to the selection o the LEMI-011Bfluxgate magnetometer as a candidate or use in spacemagnetometry or small satellites, and the necessarymodifications required to the instrument.

2. Description and Selection of FluxgateMagnetometer

Fluxgate magnetometers, which are vector mag-netometers capable o measuring the strength and thedirection o the magnetic field, generate an electrical

1French South Arican Institute o echnology (http://active.cput.ac.za/sati/)

Matandirotya E., et al.

JoSS, Vol. 2, No. 1, p. 134


3/14


4/14


eter, namely the so-called linearity, thin shell, and noisetests. Te Earths magnetic field was recorded duringthese tests to monitor any geomagnetic changes thatmay be detected during the test, and dynamic correc-tions were applied to the coils. Te control algorithmso the different tests will not be discussed in this paper.

Linearity est: A linearly varying magnetic field isgenerated in the coils system axes in 1000 n step sizesat five seconds intervals rom 60000 n to 60000 n.Te result o this test is the difference between the dataset o the applied magnetic field and the field measuredby the magnetometer. Te non-linearity (NL) o themagnetometer was calculated as the maximum differ-ence between the applied and measured magnetic fielddata sets.

Tin Shell est: Te thin shell test is executed on a

sensor placed in the Helmholtz coils to evaluate the sen-sors response to the applied magnetic field. While themagnetometer is in the Helmholtz coil, a specific num-ber o magnetic field vectors are applied, depending onthe capabilities o the coils system and the desired levelo accuracy. Te vectors all have a constant magnitude,but random orientations, so as to create a three dimen-sional sphere o vectors. A spherical harmonic analysisalgorithm is applied that gives the spectral represen-tation o the response o each magnetometer axis at agiven field magnetic strength (Risbo et al., 2000). Te

resultant matrix solution obtained rom the algorithmcontains inormation relating to the calibration param-eters, such as the sensitivity, offsets, non-orthogonalityand misalignments o each channel o the sensor. Teseparameters are applied in the calibration matrix or er-ror compensation. For the particular test perormed onthe magnetometers in this paper, 200 vectors o 50000

n magnitude were applied in the Helmholtz coils sys-tem. A detailed description o the Tin Shell Teoremand how the spherical harmonic model is applied dur-ing magnetometer calibration is described by Risbo etal. (2000) and also by Sipos et al. (2011).

Noise est: Te magnetometer is placed in a mag-netically clean building with low magnetic noise or in-tererence. No external magnetic field is applied to themagnetometer, and the output o the magnetometer isrecorded at a requency o 12.83 Hz (this requency isnot a documented specification, but what the test acili-ties are capable o executing). Te standard deviationand the peak-to-peak noise o the recorded data arecalculated. Te Earths magnetic field is recorded as areerence or any changes in the background magneticfield during the test.

3.2 Correction Matrix

Te correction matrix consists o: Sensor sensitivity coefficients, which convert

the measured output voltage o each magne-tometer axes to a magnetic field strength in nTey orm the diagonal elements o the correc-tion matrix.

Sensor non-orthogonalities and misalignmentsorm the off-diagonal elements o the correc-

tion matrix and correct or non-orthogonalitiesbetween the sensor axes and misalignment withthe reerence axes.

Te calibration parameters are applied as indicatedin Equation (1):

able 2. Parameters o the three evaluated magnetometers.

MagnetometerMass (g)(Sensor + electronics)

Power supply(V)

Range (n) Power (W) Sensitivity Dimensions(mm)

Noise (ptp)(n)

LEMI-011B 120 5 0.25 60 000 0.03 27.4 (n/mV) 55 x 51 2

LEMI-011 120 5 0.25 50 000 0.03 27.4 (n/mV) 155 x 20 2

MF1D 155 9 70 000 1.5 (digital output) 55 x 44 2


JoSS, Vol. 2, No. 1, p. 136


5/14


3.3 Sensor Alignment

During the calibration o the magnetometer in theHelmholtz coils system, an alignment method knownas current flipping was used. During current flipping,the field in the y-axis o the coils is flipped to a setpositive and negative value. While flipping, no fieldis applied along the x-axis o the coils. Te x-axis othe magnetometer is gradually rotated up to a pointwhere the output o the x-axis o the magnetometer isthe same or positive and negative y-fields. Tis impliesthat the x-axis o the magnetometer is measuring nofield generated by the y-coils, but only representing the

electronic offsets o the magnetometer; thus the x-axiso the magnetometer is orthogonal to the y-field o thecoils. When orthogonality o the two axes has beenachieved, the magnetometer is deemed to be alignedwith the corresponding axis o the coils.

3.4 Calibration Results

Te main criteria or the selection o the magne-tometers were based on the sensor linearity and noiselevels. Te noise and linearity characteristics o the can-

didate magnetometers are shown in Figures 2 to 4, onthe ollowing page. Te results explained in this sectionare calibration results derived beore any modificationo the magnetometers.

From Figure 2, though the linearity o the magne-tometer was not well defined, the major concern wasin the y axis, which seemed to have a cyclic variationwhen a positive field was applied. Te problem was sus-

pected to be due to a system error in the hardware, asmore than one magnetometer o the same group hadsimilar calibration results. Te noise also exhibited asinusoidal wave pattern with amplitudes o ~10 n.

As illustrated in Figures 3 and 4, the LEMI 011Band LEMI 011 showed some similarity in the linear-ity pattern. A maximum o 3 n non-linearity over60000 was recorded or the two magnetometers. Tepeak-to-peak noise o the LEMI-011B was lower thanthat o the LEMI 011 magnetometer. Te similaritieswere expected, as the sensor configuration is similar andthe only difference is in the electronics control boardsable 3 is a urther extraction o the results rom the

calibration tests. A result worth noting is the high, butstable, offsets o the LEMI 011B magnetometer afer itsfirst calibration tests. It was expected that these offsetswould decrease significantly afer the separation o thesensor and the electronics.

4. Modification

4.1 Rationale

A decisive step in the evaluation o the three mag-

netometers was the separation o the sensor rom theelectronics. Te separation was necessary to reduce themagnetic intererence rom the current flowing withinthe enclosure. Modification warrants re-evaluation otheir perormance. Challenges in the modification othe magnetometers influenced the selection o the can-didate magnetometer, as explained below.

(1)

Where:Field(X,Y,Z) = measured fields or the three magnetometer axes in n;C

xx, C

yy, C

zz= sensor sensitivity or the x, y, z axes, respectively, in n/mV;

Cxy, C

xz, C

yx, C

yz, C

zx, C

zy= non-orthogonalities and misalignments, in n/mV;

Output(X,Y,Z) = voltage readings or each axis in mV;Offsets(X,Y,Z)= the electronic offsets or each axis in n.

Evaluation o a Commercial-Off-the-Shel Fluxgate Magnetometer or CubeSat Space Magnetometry

JoSS, Vol. 2, No. 1, p. 137


6/14


Figure 2. Graphical representation o the non-linearity and noise o the MF-1D magnetometer.

Figure 3. Graphical representation o the non-linearity and noise o the LEMI-011B magnetometer.

Figure 4. Graphical representation o the non-linearity and noise o the LEMI-011 magnetometer.


JoSS, Vol. 2, No. 1, p. 138


7/14


able 3. Calibration Perormance o Selected Magnetometer.

Axis Noise (ptp) (n)Standard deviationo noise (n)

Non-linearity(n)

Outputoffsets (n)

Separation o sensorand electronics

MF1D x 22 4.631 18.5 252 Not possibley 22 4.636 300 -397

z 34 7.617 20 -457

LEMI-011B

x 4 0.542 3 -1627

possibley 4 0.484 2 -1246

z 4 0.631 2 -1445

LEMI-011

x 6 1.302 3 -114

possibley 6 1.145 3 -67

z 7 1.077 3.5 57

a) b)

Figure 5. a) Preliminary enclosure and b) boom designed or calibration purposes.

4.2 Modification Outcome

Te sensor and electronics o the MF1D magne-tometer could not be separated due to potting o thedevice, which limited urther evaluation o the mag-

netometer. It was possible to separate the sensor othe LEMI 011. However, the length o the electronicsboard (130 mm) would present integration challengeswith the CubeSat orm actor, so urther evaluation othe magnetometer was not perormed.

Te 20 g (50 x 16 x 16 mm) LEMI 011B sensor was

separated rom the electronics. Te LEMI 011B elec-tronics board was ound to be o a convenient size (51mm 55 mm), that could be fitted in either a 1U or3U CubeSat without urther structural modificationA 100 mm boom and enclosure (both made rom alu-

minum) were designed or evaluation purposes. Tedesigns are shown in Figure 5. A 9-pin connector andadditional connecting wires were used to interace thesensor and its electronics through the hollow alumi-num boom. Figure 6 illustrates the position o the sen-sor beore and afer separation.


JoSS, Vol. 2, No. 1, p. 139


8/14


Figure 6. Te sensor position beore and afer separation.

5. Environmental esting

An essential point to note when executing envi-ronmental tests is the element o misrepresentationo the space environment by the use o test chambers.Te spacecraf (in this case the CubeSat) may have anorbital period o 100 min, and yet the chamber maynot have the capacity to achieve the rapid temperature

ramps desired. Termal regions, which prescribe thetemperature limits or qualification and acceptancetesting, are determined by a dedicated thermal analysisteam. Four proto-flight qualification level tests, i.e., thethermal vacuum, thermal cycling, unctional tempera-ture and vibration tests, were proposed to determinethe robustness o the instrument to survive the launch

and the harsh orbital environment (Korepanov, 2003)Te thermal vacuum test could not be executed satis-actorily, because o technical problems experiencedwith the thermal vacuum test during the research pe-riod. Tese results are, thereore, not reported here

However, it is worth noting that the magnetometer didnot ail during the test. wo environmental tempera-ture tests were ully executed; the procedures are de-scribed in Appendix A.

5.1 Termal esting Margins

Tree temperature margins are considered duringthe thermal testing o space-based instruments. Tesemargins are cited by Gilmore (2002) and Goodman, etal. (2006) as:

Termal Uncertainty Margin (UM): Tis marginis considered a saety margin. Te thermal uncertaintymargin is applied as an additional temperature marginto the worst case predicted temperature limits.

Acceptance est Termal Margin (AM): Tismargin determines the maximum allowable temper-ature range or flight. It is added to both ends o thepredicted worst case hot/cold temperature range. Tismargin is typically 5 C.

Qualification est Margin (QM):Tis is the tem-perature above the expected operational environmenttemperature. It is a measure o the ruggedness o thedevice under test (DU) design. Design flaws are ex-posed during the test.

In the presented results, the temperature valuesused during the test were as predicted by the satellitethermal models, excluding any UM. Tese marginswere applied according to the MIL-SD-1540B stan-dards (EverySpec, 2009). Te temperature ranges pre-dicted (beore applying the margins) or electronicswithin the satellite were + 5 C to + 35 C, while 10

C to + 35 C were predicted or electronics outsidethe satellite body. Te added margins or the tests were 10 C (AM) and 16 C (QM). Te maximumqualification test temperature ranges were expected tobe 11 C to 61 C or electronics within the satelliteand 26 C to + 61 C or electronics outside the satel-lite body.


JoSS, Vol. 2, No. 1, p. 140


9/14


5.2 Vibration est

Te vibration test is usually launcher-standardised.Te main objective o the vibration test is to subject theinstrument to the vibration profile o the launch ve-

hicle in order to assess the structural integrity o thesubsystem to withstand the orces that are exerted onit during launch.

Figure 7 shows the assembly o the electronicsboard and the sensor separately on the vibration jig.For this particular test, the vibration test limits appliedto the magnetometers are shown in Figure 8. Te speci-fications are derived rom consolidating a standard thatsatisfies all specifications o potential launch vehiclesas identified by survey. Te maximum vibration am-plitude that was used was 10.26 g rms. Te sine levels

listed in able 4 were recommended at a sweep rate o2 Octaves/min. Te launch vehicles that were consid-ered in the consolidation were the Dnepr, Soyuz, Strela,Volna, Falcon and PSLV. Teir vibrations specificationsare not given in this paper, but can be obtained romthe respective operators.

able 4. Specification o the Sinusoidal Levels During the Ran-dom Vibration est.

DirectionFrequency band(Hz)

Sinusoidal ampli-tude (g or mm)

Longitudinal 4 - 1010 mm peak topeak

Longitudinal 10 - 100 3.75 g

Lateral 2 - 810 mm peak topeak

Lateral 8 - 100 2.5 g

5.3 emperature Cycle est

Workmanship, material and process flaws can be

detected during the temperature cycle test (Korepa-nov, 2003). Figure 9 illustrates the temperature profileinside and outside the chamber during the our pre-scribed temperature cycles. Several temperature sen-sors were used or temperature monitoring to enhancethe accuracy o the measurements. Te magnetometer

a) b)

Figure 7. Mounting o the equipment on a vibration test jig a) set up or the transverse random test and b) or the longitudinal andvibration tests.


JoSS, Vol. 2, No. 1, p. 141


10/14


was switched off during the temperature cycle test. Te

procedure and temperature limits are reported in ableA1 o Appendix A.

5.4 Functional emperature est

Te unctional temperature test verifies the unc-tionality o the instrument within the design tempera-ture range as determined by the thermal model. Te

temperatures o the electronic board and the sensor

were monitored. Figure 10 (next page) shows the tem-perature profile within the test chamber. Te magne-tometer was switched on and off at different intervalsincluding a hot start at maximum temperature (61 C)and a cold start at the minimum temperature ( 26 C),while veriying its unctionality. Te procedure andtemperature limits are reported in able A2 o Appen-dix A.

Figure 8. FSAI vibration test limits.

Figure 9. emperature profile in the chamber during the temperature cycle test.


JoSS, Vol. 2, No. 1, p. 142


11/14


6. Performance Analysis

o evaluate the unctionality o the magnetometerafer every environmental test, a re-calibration processwas perormed on the modified magnetometer. Tepurpose o the re-calibration process was to evaluate

the effects o the environmental tests on the peror-mance o the magnetometer.

ables 5 and 6 (next page) show the sensor param-eters afer each stage o environmental testing. Tesignificant drop in the electronic offsets o the sensorafer separation rom the electronics is notable in able5. Te reduction in the offsets can be ascribed to theincreased distance between the sensor and the elec-tronics, as the permanent magnetism o the electron-ics would affect the sensor. Te sensitivity o the sensorwas also affected slightly; however, it continued to be

stable afer sensor separation. Furthermore, it can beseen rom able 5 that there are relatively small differ-ences in the offsets o the separated sensor afer eachenvironmental test. Tese differences may have result-ed rom misalignment o the sensor axes and the cor-responding Helmholtz coil axes during the calibrationprocess. Tese differences are not critical at this stage,

as the final magnetometer offsets will be determinedduring in situcalibration o the magnetometer on thesatellite beore launch.

able 6 shows the sensor noise and the standard de-viation o the noise afer each step o the modificationand environmental tests. Te noise levels listed in able

6 show that the modification and environmental testsdid not degrade the perormance o the magnetometersignificantly; thus, harsh orbital conditions should nothave a negative impact on the noise o magnetometerwhich is a significant measure o magnetometer stabil-ity.

7. Recommendation and Conclusion

Te LEMI-011B magnetometer is proposed ornano-satellite space magnetometry or or ADCS. Te

magnetometer, which is a tri-axial sensor, was selectedas the best among the three magnetometers that wereevaluated, based on the supporting perormance pa-rameters shown in able 3. It exhibits good linearityover its measurement range, as indicated in Figure 3Te modified LEMI-011B sensor passed the environ-mental tests to which it was subjected. A space-qualified


JoSS, Vol. 2, No. 1, p. 143

Figure 10. emperature profile within the test chamber during the unctional temperature test.


12/14


able 5. Sensor Sensitivity and Offset o the Modified LEMI-011B During the Different Stages o the Evaluation.

Evaluation stage x-axis y-axis z-axis

Sensor sensitivity (mV/n)

Beore separation 27.50 27.34 27.44

Afer separation o sensor and electronics 26.47 26.50 26.50

Afer vibration test 26.47 26.51 26.49

Afer temperature cycle test 26.46 26.51 26.48

Afer temperature unctionality test 26.46 26.51 26.48

Sensor offset (n)

Beore separation 1638 1233 1461

Afer separation o sensor and electronics 89.24 83.55 32.73

Afer vibration test 87.12 117.84 75.84

Afer temperature cycle test 106.7 99.20 60.11

Afer temperature unctionality test 97.94 107.83 114.73

able 6. Noise Characteristics o the Modified LEMI-011B Sensor During Different Stages o Evaluation.

Evaluation stage x-axis y-axis z-axis Average

Peak to peak (ptp) noise (n)

Beore separation 3.995 3.421 4.194 3.870

Afer separation o sensor and electronics 4.743 3.878 3.878 4.166

Afer vibration test 4.743 3.160 3.591 3.831

Afer temperature cycle test 4.722 3.864 4.007 4.198

Afer temperature unctionality test 5.154 3.578 3.434 4.055

Standard deviation o the noise (n)

Beore separation 0.542 0.484 0.632 0.553

Afer separation o sensor and electronics 0.762 0.552 0.723 0.679

Afer vibration test 0.724 0.518 0.575 0.606

Afer temperature cycle test 0.723 0.631 0.746 0.700

Afer temperature unctionality test 0.760 0.576 0.563 0.633


JoSS, Vol. 2, No. 1, p. 144

and space weather resistant enclosure can be designedso that the sensor is well protected. Te LEMI-011Bcan be obtained rom the suppliers at less than US$600per unit. Additional modification, which may includedevelopment o a digital interace as the original mag-

netometer has an analogue interace, can be achieved

in a standard laboratory.Further research is required to assess i the sepa-

ration o 100 mm between the sensor and the satelliterame is sufficient to mitigate the influence o the mag-netic characteristics o a typical CubeSat bus.


13/14


Appendix

EMPERAURE CYCLE AND FUNCIONAL EMPERAURE ESS: PROCEDURES AND EMPERAURE LIMIS

able A1. emperature Cycle est Procedure

Procedure

1. Perorm the unctional tests at ambient temperature2. Cool down to low temperature (power off)3. Allow 1 hour dwell time at the minimum extreme temperature4. Increase temperature to maximum extreme5. Allow 1 hour dwell time at maximum extreme temperature6. Decrease temperature to ambient7. Repeat steps 2 to 6 or our cycles or acceptance8. Perorm unctional tests at ambient temperature afer completion

Pressure level: Ambient

emperature Limits: Inside spacecraf:

Qualification Acceptance

Maximum extreme +61 C Maximum extreme +51 C

Minimum extreme -21 C Minimum extreme -11 COutside spacecraf:


Minimum extreme -36 C Minimum extreme -26 C

Gradient: During heating: During heating:1 C/min (nominal)

During cooling:0.2 C/min (nominal)

Number o C ycles: 4 cycles (Qualification); 2 cycles (Acceptance)

able A2. Functional emperature est procedure

Procedure

1. Start at ambient temperature (power on)2. Perorm the unctional tests

3. Increase temperature to the maximum extreme (with power on)4. Switch off the power o the test item5. Allow 1 hour dwell time at the maximum extreme temperature6. Perorm a hot start and repeat unctional tests7. Cool down to minimum extreme (with power on)8. Switch off power o the test item9. Allow 1 hour dwell time at the minimum extreme temperature10. Perorm a cold start and repeat unctional tests11. Allow test item to return to ambient temperature (with power on)12. Repeat unctional tests

Pressure level: Ambient

emperature Limits: Inside Space Craf Inside Space Craf


Minimum extreme -11 C Minimum extreme -26 C

Gradient: During heating: During heating: 1 C/min (nominal)

During cooling: 0.2 C/min (nominal)

Number o Cycles: One


JoSS, Vol. 2, No. 1, p. 145


14/14

Acknowledgements

Te opportunity to use the acilities o SANSASpace Science2, ISSA3, SunSpace4 and the University o

Stellenbosch5throughout the calibration and environ-mental testing is greatly appreciated.

References

Acuna, M. H. (2002): Space-based magnetometers.Review o Scientific Instruments,Vol.73, pp. 3717-3736.

Balogh, A. (1999): Space instrumentation. Space andAtmospheric Physics. Available: www.sp.ph.ic.ac.

uk/~balogh/spinstr.htm. [Accessed 12 March2011].Ciudad, A., et al. (2010): Small fluxgate magnetometers:

development and uture trends in Spain, Sensors,Vol. 10, pp. 1859-1870.

Diaz-Michelena, M. (2009): Small magnetic sensors orspace applications. Sensors, Vol. 9, pp. 2271-2288.

Digital Magnetometer MF-1D. (2009): echnicalreerence and users manual. Available: http://www.innalabs.com/ [Accessed 14 November 2010].

EverySpec, (2009): Military standard: test requirements

or space vehicles (10 Oct 1982). Available: www.everyspec.com/MIL-SD/MIL-SD-1540_17789/[Accessed 25 March 2011].

Forslund, A., et al. (2007.): Miniaturized digital fluxgatemagnetometer or small spacecraf applications.Measurement Sci. and ech., Vol. 19. pp 1-10.

Gilmore, D. G. (2002): Spacecraf Termal ControlHandbook: Fundamental echnologies, Vol. 1. ElSegundo, Caliornia: Te Aerospace Press.

Goodman, J., et al. (2006): Environmental (thermal)testing o space instrumentation: the GLASexample. Nucl. Instruments & Methods in PhysicsResearch A,Vol. 563, pp. 377-380.

Korepanov, V. E., (2003): Te modern trends in spaceelectromagnetic instrumentation. Advanced SpaceResearch, Vol. 32, pp. 401-406.

Langel, R., et al. (1982): Te MAGSA mission

Geophysics Research Letter, Vol. 9, pp. 243-245.Lenz, J. and Edelstein, A. S. (2006): Magnetic sensors

and their application. Sensors, Vol. 3, pp. 631-648.LvivCenter o Institute o Space Research. (2009)

Available: http://www.isr.lviv.ua/lemi011.htm[Accessed 3 June 2011].

Merayo, J M G, et al. (2000): Scalar calibration o vectormagnetometers.Measurement Sci. and. ech. l. Vol11, pp. 120132.

Munakata, R. (2009): CubeSat design specifications

Available: www.cubesat.org/index.php/documents/develpers [Accessed 18 March 2013].Moldwin, M. (2008):An introduction to space weather

New York: Cambridge University Press.OBrien H., et al. ( 2007): A radiation tolerant digita

fluxgate magnetometer. Measurement Sci. andech.. Vol. 18, pp. 3645-3650.

Olsen, N., et al. (2003): Calibration o the rsted vectormagnetometer. Earth Planets Space, Vol. 55, pp. 11-18.

Quintero, A.H., et al. (1999): Perceptiveness o thermal

vacuum testing presented at the 18th Aerospaceesting Seminar, 16-18 March. Los Angeles, CA.

Ripka, P. (2000): Advances in flux-gate sensors. Sensorsand Actuators.A. Vol. 106, pp. 8-14.

Risbo, ., et al. (2001): rsted calibration missionthe thin shell method and spherical harmonicanalysis. Ground and In-Flight Space MagnetometerCalibration echniques, ESA SP-490.

Sipos, M., et al. (2011): Improvement o electroniccompass accuracy based on magnetometer andaccelerometer calibration, in Proc. 10th SymposiumMagnetic Measurements, Warsaw.


2 South Arican National Space Agency: Space Science, Hermanus, South Arica http://www.sansa.org.za)3 Institute or Satellite and Sofware Application, Houwteq, Grabouw, South Arica4 SunSpace, Stellenbosch, South Arica5 University o Stellenbosch, Stellenbosch, South Arica

Documents

0201 Evaluation of a Commercial-Off-The-Shelf Fluxgate Magnetometer for CubeSat Space Magnetometry