Apollo Experience Report Flight Instrumentation Calibration

8/8/2019 Apollo Experience Report Flight Instrumentation Calibration

1/14

A73 - 4 6 8%N A S A T E C H N I C A L N O T E N A SA TN D-7144

APOLLO EXPERIENCE REPORT -FLIGHT INSTRUMENTATION CALIBRATIONby J . F. DeMossManned Spmecrufi CenterHonston, Texus 77058 .

,N A T I O N A L A E R O N A U T IC S A N D S PA CE A D M I N I S T R A T I O N W A S H I N G T O N , D. C. M A R C H 1973


2/14

1. Report No.NASA TN D- 7144 2. Government Accession No. 3. Recipient's Catalog No.4. Title and SubtitleA P O L L O E X P E R I E N C E R E P O R TFLIGHT INSTRUMENTATION.CALIBRATION

5. Supplementary NotesTh e MSC Di rec tor waived the u se of the Internat ional Sy stem of Un i ts (SI) fo r this Apol lo Ex-p e r i en ce R ep o r t b ecau se , i n h i s j ud g m en t, t h e u s e of SI Uni t s would im pa i r t he usefu lness of t her e p o r t o r r e s u l t i n e x c e s s i v e c o s t.T h r ee t y p es of i n s t ru m en t a t io n -ca l i b r a ti o n d a t a w e re u sed i n t h e A po ll o P ro g ram t o p ro v i d e th eco r r e c t en g i n ee r i n g d a t a fo r t e s t s an d m i s s i o n su p p or t. T h e co m m an d an d s e rv i ce m o d u leins trumenta tion-component p roc urem ent spec i f ica t ions req ui red ind ividua l -component ca l ib ra tion ,add ca l ib ra t ion da t a for t h ese ind iv idua l components (convent iona l -cal ib ra t ion da t a ) we re a lw aysu sed fo r m i s s i o n d a t a su p p o rt . A me an sta nd ard type of cal ibr at ion data d e r i v ed f ro m a s t a t i s t i c a ls am p l i n g of convent iona l -ca lib ra tion da t a wa s used for test and checkout dur ing the l a t t e r p ar t oft h e A po ll o P ro g ra m . T h e l u n a r m o d u le i n s t ru m en t a t io n p ro cu re m en t sp ec if ic a ti o n p e rm i t t ed t h eu s e of s t an d a rd -c a l i b r a t i o n data. T h ese d a t a w e re ap p l ic ab l e t o s i m i l a r l y i n s tru m en t ed m easu re -me nts. The defini tion, m er i t , and appl icat ion of each type of data a r e d is c u ss e d.

6. Abstract

5. Report DateM arch 1 97 36. Performing Organization Code

7. Author(s)J. F. DeMoss, MSC

9. Performing Organization Name and AddressManned Spacecraf t CenterHouston, Texas 77058

* F o r s a l e by t h e Na t io n a l Te c h n ica l I n fo rma t io n Se rv ice , Sp r in g f ie ld , V i rg in ia 22151

8. Performing Organization Report No.MSC S-35910. Work Unit No.

914-11 -00-00-7211. Contract or Grant No.

2. Sponsoring Agency Name and AddressNat iona l Aero naut i cs and Space Admini s t ra t ionWashington, D. C. 20546

13. Type of Report and Period CoveredTech nica l Note

14. Sponsoring Agency Code

7. Key Words (Suggested by Author(s))In s t ru m e n t a t i o n ' D a t a P ro ces s i n g

' C al i b ra t i n g 'Data R educt ion' n s t r u m e n t s' A p o ll o P ro j e c tIn s t ru m e n t C o m p en sa t io n

18. Distribution Statement

19. Security Classif. (of this repor t)None

20. Security Classif. (of this page) 21 . NO . of Pages 22. Price'None 1 3 $3.00


3/14

APOLLO EXPERIENCE REPORTF L IGH T IN S T R U ME N T A T ION C A L IB R A T IO N

B y J . F. DeMossMa n n e d S p a c ec ra ft CenterS U M M A R Y

Th ree types of instrumentation-calibration data were used for minimizing e r r o r sin data accru ing from Apollo spacecraft instrumentation syst ems . The type usuallyspecified was unique to individual instrumentation components and was termed conven-tional. Mean-standard-calibration data were used fo r most command-service moduletest and checkout purposes during the latt er stages of the program. Advantages in theus e of standard-calibration data were la te r recognized, and this type w a s permitted bylunar module instrumentation specifications to minimize the us e of conventional-calibration data. It is concluded that the use of standard-cal ibrat ion data is preferablefor most applications; however, the requirement fo r the use of conventional- calibrationdata for critical measurements may never be eliminated.IN T R OD U C T ION

Instrumentation s ys te ms sense such physical conditions as the pressure in a fueltank, the acceleration of a mass, o r the mas s flow of a moving fluid, and the systemsrepresent them as numerical values. When using instrumentation system data, th e an-alyst is concerned with how closely the measured values represent the actual values.Ideally, the output of the measurement system should be directly proportional to the in-put and in the same units. This condition w a s seldom achieved, however, because ofe r ro rs associated with the measurement of the physical stimuli and the conversion ofthe measu remen ts to elec trical signals. Therefore, calibration data were required toconvert the numerical values into usable units t ha t could be interpreted by the analyst.The instrumentation e r r o r was normally determined by comparing the measured valueto the sa me measurement made with an established and accepted measurement system.The te rm s and accuracy of the calibration data provided had to be compatible with therequirements of the analyst when interpreting test results.

TYPES OF INSTRUMENTATION SYSTEMSA sim ple fo rm of an instrumentation system is a meter that is read directly inthe units of the measured stimulus. This kind of sys tem requires minimal calibration


4/14

100data. Complex instrumentation systemsconsist of several components, includinga transmission system, and requiremore comprehensive calibration data sothat the e rr or s ar e determined and theco rr ec t compensation is applied. Cali-curve, compare the output of the in-strumentation system with the measuredvalues. For a simple meter display, the 2 4ccurve would show actual data comparedwith indicated data (fig. 1).

ACCURACY

80

bration data, usually presented as a M:2-J-5

2(

cThe procurement of Apollo instru- Indicated valuementation components w a s based on anominal instrumentation range that could Figure 1. - Simple calibration curve.determine system status and perform-ance. The accuracy of each Apollo in-strumentation measurement f or flight evaluation w a s obtained by a root- sum-squaresummation of the allowable system-component e r ro rs . These er r o r s were separatedbasically into random errors and systematic o r repeatable er ro rs . Instrumentatione r r o r s were also influenced by environmental conditions; however, for brevity, onlysystematic er ro rs at ambient conditions w i l l be discussed in detail. Other e r r o r swere considered in qualifying instrumentation and in defining overall accuracy.

A calibration curve that illustratesvarious t ype s of errors is shown in fig-ure 2, in which the data points from anumber of calibration tests are repre-sented by X's . A faired curve that bestfits the data points is a mean-calibrationcurve. Boundary lines drawn throughthe outer data points define the random-e r r o r band, and the outermost dashedlines show the boundaries of the specified-e r r o r band. The offset of the mean-calibration curve from t h e ideal- calibrationcurve is the systematic er ro r. The sys-tematic er ro r may be a constant bias o rmay vary i n magnitude throughout themeasurement range. The important pointsa r e that the er ro r is repeatable and thatthe instrumentation-calibration data com-pensate for the error .

UIc3m

-al Mean-calibration curve5cmc-

Ideal -calibration cur veRandom-error band

Specified-error bandI 100System output, percentFigure 2.- Random and systematice r ro rs .2


5/14

Instrumentation accuracy w a s improved in two different ways: through bet terquality hardware and through repeated calibration of instrumentation components. How-ever, a point was soon reached at which it w a s necessary to compromise to providecalibrations that best served the purpose within the pract ica l limitations of cos t andscheduling. In addition, the e r ro r magnitude was rel ated to the design sta te of the artof measurement sens or s and of signal-conditioning equipment.CONVENT IONAL -CAL I BRA T ION DAT A

Individual-component calibration was required by procurement specificat ions forthe Saturn launch vehicle and the command and serv ice module (CSM). The type ofcalibration data usually specified for Apollo instrumentation sy st ems w a s individual-component data and was, therefore, defined as conventional-calibration data. Apolloscientific equipment also generally requ ired the us e of conventional-calibration data.The specified instrumentation range w a s equal to or gre ate r than that requiredto determine system performance. Actual component-calibration data were used toeliminate systematic e r ro rs , thereby improving the accuracy of the measurement.Fo r linear curves, end points were determined for the specific transducer, and astr aight line was drawn between them. The maximum allowable deviation fr om th isline was defined by the transducer specification. The data points for each specifictransducer taken during testing were used to produce conventional- calibration curves.Systematic e r r o r s were verified by using actual data from vendor acceptance and veri -fication testing. A l l these tests were conducted under ambient pr es su re and tempera-ture conditions.Each component was calibratedthroughout its operational range in incre-ments of 10 o r 20 percent. The results oftwo tests that were conducted over the full

tu res are shown in f igure 3, in which va-rious components of instrumentatione r r o r are exaggerated. The number ofdata points fo r each instrumentation sen-sor va ri es with the instrumentation typeand the ease of calibrat ion. For example,it was ea sie r to calibrate a spacecraft-cabin temperature sen sor than a cryogenic-helium-tank temperature se ns or becausethe la tt er data points ranged fro m - 425"to -200" F, and sensor characteristicschange at these low temperatures . Whenmultiple calibration cu rves were suppliedand annotated in t e rms of the secondarynormally a function of the environmentalconditions; temperature was the mostcommon variable . (exaggerated).

-

range of the sensor at ambient tempera- First increasing

-

off-ambient calibrations were conducted, calibrationVariation in repeatabilityI

0 100variable. The secondary variable was S y s t e m output, percentFigure 3. - Calibration e r r o r s

3


6/14

Ehcerpts f ro m Apollo CSM instrumentation-component specificat ions are in-cluded i n this report to demonstrate how conventional-calibration data were specified.The following pressure-t ran sducer specificat ion MC 449- 005 (dated May 11, 1964,and revised February 12, 1968) s presented as an example.3.3 PERFORMANCE

3.3.10 Output Voltage. - The output voltage sha ll be ze ro volts (plus 0.15 volts,minus 0 volts) to 5.0 volts (plus 0 volts, minus 0.15 volts) dc floating and di-rectly proportional to the specified pr es su re range. The output noise shall notexceed 10 millivolts peak-to-peak to 10 000 cps.3.3.14 Long-Term Stability. - Combined sensitivity and long-term zero driftunder continuous operation for 360 hours, at any point within the specified pres -su re range, shall be less than plus o r minus 0.5 percent of full scale. Fo r in-termittent operation over a period of 90 d a y s , the above dri ft shall be less thanplus or minus 1.0 percent full scale.3.3.15 Hysteresis. - Maximum hysteresis shall not exceed plus o r minus0.15 percent of full scale.3.3.16 Repeatability. - Repeatability er r o r shall not exceed plus o r minus0.1 percent full scale.3.3.18 Error Band. - The algebraic sum of the total combined e r r o r s fr omhysteresis, linearity, repeatability, regulation, and all environmental param-eters shall not exceed 4.5 percent full scale.

(End of specification)In an interpretation of these specifications, the e r r o r bands in figure 4 a r e 3 per-

cent of full scal e for the end-point tolerance and 4.5 percent for the specified-errorband. The instrumentation-component-procurement acceptance te st required that twocalibration tests, each consisting of 11 data points, be provided ove r the inst rumenta-tion range. These test data confirmed acceptable performance and providedconventional- calibra tion data points.These data points were then used to generate calibration-curve expressio ns fo rdata processing. Using x for the independent variable (instrumentation output) andy fo r the dependent variable (engineering units), the polynomial expression f o r thecalibration curve is

2 ny = a O + 3 x + a $ +... + a xnFor a linear curve, the data end points establi shed a straight-line curve expresse d bythe polynomial

4


7/14

Here, the constant a. establishes a bias and al establishes a slope. These datapoints were tested to establi sh whether the variance fro m a stra ight line was equal toor less than 0.5 percent. If the variance was less than 0.5 percent, a straight linew a s used. A polynomial approximation curve was fitted through th e calibrated datapoints when the variance was grea ter than 0.5 percent. This curve was based on anorthogonal polynomial least- squares fit. A second- through fifth-o rder polynomialw a s tried, and the residuals of each order were established. The proper polynomialw a s then selected on the basis of a statistical test to establish the best fi t (F-ratio testof 1-percent significance). The selected polynomial w a s compared with the test-calibration data to ensure that the variance did not exceed 0.8 percent.

If a curve exceeded the 0.8-percent criterion, a piecewise linear fit w a s thenused. This type of curve is shown in figure 5. The curve is produced by vectoring aline between calibrat ion data points. This type of data fit was used for vehicle testingwith the Apollo acceptance checkout equipment. The number of acceptance-checkout-equipment input data points was held to six (20-percent increments) because of equip-ment limitations. Additional data points (as many as 14) were used w i t h specialequipment modifications.

Data proces sed in support of mission control and postmission data evaluationused a polynomial fit whenever possible. A limited number of measurements couldnot be fitted to a polynomial. These measurements were expressed as piecewise fitsand were the least accurate.

Specified-error bandictual -erro r band

Ideal curve-%1 J

5System output, VFigure 4 . - System calibration curve. System output, percentFigure 5. - Vector- and polynomial-curve-fit comparison.

5


8/14

Conventional-calibration data were used for preflight t es t and checkout operationsat the launch site, at mission control, and during postmission-evaluation data process-ing. The data-presentation technique is shown in figure 6 and table I. The calibrationcurve (fig. 6) includes information concerning the measurement identification, title,equation coefficients, and acceptance-checkout-equipment data points that were used.This presentation indicates the general curve chara cteri stic (nonlinearity) andis alsoused for approximate scaling of st rip-chart recordings and mete r displays. Thedigital-to-analog conversion fo r each telemetry count value is shown in table I. Thistabulation provides specific engineering-unit values and indicates the resolution capa-bility of the instrumentation system.

Figure 6. - Typical calibration data used fo r mission-evaluation data analy sis.

6


9/14

i-30i-za4i-4z

znEdmel4UWela

n

z;TIW4m4i-

da

PzV

i-zV

da

i-zV

p:W4XV

dB aW

0 i-z 7.W VB

42i-zV

da

7


10/14

MEAN-STANDAR D CAL I RAT ION DATAMean-standard-calibration data are derived from a statistical sampling ofconventional- calibration data. The us e of mean- standard- calibration data simplifiedt h e test and checkout of each vehicle because individual instrumentation rec or ds werenot required. Replacement of hardware did not requi re the suspension of testing fo r

the specific purpose of updating calibration data. The time and effort required forcalibration updating and data-processing verification were usually gre ater than thetime and effort required for actual component replacement.Mean-standard-calibration data were derived fo r each specific type of hardwarecontained within each measurement system. Data points were compiled and combinedstatistically by a computer to obtain the calibration-data curve. A s additional instru-mentation w a s procured, these data samples were used to update the mean-calibrationdata. Though us e of mean-standard-calibration data reduces test and checkout time,additional computer programing and data-processing time are required. If la rg e num-bers of components are involved, the resulting mean standard curve will approach thespecification curve. In that case, the interim step of generating mean data should beavoided because of the expense.The u s e of mean- standard-calibration data relieved the computer updating prob-lem fo r test and checkout during initial vehicle buildup, when the incidence of ins tru -mentation failure was highest. A s more vehicles were fabricated, the instrumentationinstallation and test procedures wer e improved, result ing ina decreased instrumentation-component failure rate. Computer updates became less frequent, and mean-standard-calibration data were no longer desirable.Mean-standard-calibration data were used for factory checkout of analog meas-urements on the CSM on which the tolerance band (nominally ? 5 percent) was notcritical. For those f ew critical measurements, such as the amounts of consumablesand flow rates, conventional- calibration data were used.

STANDARD-CALI BR ATIO N DATAThe procurement of instrumentation components fo r the lunar module (LM) wasmade to minimize o r eliminate individual- component ca libration data. This effort wasachieved by instrumentation-component specifications that specified a great er accuracyof data end points and greater linearity of instrumentation-component performance.The use of standard-calibration data was based on the assumption that the e r ro rsassociated with a measurement are all random err or s. Standard calibration was de-fined in the instrumentation hardware specification by a curve that connected the zeroand full -scale points. The calibration was bounded by a specified-error band that wasderived by a root- sum-square calculation of the typical se nsor and signal-conditionererrors. Test data were not used in these calibrations and correction for systematice r r o r s was not made.

8


11/14

The following excerpt from the LM instrumentation component specification in-dica tes how standard- calibration data were obtained. The example used is pressure-tr ansducer specification LSP-360-624A (dated A p r i l 25, 1966, and rev ised May 5, 1967).

3.3 PERFORMANCE3.3.4 Signal Output Requirements. -

(a) Fo rm and Mode. - The signal output shall be a n analog voltage, uni-polar and ungrounded. The magnitude shall vary between ze ro (0) andfive (5) Vdc and shall be directly proportional to the p re ss ur e over therange and within the accuracy specified.(b) Ripple and Noise. - The internally generated ripple and noise contentof the output shall not exceed 5 millivolts peak-to-peak into a load of1 megohm or greater .(c) Noise Feedback. - The ripple o r noise feedback into the primarypower so urce shall not be grea ter than 1 0 millivolts peak-to-peak meas-ured acr oss a network consisting of a 0.5-ohm resistan ce in ser ie s witha 20-microhenries inductor over a frequency range of 20 cps to 1 5 KC/s.

3.3. 5 Theoretical Curve. - The theoretical curve used to dete rmine the magni-tude of e r r o r s shall be a stra ight line terminated by 0.000 volt s and +5.000 vol tsand shall be directly proportional to the pre ssu re f rom 0.0 percent to 100.0 per-cent of the measurand. Any deviation from thi s theoretical straight line is theunit output error.3. 3.6 Static Er ro r Band. - Any data point shall not be g reat er than the percentof full scale as specified fro m a corresponding parameter point on the theoreticalcur ve and shal l include the effects of linearity, hys teres is , repeatability, excita-tion regulation, and end points. The static e r r o r band shall be determined at thestandard ambient conditions specified herein.3.3.7 Total Er ro r Band (Dynamic Er ro r Band). - The total error band shallinclude all deviations from the theoretical curve due to environment, elec tric alcharacteris tics, unit performance, and any other requirements stated hereinthat would contribute to the er r o r s in the unit. Any data point shall not be gr ea te rthan the percent of full scal e specified from its corresponding parameter pointon the theoret ical curve.

9


12/14

A C C U R A C Y SPECIFICATIONContractor Part Number

LSC360- 24- 07LSC360-624-205LSC 60- 24- 03LSC 360-624-1 7LSC 360- 24-1 5LSC 360- 24- 1LSC 360- 24- 09LSC360-624-103LSC360-624-101LSC 360- 24-1LSC360-624-

Static Er ro r Band(3. 3.6)

+1.5% FS+1.5% FS+1.25% FS*l.O% FS+l.O% FS+l.O% FS+l.O% FS+l.O% FS+l.O% FS+l. 0%+l. 0%

Total Error Band(3.3.7)

-12.5% FS+2.5% FS+2.0% FS+1.8% FS+1.8% FS~ 1 . 8 % S+1.8% FS+1.8% FS~ 1 . 8 % Sf 2.0% (0-200"F)+ 2.0% (0-200"F)

(End of specification)In a comparison of standard-calibration data with conventional-calibration data(fig. 4), the end-point tolerance is ze ro and the specified-er ror band is the total error.The us e of standard-calibration data offered two advantages. First, very littlecalibrat ion-data updating was required . During the vehicle test cycle, any componentin the instrument system (sensor o r signal conditioner) could be replaced with a likecomponent and no calibra tion change was required. The second advantage benefitedthe analyst. Because numerous standard calibrations wer e straight lines, the instru-mentation output w a s directly proportional to the input over the full-scale range of themeasurement. The analyst made simple, di re ct conversions fr om percent of full -scal e

output to engineering units. For example, i f a measurement with a range of 0 to500 psia indicated 10-percent deflection on a meter or st ri p chart, the correspondingengineering-unit value was 50 psia.The disadvantage of using standard-calibration data is that no corrections aremade to the measurement output fo r systematic er ro r. When actual test-calib rationdata were compared with standard-calibration data fo r a representati ve sample ofmeasurements, the systematic e r r o r s were usually less than 1 percent.

10


13/14

Because of t h e advantages previously discussed, standard-calibration data wereused for the majority of LM measurements . However, cer tain LM measurements re -quired correction for systematic e rr o rs because of the need to achieve the greatestdegree of obtainable accuracy for mission-evaluation purposes (for example, measure-ments used to calculate critica l systems performance or determine consumablesstatus ). These components were individually calibrated and conventional calibrationdata provided.

CONCLUDING REMARKSThe use of standard-calibration data is preferable to the use of mean-standard-and conventional-calibration data fo r most applications. The effective use of standard-calibration data, however, may be achieved only when the instrumentation procurementspecifications ar e designed to require standard-calibration data. When requirementsof th is type ar e levied, the procurement cost will be greater, but the overall programcos t may be l e s s because of the reduced data-processing requirements . Mean-

standard-calibration data offer no advantage over standard-calibra tion data. The re-quirement for the use of convent ional-calibration data for a critical measurement maynever be eliminated, but the number of applications can be minimized. Scientific andexperimental equipment will still require the use of conventional- calibration data be-cause of the limited number of such equipment, the high degree of accuracy required,and peculiar design character istics .Manned Spacecraft CenterNational Aeronautics and Space AdministrationHouston, Texas, September 25, 1972914-11-00-00-72

NASA-Langley, 1973- 1 s- 359 11


14/14

AERONAUTICS AND SPACE ADMINISTRATIO NWASHINGTON, D.C. 2 0 5 4 6

OFFICIAL BUSINESSP E N A L T Y FOR PRIVATE U S E $300 S P E C I A L F O U R T H - C L A S S R A T E 4 S (ZjSMAILP O S T A G E A N D F E E S P A I DN A T I O N A L A E R O N A U T I C S A N DS P A C E A D M I N I S T R A T I ONBOOK

POLITMASTER : If Undeliverable (Secti on 158Poatal Mnnunl) Do Not Return

The aeronautical and space activities of the United States shall beconducted so as to contribute . . . o the expansion of human knowl-edge of phenomena in the atmosphere and space. The Administrationshall provide for the widest practicable and appropriate disseminationof information concerning its activities and the results thereof.-NATIONAL AERONAUTICSND SPACE ACT O F 1958NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS

TECHNICA L R EPORTS: Scientific and ,technical information considered important,complete, and a lasting contribution to existingknowledge.TECHNICAL NOTES: Information less broadin scope but nevertheless of impo rtance as acontribution to existing knowledge.TECHNICAL MEMORANDUMS:Information receiving limited distributionbecause of preliminary data, security classifica-tion, or other reasons. Also includes conferenceproceedings with either limited or unlimiteddistribution.CONTRACTOR REPORTS: Scientific andtechnical information generated under a NASAcontract or grant and considered an importantcontribution to existing knowledge.

TECHNICAL TRANSLA TIONS: Informationpublished in a foreign language consideredto merit NASA distribution in English.SPECIAL PUBLICATIONS: Informationderived from or of value to NASA activities.Publications include final reports of majorprojects, monographs, data compilations,handbooks, sourcebooks, and specialbibliographies.TECHNOLOGY UTILIZATIONPUBLICATIONS: Information on technologyused by NASA that may be of particularinterest in comm ercial and other. non-aerospaceapplications. Publications include Tech Briefs,Technology Utilization Reports andTechnology Surveys.

Details on the avoilability of these publications may be obtained from:SCIENTIFIC AND TECHNICAL INFORMATION OFFICE

N A T I O N A L A E RO N AU T IC S A N D SPACE A D M I N I S T R A T I O NWashington, D.C. 20546

Documents

Apollo Experience Report Flight Instrumentation Calibration