TIMING ERRORS THEIR DETECTION AND CORRECTION IN THE …€¦ · TIMING ERRORS-THEIR DETECTION AND...

X-612-64-328

I J , -

-

TIMING ERRORS - THEIR DETECTION ' AND CORRECTION

IN THE IMP INFORMATION- PROCESSING[ SYSTEM

$- I GPO PRICE

h ,

-

B Y WILLIAM H. MISH

\

OCTOBER1964

GODDARD SPACE FLIGHT CENTER GREENBELT, MD.

/

I

1-

https://ntrs.nasa.gov/search.jsp?R=19650008663 2020-07-15T03:47:42+00:00Z

c

TIMING ERRORS-THEIR DETECTION AND CORRECTION

IN THE IMP INFORMATION PROCESSING SYSTEM

by W i l l i a m H. Mish

October 1964

TIMING ERRORS-THEIR DETECTION AND CORRECTION IN THE IMP INFORMATION

PROCESSING SYSTEM

SUMMARY

One of the major problems encountered by many experimenters in the analysis of data retrieved by experiments flown on board satellites has been that of obtaining correct Universal Time in conjunction with this data.

This paper presents the scheme that was used by the IMP Infor- mation Processing System in the processing of the data f rom IMP-I which provided a means of eliminating most of the timing e r r o r s before they reached the experimenter. In addition this scheme proved valuable in the initial debugging of the I M P analog to digital line and most s i g - nificant in evaluating the stability of and providing a check on the IMP-I space c raft telemetry s y stem.

iii

CONTENTS

Page

SUMMARY iii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

APPLICABILITY O F THE IMP-I TELEMETRY FORMAT TO TIME ERROR DETECTION AND CORRECTION. . . . . . . . . . 2

IMP INFORMATION PROCESSING SYSTEM (IMP-IPS). . . . . . . 4

THE TIME ERROR DETECTION AND RECTIFICATION. . . . . . 7 SOFTWARE .................................

19 CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

A-1 APPENDIXA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1 APPENDIXB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . c - 1 APPENDIXC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D- 1 APPENDIXD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

APPENDIXE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E - 1

F -1 APPENDIXF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G- 1 APPENDIXG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

V

TIMING ERRORS-THEIR DETECTION AND CORRECTION IN THE IMP INFORMATION

PROCESSING SYSTEM

INTRODUCTION

The generation of Universal Time at the ground station, the merging of this time with recorded spacecraft data, and the subsequent processing of this t ime through analog to digital conversion equipment has, historically, had a multiplicity of problems associated with it. It would appear that a certain amount of these problems a r e unavoidable when one considers the complexity of the over-all situation.

The "aggregate" timing of data f rom most spacecrafts is a product of a relatively large number of Pr ime Minitrack Time Standards located throughout a world wide network of Space Tracking and Data Acquisition Stations. base for the spacecraft data that is in agreement with Universal Time, i.e., WWV, to within one millisecond when the time standard has been synchronized correctly with W W V and the propagation delay t imes between W W V and the stationhave beencompensated (see reference 12, page 2-95). The pr ime function of these t ime standards is to provide a t ime code that can be written on analog tape simultaneously with the recording of the spacecraft data at the various telemetry receiving stations. This analog tape is then shipped to a central information processing facility where it provides the input for analog to digital conversion equipment, one function of which, is to decode the t ime and convert it to a digital format in conjunction with the data, and finally production of a digital magnetic tape which can subsequently be processed with computers.

Each of these t ime standards is designed to provide a t ime

As the practice of flying scientific experiments which required high resolution of the ground station time became prevalent it became obvious to the experimenters that they were not - always realizing the design capabilities of the timing standards, i.e., one millisecond agreement with W W V when propagation effects have been accounted for , due to the degradation of the timing a s it t raversed the above mentioned se r i e s of operations. so frequent in some cases as to require a considerable expenditure of effort on the par t of the individual experimenters to verify the integrity of the timing associated with their data.

The occurrences of these "timing e r r o r s " was

1

One solution to this problem was to incorporate a s par t of the information processing facil i t ies a systematic method, necessarily computer orientated because of the large volumns of data, of detecting and correcting timing e r r o r s before they reached the experimenter.

Up to the advent of the IMP-I spacecraft and its crystal controlled telemetry system a scheme for the detection and correction of timing e r r o r s of the nature described in this paper was, in most cases , im- practical because the telemetry systems, even those that telemetered "pseudo clocks", (see reference 11, page 9) were not stable enough with t ime to allow the telemetry patterns o r pseudo clocks to be used as an independent "clock" against which ground station t ime could be accurately compared. However, with the successful launch of IMP-I on November 27, 1963, a satell i te became available which had a telemetry format consisting of a repeating pattern of telemetry sequences with a relatively stable period that could be used to perform an independent check on ground station time. This aspect of the IMP-I telemetry system was taken advantage of in the design of the IMP Information P r o c - essing System (IMP-IPS). (See reference 13 . )

This paper gives a brief summary of the IMP-IPS, discusses in detail those phases of the IMP-IPS which a r e concerned with the detection and correction of timing e r r o r s and how this aspect of IMP-IPS was employed to trouble-shoot the IMP analog to digital line and monitor the IMP-I spacecraft telemetry system throughout the useful life of the spacecraft.

APPLICABILITY O F THE IMP-I TELEMETRY FORMAT TO TIME ERROR DETECTION AND CORRECTION

The IMP-I telemetry system is a modification of the basic type of PFM telemetry system flown on some of the past satell i tes, e.g., Ex- plorer XI1 and UK-1. references 1, 15 and 17.

For a discussion of this telemetry system see

There a r e two significant features of the IMP-I telemetry format that permit the effective implementation of a t ime e r r o r detection and correction scheme of the type described in this paper:

-

A. The telemetry format is a synthesis of a repeating pattern of four telemetry sequences each approximately 81.914 seconds in length (see Figure 1 ) . ception of the fourth, consists of an a r r a y of sixteen channels by

Each of these sequences with the ex-

2

.

NORMAL SEQUENCE IS 16 CHANNELS x 16 FRAMES. THE SATELLITE CLOCK IS READ OUT IN CHANNELS II, 12,13, 14 e 15 - FRAMES 4

AND 12. / 3

s 9 ZE

W v)

SEQUENCE 4 IS

SEQ. 4 CONTINUOUS ANALOG SIGNAL. NO SATELLITE CLOCK IS PRESENT.

REPEATING PATTERN OF 4 SEQUENCES EACH SEQUENCE IS APPROXIMATELY 81.914 SECONDS IN LENGTH. TOTAL PATTERN IS APPROXIMATELY 4 x 81.914 = 327.656 SECONDS IN LENGTH.

Figure 1 - The IMP-I Telemetry Format

3

sixteen frames. The f i r s t three sequences contain an octal satellite clock reading in channels 11 through 15, frames 4 and 12, which increases by one decimal unit each sequence including the fourth. The low order octal digit of the satell i te clock reading is decoded by the analog to digital line into a 1, 2 o r 3 depending on whether it is the first, second, o r third telemetry sequence respectively. as par t of the digitized data and is utilized by the t ime e r r o r detection software. netometer sequence) contains continuous analog data and consequently is missing the satell i te clock reading. clock sti l l up-dates itself by one decimal unit in this sequence even though it is not read out.)

This information i s included

The fourth sequence (Rubidium vapor mag-

(The satell i te

B. The periodicity with t ime of the channel ra te is insured through the use of a 20kc c rys ta l as a basic component of the spacecraft telemetry system. down t o obtain the desired sampling rate and i s backed up with a f r ee running multi-vibrator synchronized to one of the subdivided frequencies f rom the crystal . When not synchronized with the crystal the multi-vibrator runs some 25 percent lower in frequency than when in synchronization with the crystal. The significance of this feature becomes apparent in la ter discussions in this paper.

This basic crystal frequency is divided

IMP INFORMATION PROCESSING SYSTEM (IMP-IPS)

The IMP-Information Processing System i s an integrated set of analog to digital conversion equipment and computer programs which receive a s input information in an analog format and output information in a standard Binary Coded Decimal format. system is extensive and details can be found in References 6, 7, 9, 13, 14, 16, and 18.

Documentation of this

Because information is processed serially through the system a brief explanation of each of the steps is appropriate to allow the p r e s - entation of the Time E r r o r Detection and Rectification Software in its proper place with respect to the over-all s e r i e s of operations ( see Figure 2).

The flow of information through the IMP-IPS commences with the reception of the analog tapes f rom the network of stations covering the spacecraft. These analog tapes a r e processed through the IMP analog

4

5

to digital (A/D) line, ( see references 7, 9, and 18) which digitizes the analog data and time and crea tes various f l a g s and symbols that a r e a function of the time-data a r r a y at 16 t imes "real time". The digitized output of the A/D line is writ ten on tape in a BCD format, this tape is called a Raw Data Tape (RDT) and contains f rom 30 to 40 hours of digitized "real t ime" data. This 30 to 40 hours of data constitutes many satell i te passes* f rom a number of different stations. with control cards punched by the A/D line operators, serve as input to the I M P Edit Program (see reference 16) . is operationally broken into two passes on the computer, performs in the most general sense the following functions:

The RDT, along

This edit program, which

A. The Program checks for and cor rec ts operational and equipment A/D line e r r o r s , i.e., "cleans up" the RDT.

B. Uses the IMP-I satell i te clock' to tag each telemetry sequence with a monotonically increasing number for each 90 days of satellite ope ration.

C. Uses the e r r o r detecting and correcting properties of the IMP-I satellite clock to create a flag for each telemetry sequence which gives a measure of the quality of the data in the sequence.

The final output of the IMP Edit P rogram is a Master Data Tape (MDT) which contains 120 to 160 hours of rea l t ime data, and control cards that a r e punched by the program. 3 Control Card which, after being operated on by the Time E r r o r De- tection and Rectification Software, se rves as the timing interface between the MDT, which contains "raw time", and the Experimenter Data Tapes (EDTs) which contain "corrected and smoothed" time.

Of special interest is the Code

At this point in the se r i e s of computer operations the MDTs a r e allowed to backlog until two consecutive orbits ' worth of data have accumulated. about eight days of rea l t ime data on each MDT.

As each complete orbit takes about 93 hours this constitutes

* A pass i s defined as one uninterrupted data recording made at a single station (average pass on

'The spacecraft "clock", a 15-bit accumulator, can accumulate a maximum of 32,767 counts, Le., 215 counts, before it returns to zero. To fill th is accumulator takes approximately 30 days. This clock, through the use o f programming techniques on the ground, has been extended to a 90-day clock in the data that the experimenter receives ( s e e Appendix G).

IMP-I is about 4 hours).

6

The two orbits ' worth of data a re then processed through the Time E r r o r Detection and Rectification Software system. In this se r ies of computer operations, which will be covered in detail in the following sections, e r r o r s in t ime a r e detected and provision i s made for the subsequent correct ion of these e r r o r s and "smoothing" of the t ime through the creation of the Pseudo Code 3 Cards. 3 Cards and the MDTs then provide the input t o the decommutation pro- g ram (see reference 6) which "smooths'' and cor rec ts , i f necessary, the t ime for each pass contained in the two orbi ts ' worth of data being processed, and crea tes the experimenter tapes containing each individual experimenter ' s data. into the Time E r r o r Detection and Rectification Software where the a l te r - ation made to the timing through the use of the Pseudo Code 3 Cards is reviewed.

These Pseudo Code

Finally, one of the experimenter tapes is fed back

THE TIME ERROR DETECTION AND RECTIFICATION SOFTWARE

The Time E r r o r Detection and Rectification Software consists of four separate IBM 1410 computer programs which have been represented by a single block in Figure 2. detail in Figure 3, where the four programs a r e ser ia l ly numbered f rom 1 to 4.

The contents of this block a r e shown in

This s e r i e s of programs is designed to accomplish the following objective s :

A.

B.

C.

The MDT Time Verification P rogram (program 1 in Figure 3 ) produces a plot tape, which, when sor ted chronologically with respect to time and plotted, provides a means of visually detecting timing e r r o r s a s well as providing information on the performance of the spacecraft te lemetry system and the A/D line, and when input to the Pseudo Code 3 Card P rogram (program 3 in Figure 3 ) allows the timing to be analyzed, corrected and smoothed.

The Time Verification Sort Program (program 2 in Figure 3 ) insures that the data sent to the experimenter is in chronological o rde r and provides a sorted input tape for the Pseudo Code 3 Card Program.

The Pseudo Code 3 Card Program per forms an analysis on the timing and cor rec ts timing e r r o r s that may exist by altering the Code 3 Cards to Pseudo Code 3 Cards which a r e used to control

7

I

W t I-

8

Q)

C 0

0 U

.- c

.- Y .- L

V Q) CY -c1 C 0 C 0 .-

I-

I- I c)

Q) r

the timing that i s generated by the decommutation program for the expe rimente r s ' tape s .

D. The California EDT Time Verification Program (program 4 in Figure 3) again produces a plot tape which when plotted i l lus- t r a t e s the alterations that have been made to the timing by the use of the Pseudo Code 3 Cards.

The MDT Time Verification Program - A Tool for Detecting Time E r r o r s and Monitoring the A/D Line and Spacecraft Telemetry System Pe rf ormanc e

Referring to Figure 1, the MDT Time Verification P r o g r a m (pro- g ram 1 in Figure 3) takes advantage of the fact that channel 0 of f r ame 0 of each telemetry sequence 1 occurs approximately every 327.6 sec - onds. This enables the program to assign a unique number defined a s N to each sequence 1, Le., the program counts the number of sequence 1 s that have occurred since some chosen reference sequence 1 (a t the reference sequence 1, N equals zero). N is determined by the program as follows:

- t ( 1 ) (Note that the brackets N = [ t n ~ t O ]

indicate that N has been rounded to an int e g e r . )

Where:

to i s defined a s the t ime for channel 0 of f rame 0 of the r e fe r - ence sequence 1: t, i s the t ime for the current channel 0 of f rame 0 of the sequence 1 being counted, and At i s a chosen supercommutated sequence ratef and is equal to 327.656 seconds. sequence rate was chosen at the same time that to was chosen and does - not represent the average supercommutated sequence ra te f rom to to t,.)

(It should be noted that the above supercommutated

*It has been necessary to change this reference point once during the useful life of the spacecraft. (See Appendix A orbit 25.)

tThe sequence rate i s assumed to be 81.914 seconds ( s e e Figure 1).

9

The above quantities a r e used by the program to compute the param- e t e r Delta f o r each sequence 1 that is counted. following formula:

Delta is defined by the

Delta = ( t , - to) - N ( A t ) ( 2 )

Where:

(t -t ) represents the actual elapsed t ime for N sequence I s , NiAtrrepresents the elapsed time that would be predicted for N sequence 1s i f the actual supercommutated sequence rate was a constant, and Delta represents the algebraic difference of the two for e going quantitie s . A sample computation of Delta is presented in Appendix F. Note

that the computation of N is accurate as long a s t, is in e r r o r by l e s s than 50 percent of 327.6 sec. minus the accumulated e r r o r due to the slightly improper choice of the supercommutated sequence rate.

As previously mentioned the parameter Delta is computed for each sequence 1 that is counted by the program. This parameter Delta along with i ts corresponding N is writ ten on magnetic tape by the program in a format that allows, after being sorted chronologically with respect to time (program 2 in Figure 3), Delta versus N t o be directly plotted with an EAI::' data plotter. Appendix A contains the complete l ibrary of these plots for the useful lifetime of the spacecraft.

These plots a r e an extremely useful tool i n verifying the integrity of the timing associated with each pass of data that has been recorded and processed through the A/D line during any particular orbit. addition, a great deal of information about the performance of the spacecraft telemetry system, i.e., stability of the bit rate, can be derived f r o m a review and analysis of these plots.

In

The following paragraphs treat the applicability of these plots t o the detection of timing e r r o r s originating at the station and during analog to digital conversion. In addition the plots proved valuable in the initial debugging of the A/D line. of the plots in the evaluation of the performance of the spacecraf t t e lem- e t ry system i s a l so included.

A discussion encompassing the usefulness

*Electronic Associates, Inc. Model 3440.

10

Detection of Timing E r r o r s Originating at the Ground Station

A. The Ground Station clock s e t incorrectly with WWV during the recording of a pass.

This relatively common e r r o r ( see Appendix E) is reflected in the plot of Delta versus N as a positive o r negative displacement in the ordinate values, Le., Delta, f rom the nominal curve during the pass. Appendix A, orbi t number 1, contains a number of passes recorded at Rosman, North Carolina which contain this par t icular e r ro r . t ime code and the W W V signal simultaneously s t r ip charted to i l lustrate the discrepancy in the two time codes. char t s a r e displayed on the same page with the plot of orbit Number 1 in Appendix A.

Two of these passes have had the BCD

These s t r ip

B. E r r o r s introduced by the initial disagreement of the station clocks.

During the life of IMP-I the tracking stationsJwere - not cam- pensating for the propagation t ime f rom WWV to the station site. This resul ts in all the station clocks initially being in disagreement with each other and with WWV. Appendix D con- ta ins a tabulation of these delay t imes.

This initial disagreement of the station clocks combined with the fact that the distance from the spacecraft to the tracking stations changes abruptly as coverage passes f rom station to station is readily apparent i n all the plots appearing in Appen- dix A a s small positive and negative s teps in the curve of Delta versus N.

C. Analog to Digi ta l line could not synchronize properly with the data because of a weak o r noisy te lemetry signal. versely effected the timing.

This ad-

This anomaly is reflected in the plot of Delta ve r sus N a s a predominant scattering of the points during the pass . A, orbit 5 around sequence 1 counts of 4189 and 4789 are good examples.

Appendix

11

D. Both BCD and Serial Decimal t ime were found to be unacceptable and the pass was processed with relative t ime?.

This type of e r r o r is generally reflected in the plots as a gap in the data where the pass would, under normal circumstances, be located. N will generally be completely erroneous during the pass and consequently a r e off scale when the plots a r e made.

This is due t o the fact that the values of Delta and

E. Incorrect ID

Because in the computation of t, the program utilizes the ID of the file ( see sample calculation in Appendix F) an incorrect ID will also result in completely erroneous values of Delta and N. Again when the data is plotted a gap will be present where the pass would, under normal circumstances, be located.

This ID check is an important feature of this scheme, as in- cor rec t IDS a r e very common (particularly when the recording of a pass at the station commences very close to midnight) and when not detected and corrected can cause the experimenter considerable difficulty in rectifying the proper date.

Trouble-Shooting the IMP Analog to Digital Line

As the A/D line had completed the final stages of construction just prior to the launch of the spacecraft there were a certain amount of un- discovered problems that remained to be resolved and improvements made during the processing of the first orbits of data.

A. At the t ime that data f rom the sixth orbit was being processed a refinement was made to the line in the fo rm of a circuit that served a s a window to look for phase e r r o r s a t the end of each channel between the system flywheel frequency, which determines when time is sampled, and the 50 cycle channel ra te derived f rom the recorded data on the analog tape. this window was mistakenly inverted. N reflected this malfunction as small "ramps" in the plots of each pass.

During installation The plot of Delta versus

(See Appendix A orbi ts 6 through 13; orbit 1 0 around

*Relative t i m e is defined as t i m e that i s generated by the A/D l ine accumulator after the accumulatorhas been set to zero. (See reference 18.)

12

I . sequence 1 counts of 9709 and 9909 is a particularly good example. )

The chances of seeing an e r r o r of this magnitude would have been remote if a scheme, such a s described in this paper, for detecting t ime e r r o r s was not being employed. for the ramps in the plots was investigated and subsequently discovered and corrected (see reference 6) a s can be seen in Appendix A in the plots after the 13th orbit.

The reason

B. Another interesting and yet incompletely explained s t ructure in the plots of Delta versus N a r e the "tails" observed on many of the passes. (See Appendix A orbits 2 and 3 . ) It is conjectured by the designers of the A/D Line that these tails a r e due to a disagreement between the frequency of the flywheel, which determines the rate at which time is sampled, and the 50 cycle channel ra te derived f r o m the recorded data on the analog tape at the s ta r t of processing of the pass. As can be seen in the plots a s t ime progresses the flywheel slowly "syncs" in on the 50 cycle channel rate. itself during the processing of data in the la te r orbits but again appears in plots after orbit 38.

This malfunction seems to have remedied

Monitoring the Spacecraft Telemetry System

Because the program i s looking at the difference between the actual t ime for each sequence 1 and the predicted t ime for each sequence 1 (assumed supercommutated sequence ra te of 327.656 sec.) a considerable amount of information is made available about the periodicity with t ime of the sequence rate. the fact that a change in the sequence period will be reflected in the curve.

One of the distinct advantages of this analysis i s

A. In Appendix A, orbit 1 a sudden change in the period of a single supercommutated sequence introduced an approximate 71 second discontinuity in the curve ( see Appendix A, orbit 1, sequence 1 count of 800). As explained in the section titled Applicability of the IMP-I Telemetry System to Time E r r o r Detection and Correction the stability of the sampling rate of the spacecraft i s assured through the use of a 20kc crystal which i s backed up with a f ree running multi-vibrator synchronized to one of the subdivided frequencies from the crystal. Apparently this discontinuity was caused by an overload on the satellite power

13

system which prevented the multi-vibrator f rom synchronizing to the crystal . Consequently it "free ran" about 25 percent low for about 5 minutes. (See reference 16, Appendix F, page 8.)

Orbit number 42 i n Appendix A covers that period when the spacecraft was eclipsed by the Ear th for 7h 57m, as is evident from the plot the spacecraft t ransmit ted for lh 17m after entering the Ear th ' s shadow. It was completely shut off for a period of 15h 17m and became operational again on May 7 at 07h 38"' U.T. (see reference 2 and 4). It is estimated that during the period of time that the spacecraft was not operational the temperature of the satellite electronics dropped to temperatures between -60" and -80°C (see reference 8, page 3) . evidently changed the operating frequency of the c rys ta l and consequently the sequence rate.

This "cold soak"

The te lemetered temperature of the spacecraf t battery, which i s physically located close to the encoder and thus the crystal , is shown plotted against t ime following spacecraft tu rn on af ter the shadow in Figure 4. N for this same period of t ime has a l so been included to i l lustrate the changing sequence rate. it took about 19.5 hours before the c rys ta l frequency stabilized close to its original value. The graphs of Average Sequence Rate per Pass ve r sus Satellite Clock Number in Appendix C also reveal this ra ther "radical" change in the sequence rate.

The "raw time' ' plot of Delta versus

As can be seen f rom Figure 4

Appendix C provides a summary of the stability charac te r i s t ics of the IMP-I te lemetry system during the useful life of the spacecraft*. Graph 1 was generated by plotting the apogee value of Delta against the appropriate orbit number. t o equation ( Z ) , i.e.,

Referring

Delta = ( tn - to) - N ( A t )

it would be expected that if A t was chosen incorrect ly it would introduce a constant slope in the plot of the Apogee Values of Delta ve r sus the Orbit Number.

*With two exceptions the spacecraft operated continuously from launch on November 27, 1963 lBm U.T. ( s e e reference 3). After May 30 the satellite became ex- h until May 30, 1964 at 09

tremely intermittent.

14

4 126.528 3 W 0

125.528

- -

"RAW TIME" PLOT FROM APPENDIX A, ORBIT 42 AFTER EXTENDED EARTH

SHADOW

- - /

#' /

- a** # z ..'

f 1 I I I I I

Figure 4 - Telemetered Temperature of Battery after Spacecraft Turn On Fol lowing the Extended Earth Shadow Compared with the "Raw Time" P lo t During th is Time.

+ 2 0

+ l e -

y + 1 2 -

t 8 + 4 -

z! a 8 w - 4 - z

*- a I- + 8 - m

ne

W

0 -

I-

e - 8 -

-12

-16

- 20

1 5

-

TELEMETERED TEMP. OF SPACECRAFT BATTERY VERSUS TIME AFTER EXTENDED EARTH SHADOW

-

- I 1 I I 1 I

OTh2Im IOh0Tm 12%3" 1!ih40'' 16%" 21'13' OOhOOm 02h46m 05'

As it turned outD t equal to 327.656 was not precisely the ave r - age supercommutated sequence rate. the plot of the Apogee Values of Delta ve r sus the Orbit Number in Appendix C.

- This fact is reflected in

The interesting aspect of this plot is not the slope itself but the rate of change of the slope which gives a measure of the long range stability of the te lemetry system. tangent to a least squares quadratic f i t of data f rom the 5th to the 24th orbit at the 5th orbit indicates the sequence rate had drifted on the order of 4 seconds in 6.4 million seconds through the 24th orbit. f o r an extended length of t ime in the 42nd orbit the stability undergoes some deterioration.

Construction of a

F r o m the 25th orbit until the satellite was eclipsed

D. In addition to the Apogee Values of Delta plotted against the Orbit Numbers, Appendix C contains a plot of the Average Sequence Rate P e r Pass plotted against the Satellite Clock Reading at the beginning of the pass. The instabilities that were mentioned in the previous paragraph a r e particularly evident in this plot.

The Pseudo Code 3 Card Program - A Means of Correcting and Smoothing Time

The Pseudo Code 3 Card P rogram (program 3 in Figure 3) u ses the sorted MDT Time Verification Tape, generated by the MDT Time Verification Program (program 1 in Figure 3) and sor ted by the Time Verification Sort Program (program 2 in Figure 3), as input. contains the tabulation of Delta ve r sus N (the sequence 1 count) for the two consecutive orbits, henceforth r e fe r r ed t o a s the Delta-N data a r r ay , of data being processed.

This tape

At the outset the program operates on the Delta-N data a r r a y a l - gebraically subt,racting the delay t ime of W W V propagation to the station site ( see Appendix D) f rom each of the Deltas computed for each sequence 1 that comprise a pass. This compensation for the delay t ime of WWV synchronization to the station site is c a r r i e d out on each of the passes that comprise the orbit being processed.

As an example in Appendix F the re is a sample calculation f o r the Delta corresponding to the f i r s t sequence 1 in a pass that was recorded a t Woomera, Australia on January 8, 1964. In this calculation Delta is

16

computed to be t86.697 sec. Now referring to Appendix D the delay t ime f rom WWV to the station site at Woomera is 59.1 milliseconds. Thus the adjusted Delta would be:

t86.697 -.059 = 86.638

Next the program operates on the adjusted Delta-N data a r r a y for each individual orbit fitting it, using least squares, to a quadratic equation. It i s readily apparent from the plots in Appendix A that a second degree equation of the form:

DELTA = A , + A,N t A ~ N * (3)

will approximately f i t the data i f we proceed f rom perigee to perigee.* Appendix B shows a comparison of the raw time and the adjusted and least squares smoothed and corrected t ime for orbits 36 and 37. Note that the adjusted and smoothed t ime curve is below the raw time curve during most of the orbit due to the propagation delay t ime adjustments that have been made. After making the least squares f i t the program operates on the Code 3 Cards which were generated by the IMP Edit p rogram (see Figures 2 and 3) . containing the following information:

There is one of these cards per pass,

1. The s ta r t t ime of each pass 2. The corresponding satellite clock reading for the s t a r t t ime 3. The stop t ime of each pass 4. The corresponding satellite clock reading for the stop t ime 5. "Logging" information, e.g., ID of the file, MDT tape number, etc.

(A complete write up on the Code 3 Cards can be found in reference 16, Appendix B.)

In this operation the program obtains f rom the sor ted MDT Time Verification Tape the ground station time for the f i r s t sequence 1 in each pass and the corresponding N and Delta and uses this N in the second degree equation that it has just generated fo r the orbit, Le., equation 3, to compute a theoretical Delta. It then subtracts the actual Delta taken f rom the sorted MDT Time Verification tape f rom the

*Other mathematical expressions have been suggested by Cyrus J. Creveling of the Data Systems Division, GSFC. In particular, he fee ls that f i t t i ng the time curve to an analytically-generated curve of a cycloid shows promise of increasing the accuracy of the operation with an accompanying saving in computer t ime ( see reference 5).

1 7

theoretical Delta and algebraically adds this difference to the ground station time of the first sequence 1 in the pass.

In a s imilar manner the t ime for the last sequence 1 of the pass is adjusted to agree with the second degree equation. and stop t imes and corresponding satell i te clock readings* a r e then punched into the Pseudo Code 3 Cards along with the logging information.

The adjusted s t a r t

Use Of The Pseudo Code 3 Cards In The Decommutation P rogram

I The Pseudo Code 3 Cards, along with the MDTs, now provide the inputs to the Decommutation P rogram (see Figure 3) where information f rom the Pseudo Code 3 Cards is used to compute the average sequence rate per pass which in turn is used to increment the t ime that goes on the experimenter tape.

The average sequence rate per pass is computed by dividing the difference between the start time on the Pseudo Code 3 Card and the stop time on the Pseudo Code 3 Card by the difference between the two 30-day clock readings associated with these s t a r t and stop t imes. This computation results in an average sequence rate for each pass. This average sequence rate is then used by the program to update the time for each sequence throughout the pass using the start t ime as the reference time. Thus the t ime that is writ ten on the experimenter tape for each pass is a l inear approximation to that particular segment of the least squares fi t where the pass occurs.

I

~ Actually then, the smoothed time plots which appear in Appendix B a r e made up of straight line segments each of which represents one pass. because the curvature i s very small.

This feature is not readily apparent in the plots in Appendix B

The California EDT Time Verification P rogram

This program i s identical to the MDT Time Verification P rogram except that it i s written to accept the California Experimenter Data Tape instead of the Master D a t a Tape. Its purpose is to provide a means of checking the alterations that have been made to the t ime

*The satellite clock (30-day clock) and N (the sequence 1 count) are related such that one can easily be computed from the other ( see Appendix G).

18

through the use of the Pseudo Code 3 Cards in the Decommutation P r o - gram (the same time goes on al l experimenter tapes, thus it i s neces- sa ry to check only one of them). The smoothed t ime plots that appear in Appendix B a r e a result of plotting the plot tape that is created by this program.

This final check on the timing i s an important feature of the entire scheme as the t ime that actually is present on the experimenters’ tapes i s being reviewed in this operation.

CONCLUSIONS

As i s evident f rom Appendix A the scheme just described for detecting and correcting t ime e r r o r s was successfully applied throughout the useful life of IMP-I::, i.e., f o r almost 48 orbits, with the following resul ts :

A.

B.

C.

D.

A large percentage of the timing discrepancies were fi l tered out before the data was received by the experimenters. This has saved the experimenters considerable t ime and effort in the analysis of their data and permitted rapid evaluation of the scientific significance of the data.

At least two malfunctions of the A/D line were detected that otherwise would have probably gone unnoticed.

The scheme provided a means of continuously keeping t rack of the spacecraft telemetry system.

Finally, it provided a relatively easy way in which statist ics could be compiled on the timing failures when they did occur. Appendix E contains a tabulation of these timing failures through the 23rd orbit.

This scheme i s currently being employed to analyze the timing associated with the data f rom IMP-I1 which was launched on October 4 of 1964.

*The Pseudo Code 3 Card Program w a s written after the launch of IMP-I when the init ial “raw t ime” Dlots from the f i rs t few orbits of data indicated that the procedure for correcting and smoothing the time d iscussed in th i s paper would be a feasible approach to the problem of t ime errors, consequently, the first eleven orbits were not fit. In addition, orbits 42 and 43 were not fit due to lack of trans- mission from the spacecraft during portions of these orbits.

19

References

1. Beauchamp, K. C., The Encoding, Telemetering and Processing of Data F rom the U. K. Satellite "Ariel", Space Research Man- agement Unit-RAE, Farnborough, England, September 1963.

2. Butler, Paul, IMP Weekly P rogres s Report f o r the Period Ending 13 May 1964, Greenbelt, Maryland, Goddard Space Flight Center, May 14, 1964.

3. Butler, Paul, IMP Weekly P rogres s Report for the Period Ending 3 June 1964, Greenbelt, Maryland, Goddard Space Flight Center.

4. Car r , Frank A., IMP-I Shadow Fact Sheet, Greenbelt, Maryland, Goddard Space Flight Center, May 7, 1964.

5. Creveling, C. J., Time Correction P rograms for the IMP Information Processing System, Memorandum, Greenbelt, Maryland, Goddard Space Flight Center, May 13, 1964.

6. Haney, Herbert E., and Edward Schoonover, IMP Master t o Experi- menter Data Tape FP OlOA and FP OlOB, Greenbelt, Maryland, Goddard Space Flight Center, November 1, 1963.

7 . House, Clarence B., "Magnetic Core Flywheel and Synchronizer", Transactions of the 1964 International Space Electronics Symposium .

8. International Geophysics Bulletin, Initial Results f rom the First Interplanetary Monitoring Platform (IMP-I), National Academy of Sciences, Number 84. June 1964.

9. Lokerson, Donald C., "Synchronization and Control of the IMP Information Processing System", Transactions of the 1964 In- ternational Space Electronic s Symposium.

10. Longanecker, Gerald W., IMP, S-74 Weekly P rogres s Report fo r Period Ending 11 December, 1963, Greenbelt, Maryland, Goddard - Space Flight Center, December 12, 1963.

11. Meyerson, Herbert, Fourth Fact Sheet, Explorer XIV, Greenbelt, Maryland, Goddard Space Flight Center, October 17, 1963.

20

12. Minitrack System Instruction Manual For 136 Mc Minitrack Inter- ferometer System, Volume I of 11, NASA.

13. Mish, William H., A Description of the IMP-A (S-74) Ground Station Time E r r o r Detection and Rectification Software FP -026. 030. 033 and 040, X-612-64-128, Greenbelt, Maryland, Goddard Space Flight Center, March, 1964.

14. Ness, Norman F., IMP Information Processing, Greenbelt, Maryland, Goddard Space Flight Center, June 29, 1962.

15. Rochelle, R. W., Pulse Frequency Modulation, NASA Technical Note D-1421, Greenbelt, Maryland, Goddard Space Flight Center, October 1962.

16. White, Hosea, Description of IMP Edit P rogram FP 08A and FP 08B (Prel iminary) and Appendices, Greenbelt, Maryland, Goddard Space Flight Center, January 30, 1964.

17. White, Hosea, On the Design of PFM Telemetry Encoders, NASA Technical Note D-1672, Greenbelt, Maryland, Goddard Space Flight Center, June 1964.

18. Wolfgang, John L., Jr., "Design Philosophy of the IMP Information Processing System", Transactions of the 1964 International Space Electronics Symposium.

21

APPENDIX A

"Raw" time plots of DELTA versus N (The Sequence 1 Count)

for the useful life of IMP-I

A- 1

3

w 4

I

P

2000

ORBIT I PERIGEE OCCURS AT 12/01 dad"

SEQUENCE I COUNT BEGINS ON 11/27/63 AT 03h53'"46.175s

ABOVE TIME (SEE TEXT FOR EXPLANATION OF 1.1

ON 11/30/63 BETWEEN 04h56'"54531s UT AND - UT, le, to IS EQUAL TO THE 05h05'"33026s UT A SEVERE OVERLOAD AP- 7 2 0 0 0 - ROSMAN 007 PEARED ON THE SATELLITE POWER SYSTEM THIS SEE STRIP

MALFUNCTION IS VIVIDLY REFLECTED IN THE CHART AND

These strip charts were run on two of the seven Rosman tapes recorded during the first orbit. Note the BCD

time code leads the WWV signal ( t op trace on each chart) by 95 to 100 milliseconds. A l l seven tapes recorded at

Rosman during the first orbit exhibit this error.

This displacement i s reflected in the plot of Delta versus N ( the Sequence 1 Count) above as a positive

displacement of approximately 100 milliseconds in the ordinate values from the nominal curve for these particular

tapes.

Rosman 003, which also contained the above mentioned error, exhibits a negative displacement on the above

plot. This error was introduced by the A / D line when time was decoded.

1- PLOT OF DELTA VERSUS THE SEQUENCE I COUNT AS AN APPROXIMATE 71 SECOND DISCONTINUITY IN THE CURVE

ROSMAN 001 ROSMAN 005 71000- 8 006

ROSMAN 004

COMMENTS BELOW +

\* \\

7300

72oc

noc

73K

72s

71M

ORBIT 2 PERIGEE OCCURS AT 12/04 &ISm

)

)

r 1

. \ . . L+W-"; - . .: " - L-x,

' Y . . >- .

>

TIME 12/01 12102 12n3 1 2 m 12104 I@ w 12"4Zm 06h 54m 01"P 19 19"

I 1 I I I i I I I I 1215 1315 1415 1515 1615 1715 1815 I915 2015 1115

PERIGEE

12/08 18h 25-

N (SEWENCE I UXINT)

ORBIT 3 OCCURS AT

> > i J

5s----*.- . ' _, \ ,../e'

Ps- ' . . . . --/

/ .

TIME /:+:, 15h 12a5 5@' 12/06 Io"11' '/ 04h I2107 23m 22" 12/07 35' 16'47'O 12/08

i' I I I I I I I I I I

042 2142 2242 2342 2442 2542 2642 2742 2842 2942 w 2 N(SE0UENCE I COUNT)

A- 5

74700

73700

72700

7590(3

74%

7390

ORBIT 4 PERIGEE OCCURS AT 12/12 I5'w

D G '--*...d.

. . . . .-. . .

-'/" ,

8 , .

.* *,..** -.\ . .

;./

TlME 12/ 12 12/10 12Al 12/11 -

07h 39" or 5~ 2@W' 1 4 9 6 ~ I I I I I

3669 3769 3069 3969 4069

1 2 m 13027"'

/jFY'

,/: . . .. . I I I I

369 3269 3369 3469 3569 N (SEQUENCE I COUNT) PERIGEE

OCCURS AT 12/16 ORBIT 5

l3W

A-6

77 200

76 200

75 200

79 200

78 200

77200

ORBIT 6

.- rir. *'. *-.:-.* --.\ ~

/,G,w--'--- *.

./ V %

.L-- g ,..v . w 0 /-+y'. . .

-y- + . .f'.

. - c /--= I' PERIGEE

OCCURS AT . ....?

i p- 12/20 <.. .> f . IOh 15'

TIME 12/!7 12/18 12/18 12/19 12/20

0 ? 4 a 01h5T 2cr 04" Kh 16" 08" 2 9 I I I I I I I I I I

5215 5315 5415' 5515 5615 5715 5815 5915 6015 6115

N (SEOENCE I COVNT)

ORBIT 7

'4 PERIGEE

OCCURS AT 12/24

07h 30" TIME

12/21 12/21 12/22 12/23 12/24 x /:/: 04h 57" 23h IO"' 17'22" I Ih 34m 05h 46"

I I I I I I I I I I 6140 6240 6340 6440 6540 6640 6740 6840 6940 7040 7140

N (SEQUENCE I COUNT)

A-7

80 80C

79 80(

78 ea

82 60

81 M)

806c

ORBIT 8

.. . --,+*.;.

.V.G +.-: .- .. .+- f . +*F*--.-

0 /'.? I . :/*.

/ %!

1 .,/"- PERIGEE ,/'?7 OCCURS AT '..=.--. 12/28 04h 50'"

TME 12/25 12/25 12/26 12/27 12/28

02h 04- 20h 16" 14h 2Sm CSh 40' 02" 52m I I I I I I I I I I

7263 7363 7463 7563 7663 7763 7863 7963 8063 8163

N (SEWENCE I COUNT1

ORBIT 9

J- . :

TIME 4 J '1, .I -

,,.' 12/28 12/29 12/30

18 8288 8388 8488 8588 8688 8788 ,8888

23h 21- 17h 34- 1Ih46- I I I I I I I

N (SEOUENCE I COUNT1

PERIGEE OCCURS AT

01/01 02" 0 6

12/31 01/01 05h 56"' 00" 10- - I I 8988 9088 9188

1 ~-

I

A - 8

84600

83600

82 600

86 600

85600

84 600

ORBIT IO

f

' , ;/,>.P: VI

..-:> L?

L$ W 0 /:"?*

. #."'

,..P - *d. .

&.-+ PERIGEE OCCURS AT

01/08 20' 2 5 .:/,

.;/ TIME /. 01/05 01/06 01/07 01/08 01/08 Is" 401 12' 5 P 07" 04- O P l F Is* 29-

I I I I I 1 I I I I

PERIGEE OCCURS AT

01/04 \,.- 23h 5 P " TME .i' 01/01 01m2 01/03 01/04 01/04

I 2c? 17- 14" 29" 08' 41' 02' 54' 21'06' I I I I 1 I I I I

9409 9509 9609 9709 9809 - Kxx)s 10109 10209 9309

N (SE(xENCf I CWNT)

ORBIT II

A-9

88600

87600

866OC

90 8 o C

e9 Bo(

88 80(

ORBIT 12

PERIGEE OCCURS AT

01/12 17%V

01/12 I@ 14"

1 1 I 12059 12159 12259

TIME 01/10 01/11 01/11 01/09

14h5T 0S"w 03h 16" 2P 291 I I I I I I I

11359 11459 11559 11659 11759 11859 11959

N E€WENCE I CCUNTI

ORBIT 13

/--2a- e----

i a- 5 n w

PERIGEE OCCURS AT

01/16 l b 5 P

01/16 TIME 01/14 01/15 01/15

00" 39' 18" 52" 13h 04"

12785 12885 12985 Ix)B5 13185 13295

lp 15"' 06h 2 r

35 12385 12485 12585 12685

/f101A3

N (SEWENCE I COUNT)

I I I , , I

A-1 0

92 9OC

91 9OC

9 0 9oc

95 IOC

941M:

93 Ioc 1,

+.-=.2=.2-~ / &-

A’

PERIGEE OCCURS AT

01/20

8 g w n

12h 00“ 01/18 01/18 01/19 01/20 09 w 2 r 51‘ Id 04- iomw

I I I I I I I

TIME OW7

I I I 0927-

, ,/’

14309 I3409 13509 13609 13709 13809 13909 14009 14109 14209


PERIGEE OCCURS AT

01/24 09h 15’

TIME

01/21 01/22 01122 OV23 01/24 @Ur” Oo”40- 18h 53” 13’05’ 07h 17‘

14431 14531 14631 14731 14831 14931 15031 15131 1523l 15331 N (SEOUENCE I COUNT)

I I

/- 4

I I I I I I I I

A-11

97300

96 3oc

95 XK

998a

9880

97 fa I

7 ii /‘

g z #/.

/

PERIGEE OCCURS AT

OV28 O d 2Sm

/’ TIME

01/25 01/25 OV26 01/27 01/28 0*4V z r s p @or ldlF 041 29-

/ I I I I I I 1 I

15455 15555 15655 15755 I5855 15955 16055 16155 16255 16355


:/ /” PERIGEE OCCURS AT

02/01 09 25-

TIME 01/29 01/32 01/31 02/01 I&?+ I*?# 07h JP 02* O F

16783 I 1 I 1 I I I

16883 1696.5 17083 17183 17283 17383 - 93 16483 16583 is683

N (SEWENCE I COUNT)

A-12

D2.100

103400

102 400

-

Y v)

$ W n

W ZI n

/ / ’ - PERIGEE

OCCURS AT 02/08 22”m

TIME

/! /” /

0

02/05 02/06 U2/07 02/08 02/08 I’

Id 5 6 13h 07m 07b 01~‘ 31- 19“ 45m L I I I I I I I 1 I I

’ /

./’

I 3‘

ORBIT 18

PERIGEE OCCURS AT

02/05 0@48

TIME 02/a ozmz 02103 02/04 02/04 2?5P 16’1F Id 23- o# 35- 2P 4g

1 I I I I I I I I I 17502 17602 17702 we02 17902 18002 18102 m m 18302 18402 100.100

N (SEOUENCE I COUNT)


A-1 3

/ i

02/10 11’021

TME 02/11 059151

02/11 23’ 27-

cwn 17’ 3(F

// ORBIT 21

A- /’ w $ /

/ x */

PERIGEE OCCURS AT

/ 02/16 i6m 15m

/’

./

02/16 TME

02/14 02/15 02/15 / - e

08’25” 02’ 3e“ &50“ *15’@ 02/13 l 4 h Is”

I I I I I 1 I I I 2081 21KU 21281 2Mu 21481 981 20581 20681 20781 20881 20981

A-14

Ill 200

110.200

139.2CC

113.500

II 2.500

111.5oc 2:

ORBIT 22

s 1- / g

8

PERIGEE OCCURS AT

02/20 13h 301

TME 02W 02m 02/19 02/20

V’ I , , I T 11’25”

215% 21696 217% 21896 21996 220% 22196 222% 22396 224%

/ , ldW 04.e I I 23‘ I , ou

N (SEOUENCE I CWM)

ORBlT 23

/- - w !5 / /

P / $;” 02’ 27“ 20’ 40- 14’521 09”-

/ PERIGEE

OCCURS AT 02/24 d 45”

TlME 02/22 02/22 02/23 02/24

r” I I I 1 I I I

5 22625 22725 22825 22925 23025 23125 23225 23S5 23425 -5

N (SEWEKE I COUNT)

A-15

115 400

114 400

113 40(

\ v

1173M:

116 301

11530

ORBIT 24

../ --I ../ --I

/ Y In J / W

PERIGEE OCCURS AT

02/20 07h 45"

TIME 02/26 02/27 02/28

0 6 h 00" I1'48" 02/25 02/25 0SmIl" 23'23" 17'35"

I - _ - 1 - I I I

5 24046 24146 24246 24346 24446 24546

,/" I I I 1

23646 23746 23846 239

u W v)

g W n

-

PERIGEE OCCURS AT

02/20 07h 45"

TIME 02/26 02/27 02/28

0 6 h 00" I1'48" 02/25 23'23" 17'35" I - _ - 1 - I I I

5 24046 24146 24246 24346 24446 24546

IOOO-


ORBIT 25 THE SEOUENCE I COUNT REACHED THE VALUE OF 24,979 CORRESPONDING TO A 90 OAY CLOCK READING OF 100.000, l e , 90 DAY CLOCK = 4 (SEO I COUNT) + 84 AT THIS POINT A NEW REFER- ENCE TIME 1. WAS CHOSEN (SEE TEXT FOR EXPLANATITION OF 1.) THIS NEW REFERENCE TIME IS ON 02/29/64 AT 21m24m22114' UT

PERIGEE OCCURS AT

05h 05" I 03/03

TIME 03/0b 03/02 03/03

09n 4 T 04h 01"

200 300 400 500 600

02/29 02/29 20" 24" 15h 37

I I I II I I I --

o r 12"

100

/ /'

568 24668 24768 24868 24968 0

N (SEOUENCE I COUNT1

A-16

2

I300

0300

- /-.- --,

0 W -/*-

g In

W

n /- '-

PERIGEE OCCURS AT

03/07

-

0zh w / /-

/ TIME

03/04 0 3 D I 03/05 03/06 03/07 o o h O F larn 15" 12h2F 0 6 b 3 Y ob 5111

I I I I I I I I I I 720 820 920 1020 1120 1220 1320 1420 1520 1620

N (SEWENCE I COUNT)

ORBIT 27

PERIGEE OCCURS AT

23h 35m ?/ 03/10

TIME 03/08 0 3 m 03/10 03m 15*W Orr 28- OJb 40" 2P 52"

/- 2r o 3 m 03-

0 W In 4 5 n W

2200

c

A-1 7

5 5oc

45oc

3 501

6 901

5%

4901

ORBIT 28

PERIGEE OCCURS AT

03/ I4 20h 45”’

TIME 03/1 I 03/12 03/13 03/14 03/14

18h 15” 12“ 27- 06m40m OO”52- 19h04m /

I I I I I I I I I I 2766 2866 2966 , 3066 3166 3266 3366 3466 3566 3666


ORBIT 29

c 0 W v) - g \-f*

1 /- ”** x

PERIGEE OCCURS AT

03/18 leh 05”

TIME 03/16 03/17 03/17 03/18

/ 15’22T os“ 34- 03’446- 21” 59- 16h111m

, = 03/15

I I I I I I I I I I 3 3789 3889 3989 4089 4189 4289 4389 4489 4589 4689

N KEWENCE I COUNT)

A-18

Y

7800

ORBIT 30

-

9300 I 6800

5800.

N (SWENCE I COUNT)

ORBIT 31

0 w u)

g 8

PERIGEE OCCURS AT

03/22 15'25'"

- / /-/ TME -- 03/19 03/20 om1 03/21 03/22 IP 34" & 46" 00%- dir 1923-

I I I I I 1 I 1 I I 4913 5013 5113 5213 5313 5413 5513 5613 5713 4813

8300-

A-19

PERIGEE OCCURS AT

/- ,"

10 70

9 70

8 70

12 30

II 30

10 30.

ORBIT 32

PERIGEE OCCURS AT 03/30 os" 5(r

TIME 03/28 03/29 03/30 ,' O@ 5 P 01'05" 19 17- 13m291 07'441'

/'- 03/28

03/27

I 1 I I I I I I I I

6860 6960 7060 7160 7260 7360 7460 7560 m60 7760


ORBIT 33 b

/

PERIGEE OCCURS AT 04/03 07h 25m

/" V W v)

g W a

TIME 03/31 04/01 04/02 04/03 22'33" 16'45" I@ 58- 05h lo"

/"- / 04' 03/51 21"

I I I I I I I I I I

7887 7987 8087 8187 8287 8387 8487 8587 8687 8787

N (SEWENCE I COUNT)

A-20

14.300

13 .m

123oc

16.W

15W

148Y

ORBiT 34

0 W In

4 5 n W

PERIGEE OCCURS AT

04/07 0 4 h 3 6 '

04/07 TIME

04- 04m 04106

1 I I 14"W 0e"W 02%-

I 1. -/A< o r e & O T

9114 9214 9314 9414 9 5 1 4 / 9714 9814

/ p/ " 04104

8914 9014


/' ; /./'/ ORBIT 35

a 4 W

PERIGEE OCCURS AT

04/1 I O l h 5 P

TIME 04/07 04108 04/09 04/0 04/10

2 2 b 561 17'oe" 1r2o" 05' 33" 23' 45- I I I I I I I I I

10637 I0737 10837 10037 10137 I0237 10337 0437 10537

/' / 37 9937

N (SEWENCE I COUNT)

A-21

21.000

24600

23600-

22600

ORBT 36

ORBIT 37

4- -

Y

8

v)

6

PERIGEE OCCURS AT

04/18 2Oh 35m

TlME /'

-

PERIGEE OCCURS AT

04/14 23" 15"'

Y

8

u)

2oooo- g

TIME 04/12 04/13 04/14 04/14 14 26"' 0S"w 02h w 2 1 " W

I I I I I I 10962 11062 11162 11262 11362 11462 11562 11662 11762 11862


A-22

27800

26 800

25800

24800

30600

29600

2860C

276K I:

ORBIT 38

----j--

PERIGEE OCCURS AT

04/22 17h45m

TIME 04/19 04/20 04/21 04/21 04/22 14"59" , Wh 12" 03h 24" 21h36" 15'4e

I I I I I I I I I I 13114 13214 13314 13414 13514 13614 13714 13814 13914

/A Y ul

g W n

/ 13014


ORBIT 39

\, - - - --

0 ,--- 5 //

/

/- - w ul

4 r" W 0

PERIGEE OCCURS AT

04/26 15h lo"

TIME 04/26

/ /' 04/23 12h 12" 06h 24" ooh 36m 18h 4e" 13h OW"

14338 14438 14538 14638 14738 14838 14938 8 I4038 14138 14238

J/ A'' 04/24 04/25 04/25

I I I I I I I 1 I I

N (SECUENCE I COUNT)

A - 2 3

32800

31800

30 800

29800

34900

33 900

32 900

31 900 I 5l

ORBIT 40

1- /' ,7!

04/28 14' 29'"

t /' 04/27 09'W

TIME 04/28 22' 04"

I I I I I 4-- 15065 15165 15265 15365 15465 15565


04/29 16'117'

PERIGEE OCCURS AT 04/30 12h 5om

04/30 10'22p

L- 1- L 1 - 15665 15765 15865 15965

ORBIT 41

8

x m

g-

PERIGEE OCCURS AT

05/04 l 0 h O b

TIME 05/02 05/02 05/03 05/04 d 2 6 m @ 38" 13'55a" 08h 02-

I I I I I I I I I I 5 16093 16193 16293 16393 16493 16593 16693 16793 16893 16933


A-24

36600

35600

34600

u600

WCQO

129 OOO

lz88ooo

127000 II

ORBIT 42

126528

125 528

3 X

07h38"' UT

PERIGEE OCCURS AT

05/00

SATEWTE UNEXPECTEDLY STOPS TRANSMSSION ON MAY II AT &37"UT

/

PERIGEE OCCURS AT

05/12 O b 4 s

05/11 oe" 2P

A-25

50000

49000

48 000

4700c

5290C

51 90(

5090(

49 901 2

Y In

W n

ORBIT 44

SATELLITE RETURNS TO ,>/-- /' /

LIFE AFTER AN UNSCHEDULED SHUT-DOWN OF 12' 39" /-

//* . <-~/

'"/

PERIGEE OCCURS AT

05/16 0 2 h IO"

TIME 05/14 05/15 05/15 05/13

22' 17"'

19150 19250 19350 19450 19550 19650 19750 19850 19950 m 5 0

05/12 21'28" 15'40"' 09'53"' 04' 05"

I I 1 I I I I I 1 I


ORBIT 45

8 u 5 n

Ln

W

OCCURS AT 05/19

23h 2om TIME

05/19 1'

05/18 05/19 21' 18"

05/17 03'06"

,' 05/16 ,/' 20n 29" 14'4p 08' 54"

I I I I I 1 I 1 I I

14 20194 20294 20394 20494 20594 20694 20794 20694 20994 21094


A - 2 6

ORBIT 46

54Doo

53ooo-

2. .. Q _C_--.rT : ' .' --J

-E y-- /-- , .:. ;;y

in , ! =/ PERIGEE .:' '4 05/23

OCCURS AT 6 ' '.

ORBIT 47

5uxx)- I L I , 21224 21324 21424 21524 21624 21724 21824 21924 22024 22124

N(YOVENCE I COUNT)

A-27

56mo

55.000

0 w u!

-x - 5 4 --,

-.------/

/--*/

PERIGEE OCCURS AT

05 /27 /A TIME I Eh 25m

P,.?. - - ' ,. /' .

05/26 05/26 05/27 16%4-

05/25 2 r 5 ~

05/24

1 ORBIT 48

SATELLITE BECAME EXTREMELY INTERMITTENT ON 05/30 AT os"l8'" UT

57100- /'/.

0 / TIME

05/30 &*Im 23768 23060 23968 24068 24168

N(SEPUENCE I COUNT1

05/28 12116- 1 I I L

2350 23568 23468 23% 23668 1

/I- 56.100 -

55567M).

A-28

APPENDIX B

llRaw" time plots of DELTA versus N for orbits 36 and 37 compared with the

corrected and smoothed plots

B-1

ORBIT 36

ORBIT 37

22MK) t 21600 I I I I I 1 1 1 1 1

I2012 12112 12212 12312 12412 12512 12612 12712 12812 12912 N(SEOUENCE I COUNT)

B - 3

APPENDIX C

Telemetry system stability plots

.

c-1

118.0

114.0

I 10.0

106.0

6102.0 W v)

a- 2 98.0

W

-I 3 94.0

s W W 8 90.0 %

86 -0

82.0

78.0

74.0

70.0

a

a

a

a

a

a

a

a

a

a

a

a

a

a

a

?

a

a a

a a

a

4 8 12 16 20 24 28 32 36 ORBIT NUMBER

c-3

0

e

e

e

e

- e e

e 8.0 - - e

e 4.0 e -

- 0

1 I I I 1 I I I I I 1 1 I I I I

'24. ' 28 32 36 40 44 48 52 58 62 ORBIT NUMBER

c-4

7

(P cu v)

QI

0

t a a e is 3 0 c LL 4 a B v)

'33s 'SSW 113d 31W 33NMd3S 39W3AV '33s 'SSVd 113d 31W 33NYl03S 39VMAV

c-5

APPENDIX D

Propagation delay t imes from WWV to the Space Tracking and Data

Acquisition Station Site

D- 1

.

PROPAGATION DELAY TIMES FROM WWV TO THE SPACE TRACKING AND DATA ACQUISITION STATION SITE*

STATION #

61

03

05

166

07

08

12

13

14

15

16

17

18

19

26

55

56

61

STATION

BLOSSOM POINT, MARYLAND

FORT MYERS, FLORIDA

QUITO, ECUADOR

LIMA, PERU

ANTOFAGASTA, CHILE

SANTIAGO, CHILE

ST. JOHNS, NEWFOUNDLAND

FAIRBANKS, ALASKA

EAST GRAND FORKS, MINNESOTA

WINKFIELD, ENGLAND

JOHANNESBURG, SOUTH AFRICA

GOLDSTONE LAKE, CALIFORNIA

WOOMERA, AUSTRALIA

GILMORE CREEK, ALASKA

ROSMAN, NORTH CAROLINA

SOUTH POINT, HAWAII

ASCENSION ISLAND

COLLEGE PARK,MARYLAND

'ROPAGATION

dILLISE CONDS TIME+

.5

5.8

16.1

26.2

24.6

28.7

8.1

27.4

7.2

213.8

45.7

12.6

59.1

27.4

3.6

28.2

30. 1

0.0

*Propagation times calculated by the Network Timing Section, Network Engineering and Operations Division, Goddard Space Flight Center.

tIncludes .3ms for WWV Receiver; Accurate to within t 5 m s .

D- 3

A P P E N D I X E

Timing failure rates on IMP-I for the first 23 orbits

E-1

I.

II.

111 .

IV .

V.

VI.

TIMING FAILURE RATES ON IMP-I FOR THE FIRST 23 ORBITS*

STATION CLOCK SET INCORRECTLY WITH WWV.

141855 = 1.6%

BCD TIME PRESENT AND ACCEPTABLE AT START O F PASS BUT FOUND DEFICIENT AT SOME LATER TIME I N THE PASS.

101855 = 1.27'0

BCD TIME CODE FOUND UNACCEPTABLE - PASS PROCESSED WITH SERIAL DECIMAL TIME.

11855 = 0.270

PASS CONTAINED NO USABLE TIME CODE - PASS PROCESSED WITH RELATIVE TIME.

81855 = 0.9%

A I D LINE COULD NOT SYNCHRONIZE PROPERLY WITH DATA

ADVERSELY EFFECTED THE TIMING. BECAUSE O F A WEAK OR NOISY TELEMETRY SIGNAL - THIS

131855 = 1.570

CAUSE O F UNACCEPTABLE TIMING UNKNOWN.

41855 = 0.570

TOTAL NUMBER O F PASSES WHICH CONTAINED A TIMING FAILURE.

501855 = 5.8%

*Stat is t ics compiled on a p a s s by p a s s bas i s , i.e., a p a s s i s defined as one uninterrupted data recording made a t a single station, not by analog tape. This policy was adopted because one analog tape can and often does contain more than a single pass .

Stat is t ics on timing fai lures were compiled during the entire time the F ie lds and Plasmas Branch had operational responsibility for IMp-IPS, this consti tuted approximately the first 23 orbits.

E - 3

APPENDIX F

Sample calculation for DELTA and N

F-1

Sample Calculation f o r DELTA and N

Sample sequence 1 recorded at Woomera, Australia on January 8, 1964 (this information obtained f r o m the ID of the file) at 20h 39m 51.776' U.T. during orbit number 12.

F r o m equation (1) in text:

(Note that the brackets indicate that N has been rounded to an int ege r . )

Where:

At = 327.656 sec.

t o = November 27, 1963 at 03h 53m 46.175'

t, = January 8, 1964 at 20h 39m 51.766s

All time is referenced f r o m January 1, 1963 thus:

x (330 Days) Sec. t o = 86,400 - Day

x ( 3 Hours) Sec. t 3,600 - Hour

x (53 Min.) Sec. Min

t 60 -

t 46.175 sec.

t o = 28,526,026.175 sec.

t, = 86,400 - x (372 Days) Day

x (20 Hours) Sec. t 3,600 ~

Hour

x (39 Min.) Sec. Min

+ 60 - + 51.776 Sec.

F -3

tn = 32,215,191.776

N = [ 327.656 1 32, 215, 191.776 - 28,526,026.175

N = 11,259

From equation (2) in text:

DELTA (tn - to) - N ( A t )

DELTA = 3,689,165.601 - 11,259. X (327.656)

DELTA = t86.697 Sec.

F-4

APPENDIX G

A comparison of the 30-Day Clock, 90-Day Clock and the

Sequence 1 Count f o r approximately 6 months of satellite operation

G-1

'\

THE IMP-I ON BOARD '3MAY CLOCK'+

THE '90-DAY CL& RECEN ED BY THE EXPERIMENTER

THE SEQUENCE I COUNT, N

3 3 3 3 3 3 2 2 2 1 1 2 2 2 3 3 7 7 7 6 6 7 7 7 3 3

8 6 6 6 9 9 6 6 6 8 9 4 7 7 7 3 6 7 7 7 9 2

1 ; 0 0 0 0 0 0 0

I I I t I I I 1 I 1 I I

3 6 9 9 9 9 9 2 5 8 9 9 9 9 7 5 3 9 9 9 9

8 6 3 0 9 9 9 9 4 7 5 3 6 9 6 9

t I 0 0 0

I I 1 1 1 1 I l l I I I l l

2 2 4 4 9 9 7 7 8 8

I I I I I I I I I *AS THE CLOCK UP-

0 0 DATES ITSELF I DECYAL 0 UNIT PER SEQUENCE a IT TAKES APPROXIMATELY - 81.914 sc . A SEQUENCE 90-DAY CLOCK=4(SEQ.I C0UNT)tM 9O-DAY CLOCK=4(SEQ.I COUNTItO

I

TO READ OUT, TO COUNT TO 32,767 TAKES ABOUT 30 DAYS.

G-3

NOT TO SCALE

Date post:	27-Jun-2020
Category:	Documents
Upload:	others
View:	8 times
Download:	0 times

TIMING ERRORS THEIR DETECTION AND CORRECTION IN THE …€¦ · TIMING ERRORS-THEIR DETECTION AND...

Documents