In Search of the 10-Centimeter Range Measurement · Figure 1. (a) Long-loop test setup; (b)...

1

IPN Progress Report 42-182 • August 15, 2010

In Search of the 10-Centimeter Range Measurement

Mick Connally,* Scott Bryant,* Sherill Hampton,† and Bruce Klein†

* Communications Ground Systems Section. † ITT Corporation.

The research described in this publication was carried out by the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration. © 2010 California Institute of Technology. Government sponsorship acknowledged.

The BepiColombo mission to Mercury desires more accurate radio metric ranging measure-ments to study relativistic effects on the planet’s orbit around the Sun. Ranging measure-ments as accurate as 10 cm have been proposed for this purpose and will require improve-ments in spacecraft hardware and calibration of effects from the transmission media. But there are also improvements needed to the Deep Space Network (DSN) ranging system to achieve this goal. Delay through the equipment must be either controlled or calibrated be-yond the current 2-m requirement. In 2009, a series of tests was performed to characterize the stability of range delay through the DSN uplink and downlink equipment. It was shown that stability is not consistent with the higher accuracy and that a more advanced calibra-tion technique is needed.

I. The Deep Space Network Ranging System

The Deep Space Network (DSN) ranging system supports spacecraft navigation and radio science observations. A particularly challenging set of radio science observations has been proposed for the BepiColombo mission that will require 10-cm ranging accuracy, or about a factor of 10 improvement over current performance [1]. The Deep Space Network (DSN) ranging system consists of five main components:

(1) The uplink (UPL) subsystem, which modulates the uplink carrier with ranging tones.

(2) The transmitter (TXR) subsystem, which amplifies the uplink signal.

(3) The antenna subsystem, which focuses the RF energy on both the uplink and downlink.

(4) The low-noise amplifier (LNA) subsystem, which amplifies the received downlink signal.

(5) The downlink tracking and telemetry (DTT) subsystem, which demodulates the ranging tones from the carrier and performs the ranging measurements.

2

The ranging system also utilizes a test translator that translates the uplink signal frequency (modulated carrier) to the downlink frequencies, effectively emulating a spacecraft tran-sponder. This capability is used to conduct testing and obtain calibration data, as shown below. A more detailed description of the DSN ranging system and its capabilities can be found in the DSN Telecommunications Link Design Handbook [2].

A. Test Configurations

Two methods of measurement of the DSN tracking station internal range exist: (1) short-loop and (2) long-loop. In short-loop testing, the carrier with ranging modulation output from the UPL passes directly into the test translator. The test translator then translates the signal to the downlink frequency, it is downconverted to IF, and then input to the DTT where ranging measurements are generated.

In the long-loop configuration, the UPL provides the signal to the transmitter. The ranging calibration coupler on the transmitter provides the input to the test translator. Again, the signal is then downconverted to IF and input to the DTT.

Diagrams showing both test configurations are shown in Figure 1.

Figure 1. (a) Long-loop test setup; (b) short-loop test setup.

In normal ranging operations with a spacecraft, the range delay is measured prepass using the long-loop test configuration. This measurement is called precal or station delay. Delays through the antenna subsystem, called Z-heights, are measured separately and provided to users. The ranging users (e.g., spacecraft navigation or radio science) calibrate the range measurements taken during the tracking pass by removing the station delay and Z-height measurements. It has been assumed that the range delay thus calibrated remains constant throughout the tracking pass.

B. The Starting Point

The starting point for the study was a measurement of DSN internal range made simultane-ously with Cassini spacecraft ranging at DSS-14 made by Dong Shin. The measurements were made by setting the transponder turnaround ratio on the test translator to provide a calibration signal out of the ranging coupler on the transmitter that was shifted in frequen-

UPL UPL

x xDTT DTT

TXR

LNA LNA

TestTranslator

TestTranslator

MeasuredRange & CarrierPhase

MeasuredRange &CarrierPhase

X-band X-band

X-band X-band

IF IF

(a) (b)

3

cy away from the Cassini downlink. A second DTT was used to measure the station delay. A plot of the data is provided in Figure 2.

18:04:00

4781

4780

4779

4778

4777

4776

4775

477420:17:00

DOY 215

DSS-14 Station Delay During CAS Track

22:30:00

ns

Figure 2. DSS-14 station delay during the Cassini track.

Peak-to-peak variations of 5 ns were observed. The subsequent observation program was designed to validate these results and attempt to discover the cause, and propose methods for control or calibration of this varying ranging term.

C. Approach

Characterizing the internal range delay using X-band uplink/X-band downlink ranging was chosen since the BepiColombo mission will use these links. The observation program proceeded by first establishing the baseline DSN equipment performance. Two facilities were used: (1) the Receiver/Exciter Laboratory in JPL’s Building 238, room 527; and (2) the DSN Test Facility (DTF-21). These facilities contain replicas of DSN equipment found at the DSN’s Signal Processing Centers (SPCs) and antennas, with the exception of the antenna microwave equipment, including the transmitters. The 238-527 laboratory does not have a fiber-optic run simulating the signal distribution at the Deep Space Communications Complexes (DSCCs), but DTF-21 does. In addition, the 238-527 laboratory is not tempera-ture controlled as stringently as SPC-10, but DTF-21 is temperature controlled to the same specification as SPC-10.

Following the establishment of the equipment baseline, the next step was to characterize the internal range variations at DSS-25. This station was selected because (1) it was specially equipped for superior phase stability; and (2) its beam-waveguide design meant that the transmitter and test translator are in the pedestal so the carrier and ranging signal do not travel through the cabling and waveguide required to reach the feedcone and back, as it would at DSS-14 or DSS-15. The underlying assumption was that DSS-25 should produce the most stable ranging calibration results available at the Goldstone DSCC.

4

Internal ranging data would be taken at other DSN antennas as opportunities became avail-able for comparison and contrast with the DSS-25 results.

D. Laboratory Measurements

The first laboratory results from 238-527 were obtained in December 2008. Ranging from the UPL was passed through the test translator directly to the DTT; there was no LNA or TXR in the test configuration. Static ranging data was acquired for about 30 hours. The results are shown in Figure 3.

1610

1608

1606

1604

Ran

ge, r

u 1602

1600

1598

1596

1594

1592

159019 21 23 25 27 29 31 33

Hours of Day

Restarted Ranging

Opened, Then Closed RRP Door

Lab Door Left Open,

7–8 am PST

Lab Door Closed,

1 pm PST

35 37 39 41 43 45

?? Decreased Pr/No, 9:20 am PST?

?? Decreased Pr/No, 10 pm PSTRange through TRI EXC

238-527, DOY 353, Dec18, 2008 Sequential Range Test 7, 162, 312, 500 Hz

These plots show range variation in range units (ru). A range unit and a nanosecond can be seen as equivalent for the purposes of this article. An exact definition of a range unit can be found in [2]. As can be seen, the variations in range can be correlated to events affecting the thermal environment of the equipment. The cause of the anomaly resulting in the loss of receiver lock at about hour 30 is unknown.

The second set of laboratory measurements is also from the 238-527 laboratory. This data set covers almost 160 hours, or almost a week. The results are shown in Figure 4.

Again, variations in the measured range all correlate to events affecting the thermal envi-ronment of the equipment. For example, the large, 5 ru increase at about hour 70 correlates with the air conditioner being turned off for servicing. It is estimated that the temperature increased by about 80 deg C (150 deg F), or about 1 ru/deg C.

Finally, results were obtained from a ranging systems performance test (SPT) performed at DTF-21 in April 2009. The test was run over 60+ hours with the equipment in a thermally controlled environment consistent with the environment at SPC-10. The peak-to-peak rang-ing variation observed was 1.77 ns.

Figure 3. First laboratory results from 238-527, December 18, 2008.

5

1600

1590

1598

1597

Ran

ge, r

u 1596

1595

1594

1593

1592

1591

15900 12 24 36 48 60

Hours of Day

72 84 96 100 120 132 144 156 168

Saturday, Apr 18, 5 am PDT. Air conditioning off for servicing this day; 238-527 Lab temp probably increased 80 °C (150 °F)

238-527 Lab, DOY 106–112, Apr 15–21, 2009 TRI EXC, Constant 2295 MHz (Chn 14)

Figure 4. Second set of laboratory results from 238-527, April 15–21, 2009.

The conclusion reached from the laboratory testing is that provided an environment with a stable temperature, the internal range of the DSN equipment will produce very stable rang-ing results.

II. DSS-25 Results

For the May 13, 2010 test, the long-loop configuration was used, but the IF was also pro-vided to the Radio Science Receiver (RSR). The RSR was configured to digitally record the translated downlink carrier and both the ranging clock tones. This test setup is shown in Figure 5.

Figure 5. DSS-25 test setup, May 13, 2010.

UPL

xDTT

RSR

TXR

LNA

TestTranslator

MeasuredRange

X-band

X-band

IF

Open-Loop Carrier & Ranging Clock

A. DSS-25 May Tests

On May 13, the test setup that included the RSR was used to collect internal range measure-ments. The test was repeated without the RSR on May 14 by Bruce Klein. The results from the DTT are shown in Figure 6.

The results from the analysis of the May 13 RSR data by Meeyong Paik and Jim Border are shown in Figures 7, 8, and 9. RSR data collection was performed by Gene Goltz.

6

Ran

ge, n

s

157835

157834

157833

157832

157831

157830–1 0 1 2

Hours, UTC

3 4 5

DOY 133, Constant 8409.9 MHz DOY 135, Constant 8409.9 MHz

Range Test Xltr Tests, DOY 133–135, 2009

Figure 6. DSS-25 range tests, DOY 133–135, 2009.

Static Range Test, DSS-25, DOY 133–134, 2009

Seconds from First Data

Upper Tone Carrier Lower ToneR

esid

ual P

hase

, Cyc

le

× 1040 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

0.3

0.25

0.2

0.15

0.1

0.05

0

–0.05

–0.1

Figure 7. Static range test, DSS-25, May 13, 2009. The plot shows the residual phase

of the carrier and the higher and lower range clock tones.

7

Figure 9. Static range test, DSS-25, May 13, 2009. The plot shows the frequencies of the

higher and lower frequency range tones relative to the carrier.


Seconds from DOY 133

Drif

t of

Ran

ge T

one

Pha

se, r

u512x(Lower-Upper)Cubic

× 104

8.5 9 9.5 10 10.5

180

179

178

177

176

175

174

173

172

171

170

Figure 8. Static range test, DSS-25, May 13, 2009. The plot shows the

range measurement derived from the RSR data.


Seconds from DOY 133

0.5x(Lower Tone + Upper Tone) – CarrierLinear

× 104

8.5 9 9.5 10 10.5

–0.282

–0.284

–0.286

–0.288

–0.29

–0.292

–0.294

–0.296

–0.298

–0.3

–0.302

B. Conclusions from the May Tests

The DTT measurements show the internal range increasing by about 3 ns over a period of 3 hours on both nights. The amount and rate of increase appear consistent on both May 13 and 14.

The measured range by the DTT is consistent with the results of the RSR data analysis: the variation is not due to measurement by the DTT. The carrier phase is also stable throughout the test period. The variation in measured range seems to be caused by the spectral move-ment of the range tones toward the carrier frequency.

8

C. DSS-25 July 2 Test

A short-loop test was conducted by Bruce Klein on July 2. The ranging measurement results and the carrier phase measurement results are shown in Figures 10 and 11.

These results show very stable range measurements and carrier phase in the short-loop con-figuration spanning the evening hours that produced range variations at DSS-25 in May.

Figure 11. DSS-25 short-loop test, July 2, 2009: carrier phase measurement.

Range Test Xltr Tests, DOY 183, 2009

DOY 183, DSS-25, DCC07, Constant 8409.9 MHz (Chn 10)157706

Hours, UTC

Ran

ge, n

s

157705

157704

157703–2 0 2 4 6 8 10

Figure 10. DSS-25 short-loop test, July 2, 2009: ranging measurement.

Range Test Xltr Tests, DOY 183, 2009

Hours, UTC

Pha

se R

esid

ual,

ns

0.1

0.05

0

–0.05

–0.122 24 26 28 30 32 34

Phase Residual, DOY 183, DSS-25, DCC07, Constant 8409.9 MHz (Chn 10)

9

D. DSS-25 August Tests

The results from the May tests and the July 2 test suggested that the range variations may not be the result of any equipment in the short-loop configuration. To test this, three tests were run at DSS-25 on consecutive nights in August — the 27th, 28th, and 29th. Scott Bry-ant created a command script to switch the ranging configuration from short-loop to long-loop every five minutes. Range measurements were generated every 30 seconds. The results are shown in Figure 12.

Figure 12. DSS-25 tests in August 2009: (a) short-loop configuration; (b) long-loop configuration.

98

7

6

5

4

3

2

1

0

–1

–2–3

DOY 240, DSS25, DCC04, Constant 8409.93 MHz DOY 241, DSS25, DCC07, Constant 8409.93 MHz DOY 242, DSS25, DCC07, Constant 8409.93 MHz

DOY 240, DSS25, DCC04, Constant 8409.93 MHz DOY 241, DSS25, DCC07, Constant 8409.93 MHz DOY 242, DSS25, DCC07, Constant 8409.93 MHz

Hours, UTC Hours, UTC

Test Translator in Short Path, TRI EXC

Range Test Xltr Tests, DOY 240–242, 2009, X-band Range Test Xltr Tests, DOY 240–242, 2009, X-band

Test Translator in Long Path, TRI TXR

Ran

ge R

esid

ual,

ru

Ran

ge R

esid

ual,

ru

–133

–131–132

–134

–135

–136

–137

–138

–139

–140

–141

–142–143

0 02 24 46 68 810 1012 1214 1416 1618 18

(a) (b)

The increase in scatter of the short-loop measurements is due to the poorer signal-to-noise ratio (SNR). The results show that the range variation in the short-loop equipment is rough-ly equivalent to the variation in the long loop — the opposite of the hypothesis tested. Also surprising is that while the increase in range looks very much like that observed in May, the increase starts about midnight local time, not at sunset as in the tests performed in May. The timing of the increase in range also starts later in the test period than in the May tests. This seems to rule out diurnal effects or “warm-up” issues.

III. Results from Other Beam-Waveguide Antennas

A. DSS-24 Test Results

On June 17, a long-loop test was conducted by Bruce Klein. The results are shown in Fig-ure 13. These data taken during midday show good range and carrier stability.

B. DSS-26 Test Results

Two tests were conducted by Bruce Klein at DSS-26. On May 24, long-loop testing was per-formed and on July 2, short-loop testing was conducted. The results are shown in Figures 14 and 15. Both tests show good range measurement and carrier phase stability.

10

Ran

ge, n

s

Range Test Xltr Tests, DOY 144, 2009 Range Test Xltr Tests, DOY 144, 2009


DOY 144, DSS-26, DCC09, Constant 8417.7 MHz (Chn 16)


161411 0.2

0.15

0.1

0.05

0

–0.05

–0.1

–0.15

–0.2

161410

161409

16140816 16.5 16.517 1717.5 17.5 19.518 1818.5 18.519 1919.5

Pha

se r

esid

ual,

Cyc

les

(a) (b)

Figure 14. DSS-26 long-loop test, May 24, 2009: (a) ranging measurement; (b) carrier phase measurement.

Figure 13. DSS-24 long-loop test, June 17, 2009: (a) ranging measurement;

(b) carrier phase measurement.

21 22 23

Range Test Xltr Tests, DOY 168, 2009Range Test Xltr Tests, DOY 168, 2009




588 0.2

0.15

0.1

0.05

0

–0.05

–0.1

–0.15

–0.2

587

586

58516 1617 1718 1819 1920 2021 22 23

Ran

ge, n

s

Pha

se R

esid

ual,

Cyc

les

(a) (b)

Figure 15. DSS-26 short-loop test, July 2, 2009: (a) ranging measurement; (b) carrier phase measurement.

(a) (b)





161289 0.1

0.05

0

–0.05

–0.1

161288

161287

161286–2 230 242 254 266 278 2810 2912 30 31 32

Ran

ge, n

s

Pha

se R

esid

ual,

ns

11

IV. Non-Beam-Waveguide Antenna Results

Long-loop testing was performed by Bruce Klein at DSS-14 and DSS-15. The DSS-14 test was conducted on June 4 and the DSS-15 test was conducted on May 21 and June 4. The results are shown in Figures 16, 17, and 18.

Figure 16. DSS-14 long-loop test, June 4, 2009: (a) ranging measurement; (b) carrier phase measurement.



DOY 155, DSS14, DCC06, Constant 8409.9 MHz (chn 10)


0.24585

0.15

0.14583

0.05

0

4581

–0.05

–0.1

–0.15

–0.2

4579

4577

4575

4573

457116.5 16.517 1717.5 17.518 1818.5 18.519 1919.5 19.520 2020.5 20.521 2121.5 21.5

Ran

ge, n

s

Pha

se R

esid

ual,

Cyc

les

(a) (b)

(a) (b)



DOY 141, DSS-15, DCC06, Constant 8409.9 MHz

Phase Residual, DOY 141, DSS-15, DCC06, Constant 8409.9 MHz

7250

0

7251

0.05

7252 0.1

7249

–0.05

7248

–0.1

7247

7246

7245

72447243

20 2020.5 20.521 2121.5 21.522 2222.5 22.5

Ran

ge, n

s

Pha

se R

esid

ual,

Cyc

les

Figure 17. DSS-15 long-loop test, May 21, 2009: (a) ranging measurement; (b) carrier phase measurement.

These results show that range variations at non-beam-waveguide antennas are much more severe that at the beam-waveguide antennas. This is consistent with the fact that the signal at DSS-14 and DSS-15 must travel up to the feedcone rather than just to the antenna pedes-tal. Still, the magnitude and rate of change were surprising.

V. Range Variations with Frequency

A concern in the beginning was variations in the measured range with frequency. Scott Bryant first noticed these in results of testing in the 238-527 laboratory with a transmit-

12

ter, long-loop, and short-loop. During the test, a range calibration (five independent range measurements, validated to be consistent within the predicted measurement error) was performed across the X-band deep space allocation at 0.5-channel steps. The sweep was repeated. The results are shown in Figure 19.

Discussions with Tim Cornish, the Transmitter Subsystem Engineer, determined that the variation was the result of voltage standing waves in the cabling and the resonance cavities in the transmitter.

Figure 18. DSS-15 long-loop test, June 4, 2009: ranging measurement.

Hours, UTC

–4 –2 0 2 4 6 8 10 12 14

Ran

ge, r

u

Range Test Xltr Test, DOY 215, 2009

DOY 215, DSS-15, DCC02, Constant 8409.9 MHz (Chn 10)7608

7607

7606

7605

7604

Figure 19. X-band range group delay tests in 238-527, DOY 239–247, 2003.

DSN Channel Number

0 4 8 12 16 20 24 28 32 36

MGSM01

40

Ran

ge, r

u

ns, T

XR

DOY 239-247, 2003: X-band Range Group Delay Tests in 238-527

1524 1662

1522 1660

1520 1658

1518 1656

1516 1654

1514 1652

1512 1650

1510 1648

1508 1646

DOY 239 TRI=EXC DOY 240 TRI=TXR DOY 246 TRI=TXR

13

A similar test was conducted at DTF-21. Here, the X-band frequency allocation was sampled at every channel. The results are shown in Figure 20.

The DTF-21 test script was used at DSS-25 on two consecutive nights, May 13 and 14 (Fig-ure 21). The second night the test was conducted by Bruce Klein. The two test results are consistent.

Figure 20. X-band range group delay tests in DTF-21, DOY 035–036, 2009.

Figure 21. DSS-25 test, May 13–14, 2009.

DSN Channel Number00

8395 8400 8405 8410 8415 8420 8425 8430 8435 8440 8445 8450

02 04 06 08 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

STD

L, n

sDOY 035–036, 2009, X-band Range Group Delay Tests in DTF-21

Downlink Frequency, MHz

2316

2314

2312

2310

2308

2306

2304

2302

2300

DOY 035 Test at 0.5 Channel Spacing DOY 036 Test at 1 Channel Spacing

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

Ran

ge, n

s

Range Test Xltr Test, DOY 133–135, 2009

DSN Channel Number

157705

157704

157703

157702

157701

157700

157699

157698

157697

157696

157695

157694

157693

157692

157691

157690

DOY 134, DSS-25, DCC07, TRI=EXC

DOY 135, DSS-25, DCC07, TRI=EXC

14

Because even severe Doppler does not move the downlink over a significant part of the deep space allocation and the result seems repeatable, this was not pursued further.

VI. Conclusions

Variations in the measured internal range were observed at all DSN tracking stations at Goldstone over a period of hours. Current DSN practice of calibrating internal range once, at precal or postcal, is inadequate to achieve 10-cm ranging accuracy even when consider-ing just the DSN contribution. Changes at the non-beam-waveguide stations, DSS-14 and DSS-15, were by far the largest.

At DSS-25, changes of 3 ns were observed over 4 hours on consecutive days in May. Analy-sis of RSR data shows that the increase in the measured range was caused by the ranging modulation moving in the frequency domain with respect to the carrier, not an increase in physical delay through the tracking systems. In August, changes of over 4 ns were observed over 6 to 12 hours on 3 consecutive days. The observations in August included both “short-loop” and “loop-loop” (without and with the transmitter, respectively) and were consistent within the limits of the data collected. Several causes were investigated but the root cause remains a mystery.

Here is a summary of the conclusions from the study:

(1) Ranging calibration as currently performed in the DSN is inadequate to support 10-cm ranging.

(2) Ranging variations at DSS-25 are repeatable on a daily basis but change over months.

(3) The ranging variations at DSS-25 are not correlated to local time-of-day or equip-ment “warm-up.”

(4) Precal range calibrations are consistent at each antenna. This is verified by the automated range calibration software, which records the measured value and com-pares it to measurements from previous tracks.

(5) The cause of the range variations is not obvious:

• The variations in measured range are not caused by a physical increase in the signal path, such as thermal expansion, since we see very constant carrier phase.

• The variations are not caused by either the modulation or measurement of the ranging signal since no variation shows up in DTF-21 testing using identical equipment.

• The variations are not caused by the DSN transmitter since we see identical variations, short-loop and long-loop, at DSS-25.

• The variations are not caused by modulation or demodulation onto the fiber-optic cables since these are included in the DTF-21 configuration.

15

• The variations are more severe at the non-beam-waveguide antennas than at the beam-waveguide antennas.

VII. Next Step

In order to attain 10-cm ranging, the internal range delay in the DSN ranging system must be measured during tracking so that range delay variations can be calibrated. In addition, simultaneous ranging at two uplink/downlink frequencies would allow calibration of charged particle effects on delay, an important consideration for a mission to Mercury, near the Sun. An Advanced Ranging Instrument (ARI) has been proposed that meets these needs. It provides constant measurement of internal range delay and provides simultaneous range measurement at both X-/X-band and Ka-/Ka-band.

Acknowledgments

The authors would like to thank Sami Asmar for initiating this task and reviewing the report and James S. Border for work in analyzing and interpreting the RSR data and his overall technical guidance. The approach this study took was based largely on his ideas. The authors would also like to thank Gene L. Goltz for RSR data collection, Meegyeong Paik for analysis and interpretation of the RSR data, and Dong Shin for providing background on his initial results and for his advice. Finally, the authors thank Lawrence Teitelbaum for his support of this work.

References

[1] Luciano Iess, Sami Asmar, and Paolo Tortora, “MORE: An Advanced Tracking Experi-ment for the Exploration of Mercury with the Mission BepiColombo,” ACTA Astronau-

tica, vol. 65, pp. 666–675, September/October 2009.

[2] DSN Telecommunications Link Design Handbook, DSN No. 810-005, Module 203, “Se-quential Ranging,” Rev. C, Jet Propulsion Laboratory, Pasadena, California, October 31, 2009. http://eis.jpl.nasa.gov/deepspace/dsndocs/810-005/

In Search of the 10-Centimeter Range Measurement · Figure 1. (a) Long-loop test setup; (b)...

Documents

Transcript of In Search of the 10-Centimeter Range Measurement · Figure 1. (a) Long-loop test setup; (b)...