ELEKTROTECHNIEK RUIMTEV AART. IEEE Student Branch Eindhoven

SYMPOSIUM

ELEKTROTECHNIEK IN DE RUIMTEVAART Proceedings

IEEE Student Branch Eindhoven

Uitgever:

IEEE Student Branch Eindhoven Postbus 513 5600 MB Eindhoven

Aile rechten voorbehouden. Niets uit deze uitgave mag worden verveelvoudigd, opgeslagen in een geautomatiseerd gegevensbestand, of openbaar gemaakt, in enige vorm of op enige wijze, hettij elektronisch, mechanisch, door fotocopieen, opnamen, of op enige andere manier, zonder voorafgaande schriftelijke toestemming van de uitgever.

All rights reserved. No part ofthis publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronically, mechanically, by photocopying, recording, or otherwise, without a prior written permission of the publisher.

Inhoudsopgave

Inhoudsopgave bIz.

0.1.

Inleiding Elektrotechniek in de Ruimtevaart C.H.F. Lennartz, Chairman IEEE Symco'93

1.1

Hoogspanningsproblemen in satellieten dr.ir. J.M. Wetzer, Technische Universiteit Eindhoven

2.1.

Satellietcommunicatiesystemen: techniek ell technologie ira R. Suttels, Alcatel Bell Space Systems

3.1.

De tracking van de ERS-! satelliet en de verwerking van ERS-! radarhoogtemetingen ira B. Arrlbrosius, Technische Universiteit Delft

4.1.

Satellite communication systems planning in an interference environment ira R. Hekmat, PTT Research

5.1.

Space Power Electronics • Design Drivers D. O'Sullivan, B.E., ESA·ESTEC

6.1.

De betekenis van robotica in de ruimtevaart ira W. de Peuter, ESA-ESTEC

Z1.

Space-qualified optical memory for the Columbus Pressurized Module ira T. Algra, Nationaal Lucht- en Ruimtevaart Laboratorium

8.1.

A modular instrumentation concept for experiments under microgravity ing. L.J. Aartman, Nationaal Lucht-en Ruimtevaart Laboratorium

9.1.

Nawoord J.M.W.M. Janssen, Chairman IEEE Student Branch Eindhoven

10.1.

Bestuur IEEE Symco'93

11.1.

Lijst van adverteerders

12.1.

Commissie van aanbeveling

13.1. Woord van dank

Chris Lennartz: In/eiding Elektrotechniek in de Ruimtevaart

Inleiding

Elektrotechniek in de Ruimtevaart C.H.F. Lennartz

De Eindhoven Student Branch van the Institute of Electrical and Electronics Engineers, Inc. is een organisatie die haar leden onder andere in de gelegenbeid stelt onl toegang te hebben tot dezelfde informatiebronnen als degenen die de ingenieurstitel al verworven hebben. Een manier van informatieoverdracht die altijd tot de verbeelding heeft gesproken en dat waarschijnlijk ook altijd zal blijven doen is het symposium, zeker als het de verbeelding van de student betreft. De grootste kracht van een symposium ligt in het feit dat allerlei mensen van verschillend pluimage, zowel academisch als industrieel, met elkaar van gedachten kunnen wisselen over het onderwerp. Dit onderwerp en de invalshoek waaronder het tijdens het symposium belicht zal worden moet dus zeer bewust gekozen worden. Het criterium voor de onderwerpkeuze van afgelopen jaren was dat het onderwerp een "hot item" in de elektrotechniek behoorde te zijn en titels als Fuzzy Logic: Systems and Design, Multimedia, Telematics: Technology and Applications en Mobiele Communicatie waren afgelopen jaren dus niet van de lucht. Met deze onderwerpen is de voorraad hot items voorlopig uitgeput, en het nu houden van een symposium over nieuwe ontwikkelingen in een van deze gebieden zou hetzelfde (tegen)effect hebben dat een sequel, een geforceerd vervolg op een succesfilm, doorgaans sorteert. Vandaar dat dit jaar gekozen is voor een ander criterium: aspectintegratie. Dit houdt in dat een onderwerp gekozen wordt waarin vele aspecten van de elektrotechniek vertegenwoordigd zijn, zodat de specialisten op een bepaald gebied hun vakgebied in een bredere context geplaatst zien. Een ideaal onderwerp dat aan dit criterium voldoet is natuurlijk "Elektrotechniek in de Ruimtevaart". Bovendien heeft ruimtevaart altijd al tot de verbeelding van de. mens gesproken, vooral vanaf het begin van de bemande ruimtevaart met als "kleine stap voor een mens, maar grote stap voor de mensheidt, Neil Armstrong's eerste schreden op het maanoppervlak. Het is geen toeval dat de vakgebieden van de elektrotechniek en de ruimtevaart vanaf het begin der zestiger jaren een parallelle explosieve groei hebben doorgemaakt. De wisselwerking tussen beide disciplines heeft duidelijk aan deze groei bijgedragen. De meest in het oog springende toepassing van elektrotechniek in de ruimtevaart zag het levenslicht toen Arthur C. Clarke in 1945 in Wueless World een 0.1

Chris lAmartz: Inleiding Elektrotechniek in de Ruimlevaart

artikel publiceerde waarin hij de geostationaire baan van een kunstmaan aantoonde; de basis voor wereldomvattende satellietcommunicatie was gelegd. Om de betrouwbaarheid van zulk een communicatielink te garanderen zal de satelliet zeer nauwkeurig genavigeerd moeten worden, wat geschiedt met behulp van geavanceerde regelsystemen. Het spreekt natuurlijk vanzelf dat de apparatuur aan boord van de satelliet voor het merendeel bestaat uit electronica in de vorm van hoog-technologische telecommunicatieapparatuur die ervoor moet zorgen dat de "ruimtelijke ordening" zo min mogelijk voor communicatiestoringen zal zorgen. Om deze apparatuur te laten functioneren is een efficiente energievoorziening in de vorm van bijvoorbeeld zonnepanelen noodzakelijk. Het voeden van de Traveling Wave Tube Amplifiers brengt een aantal hoogspanningsaspecten met zich mee die noodzakelijk zijn voor een goede werking. Blijkbaar zijn er alleen al in de satellietcommunicatie genoeg aspecten van elektrotechniek om een symposium aan te wijden, maar in andere disciplines doet elektrotechniek evenzeer van zich spreken. Bij de verkenning en observatie van hemellichamen speelt remote sensing m.b.v. radar en/of radiometrie een vooraanstaande rol, voordat wordt overgegaan op meer fysieke verkenning waar de robotica van wezenlijk belang is. Verder is de radiocommunicatie in alle vormen van het bedrijven van ruimtevaart van essentieel belang, wat hopelijk geen uitleg behoeft. Deze voorbeelden geven blijk van het feit dat het gehele vakgebied der elektrotechniek onmisbaar is bij het bedrijven van ruimtevaart. Dit geldt echter ook vice versa: vele producten worden eerst in de ruimte getest om na te gaan of ze onder extreme atmosferische omstandigheden nog adequaat functioneren. Concluderend kan er dus gesteld worden dat een symposium over Elektrotechniek in de Ruimtevaart uitermate informatief kan zijn voor aldiegenen die vanuit welk oogpunt dan ook enigermate geinteresseerd zijn in ruimtevaart. Ik hoop dan ook dat deze proceedings zullen bijdragen aan het verwerken van die informatie. C.R.F. Lennartz Chairman Symposiumcommissie

0.2

Hoogspanningsproblemen in satellieten

de..•·W • •·F:indh9ven. •1I1.J?84. •. pr()tl1?y.e~rd~.•.•hiJgI)~¢1l: ?n4rr~9.¢~ : Aaa~ ·•..9ptische

maanden.•·door. • • bij.•·8ntaTjo•.· IIydf()Jtes~.a~fh .pivi~i?I1..•;iI1/.T9ront();.·. •
werkt~

daaraang~geis()leerd~~~hakelsyst~fu~l1i· ···:··

..... .

'. .....

en

Zijhhaidig¢~ef~~~~~~li~~~g~I~~~i~~~ de~~~~~~~~~~~t1~dingen

en ..i~OI(lti~,:. ·.•.Il1et~arn.~j~yaCll~Jll. eIJ.!n.g~S:y?rtl1ig~:Ine
t~di~~~~~?!E~!M~~;ll~'~/lil~~~r~~f~f~r~=~l~~!~~

uitgevoerd in .saIllellWerlting . niet::a~ .E4rppes~ . ~qiIl1tey~(lt.tOrg~nisati~ESA..In

.1992·· ontvinghiJvpor 41t.Werlc.v~r•. ·~pj~£H#rg¢s·riI1d~!f¢~~icaI·· rriSUHiti6Hliitvacuthp.". . • •. ·.: :...:.

.

1.1

I.M. Wetzer: Hoogspanningsprob/emen in satellielen

Hoogspanningsprohlemen in satellieten Dr.ir. J.M. Wetzer Vakgroep Hoogspanningstechniek en EMC Technische Universiteit Eindhoven Satellieten herbergen een veelheid aan elektrische en elektronische apparatuur. Soms spelen hoge spanningen daarbij een belangrijke rol. Hoge spanningen zijn bijvoorbeeld essentieel voor het functioneren van lopende-golf zendbuizen (travelingwave tubes of TWTs). Deze buizen worden gebruikt om hoogfrequente signalen, continu dan weI gepulst, te versterken. Versterking van CW (continuous wave) signalen is van belang voor satelliet-communicatie, versterking van gepulste signalen speelt onder andere een rol bij radar (denk aan de European Remote Sensing satelliet). Bij CW-buizen gaat het om frequenties in de range van 10-40 GHz en vermogens van 10-250 W, bij gepulste buizen liggen de frequenties in de orde van 5 GHz en bedraagt het puls-vermogen ongeveer 5 kW. De hierbij gehanteerde spanningen varieren van 1 tot 25 kV, corresponderend met veldsterkten in de range van 10-100 kV Icm. De genoemde spanningen en velden moeten betrouwbaar geproduceerd en gehanteerd worden in vacuum, hetzij het vacuum in de zendbuis (ongeveer 10-6 mbar), hetzij het satelliet-vacuum (ordegrootte 104 mbar). Vacuumisolatie wordt overigens niet alleen in de ruimtevaart bedreven, maar ook op aarde. Hierbij kunnen we bijvoorbeeld denken aan beeldbuizen, rontgenbuizen en vacuumschakelaars. Toepassing in satellieten stelt echter hogere eisen aan de isolatie vanwege: 1. de gewenste betrouwbaarheid (in verband met de continuiteit van transmissie en de hoge "voorrijdkosten"), 2. de agressieve omgeving (hoogenergetische fotonen), en 3. de variatie in de actuele druk; satellieten worden op aarde getest bij atmosferische druk maar moe ten, na lancering, opereren in het satelliet-vacuum; bovendien is dit satelliet-vacuum veranderlijk van samenstelling en druk. De vakgroep Hoogspanningstechniek en EMC van de TU Eindhoven verricht sinds 1983 onderzoek aan hoogspanningsproblemen in satellieten, in nauwe samenwerking met het onderzoekinstituut ESTEC van de Europese Ruimtevaart Organisatie (ESA). Dit werk is zowel fundamenteel als toegepast van aard. Het fundamentele onderzoek houdt zich bezig met de vraag welke mechanismen en processen verantwoordelijk zijn voor doorslag. Het toegepaste werk heeft tot doel deze inzichten te vertalen naar richtlijnen voor het ontwerp en bedrijf van toekomstige componenten, en deze richtlijnen te verifieren. Deze ontwerpregels zijn onder meer met succes toegepast bij het ontwerp van de ERS-l radar-zendbuis. In deze presentatie wordt ingegaan op de geschetste problematiek en worden de bevindingen besproken van een in 1991 afgerond onderzoek. Als leidraad bij deze presentatie is de "Executive Summary" van het eindrapport bijgevoegd. 1.2

1M. Wetter: Hoogspanningsproblemen in satellieten

EUROPEAN SPACE AGENCY CONTRACT REPORT

High-Voltage and EMC Group Department of Electrical Engineering Eindhoven University of Technology ESTEC CONTRACT 7186/87/NL/JG(SC)

1.3

I.M. Weller: Hoogspanningsproblemen in satellieten

November 1991 EUROPEAN SPACE AGENCY CONTRACT REPORT The work described in this report was done under ESA contract. Responsibility for the contents resides in the authors or organisation that prepared it.

EXECUTINE SUMMARY OF THE STUDY ON HV-DESIGN ASPECT'S OF MICROWAVE TUBES EHC/JW/MV/RAP91010

ESA Technical Management

Project team EUT

Dr. S.J. Feltham

Dr.ir. J.M. Wetzer Dr. P.AA.F. Wouters Ing. A.J.M. Pemen Dr. M.G. Danikas Prof.dr.ir. P.C.T. van der Laan A.J. Aldenhoven High-Voltage and EMC Group Eindhoven University of Technology P.O. Box 513 5600 MB Eindhoven The Netherlands

ESTEC/Contract no. 7186/87/NL/JG(SC)

Front cover:

1.4

Example of optimized insulator design for concentric cylindrical conductors

I.M. Wetter: Hoogspanningsprob/emen in sateUieten

INTRODUCTION The present generation of traveling wave tubes (TWf's) suffers from spurious discharges and tube failure as a result of high voltage breakdown. Future TWf's for use in spacecraft will operate at increased frequency and power. As a consequence the designs should be made more compact, and yet able to withstand higher voltages. Various microwave tube designs typical of future application at frequencies above 11 GHz have been studied within the frame work of the European Space Agency's "Advanced Space Technology Programme". It was reported that spurious switch offs can occur on a seemingly random basis. Insulation breakdown in the region of the high voltage feedthroughs was mentioned as the prime suspect as the source of these SSO's. It was concluded that a detailed investigation of high voltage design aspects was required, in order to achieve a dramatic improvement of the high-voltage performance. In this study the high voltage (HV) design concepts used in microwave tube technology are investigated. The objective of this work is to provide guidelines for the improvement of the HV performance in order to meet future requirements.

APPROACH Means to improve the high-voltage performance of microwave tubes involve the choice of materials and material treatments, field control techniques and conditioning. The present state of the art with regard to materials and material treatments does not offer opportunities for significant inlprovement, mainly due to the lack of information on long term behavior and stability. Much better opportunities for improvement are given by field control and conditioning techniques. Based on a literature survey, on an evaluation of existing designs and on earlier work, problem areas are defined. These problems areas are implemented in a number of test geometries that are subjected to various tests. The complete set of insulator shapes is shown in Figure 1. The tests include DC-current measurement, partial discharge measurement, breakdown voltage measurement, and time-resolved measurement of current and optical emission during breakdown. Also different conditioning procedures are studied experimentally. A theoretical study on surface charging is performed to support the evaluation of the experimental results. From the theoretical and experimental results guidelines are derived for the design and operation of vacuum high-voltage components. Specifically, guidelines are formulated regarding the design of insulators, feedthroughs, high-voltage cables and tube assemblies, and regarding procedures for electrode treatment, conditioning, potting and pre-flight testing at ambient pressure.

1.5

I.M. Wetter: Hoogspanningsproblemen in saJeUieten

J.J_

2

*

3 4

-----

s"

*

*

-_6..

- .. -

7_ _---_ 8 _ _9- 10 ..

..

...

..

..

INSULATOR GEOMETRIES FiguTf! 1 Insulator shapes used in the present study. For the asymmetric shapes denoted • (4, 5, 6 and 10), extension "a" indicates that the cathode is at the right side (smallest contact surface), extension "b" indicates that the cathode is at the left side (largest contact surface).

PROGRAM DESCRIPTION The contract was divided into two phases. Phase I was split up into two workpackages. The following program was adopted: PHASE 1, Workpackage 1 update of the literature survey performed in 1983 under ESA contract 5419/83/NL/GM(SC); evaluation of existing HV tube sub-assembly designs with respect to geometry and materials (electrodes/insulators); reduction of the complex HV design configurations into basic design elements; investigation of available insulator and electrode materials. PHASE 1, Workpackage 2 design and procurement of an experimental set-up to investigate the HV performance of the designs analysed in workpackage 1; definition and procurement of test samples to be investigated.

1.6

I.M. Wetter: Hoogspanningsproblemen in satellieten

PHASE 2, Workpackage 3 experimental program; evaluation of the results in terms of the mechanisms determining the voltage holdoff capability; formulation of guidelines for the High-Voltage design of future tubes and for the conditioning procedures to be applied; continuous update of the literature survey; preparation of the final report.

BASIC FINDINGS The two key mechanisms, important for the design of vacuum high-voltage components, are primary electron emission and surface charging. These mechanisms drastically affect both the holdoff performance and the conditioning process. Primary electron emission from negative electrodes can be reduced by field control, in particular by concentrating the field at the positive electrode. Surface charging should either be avoided or controlled: properly shaped insulators can trap charges such as to reduce the cathode field, proper conditioning can produce a beneficial charge distribution. The choice of conditioning procedure affects the choice of insulator geometry. Conditioning can drastically improve the voltage holdoff capability. Effective conditioning requires a number of breakdowns. It is therefore important that future designs permit breakdowns, with limited energy in order to avoid damage. The number of breakdowns required to establish a high breakdown voltage depends on the insulator geometry. Because conditioning is related to the deposition of surface charge, the conditioning effect may be lost as a result of charge leakage or exposure to gases.

CONCLUSIONS

Materials and material treatments 1.

A literature survey of new, extensively studied, electrode and insulator materials does not reveal alternatives which meet all requirements and perform significantly better than the materials used in practical designs.

2.

New materials and material treatments such as laminated electrodes, and insulators with surface coating and bulk doping, have not been sufficiently studied with respect to their stability and can therefore at present not be recommended as reliable alternatives. Such materials and treatments could form the basis for future study.

3.

Mechanical polishing of electrode surfaces is not recommended because impregnated particles can serve as electron emission sites which might initiate

1.7

JM. Wetter: Hoogspanningsproblemen in satellieten

a breakdown. 4.

The grinding of a sharp recessed corner into an alumina spacer can cause the formation of small cracks, which may lead to electrical breakdown. Recessed corners should preferably be rounded off in order to mimimize mechanical stress concentrations.

5.

Next to the choice of insulating material, the quality of the manufacturing process is of crucial importance for the insulator hold-off performance.

Field control 6.

Field control techniques offer good opportunities for the improvement of the HV-performance of TWTs. The first objective should be the prevention of primary electron emission from negative electrodes.

7.

The Finite Element Method (FEM) is a powerful tool to locate high field regions and to study design modifications. For DC, the applicability is sometimes limited by leakage of the insulator and by prebreakdown currents. Both processes result in charges on the insulator surface.

Diagnostics 8.

Sensitive pre-breakdown current measurements, such as (pA-range) DC current and (pC-range) partial discharge measurements, provide insight into the mechanisms responsible for insulator flashover. They give a reliable indication of the voltage hold off performance of an insulator or component only at voltages close to the breakdown voltage.

9.

Breakdown voltage measurements reveal important information on the conditioning process and on the role of surface charge. The effect of surface charge can be observed very directly by performing breakdown measurements before and after an insulator has been exposed to low pressure dry N2•

10. The time-resolved measurement of the breakdown current shows that the electrodes are fully discharged upon a flashover. More charge may be involved, depending on the external circuit. Further, the time-resolved measurement of current and optical emission during breakdown shows how the breakdown process is influenced by the insulator shape.

Surface charging 11. Surface charging determines to a large extent both the voltageholdoff performance 1.8

I.M. Wetzer: Hoogspanningsproblemen in satellieten

and the conditioning process. It should therefore be incorporated in the insulator design.

Conditioning 12. With an effective conditioning procedure it is possible to attain a dramatically improved and reproducible breakdown voltage. 13. For most insulator geometries, effective conditioning requires at least a few breakdowns. It is therefore important that future components are designed in such a way that breakdown conditioning is permissible. This requires a limitation of the breakdown energy. A value of around 30 mJ appeared to be safe and effective in this work, but is not necessarily the optimum value. 14. Of the conditioning procedures tested, step-conditioning provides the fastest rise of breakdown voltage, and the smallest spread. The step-conditioning procedure is illustrated in Figure 2. 50~------------------------------------~ 80

40

30 -

20

10

o

20

40

60

eo

100

VOLTAGE (kV) VERSUS TIME (minutes) (BD =breakdown) Figrue 2: Applied voltage versus time for step-conditioning procedure.

15. The choice of conditioning procedure should influence the choice of insulator geometry, and vice versa. 16. With a sufficient number of breakdowns, critically designed insulators may still attain

1.9

J.M. Wetter: Hoogspanningsproblemen in salellieten

a high breakdown voltage. 17. The conditioning effect may be largely lost when insulators loose their surface charge. The loss of charge can be the result of charge leakage (when the voltage is switched off for long periods of time) or of exposure to low pressure N2, or any other gas at low pressure. 18. For space applications, conditioning should be required only once, before launch, preferably at the pumpstand before sealing. This requirement can only be met for well-chosen geometries.

Operation in vacuum and air 19. The design requirements with respect to operation in air and in vacuum are different and sometimes conflicting. Pre-flight tests in air are not representative for the behavior in vacuum, and may even cause insulation degradation.

GUIDELINES General guidelines 1.

a.

2.

Mechanical polishing of electrode surfaces is not recommended because impregnated particles can serve as electron emission sites which might initiate a breakdown.

3.

It is not justified, and even hazardous, to quote safe local cathode-field values. The macroscopic (design) field depends on materials and geometry whereas the actual field also depends on microscopic enhancements and contamination. A better approach is to quote the average field (voltage over electrode distance) for a given geometry (see also Table 2).

1.10

Reduce the field at the negative electrode (the "cathode") by shielding, and by eliminating protrusions as well as contamination by polishing (see guideline 2). b. Reduce in particular the field at the cathode triple junction by an appropriate shaping of electrodes and insulators eliminating voids and imperfect joints at the triple junction. As an alternative, voids or imperfect joints can be made field free by metallization techniques enhancing the field at the anode

1.M. Wetter: Hoogspanningsproblemen in satellieten

Insulator shape 4.

The choice of an insulator geometry depends on the conditioning process (with or without breakdowns, number of breakdowns), and on the operating conditions (regular exposure to gases, regularly switched off for long periods of time).

S.

Important insulator design parameters are: the minimum breakdown voltage the conditioning speed the sensitivity to charge loss by gas exposure or leakage. For each parameter, or combination of parameters, a ranking of insulator shapes can be derived. As an example, Table 1 gives the ranking with respect to the combination of minimum breakdown voltage and conditioning speed. Ranking of insulator shapes, on a scale from 0 to 10, with respect to the combination of minimum breakdown voltage and conditioning speed (DC voltage). The upper side of the samples shown is the cathode side.

Table 1

Score

Judgement

Remarks!

-:7

10

EXCELLENT

=:J

9.2

AE,SS AE,NS AE,SS AE,NS

Sample 6b lOb Sb 4b Sa

2 6a 7 lOa 9 1

8.S 8.0 ~

=> ~

~

=:=J

4a

8 3 1)

6.

=s

·

7.3 7.1 6.4 6.3

·

S.2 S.O

INSUFFICIENT

4.7 3.2 2.9 1.5

REASONABLE

·

· BAD

·

CE,SS BE,SS CE,SS BE,NS CE,NS BE,NS NS CE,NS BE,NS

AE =anode enhanced CE =cathode enhanced BE =both enhanced SS =stepped shape NS =non-stepped shape In terms of breakdown voltage, conditioning speed and sensitivity to charge loss, geometries with field enhancements at the anode are superior. 1.11

1M. Wetter: Hoogspanningsproblemen in sateUieten

7.

If cathode field enhancements are unavoidable, stepped shapes are recommended.

8.

The insulator shape dramatically affects the allowed averaged fieldstrength (see Table 2). The minimum breakdown fields observed are, of course, not safe design values but are subject to normal derating for space applications. In our opinion the insulator shapes at the bottom of the list ought to be derated more than those at the top of the list.

Table 2.

Averaged breakdown field (breakdown voltage over distance), the number of breakdowns required to reach 10 kV/mm, and the breakdown energy used in this work. Note that in the design the maximum rated fields are subject to normal derating for space applications. The upper side of the samples shown is the cathode side.

Sample

10.2 9.0 8.2 7.4

>12 >12 >12 >12

0 1 8 5

26·36 20-36 17-36 14-36

=s =>

7.2 7.0 6.6 6.2

>12 >12 >12 >12

36 24 45 40

13-36 12-36 11-36 10-36

=:=J

5.4 5.2 5.0

>12 >12 >12

78 61 188

7-36 7-36 6·36

4a ~ 8 3

4.0 3.8 2.8

>12 >12 >12

25 109 71

4·36 4-36 2-36

lOa 9 1

=>

1.12

Breakdown energy (mJ)

6b ~ lOb =:J 5b 4b J Sa 2 6a 7

9.

Averaged Number of Averaged breakdown field breakdown field breakdowns to (kVjmm) reach 10 (kV/mm) MINIMUM MAXIMUM kV/mm

Especially for space applications, insulator shapes with enhanced anode fields (6b, lOb, 5b,4b), which do not rely on surface charge, are advised because: hardly any breakdown is required in the conditioning procedure, and conditioning is required only once; repeated conditioning after long time switch-off is not necessary.


10. A proposed optimized insulator shape for DC voltages is shown in Figure 3. The advantages include breakdown voltage, conditioning speed and insensitivity to charge loss. Further, it is taken into account that triple junctions are never perfect and should be made harmless by surface charging. Other design examples based on the guidelines presented are discussed in the Final Report. 11. Based on the first breakdown voltage, a preliminary guideline would be to use stepped insulator shapes for AC applications. Before a definite guideline can be given, the adverse effects of surface charge at AC voltage should be further investigated.

8

Fi8f.W 3:

.An example of an optimized insulator design for DC

Feed through 12. Cathode field concentrations can be minimized by the shape of electrodes or insulators metallization of the inside of the tubular insulator a larger clearance between central conductor and insulator

13. Feedthroughs should be specifically designed either as a vacuum feed through or as a potted feedthrough. A well-designed vacuum feedthrough looses quality if potted. 14. Feedthrough connections should be well shielded (e.g. by a stress cone) in order to 1.13


avoid field concentrations.

High-voltage cable 15. In order to avoid partial discharge activity, and subsequent damage, it is advised to: surround each insulated wire with a (semi-) conductive layer, and use extruded, rather than tapewound and sintered, dielectrics. Such cables do not yet exist and should be a candidate for future development. 16. Cable ends should preferably not be potted, because potting impedes cable outgassing and introduces field enhancements.

Potting 17. Potting can give reasonable results if done properly (clean surfaces, potting under vacuum and outgassing of potting material). Improper potting causes partial discharge activity in voids (at the interfaces or in the bulk). Literature usually reports on bulk properties of potting materials, and not on applications and manufacturing techniques, whereas these are essential for the quality. Potting may prevent proper outgassing of the potted component and thereby makes the inside pressure uncontrolled (see high-voltage cable). Potting techniques must follow a strictly controlled and qualified procedure.

Tube assemblies 18. Because an effective (i.e. breakdown) conditioning procedure results in a dramatic improvement of the breakdown voltage, future components should be designed in such a way that breakdown conditioning is permissible. This requires control of the breakdown energy (or capacitance), and can be realised by using subdivided assemblies (if the energy is too large) or additional capacitance (if the energy is too small). Subassemblies should be decoupled only during the (high frequency) breakdown event, for example by means of inductances. An example of such a subdivided assembly is shown in Figure 4.

1.14

.1.

I:

:L

.'j

_0'· ~

distributed

..

'

.,

·........... C ~i--I~'--,

/"'"f"...'i,.....r,l"~

\.. ..... ' 1,-,,-,,,

(deC0upled ':'1 :nauctance

L-;

Figure 4: Schematic view of design with subdivided assemblies with low partial capacitance, interconnected by inductances which block the high-frequency cun-ent during a conditioning breakdown.

1.15

Satellietcommunicatie systemen: techniek en technologie . .. .

' .

~.:

.....

1 s me .::. •:.·•:. :•.:•:.·:,i. :M .:• ".• 1•·..·:a!:.• .:·.:•.n,•.••:.•.•·.:.u.ta•:,.·•.•e.•. •.••.•t}•.e.e.,:.·•:.,:••..n!.:.:.·.•.s:•.:.•;E n1 ca ~.••;.:·.;•l. .·:u.l:·.·.•.•:.•.••..•s:·.•:.•:•.·pP.·.•.:·.•a.·..:·.••. t·.· .• •• •. .•.••.. •

:.,·~r~,·

s.• i•.

!::.:.c.·:,.e :.· .•.e :.·. , :. .n .·.: . • ·., ...:.·.:•.:· .• :..•• n :· •..,

·.. .·.::·.:·.•..

.••..•.•• .·:•. m

·t:•.·:e .·.• n·..• . :e :·,•.: . :

·Q ·. ,·. .·.s· .•• · .... • ...•. ..,·: p .•. .: •.:....·.::.· .•.•.e .• •.• .•.•..: • •.

~

'"'

.....

,i.·•.:.i.•.•.•.•,•,••.••.:.[\

t. !.p.·•.••. •.•

r.:. · •..·.a·..•.•..•.•. .

..

,... ..

..

,',

....... ...... .

...... .. ... .... •••.••.••••• • .•••••..•.••••••.•••••.•.

.

•..•....::.::.·/:t:{.>

::.: .:.:•.:.• :.•:. •!. .•

{:;i~.: ?~ ~.· : i···

i.i.: .•••.:.•.•:!:(.;.••:.• :. •: :i.•: :. :. • •. :.•.• •:.•.••.:.•.• :..• •.::••..:

.•.•.•.:: .•.:.: • .• •:.; .•.•.:,: ...••.•.•.• :••. .••:. ·:.:..••. : •• ..•• .:.·:·. ..•••.

........ .:.:-:... :.:.: ..... ::->:.:.:.:.:-:-:-:.::... ;.

. .·.·i :ij...I; '. ::i;:.•.·.'!.; .;. . i .••·..

.... .

1 1. . . . . . . . . . . . . ( · · · .·.·.· · · · · ·

}\)U)~}/t~~~»::::)<::'::::;':<··· ·

'V{ .· . . /·./

2.1

ALe

'.T E: L'

BELL TELEPHONE

ADVANCED TECHNOLOGIES IN SATELLITE TELECOMMUNICATIONS

... TECHNOLOGICAL DRIVERS IN TERMS OF : - MISSION COMPLEXITY;

-AFFORDABLE PRICE; - ON-BOARD INTELLIGENCE;

- HIGHER FREQUENCIES. ... SPACE APPLICATION SPECIFIC TECHNOLOGIES : - LAUNCHER;

- PLATFORM; - PAYLOAD.


T

MISSION COMPLEXITY; APPLICATIONS: - FIXED SERVICE SYSTEMS E.G. FOR BROADBAND APPLICATIONS - DIRECT BROADCAST SYSTEMS - MOBILE SERVICE SYSTEMS. E.G. GLOBALSTAR

LC.ATEL


~

MISSION COMPLEXITY; EVOLUTION : - TRANSPARANT PAYLOAD BASIC ARCHITECTURE; - COMPLEX TRANSPARANT PAYLOAD CONFIGURATIONS; - COMPLEX REGENERATIVE PAYLOADS - ACTIVE MULTI-BEAM ANTENNAS - BASEBAND SWITCHING - ON-BOARD DATA PROCESSING - SATELLITES IN CONSTELLATIONS (E.G. GlOBALSTAR); - INTER-SATELLITE COMMUNICATIONS.

ALCATEL BELL TELEPHONE


• AFFORDABLE PRICE : - TELECOMMUNICATION CONCEPT ALLOWING A PRICE SETTING COMPATIBLE WITH THE ADDED VALUE PERCEIVED BY THE USER; - COST REDUCTION THROUGH MINIATURIZATION; - SATELLITE LIFETIME EXTENSION BY MINIMIZING VOLUME, MASS AND POWER CONSUMPTION OF THE HARDWARE.



to ~

~ ~

-------------------------------------------------------- t

I

• ON-BOARD INTELLIGENCE :

t I ~

- REDUCE GROUND SEGMENT OPERATIONAL EFFORT;

l9

- AUTONOMOUS FAILURE DETECTION, IDENTIFICATION AND RECOVERY;

i

- FLEXIBLE RESPONSE TO SERVICE REQUIREMENTS.



... HIGHER FREQUENCIES: -

FREQUENCY SPECTRUM OCCUPATION;

- WIDER FREQUENCY BANDS ALLOW FOR 8-ISDN COMPATIBILITY;

-

NARROW ANTENNA BEAMWIDTHS POSSIBLE WITH ACCEPTABLE GEOMETRY.

ALe .. TEL BELL TELEPHONE


ttl

r R'

Q'

t I. y

LAUNCHER SPECIFIC TECHNOLOGIES: -

PROPELLANT WITH HIGHER PERFORMANCE ON THRUST;

i" I.~

1 9

- PROPELLANT NOT RELEASING CHLORINE-MONOXIDE.


'\F


PLATFORM SPECIFIC TECHNOLOGIES :

- ATTITUDE AND ORBIT CONTROL FOR STATION KEEPING AND POINTING ACCURACY WITH MINIMUM FUEL CONSUMPTION; -

SOLAR ARRAY WITH HIGH EFFICIENCY, RELIABLE INTERCONNECTIONS AND ACCURATE "END OF LIFE" PREDICTION;

- POWER SUBSYSTEMS WITH HIGH EFFICIENCY AND ATTRACTIVE POWER CONVERSIONNOLUME RATIO; - SPACE BATTERY TECHNOLOGY WITH IMPROVED MASS/PERFORMANCE.

I I

i


., TELECOMMUNICATION PAYLOAD SPECIFIC TECHNOLOGIES; KEY PARAMETERS - ANTENNA FOOTPRINT;

- RECEIVER: Grr;

I I

- TRANSMITTER EIRP; - MASS;

- DC POWER CONSUMPTION.

I

B[Ll TELEPHONE

T


TELECOMMUNICATION PAYLOAD SPECIFIC TECHNOLOGIES; FUNCTIONS: - ANTENNAS

: RECONFIGURABLE MULTIFEED DIRECT RADIATING ARRAYS;

- RECEIVER

: LOW NOISE AMPLIFICATION;

- FILTERS

: SELECTIVITY POWER HANDLING;

- AMPLIFIERS

: SOLID STATE TRAVELLING WAVE TUBE AMPLIFIER;

- MICROWAVE SWITCHING; - BASEBAND PROCESSING; - BASEBAND SWITCHING.

I I

I


----------------------------, -" J. .... TELECOMMUNICATION PAYLOAD SPECIFIC ~ ..

TECflNet.OGIE·S;;~

.

~.~

TRENDS: MINIATURIZATION TO ACHIEVE COST REDUCTION, ItIGH'9II'e' FREQUENCIES AND REQUIRED COMPLEXITY; LOWER LEVEL INTEGRATION; - NEW MATERIALS; - NEW TEST PROCEDURES.

.

~ !l

I



... SPECIFIC TECHNOLOGICAL ACHIEVEMENTS OF THE ALCATELBELL SPACE DIVISION : · HIGH PERFORMANCE FREQUENCY GENERATING HARDWARE FOR COMPLEX AND FLEXIBLE BROADBAND MISSIONS IN TERMS OF ELECTRICAL SPECIFICATIONS, MINIMUM MASS AND VOLUME, AND HIGH RELIABILITY.

I I

· FURTHER MINIATURIZATION TRENDS BY MEANS OF MONOLITHIC MICROWAVE INTEGRATED CIRCUITS AND ASSOCIATED THIN-FILM TECHNOLOGIES.

I

ADVANCED TECHNOLOGIES IN SATELLITE TELECOMMUNICATIONS ~

• TELECOMMUNICATION PAYLOAD SPECIFIC TECHNOLOGIES:

f

- COMPONENT TECHNOLOGIES: E.G. P-HEMT; - LEVEL 1 INTEGRATION: - MULTI-CHIP MODULES - MONOLITIC MICROWAVE INTEGRATED CIRCUITS; -

lEVEL 2 INTEGRATION: -

FINE PITCH PACKAGES; PIN GRID ARRAY; PLASTIC PACKAGES; CERAMIC FEEDTHROUGHS; MICROWAVE PRINTED CIRCUIT BOARDS.

I I

i

BELL TELEPHONE

ADVANCED TECHNOLOGIES IN SATELLITE TELECOMMUNICATIONS ~

• TECHNOLOGICAL DRIVERS IN TERMS OF : - MISSION COMPLEXITY; - AFFORDABLE PRICE;

f Ia

- HIGHER FREQUENCIES.

1

- LAUNCHER; .......

I

- ON-BOARD INTELLIGENCE;

• SPACE APPLICATION SPECIFIC TECHNOLOGIES:

!'-l

I

- PLATFORM;

Vl

- PAYLOAD.

,ALCATEL BELL TELEPHONE


~

------------------------------------------J • MISSION COMPLEXITY; APPLICATIONS: - FIXED SERVICE SYSTEMS E.G. FOR BROADBAND APPLICATIONS - DIRECT BROADCAST SYSTEMS - MOBILE SERVICE SYSTEMS. E.G. GLOBALSTAR

I 9

I

BEll TELEPHONE

~


MISSION COMPLEXITY; EVOLUTION : - TRAN~PARANT PAYLOAD BASIC ARCHITECTURE; - COMPLEX TRANSPARANT PAYLOAD CONFIGURATIONS; - COMPLEX REGENERATIVE PAYLOADS - ACTIVE MULTI-BEAM ANTENNAS - BASEBAND SWITCHING - ON-BOARD DATA PROCESSING - SATELLITES IN CONSTELLATIONS (E.G. GLOBALSTAR); - INTER-SATELLITE COMMUNICATIONS.

I 9

i

ALe £ TEL BELL TELEPHONE

ADVANCED TECHNOLOGIES IN SATELLITE TELECOMMUNICATIONS ;:a

--------------------------------------------- I T

f TELECOMMUNICATION CONCEPT ALLOWING A PRICE SETTING f

AFFORDABLE PRICE : -

COMPATIBLE WITH THE ADDED VALUE PERCEIVED BY THE USER;

- COST REDUCTION THROUGH MINIATURIZATION; - SATELLITE LIFETIME EXTENSION BY MINIMIZING VOLUME, MASS AND POWER CONSUMPTION OF THE HARDWARE.

.

I 9

i


~

J

----------------------------------------------------~

~

ON-BOARD INTELLIGENCE :

- REDUCE GROUND SEGMENT OPERATIONAL EFFORT;

I 9

- AUTONOMOUS FAILURE DETECTION, IDENTIFICATION AND RECOVERY; - FLEXIBLE RESPONSE TO SERVICE REQUIREMENTS.

I f.

tv

N o

BELL TELEPHONE


------------------------------------------------~--

;:a

J

,.. HIGHER FREQUENCIES: - FREQUENCY SPECTRUM OCCUPATION;

I I

- WIDER FREQUENCY BANDS ALLOW FOR B-ISDN COMPATIBILITY; - NARROW ANTENNA BEAMWIDTHS POSSIBLE WITH ACCEPTABLE GEOMETRY.

I

ALC.TEL BELL TELEPHONE

y


LAUNCHER SPECIFIC TECHNOLOGIES : - PROPELLANT WITH HIGHER PERFORMANCE ON THRUST; - PROPELLANT NOT RELEASING CHLORINE-MONOXIDE.

i

Q

i

ALC.TEL BELL TELEPHONE

~


PLATFORM SPECIFIC TECHNOLOGIES : - ATTITUDE AND ORBIT CONTROL FOR STATION KEEPING AND POINTING ACCURACY WITH MINIMUM FUEL CONSUMPTION;

i I

- SOLAR ARRAY WITH HIGH EFFICIENCY, RELIABLE INTERCONNECTIONS AND ACCURATE IIEND OF LIFEII PREDICTION; - POWER SUBSYSTEMS WITH HIGH EFFICIENCY AND ATTRACTIVE POWER CONVERSIONNOLUME RATIO; - SPACE BATTERY TECHNOLOGY WITH IMPROVED MASS/PERFORMANCE.

I


.. TELECOMMUNICATION PAYLOAD SPECIFIC TECHNOLOGIES; KEY PARAMETERS - ANTENNA FOOTPRINT; - RECEIVER:

Grr;

- TRANSMITTER EIRP; - MASS; N

~

- DC POWER CONSUMPTION.

~

J

I f i 9

t

. ADVANCED TECHNOLOGIES IN SATELLITE TELECOMMUNICATIONS ~

• TELECOMMUNICATION PAYLOAD SPECIFIC TECHNOLOGIES; FUNCTIONS: ANTENNAS

: RECONFIGURABLE MULTIFEED DIRECT RADIATING ARRAYS;

- RECEIVER

: LOW NOISE AMPLIFICATION;

- FILTERS

: SELECTIVITY POWER HANDLING;

.. AMPLIFIERS

: SOLID STATE TRAVELLING WAVE TUBE AMPLIFIER;

- MICROWAVE SWITCHING; - BASEBAND PROCESSING; - BASEBAND SWITCHING.

J


9

TELECOMMUNICATION PAYLOAD SPECIFIC TECHNOLOGIES; TRENDS: - MINIATURIZATION TO ACHIEVE COST REDUCTION, HIGHER FREQUENCIES AND REQUIRED COMPLEXITY; - LOWER LEVEL INTEGRATION; - NEW MATERIALS; - NEW TEST PROCEDURES.

'.

I 9

i

BELL TELEPHONE

ADVANCED TECHNOLOGIES IN SATELLITE TELECOMMUNICATIONS ~ ~

y

TELECOMMUNICATION PAYLOAD SPECIFIC TECHNOLOGIES :

I

- COMPONENT TECHNOLOGIES .: E.G. P-HEMT; - lEVEL 1 INTEGRATION : - MULTI-CHIP MODULES - MONOllTIC MICROWAVE INTEGRATED CIRCUITS; - lEVEL 2 INTEGRATION : -

FINE PITCH PACKAGES; PIN GRID ARRAY; PLASTIC PACKAGES; CERAMIC FEEDTHROUGHS; MICROWAVE PRINTED CIRCUIT BOARDS.

i 9

I


y

SPECIFIC TECHNOLOGICAL ACHIEVEMENTS OF THE ALCATELBELL SPACE DIVISION : · HIGH PERFORMANCE FREQUENCY GENERATING HARDWARE FOR COMPLEX AND FLEXIBLE BROADBAND MISSIONS IN TERMS OF ELECTRICAL SPECIFICATIONS, MINIMUM MASS AND VOLUME, AND HIGH RELIABILITY.

I 9

· FURTHER MINIATURIZATION TRENDS BY MEANS OF MONOLITHIC MICROWAVE INTEGRATED CIRCUITS AND ASSOCIATED THIN-FILM TECHNOLOGIES.

i

De Rol van "Tracking" Systemen in de Ruimtevaart :<.•.i :.'<.,<. <.>__......'.:.: .":.:.>.:. :. : : :<..•:'. ...:':'.'...·x •.<. :. •.•. • • •:. •<• .:>< . • .• <......... "> ><.. . < > ' . ).• :...<>:{< .• •.• • • • •. .• . •.• • . .;: •••..•••.. . . . .•:. :• :.:.•. . ::<...: . .••.:. • <.:.......::·.:::::;:.<:)c. .::. . .":.: •. '(.. :••.•. •?<{••..•:/ ..•' ».::.•.'. ~....1 . .: .•...•.:.< <'.<.<: : .•••• :)i»: :.>\.::> :.:.:< .. .. . . .. . >.. .:::::.>::?":;:'.'<'}.:.':::; >•••. { .. (

1.

1«<:

1<:«.

,. . :

c••

c~ '..•:... . ...:: .::<. . ~., . . ' .:::.. ..

...

..

c:., ..

<.

.. •.• :>/ .~;<

.,..

•... :•.•.• :.;:::~<\

.. ,.

• ...

>,'.;:: « : ..<:'>:

. ""'. .

...

. . . . .,:::::: :.:::: :,:?·~t·::>·

.::.:.\:;. . . ',•.::. . •.• : . . :•.:.

•.. : c•.<••.::.::.·•••••.·•.• ::::: .•:.,: ;...~;<:.: . :)< ....•. :: : : ) ..;: .. <•.••::.:.:. .. > . . .• ; •••••• :<. ... :<:,. ••..• :-.: ••. :. . .: •. :; <:·····:···::}i :::.:,'<::::.' >.: .: . • . . i«::·:. •.•>;:.• : .... ::::::c::.: c : •••. ::<. <.;....>. ... . .• :. : . ..••:.::::;<: .: :..... : {.>:: ....:. . <:<: . :<. :.:.>>:/...••.•..:::.. .. :••••• <:.. ..«<::::.; !:"'.':<. .. <.:c:::::. : <.::>......:.:. :. .......: . .•••• :}><.... >.. ::.... <:.>. ... . ...••: . . . ..> ..........:::.,:.:.•.. "::'. :.::::... <.>.:.::.:. •. .•••.. .::: «?>::\{\.:. ::... .. . ..•.:. :. . . :<./.
1:::.>:<2.. . ,: .... :-::'. ? . .. :.r:.. .. :...: I<>~>: .:- :( <:<> ..\ c.:"'>}> .:.

. :...... :. •. . . ><. ·c ···Y··7S ·....·. , .. . :..}/ :.)<••• .:.:> :.::: .:. .

:• ; •·•· • ·• •·.:• :• •.•. (::.:. . . {.j . . . . ':'. :.:.: :. . •. .: . . . . ;:.: . . .: > . . ;:..> I::<....: .. >« .;':...,!..;: .'.: : .' •.. ·iLL::-:{«:> . .:. ::': «(·1 .> /2) .>, .::'<.::'.. :: .::: :.< :: . »< .. .} .• ·ir>:. :: .... .. .. . .:.:.. :: .. ... ... ..:. •. .: :.... . .. : :. . .. :. ::. ::...:. ..... :... . .. ::.. : . .... :• :~

:~:::.

.:.:: : :: .. :..

.:::.

. ... :.. . . :..

...

....

•:::"':'::"}.'::::7c.: .7

···~d(···

..):-:.:.

i,> >.I~'II~Ji k• .. ::.

:

/Ii~,

)(

:..

. ..<).

~r' u;

..

...

·Y:'·

: : :.

~~ ... le. :.: . •.:.:. . ..r.

;::7~L7 I:.I~· ,

.... :.:

I.:
.

>.

>:: .>.

'..~:

:<:

: :.'>': .:.» .

..... .<..

. . .

: . . . ~ •. ·to .• :·

\:::' :<. ><

7:~:J- ~. .'.

\I\::.····i

>.

•••.

> . ••.. :. :. . . . . ': ::. :::;::.:
.>::

.i.j ..

•. :. . ... ... ·i·>:r·:. ::«

.. /.

:.:.:.<..:.....

>:

?l< .. :. . : .: : .: :~

:( :::'

.. :,,~~::

·8:· } . . : : ; \ . ; . j < ••

:::.::,

::·~it?·

.

:.:.•...

.........

...

:.. •.••.

.. . < : ; · <

:.:.:.:~:

. .:.: \.,:

••

my;

""tt"«t;:,

, . . :

...

•. :••..•..•• -x:· .}' ,<-' >.' ...•.. :: ••...••... :. . ~":--•••.•••.•••.

»»T·::·t;: :...;> :.0:::

.. '. ..:>,> . . . '. •. ::. ..•.

:t·;·( .. . '

. •.

•::>":{ :::.' ~i

:«<

J2i:0'::~: ". <:>N·: ..:.~. 'r ... :

.•.•. •:.:.

·«

:-:::::..

:.> .:: • .:< '::':' .. '. . :c<:.

•. :': •. :. :.. ; - -

.·• }.... \> •. : : : . •. :.•:. :.:• • .:• • • <.• •:• • • .•:..:.:): . . > :.:. . . . . '. . ,:, . . :. . . ... t .. .,. .:.}\. . ,: :. ,::. . ::. : .: . . .:'. ::. .:> < •.••.'..: . <.,:............................... •••••• ···. :.:.•.«. '..}. ..: : .::::' . :..);.>:.:. . : . »:.' :: .......... :. .: .':CU . •....•.•..': <.'. :.:' ..:.•.:: ':':":":::<.:: .... ::;::-:::,::."":",,,: :<:: • <

. .. . : '

~:\. . . ; : : : : : 0 .>: :~~<.~::.::>;::: .• Y.l.J.: ~~c : :.:... <. . / ~>... 1 . . . ,T ....

:>< r..... . .·7111 '::..

.. . :<

.•':::

{j~:~

:. ;> ..• :::: . :.:.:

',.:.·.n I<}<·

.

.~::::'::. :)<~.,j : .... :::::<:<.. :

:;.- .. :.: ".:<.

..:,:.:

V

:«

·{7;;-:.'<0:,

..
.: :. .:::: .: : . ..:c.

. .::.::: :. . . :::

<

>

.:.: .

.::.:... .:: . .:. . . :: .». . '::1 .:--?J .:::<x<

'. ;;..F,;;:<

...

..

I:.;;,;:'" >:: .::.: ~'. .~::. :::,

I

>

.

'~T;.;·' . . . . . , / , · ....

.... ··T..; \ .. :

,.....:.

.,..

.. : I I I I . ::::: •

. {] ..•• : ."::·c)··

:: .. '........: . : '.: . . . : : : < .

: : . ::, .::.. :.. .

(Fi···

:> .'>. :.: : .: :... :.: <:. .,...... ::::> ::<::i·i'.j•. :•.: :;~<:>i::·<:·· ,.:::.: .::; . :.:: . . . ... .. :.:: :.'.:.' C.• .· · · • :.....:::.:::::::, <>.: . . •:.. . .··::.«L.. ....... :.:.:.••• <..•:...•:;. :. .•,.•..•;:.: ....::•. ::.:.• :••::::.• ..... :.. :.<:::.:<.<.:.<:: ,:,,', ... :::'•• ::.>•• ;: :.<.. ).•

<.

.• > •• : •.:..

::::':.:

••.

. ..

.

..

.:.:••::•. :.:,.:::,.:.:.....

. ....... ,

3.1

lr. B.A. C. Ambrosius: De Rol van "Tracking" Systemen in de Ruimtevaart

De Rol van "Tracking" Systemen in de Ruimtevaart Ir. B.A. C. Ambrosius Technische Universiteit Delft

Ruimtevaart en elektronica zijn onlosmakelijk met elkaar verbonden. Het succes van de lancering van de eerste satelliet, Spoetnik 1, was over de hele wereld met een korte-golf radio waarneembaar dankzij de trotse bliep-bliep signalen die door een miniatuur radiozender werden uitgezonden. Na zijn lancering is een satelliet normaal gesproken niet meer fysiek toegankelijk. De eenvoudigste manier am dan toch contact met de satelliet te onderhouden is door middel van radiocommunicatie. Bovendien, wat de opdracht van de satelliet ook is, men zal altijd willen weten waar hij is. Dit laatste is het hoofdthema van deze voordracht. Daarnaast zal worden toegelicht hoe de kennis van de beweging van een satelliet, omgekeerd, gebruikt kan worden om meer te weten te komen over de vorm en het opperviak van de aarde zeIf, en daarmee over de geofysische processen die daarbij een rol spelen. Het oudste en meest bekende "tracking" systeem is het menselijke oog. Dankzij eeuwenlange waarnemingen met dit instrument waren astronomen al vroeg in staat om de baan van onze natuurlijke satelliet, de maan, te berekenen, evenals die van de aarde en de andere planeten om de zon. Ook in de begintijd van de ruimtevaart speelde dit instrument een belangrijke rol, al dan niet aangevuld met een hulpmiddel, de verrekijker. Door waamemingen van het door de satelliet gereflecteerde zonlicht, kon de positie van de satelliet tussen de "stilstaande" sterren worden opgemeten, en daaruit berekende men de baan. Het eerste electronische "tracking" systeem, radar" , werd reeds aan het einde van de dertiger jaren ontwikkeld. Dit systeem, met als voordeel dat de waargenomen objecten "passief" zijn, wordt voornamelijk gebruikt tijdens de eerste fase van de vlucht van een satelliet, namelijk de lancering. Wegens de beperkte nauwkeurigheid en de betrekkelijk grate kosten van de gebruikte installaties, wordt radar echter nauwelijks gebruikt voor het ope ratione Ie volgen van satellieten. WeI wordt radar tegenwoordig toegepast voor aardobservatie vanuit de ruimte. II

De meeste "tracking" systemen zijn echter gebaseerd op radio signalen die door de satellieten zelf worden uitgezonden. Omdat over de ontwikkelingen in de voormalige Sovjet-Unie weinig bekend is, zal het overzicht beperkt bIijven tot Amerikaanse en Europese systemen. Daarbij zullen achtereenvolgens het "Minitrack" systeem, S-band transponder tracking, het "Transit" systeem en DORIS, PRARE en GPS aan de orde komen. Een bijzondere plaats wordt ingenomen door laser afstandsmeetsystemen die eveneens besproken zullen worden. Tenslotte zal de satellietradarhoogtemeter behandeld worden.

3.2

Ir. B.A.C. Ambrosius: De Rot van

~Trackinglt

Systemen in de Ruimtevaart

Naast de besprelcing van de meetsystemen zeIf, zal ook worden ingegaan op de principes van baanbepaling en plaatsbepaling met behulp van deze systemen. Verder zullen enige toepassingen zoals navigatie en geodetisch, geofysisch en oceanografisch onderzoek de revue passeren. Tenslotte zullen van een aantal van deze toepassingen enige resultaten worden besproken.

3.3

Satellite Communication Systems Planning in an Interference Environment

4.1

Satellite Communication Systems Planning in an Interference Environment R. Hekmat PIT Research Neher Laboratories

Introduction The possibility of offering analog and digital communications by using satellites is increasing as the satellite technology continues to develop. Telecommunications and broadcasting communities are becoming more aware of the extent to which satellites can help them to provide new reliable services. Considering these rapid developments, it will be increasingly advantageous to understand, at least in general terms, the limitations and benefits of services offered by satellites. In this paper we consider the planning of Eutelsat satellite communication systems for digital data and analog TV transmission. By this, the limiting factors in satellite communications will be illustrated. Further we will see how various parameters in a

satellite communication network (on the earth segment and in the space segment) can be adjusted to maximize the grade of offered services. In the first paragraph of this paper, satellite communication systems and services are explained in general terms. In paragraph 2 we clarify the meaning of the planning of a satellite system. Paragraphs 3 and 4 contain information about how the quality of an offered satellite setvice is determined. PIT Research has developed an easy-to-use PC based software tool for planning of satellite systems. The topics discussed in this paper are all used in this tool. Some distinctive features of this tool are summarized in paragraph 5.

1. Satellite Systems and Services In order to keep a satellite in orbit it is necessary that the gravitational force that acts on the satellite be equal to the centripetal acceleration of the satellite in motion around the earth. There are different types of satellite orbits which meet this requirement. The one most useful for communication purposes till now has been the Geostationary Earth Orbit (GEO). GEO satellites move in the equatorial plane in the same direction as the earth's rotation with an orbital period of 24 hours. For an observer on the earth's surface, the satellite appears to be stationary. Because a GEO satellite can always be visible from two widely spread points on the ground, it ensures continuous communication between these two points by using fIXed antennas.

4.2

R. HeIcmal: Satellite CommunictUion Systems Pltmning in an /1IIer{erence Environment

In the context of this paper, we restrict ourselves to GEO satellite orbits. From the geostationary altitude (35.786 kIn) almost a third of the earth's surface is visible to a single satellite. Due to the curvature of the earth, points above 81 degrees north and 81 degrees south of the equator are not visible to GEO satellites. This is however not considered as a severe disadvantage, as the communication demand in these areas is usually very moderate. Instead of covering one third of the earth surface by a single global antenna beam, most of the GEO satellites use shaped beams. This makes it possible to focus the satellite power on (separated) restricted areas with high communication capacity demand. Consequently, smaller antennas could be used to receive satellite signals and to transmit signals to the satellite. For example, the coverage area of Eutelsat satellites is restricted to Europe and North African countries (see figure 1). Eutelsat satellites, at 7, 10, 13, 16 and 36 degrees east, offer the following services: Broadcasting of analog TV signals to fixed earth stations with parabolic antennas with a minimum diameter of 1.6 m. Satellite News Gathering (SNG), (signal quality less than in the case of analog TV broadcasting), by using easy to move earth stations with antenna diameters of at least 1 m. Data transmission at 128 kbps or 512 kbps for VSATs (Very Small Aperture Terminals) using parabolic antennas with 60 em diameter. The frequency on the up-link channel (from earth to the satellite) is located in the 14 GHz band. On the down-link channel (from satellite to earth), Eutelsat uses the 11 and 12 GHz band. Each Eutelsat satellite has more than one transponder. The bandwidth of each transponder is almost 80 MHz. Figure 2 shows a typical transponder occupancy for a Eutelsat transponder used for transmission of TV, SNG and digital data signals.

2. Planning of Satellite Systems The main issue in planning a satellite communication system is to answer one of the two following questions: what is the quality and availability of a satellite communication link with a certain given set of equipment and network parameters (e.g. transmit power, receiving antenna gain patterns, satellite amplifier characteristics)? what are the required equipment and network parameters to assure a certain desired quality of the satellite communication link?

4.3

The quality of a communication link is expressed for digital communications as the bit error ratio and for analog 1V/SNG signals as the signal-to-noise ratio after the demodulator. Availability is the time percentage during which the quality is better than a certain threshold value. The quality depends directly on the carrier-to-noise plus interference ratio at the input of the receiver equipment. Deriving the communication quality from the overall carrier-to-noise plus interference ratio is called link performance calculation. The carrier-to-noise plus interference ratio depends on the network parameters and the equipment parameters. Parameters like the number of signals in the satellite transponder, number and nature of interfering signals, are considered as network parameters. Equipment parameters include transmit power, antenna gain patterns, amplifier characteristics, etc. Obtaining the value of the carrier-to-noise plus interference ratio from the network and equipment parameters is called link budget calculation. Unk budget calculation and link performance calculation are both necessary for the planning of a satellite communication system. In paragraph 3 we will discuss the link budget calculation. Unk performance calculation is the subject of paragrapb 4 in this paper.

3. Link Budget Calculation For the purpose of link budget calculation a satellite link is considered to consist of four distinctive nodes: transmitter output of the (transmitting) earth station, receiver input of the satellite transponder, transmitter output of the satellite transponder, and receiver input of the (receiving) earth station,

Unk budget calculation is the precise determination of the carrier power, interference power and the noise power at each of these four mentioned nodes. This involves the quantification of the different ways in which the wanted signal is attenuated or interfered with. Before commencing with the subject link performance, we will give a short description of ways that signals are affected in satellite communications and how the contribution of these disturbances to the link budget are calculated. Considering the scope of this paper our descriptions will be limited to the most elemental features of each subject.

4.4

3.1 Noise Distortion

Noise consistS of all the unwanted contributions of energy at the receiver input which tends to corrupt the wanted signal. In the absence of interfering signals, noise finds its origin in radiation from radiating bodies located within the field of view of the receiving antenna and the noise generated within the electronics of the receiver. For frequency bands used in Eutelsat satellite networks, the amount of unwanted noise energy can always be associated with thermal noise generation. Therefore, the amount of noise in a system is expressed in system noise temperature, T, in Kelvins. This is the temperature of a passive system (a resistor for instance) which would generate the same amount of noise as the considered source of noise. If the system noise temperature of a receiver antenna is T Kelvin, and the gain of the receiver antenna is denoted by G, the ratio G;r is called the Figure of Merit. This figure characterizes the effectiveness of the receiving end. In the design of antennas in satellite communications, great effort is made to obtain the highest possible value for the G/I' for a given size of the antenna. 3.2 Signal Attenuation Free Space Loss The power transmitted from the earth station to the satellite or from the satellite towards the earth is subjected to attenuation because of the spreading of the electromagnetic waves in space. Free space loss is defined as the signal power attenuation between two isotropic antennas in space. The spreading of power makes its loss directly proportional to the square of the distance between the transmitting and the receiving end. It also depends on the frequency of the electromagnetic waves. Due to the large distance between satellite and earth stations, free space loss is the most significant contribution to attenuation in satellite communications. A typical value for free space loss is e.g. about 205 dB at 12 GHz. Attenuation by Atmos.pheric Gases Along with the spreading loss signals travelling through the troposphere suffer from attenuation due to oxygen and water vapour. The expected values for tbe density of water vapour in the atmosphere for different months of the year are given in special tables published by the CCIR (Report 719). Attenuation of electromagnetic waves by atmospheric gases is strongly dependent on the frequency of electromagnetic waves. As it can be seen from figure 3 it also

4.5

depends on the elevation angle, E, of the earth station antenna. Elevation angle at the position of the earth station antenna is the angle at which the satellite is seen above the horizon. At lower elevation angles the path of the electromagnetic waves through the troposphere is longer. Therefore, as expected, atmospheric attenuation at low elevation angles is higher. For elevation angles higher than 10 degrees the amount of power loss is negligible compared to the free space loss. For example, at 10 degrees elevation angle, the amount of attenuation caused by atmospheric gases is about 0.2 dB at 6 GHz and 1 dB at 30 GHz for low water vapour density. Rain Attenuation Attenuation due to rain depends on the rainfall rate, the frequency and the polarisation of electromagnetic waves. Depending on the statistical data on rainfall rates, CCIR has divided the earth into 14 different climate zones (see figure 4). Planning of satellite systems requires to know the rain attenuation value exceeded for a given time percentage of the "worth month" of an average year. Once this value is determined, and the satellite system appears to offer enough quality at this attenuation value, the system planner can be sure that the system quality can only be better for the rest of the time. After determination of the climate zone to which the earth station belongs, the attenuation value exceeded 0.01% of an average year is read from the by CCIR published tables (Report 563). This value can then be used to derive attenuation for other annual percentages. 3.3 Signal Amplification

Amplifiers used in satellite communications, like any other amplifier, do not have a linear characteristic. When several signals are offered simultaneously at the input of the amplifier (power sharing), the effects of non-linearities manifest themselves at the output of the amplifier in the following ways: intermodulation products are created which can interfere with the wanted signal. The bandwidth of the output signal is also larger than the total bandwidth of the input signals (signal spreading). signals of different magnitudes are not amplified equally. A relatively weak signal can be suppressed because of the presence of other signals (signal suppression). variations in the input signal level lead to unwanted amplitude and phase modulation of the output signal (AM/AM and AM/PM conversion).

4.6

The first two effects are easily recognized in figure 5. This figure shows the output of a typical Travelling Wave Tube (TWI) amplifier· along with the corresponding input signals. The main problem for calculation of intermodulation degradation is to determine the power of the wanted signal and the sum of the powers of all intermodulation products that share the same frequency band with the wanted signal. Computation of these powers is a sophisticated mathematical problem and we will not discuss it here. But when this computation is carried out, the signal to intermodulation ratio is simply calculated by using the formula:

In this formula, Pw (in dBW) is the power of the wanted signal; and P;". (dBW) is the power of an intermodulation product inside the bandwidth of the wanted signal. The symbol (ell);". (dB) stands for the ratio of the wanted signal carrier power and the intermodulation power. 3.4 Interference

The quality and the maintenance of services delivered by satellites is limited by the amount of interference on the communication path. A major objective in designing and planning a communication link between two fixed or mobile points via a satellite is to estimate signal degradations due to interference. Interference in satellite communications can generally be separated into two categories: intra-system interferences: These are interferences caused by signals present in the satellite itself. This category includes: cross-polar co-channel interference intermodulation in the satellite (and the transmitting earth station) amplifier interference due to signals in the neighbouring transponder inter-system interferences: these kind of interferences are caused by signals which are transmitted to/from other satellites or ground based stations. This category of interferences include: interference caused by signals sent to other satellites interference from space services interference caused from signals transmitted by other satellites interference caused by terrestrial systems (e.g. Radio Relay Systems). Of the intra-system interference sources, the intermodulation has already been discussed in the previous paragraph. Cross-polar interference is caused by signals

4.7

transmitted to the satellite at the same frequency as the wanted signal but with a polarisation opposite to the polarization of the wanted signal. The last category of intra system interferences is caused by intermodulation products generated in a neighbouring transponder. The "tail" of these intermodulation products can infiltrate into the transponder carrying the wanted signal and disturb the spectrum of the wanted signal. The inter-system and intra-system interference sources are shown in figure 6. In the model for calculation of interference in Eutelsat satellite networks, a more refined interference estimation is obtained by considering the following four different types of interference in relation to the offered services: Interference caused by digital signals into TVISNG signals Interference caused by digital signals into other digital signals Interference caused by TV/SNG signals into digital signals Interference caused by TVISNG signals into other TVISNG signal

1.

2. 3. 4.

The general way to calculate these interferences is according to the following formulas: (ell). =IPFDw -IPFDI +XPI.,,+11 (GI7).,,+FSA (CII).."

=ElRPVI -EIRP,+XPI. +11 (G/7). +FSA

where CII

=

IPFD

VI

=

I1GIT

FSA=

4.8

=

=

ratio of the wanted carrier power and the interfering power in dB. The subscribes up and down refer to e/I on the up-link and down-link, respectively. Input Power Flux Density (in dBW1m2) at the satellite antenna. The subscribes w and i refer to IPFD of the wanted and the interfering signal, respectively. cross-polar isolation between the wanted and the cross-polar interfering signal. The subscribes sat and es refer to XPI of the satellite antenna and the down-link earth station antenna, respectively. difference between the value of the antenna Figure of Merit in the direction of the wanted signal and in the direction of the interfering signal. The subscribes sat and es refer to this difference at the satellite receive antenna and at the down-link earth station antenna, respectively. Frequency Separation Advantage. Addition of this term to the formulas is necessary for computation of TV interference into digital carriers. Viz., the interference of TV signals into digital signals reduces rapidly as the distance between the carrier frequency of the wanted and the interfering signal increases.

ElRP

=

Effective Isotropically Radiated Power (in dBW) from the satellite antenna. The subscribes w and i refer to EIRP of the wanted and the interfering signal, respectively.

For each of the cross-polar, adjacent satellite or adjacent transponder interference, a suitable and realistic value for the power of the interfering signal (or signals) must be inserted in the general formulas given above. This value can be calculated precisely if the interference environment is completely known. Most of the time this is not the case and an approximation must be made.

4. Link Performance Calculation The final goal in planning a satellite system or a satellite service is to find out what the signal quality will be at the receiver end. If we use the symbol Q for the signal quality, it can be expressed as:

where Q = wanted signal quality, e.g., signal to noise ratio for analog signals or error probability for digital signals S = the set of parameters specifying the modulation characteristics of signals (e.g., signal type, modulation index, and bandwidth). The subscribes w and i refer to wanted and interfering signals, respectively. The number of interfering signals can be more than one. D = the set of network design parameters. These parameters include for example the EIRP, frequency, antenna size and details of multiple access and amplification for the wanted and the interfering signals. The subscribes w and i refer to wanted and interfering signals, respectively. Again the number of interfering signals can be more than one. As we mentioned in the previous paragraphs, there are two distinct and separable aspects to the evaluation of the received signal quality. The first one is the evaluation of the wanted signal level and the interference levels at the receiver's input by link budget calculation. In satellite communication this requires a good estimation according to some model of signal attenuations, noise degradations, and interference sources. This first step leads to the calculation of the "carrier-to-noise plus interference ratio", denoted by C/(N+I).

The second step is to determine the actual signal quality from C/(N+I) by means of link performance calculation. For digital transmission the quality of the link is expressed in terms of bit error ratios. The value of the bit error ratio for a given value of carrier-to-noise plus interference ratio depends on the modulation/encoding techniques used and the demodulator characteristics. The demodulator characteristics

4.9

are usually given by tables or curves which show the bit error ratio as a function of the carrier-to-noise (Plus interference) ratio. The quality of analog signals is expressed in terms of signal-to-noise power ratio, S/N, at the demodulator output. Analog 1V signals are passed through a preprocessor before being FM modulated and sent over the satellite link. The reason for preprocessing is to spread the spectral power of the transmitted 1V signals as equally as possible over the available radiofrequency bandwidth. For FM/TV signals, tbe general formula to calculate the signal-to-noise ratio can be expressed as:

SIN=/ ('-£,B,b,4/",K) N+J

As can be seen from this formula, the quality of the FM/TV signals depends not only

on the value for C/(N+I), but also on tbe following factors: B = b =

4/.PI' = K=

radiofrequency bandwidth of FM/TV signal. video bandwidth of 1V signal (higbest video frequency). peak-to-peak frequency deviation sensitivity of video signal. improvement factor related to preprocessing.

s. Software tool In 1m PIT Research has developed a PC based software tool for planning of satellite systems. This tool is capable of calculating the quality of a satellite link (analog or digital) by considering the interference and noise effects on the total radio path plus the modulator and demodulator characteristics on the transmitting and the receiving end. It also can be used to determine the required power at the earth station or satellite for assuring a certain grade of service under different weather conditions. Some distinctive features of this tool include: the model used for estimation of interference levels is based on the by Eutelsat released values and scenarios about interference in their satellite network. However, this model can be expanded to cover any other satellite network like Intelsat. equipment parameters related to different earth stations and satellites are kept in accessible databases. A user who wants to evaluate the performance of a satellite link needs only to select the name of the desired receiving earth station, transmitting earth station and satellite from the database list. New base stations and satellites can be added in a simple way to the database. tables and graphs showing the bit error ratio for different modulation and coding techniques are accurately approximated by polynomials. Value of the bit error ratio corresponding to a given value of C/(N+I) is then easily determined. 4.10

a digital map of the CCIR climate zones is stored in the computer database. In this way the climate zone corresponding to the position of an earth station is automatically determined. Thus, for calculation of the rain attenuation and atmospheric attenuation no extra data are required except for the latitude and the longitude of the earth station. effects of power sharing (intermodulation levels) are calculated by taking the TWr characteristics into account.

4.11

~--...,..::;.."'---, ,_

.... ......- - - 1

I'

.,.;"

I'

II

•

Figure 1: Transmit coverage tUea of Eutelsat satellite at 7 degrees and 12 GHz (from Eutelsat 11 Satellite Handbook

Transponder Center Frequency

, I

.

. ~10HHz.

.,'

~18MHz I !

.. .

~15HHz

.,

:

FM/TV Carriers SNG Carriers 1111111111111 B Digital Carriers

Pol. X Pol. Y

80 MHz

Figure 2: Illustration of the frequency arrangements for FM/IV carriers and digital carriers in Eutelsat II 8() MHz transponders (from Eutelsat 11 Satellite Handbook)

4.12

R.Hekmal: Satellite Communication Systems Planning in an Interference Environment

ATT'ENUA nON AA6 (dB) Pressure at ground level ;II 1 Ibn Temperature at ground level • 20-C water vapour It ground level 7.S 91

=

Z

.J

I'

nr

10

FREQUENCY (6HZ)

Figr.ue J: Attenuation by atmospheric gases as a function of frequency and elevation angle,E

..,.

..

"'~

~

,.....

1.,

kl '~I

.sr

-

'r

...

.I

I

I•

.' I

Figr.ue 4: CCIR map of climate zones based on rainfall statistics (from CCIR Report 563)

4.13

output

input 0.7,..-----r-!--."...---."...--., 0.6 ~ ................. :................... ~ .................. ~ ......... -

0.6 ~ ................ ·f·· ...... ·

...... ~ ..................:......... -

0.5 -................. ;.................~ .................. ;......... -

0.5 ..................

0.4 ................... ;................ ~ .................. ;......... -

0.41-.................\................ .:. ..................i......... _

.

:

:

.................. ,......... -

0.31-............ ·.... i ........ .

. . .:. . . . . . . . . +. . . . -

0.2 _....·......·..·.. i· .. ·....· .... _.................. :......... -

0.21-···....····...... j..•· .. •··

····+· ..·..·.......... ·f..·..·.. ·-

0.1 ~ ................. ;....... .. . ,................... :......... -

0.1 ~ ........................ .

..: ................. :......... -

0.3 _................. ;....... ..

.....

. ~

.

o

20

40

.

40

60

frequency (MHz)

frequency (MHz)

FiRf.W 5: Input and output of a satellite amplifier with non-linear characteristics

Interference model:

contributions

SAT intennod. + interference

,,

Croapola,t

L

Off-axis

t

/es

, intennod

, f t Figure 6: Interference sources

.

H~

O~----~----~~--------~~

4.14

.

+................~ ..................;. .:.....-

Cross polar J.

60

Space Power Electronics Design Drivers . ::.;: . .

..... .

;.: .

.. .. ....................... .....): :;.' /:.:': '::::. : . . .. -..... :

'

..............•.......... ... ... ......

::: :.'.

: ~.

.:,

:.:.•... ....

'.:'..::.::...: ':. '. .•. :. ..•.':' ':: ... .•.:..

~: :.' .~::..' ::~::~: :.'

'.

:. :..'

..

.•... .•. .... . ... :... ...

.

5.1

D. O'SuIliwm: SptIU Puwer EI«tronks - Design Drivm

Space Power Electronics - Design Drivers D. O'Sullivan Power and Energy Conversion Division ESTEC, Noordwijk, the Netherlands

INTRODUCTION In order to understand the key differences between space and industrial power electronics, it is essential that those features that are unique to the space environment are clearly identified.

In the fust part of this paper those key drivers that effect the choice of design approach will be desaibed and the remainder of the paper will illustrate how these drivers effect the choice of regulator topologies and the overall approach to power system design.

1.

Conductance Modules

BASIC DESIGN CONSTRAINTS

1.1 Failure Tolerance Certain basic requirements for spacecraft are inherent to the fact that once launched, no repair is possible, hence, there is a fundamental rule which states: No slagle component 'ailure shall result in a slgnlftcant loss or spacecraft operation

This innocuous statement has a very important consequence on the redundancy and reliability aspects of the power system design. Significant loss, is interpreted differently from one mission to another, however, in the majority of cases, the implied requirement is that less than 10% of the power capability can be lost due to a single component failure. This implies for the major power system regulators, that at least 10 modules per function are required, which when they fail, must gracefully disappear from the system (See F'tgure 1). In addition, for manned missions this requirement is enlarged such that multiple failures can be tolerated without impacting the safety of astronauts.

5.2

1.2 ReUablllty The majority of application satellites (e.g. telecommunication or T.V. broadcast) have mission lifetimes in the range of 10 to 15 years, with the requirement of minimum ground intervention for failure recovery to ensure maximum payload operation.

Load Power Sources

PiJIw 1. ModulIIr Regu/IIlor Scheme This requirement for long mission lifetimes implies that components selected for space application must be of proven reliability, be manufactured on a high reliability production line and have a relatively proven experience in the industrial or military field. All these requirements lead to the selection of components which are relatively "old" in technology terms and hence, where an industrial engineer will select the right part for the task to optimise the end product, a space power electronics designer is constrained by the use of a limited number of qualified components. A further consequence of the requirement to achieve high reliability, is the need to use parts at ratings far below those used in the commercial electronics field. Typical requirements are to use semiconductors at:

D. O'SuIlivan: Space Power Electronics • Design Driven

< 65-' or rated voltage

< 75-' or rated correat < (lOtI, or rated JIC'ftI'

Judioa temperatures < 115 0 C

In conventional industrial applications the principal feature used to cool PCB mounted components is achieved through the use of forced air cooling, to allow higher power dissipation on these components.

Furthermore, the failure tolerance requirement desaibed in section 1.1, may result in additional under utilisation of the components, when the resultant modularity imposed by the "significant loss" requirement, dictates lower power handling per module and hence per component.

A typical component operating in a terrestrial environment at 4
Due to the importance associated with the reliability aspects of the space power electronics, great attention to detailed design analysis is required by the customer and all designs must be backed up by detailed reports dealing with:

The main reason for this dramatic temperature rise is the absence of the parallel thermal resistance associated with passive air convection. This thermal resistance can be between a fifth and a tenth of the thermal resistance produced by conduction and radiation. This effect is illustrated in figure 3.

Circuit design report Worst ease analysis Component stress analysis Control loop stability analysis Failure modes and elrects analysis ReliabiUty analysis Thermal analysis Engineering model test reports Flight hardware test reports

Convection 1ft :" / Radiation

.?/

T

~ nt

Anyone of these reports which indicate an "out-of-specification" can lead to a redesign of the effected equipment. p

From the above, it can be clearly seen that any comparison with industrial packaging performance cannot be made, since in the industrial field the drivers are cost and volume, whereas, the space drivers are reliability, modularity and failure tolerance. 1.3 Spaee Environment· THERMAL

Another major difference in power electronics for space applications is in the thermal design area. Many terrestrial power electronics designers do not fully appreciate the benefit of convection or forced air cooling. In the space environment, equipment is designed to operate in vacuum, hence, the only possibility for heat elimination is by deliberate conduction or by radiation.

tj

=P x tt1j-c + {eC-C ,lJc-s,lJc-r}]

l"i&In 2. 1'hermtIl Envimnment - TmesttiIJl A main conclusion of this thermal constraint and the additional limitation of junction temperatures lower than llSOC, is that many commercial, high speed components cannot be used due to their relatively high dissipation.

A typical terrestrial situation is illustrated in YJgUI'e 2 showing the effect of the 3 heat elimination features, radiation, conduction and convection.

5.3

D. O'Slllltwlll: Space Power EI«tronics • Design lJrivers

Radiation

The consequence of this radiation environment is that OLD, proven, technology must be used to circumvent the radiation susceptibility of the designs envisaged for space applications. 1.5

Space Environment

PadaIaI.. Constraints

AU the previous space related requirements impose specific constraints on the packaging of power electronic equipment, specific examples of these impacts can be given:

1.4 Space Radiation

Failure Tolenace design implies that modular concepts are employed and that additional protective features are incorporated into each module design to prevent short circuit of the essential sources and to eliminate the possibility of producing over-voltages on the output.In addition, the module design must ensure that catastrophic failures within a module do not propagate to create secondary failures in other redundant modules, nor effect the short circuit or over-voltage protection features. This failure propagation avoidance is achieved by separation, both mechanically and electrically, of critical sub circuits, using protective conformal coatings, mechanical separation walls and by providing independent services such as auxiliary power supplies and clock signals.

Space radiation effects the critical performance parameters of most semiconductors and this effect must be taken into account in the choice of component technology used and in the evaluation of the effective parametric drift associated with the actual dosage experienced in a particular orbit.

Furthermore, redundant connectors are employed for all redundant functions, due to the risk associated with the launch vibration and the fact that any removable electro-mechanical part is subject to workmanship errors.

Devices whose critical design parameters are highly susceptible to radiation are in principle never used, which of course eliminates the choice of many attractive components available in the commercial market.

ReliabUity design constraints impose the use of approved packaging concepts, which imply that the use of intensive high density packaging is impossible, with specific requirements on component lead bending radius, track separation etc.

Exceptional use of radiation sensitive parts

In addition, to achieve high reliability the modularity and redundancy requirements impact significantly the packaging of the power electronics.

p

ec-r tj = P x

[9 j-c + {8 c-s/IGc-r} ]

PitJn .1 1'hennal Environment - Space

is only possible if the devices are pre-irradiated to achieve reduced parametric drift and selected for flight application. The impact of this process is to

dramatically increase component costs, hence, making it very difficult to introduce new device technologies due to the complex test and qualification programmes required. Design analysis must take into account radiation drift of component parameters (in addition to life and temperature variation).

5.4

11aermal design requirements, taking into account the severe constraint imposed by the operation in vacuum and the junction temperature limitation of IISC , imply that unique packaging features are employed to ensure that the components work well within their space required temperature derating. Furthermore, the operational temperature range, -20 to + ()()oC, is wider than for most terrestrial applications.

Space Radiation protection also has an impact in the location of radiation sensitive parts, where maximum use of the shielding effects of the box structure can be used to limit the effective dosage on these parts.

Launeh Vibration is also a severe constraint on the mechanical packaging design of electronic components, particularly electromechanical parts and printed circuit boards.

2.

SPACE

POWER

• Power Distribution and Protection switch gear and protection devices are required to manage the payload and proted: the integrity of the power busses, • SeeoIldary Power Conversion

The unique feature of the majority of spacecraft power system designs is the requirement to provide continuous power to the spacecraft payloads and service subsystems (e.g. Attitude Contro~ Telemetry and Telecommand, Thermal, etc.), in an environment where the power source is routinely interrupted by the eclipse of the source, (usually a solar array), resulting from the passage of the satellite behind the earth's shadow. Other mission features, such as the degradation of the solar array due to radiation, alignment of the array with respect to the sun and widely fluctuating load consumption, are drivers in the defmition of the -ideal" power system design concept. Last but not least, system reliability is an essential parameter in the choice of a power system for a spacecraft which may be required to operate autonomously for more than a decade. Similarly, the system must be able to cope with payload and non-essential equipment failures and safely recover normal operation, with the minimum disturbance to unaffected spacecraft loads.

BASIC POWER FUNCTIONS

• P..... Control Uoit - power electronics required to manage and regulate the overall system,

SYSTEM

REQUIREMENTS

3.

• Eneru Stora&e - mainly Nlckel·Cadmium or Nlckel-Hydrogen Batteries, used to provide power in eclipse or to deliver additional peak power in sunlight

SYSTEM

A spacecraft power system, consists of the following basic functional elements:

• Enel'lY Source - usually a Solar Array for near earth satellite applications, sometimes a Radio-Isotope-Generator (RTG) for spacecraft at great distances from the sun.

converters required to generate additional or galvanically isolated voltages, associated with the spacecraft SUbsystems and payloads. The ideal power system is achieved when the combination of the above functions achieves the basic objectives outlined above, with optimum efficiency and minimum mass. The following sections of this paper will describe the impact of the spacecraft mission, orbit and the variability of the energy sources on the choice of power system topologies. Modem power system design techniques will be described, which, as will be shown, are radically different from their relatively straightforward terrestrial counterparts (Uninteruptable Power Supplies). In fact, spacecraft power electronics are frequently at the forefront of the majority of new control concepts being applied in the field of power control techniques.

4.

ORBIT BASICS

The fundamental problem to be solved in the spacecraft power system design is associated with the management of the charge and discharge of the batteries and, as can be easily understood, the spacecraft orbit has a major role in detenniniog how significant a task this is. Orbits are chiefly grouped in 4 main categories: • LEO - Low earth orbit - spacecraft close to the earth (height < 1000 Km), with an eclipse of about 05 hour during a 15 hour orbit of the earth, • GEO - Geostationary - fixed 36,000 Km above the earth, with 1.2 hours eclipse during a 24 hour orbit, occurring during the equinox periods each year.

5.5

D. O'Sullivan: Space Power Elec1ronks - Design Drivt!n

• HEO • HIahJy elUpdcaI - perigee near earth, apogee far away - with variable eclipse and orbit periods de~ndent on specific mission. Although these orbits have high orbit to eclipse ratios, the maximum eclipse times can be up to 5 hours, which becomes a major driver for the batteries and thermal control system.

• Deep Space - low number or DO eclipses but big variation in solar energy, since solar energy is inversely proportional to the square of the distance of the spacecraft from the sun. For distances greater than 2 to 3 AU, frequently Radio-lsotope-Generators are used as alternative energy sources. Of the above, LEO is unique in its high eclipse-to-orbit ratio (about 33%), all the others having a relatively low eclipse operation, relative to the orbit period. The impact for a LEO power system is then unique, in that a 05 hour heavy discharge must be replenished in a period of less than 1 hour. This inherent high charge rate and associated high power requirement, imply that a very careful design approach is necessary, since the batteries in such an orbit are usually essential to the mission success and the method of charge significantly effects the size of the required solar array.

The other orbits, whilst less stressing, have varying degrees of difficulty, dependent on the eclipse operational requirement (e.g., reduced or full eclipse power) and the power level and reliability requirements imposed by the mission.

• TeIeeom..unlc:atlOllS - GEO orbit, full telecommunications in eclipse at power levels up to 6 kW or Direct Television Broadcast with low eclipse power (circa 500 w) but high sunlight power of up to 10 kW and sometimes a combination of both! In addition, telecommunications spacecraft have to be designed to provide a high adaptability to new payloads and eclipse requirements and achieve a high reliability, autonomy in-orbit and last, but not least, low cost!

• Earth Resources - LEO orbits, high power (circa 6 kW) with high peak power requirements of up to 10 kW. High stress on batteries, which are key elements in determining mission lifetime and system mass. • MeteorolOlY· GEO orbits at power levels between 300 w and 15 tW. • Manned Missions - LEO Space Station or Laboratory applications or manned vehicles with power levels in the range 3 to 30 kW. These missions, furthermore, pose severe reliability, safety and maintenance requirements, which are unique.

6.

ENERGY SOURCES

In order to understand what the power system task is, it is essential to have a basic idea of the characteristics of the basic elements used in the system design: 6.1 Solar Array A

5. MISSION TYPES Again, the various missions can be loosely grouped into 5 basic categories: • Science - any orbit and power level between 200 w and 1.5 kW - scientific requirements can have major impact on power system choice, e.g., payloads with extremely low requirements on ambient electric, magnetic and electromagnetic fields, can result in the rejection of conventional regulation schemes (Pulse Width Modulation), magnetic materials and components (e.g. Nickel-Cadmium batteries and relays).

u

..

3.51=f:.....~~;j ~~

~~~~~~~~~~

2.5 2D

1.5 1.0 Q.5

o

o

10

20

30

40

51

8)

10

80

100 V

Solar AmIy CIuIrtlcteristi as IJ

function of Temperature

5.6

II)

D. O'Sullivtm: Space Powr EI«trottics - Design Drivers

Fig. 4a shows a grossly simplified equivalent circuit model of a solar array, which can be represented by a current source feeding a "diode" type of characteristic, with some imperfections simulated with resistors. The current source is essentially proportional to solar energy received by the array and hence, will vary with: • the angular pointing of the array with respect to the sun (Cosine factor) • the distance between the sun and the spacecraft (especially important for deep space missions - inverse square law), • radiation damage experienced (which is cumulative with lifetime).

Hence, the solar array, whilst being a convenient source of energy, requires a significant regulation effort, either at system level or by the spacecraft subsystems and payload equipment. U Batteries As might be expected, the batteries also pose a problem for the system designer, since the battery voltage characteristics are heavily dependent on the mode of operation, charge or discharge and ageing.

A typical battery charge/discharge profile is illustrated in Fig. S.

V-cell

V-Batt

1.6

43.2 40.5 37.8 36.1 32.4 21.7

1.4

The resultant sensitivity of the array characteristics are illustrated in FIgS. 4a and 4b.

1.0

A 3.5 ~

ao

fI

2.0

1.5 1.0

o

1.2

......

,~

iii

,

~"'"

•

"

, ,, l

:;

..... ,.,.-~

I

I I

0

10

3)

It should be noted that this "typicar CUI'\'e is further effected by battery temperature, charge and discharge rates and number of charge/discharge cycles!

l

~D

S)

8)

-, 70

II)

v

SoIIIr Anuy CIuJmcteristics lIS Q function o/lUumination cI)ladiation Hence, the resulting voltage and current exhibits large variations during the mission lifetime. Similarly, the voltage characteristic is logarithmically dependent on the current but, perhaps the most significant factor effecting this characteristic is the array temperature, which is spacecraft thermal and configuration design dependent (spinning or 3-Axis spacecraft, etc.) and in the case of deep space programmes, is drastically affected by the sun-to-spacecraft distance variation (inverse square law).

6.3 Subsystem and Payload Converters Power converters fall into 2 basic categories; those which directly convert by fixed ratio DC/DC converters (see Fig. 6) and those which can tolerate a variation in input voltage, utilising Pulse Width Modulation (PWM) regulation schemes (see Fig. 7). The fixed DC/DC converter approach presents a direct reflection of the loads and usually these present a highly variable resistive or current type of characteristic as illustrated in Fig. 6.

Li

Simple DCjDC Convener

5.7

D. O'SulIiwIn: Space Pt1We' EI«IrrMks - Daip Drlvm

The PWM type of converter results in a "constant" power type of input characteristic, since:

characteristics of the power sources, with the main power system providing "digital" intervention only, to manage major transitions (e.g. Sun/eclipse transients or peak power management). An eDlllple of this type of system is illustrated in rJg. 8.

1= P V

Typical input characteristics are illustrated in rag. 7.

27 to 42.5V Bus

For each converter category, numerous different topologies may be chosen, further compIicatiDg the situation! ) Inp~t Chara~eristic V-III

V-Ill-max

V-in -min !-in ...

Vin(mu:) -> VinC"';n)

U

err ~

r..,. 7.

~ 7. PWM Conve1ter

POWER SYSTEM TOPOWGIES

With all the variables described above and the significant mass involved in power system components, it is obvious that, to a large extent, each power system has to be custom configured for each mission due to the: • variable orbit and eclipse periods, • variation in power requirements (100 's of Watts to 10 's of kW), • payload requirements (scientific, telecommunication, earth resources, manned vehicles, etc.). However, although the size and hardware may be drastica1ly different from one mission to another, the basic configuration of power systems fall into a few fundamental types. 7.1 Unregulated Power Busses This approach is based on the idea of simplifying the main power system, where the system designer presents the users (subsystems and payloads) with the task of accepting the variable

5.8

u~n~re~g~U-la~m-d~p~owerBUS ,..,. & lhwgu/IIWJ p - Sy.rIem

Essentially, this concept is based on minimising the centralised power conditioning, using only simple switches and diodes, leaving the user to cope with the wide variations in voltage and power characteristics and is mainly used only for LEO orbit applications, where the addition of power regulators (especially, the Battery Charge Regulator (BCR) which has a power level equivalent to the total bus power!) would have significant mass impacts. However, the necessity to use PWM type converters for all bus users, handling a wide range of input voltage, significantly offsets the sa~ made on the centralised power conditioning (especially when considering the ratio between the actual usage versus the designed-in power capability of these converters, including fadors such as equipment redundancy.

It should be noted that the energy sources are never optimum with respect to the load, due to dependence of the available output power as a function of battery state-or-charge and bus loading (i.e. the battery voltage effects the delivered array power, since the solar array, being essentially a current source, the working voltage determines the available power).

D. O'SuIIivan: Space Power EIectronks - Design Drivers

7.2 Regulated Power Busses This approach is the opposite to the previous scheme, the regulation of the resultant power bus (or busses) is performed centrally, with two basic objectives in mind:

I-Arr:y

t I-Shun. .- - - -....

• Optimisation of Solar Array and battery design, which have major system mass impacts, usually far more significant than the power conditioning equipment mass, • Simplify user voltage interface, in the interest of simpler converter design, higher standardisation of equipment, simpler Electromagnetic Cleanliness Control (EMC). The basic concept of the regulated bus topology is shown in Fig. 9. In order to understand the basic concept, some essential principles need to be described here. Essentially, the major mission variability affecting the user is the variation in bus voltage and therefore, in sunlight operation, this is dependent on the voltage and current variation of the solar array (see F'JgS.4a & 4b). The simplest method of regulating out this variation is to shunt the current source, so that the resultant net current exactly matches the load demand. This is achieved by operating the shunt to "clamp" the array voltage, this method is schematically illustrated in Fig. lOa.

I-Load

•

SHU Reg.

Shunt Regulator

Pf&Iw JIk Shunt Regulation In addition, the task of replenishing the spacecraft batteries in sunlight, must be accomplished with a changing battery voltage characteristic during charge, absorbing power bus current and, at the same time, maintaining the bus voltage regulation. This function is performed by means of the Battery Charge Regulator (BCR) as shown in F'tg. lOb. Similarly, in eclipse operation from the spacecraft batteries, the variation in battery voltage must be buffered from the power bus by the addition of a PWM (Pulse Width Modulation) Battery Discharge Regulator ( BDR) in order to maintain the power bus at a constant voltage during the discharge of the batteries. Ftg. 10e illustrates this principle.

Power Distribution Unit

Solar Panels

....._ _ _ _... Shunt .....- - - ..... Reg.

•

-

..L

.•

-

..L

FIfIJIIf! 9. Regulated Power System

5.9

I-Arr8l

rig. 11 explains how the 3 types of regulators are ·integrated· into a single control loop and rJg.12 illustrates the essence of the principle of sequential control and how this is translated at Bus voltage level to a negligible voltage variation, by the utilisation of a high gain (proportional-integral control) amplifier.

I-Loa\,

Load

Battery Charge Regulator Charge Regulation

I-Loat

load

Battery Discharge Regulator JIigJft IDe.

DischtJTge Regulation

Again, as was the case for the payload and service converters, for each regulator function above there exist many competitive approaches! The criterion for an ideal choice will be dealt with later in this paper.

The three separate operations (Shunt, Battery Charge and Discharge) described above appear to be separate tasks, however, control concepts used in modem satellite designs, employ the concept known as; "3-Domain Control", whereby the 3 control loops are nested within a single overall control loop, with sequential operation of the inner power regulator modules. With this approach, each set of parallel modules operates under the conductance control principle. (See Section 9).

5.10

7.3 Regulated Power Bus - 3-DomaIa Controlled The concept employed in the multi-domain approach (some missions {Space Stations} require up to 4 control domains), is the idea of locating the linear control voltage of each individual regulator category in a logical sequence, such that, for example, the battery discharge would be a maximum for the minimum control voltage (V-control of rig. 12) and as the discharge current decreases to zero, the control voltage rises until it enters the domain of the Charge regulator, starting at zero current and progressively increasing, as the control voltage continues to rise. Similarly, when the limit of charge current is reached, the control voltage enters the minimum control value for the shunt regulation. Whilst the control vokage (V-control) ruge may be in the range of 10's of volts, this is translated to the bus voltage by means of a P-I (Proportional-Integrator) controlled main error amplifier, where the resultant voltage change is reduced practically to zero. The resulting 3-Domain control system results in a power bus which is always constant, irrespective of the sun/eclipse status, and is capable (due to the automatic sequence of the regulators) of delivering peak power to the user via the BDR and batteries without any bus voltage variation. Furthermore, the resultant maximum power bus source impedance (see fIgure 13) is typically less than 2% of the maximum load. This type of system is suitable for all mission applications, although for LEO it's use is limited (due to the high power conditioning mass associated with the LEO orbit) to applications where high voltage power busses are required.

Solar Array I-Array

V-ref Charge

SIC Load

.Marn lL Jntegnlted 3-DomtIin ContmI Loop

t

V-Control

I

I

Char~ I

I-Shunt

Max

V Bus

I-Charge

Amplication reduces OV-Control to CJV-Bus

Ov

Max

I-Disch. Max

oA

Net Bus Current

.Marn 12. 3 Domain Control - Signal SchemIItic 5.11

This is a variant of the Unregulated Power Bus, with the addition of a solar array regulator to damp the maximum bus voltage during the sunlight part of the mission. The effect of this is reduce by a small amount the vokage excursion of the bus wItage over the orbit period.

Z(IIIO)

111

11

-

11{

11K

1111(

,.

F.....-cr

PIJrn 13. kgu/IIted Power Bus Impedllnce

Essentially, this concept, illustrated in Fig. 14, is adopted as a compromise between the two extremes desaibed above and is usually based on the desire to eliminate the mass and cost of the Battery Discharge Regulator (BDR). This concept therefore delegates the problem of source regulation in eclipse to the spacecraft user, similar to the case of the UDJ'CI'Ilated bus concept and is therefore, not ideally suited to full eclipse operation.

7.4 S.......t ....Iated P...... Bu

aueA

7.5 Hybrid Power Bus

This type of power system is designed to bridge the gap between the Regulated Bus concept

and the Unregulated Bus approach, specifically to meet the requirements of a LEO application.

a_a

PIJrn 14. Sunlight kgu/Ilted Bus

Fig. 15 illustrates the basic topology, which shows how about 50% of the solar array is coanected Yia the Shunt Regu1ator to a Regulated bus, supported by a simple BDR (a high efficiency Buck regulator, 93% at 28V-DC), controlled by a 2 Domain controller, the remaining array sections are dedicated to battery charge control, which replace the BCR's used in the Regulated Bus Scheme.

Battery Charge Arrays

EOC Switch

Fully Regulated ""'-4":-~ Power BUfPn=L-

r.,. 5.12

V-Batt> V-Bus 15. Hybrid power bus

Fig. 15 also indicates that a mixed payload may be supplied, some using the Regulated Bus and other pulsed loads (e.g., Radar's, UDARS etc. ) can directly access the batteries. This separation of payload types can simplify Electromagnetic Oeanliness Control (EMC), by isolating pulsed loads from Service and wquietWpayloads.

BUS CONFIGURATION CHOICE

The apparently obvious choice here is the Unregulated scheme, where the main advantage lies in the fad that there is little power conditioning involved (1 kW for each kWon the bus as normally only a Shunt regulator is employed for Battery Charge control), however, this also has an inherent drawback when more than one battery system is required, as it is not possible to guarantee that parallel battery operation will be successful and it is difficult to individually manage batteries.

The many mission, orbit and payload requirements combined with the available power bus configurations and the vagaries of human preference and experience, imply that the probability of two system designers agreeing on the same concept is not always very high.

A further disadvantage of such a choice is that pulsed power loads cause conducted noise that has to be tolerated by all bus users, includiag the Service equipment and this fact tends to complicate the design of the user equipment, includiag the requirement to use complex PWM type converters.

In this section, an attempt will be made to highlight the advantages and disadvantages of alternative topologies for the 2 major orbit types: • LEO (Low Earth Orbit) • GEO (Geostationary) and also HEO

The Hybrid Power Configuration (FIg. 15) has some of the sophistication of the Regulated bus concept (simplified BDR's and a Shunt, associated with a 2 Domain control) and hence ina-eased mass (2 kW of power conditioning for 1 kW of load).

8.1 Low Earth Orbit AppUcations As a rule of thumb, for each Kilowatt of power delivered to the spacecraft loads, requires: • 1 kW discharged from the batteries • 1 kW to recharge the batteries (caused by relatively low W-Hour efficiency of batteries) • 2 kW from the solar array (of which 1 kW is used for battery charging) • 2 kW capability for a shunt regulator.

However, the configuration (see F18- 15) allows optimum use of the solar arrays (as 50% of the arrays work at fixed, high voltage), individual battery management, possibility to vary the number of battery cells ("linear" battery mass adaptation to load), provides a regulated bus interface (hence, simpler power converters) to Service and Wstaticw payloads and the possibility of access to the batteries for pulsed power loads, such as Radars etc., hence, simplifying EMC control.

8.

It can be easily seen from the above that a Regulated Power system (Fig. 9) would require a total of 4 kW of power conditioning equipment for each 1 kW of load (versus 2 kW for a "normalw OEO application) and hence, this topology is only selected when specific requirements dictate it, an example being the need to generate a high voltage (120 V) power bus (COLUMBUS - Space Station), whilst maintaining relatively low voltage batteries. A similar use is justified when the designer wishes to allow changes in battery conftguration or technology by buffering the batteries from the bus by means of a Battery Regulator Unit (BRU which is a combination ofBCR and BDR).

The only real contenders for conventional LEO mission are the: • Unregulated Power Bus (Fig. 8) • Hybrid Power Bus (FIg. 15)

8.2 Geostationary Orbit (GEO) Here, again, a difficult choice exists, but the main contenders can be simplified to: • Sunlight Regulated Bus (Fig. 14) • Regulated Bus (FIg. 9) Again it is tempting, to choose the configuration offering the simplest hardware; the Sunlight Regulated concept, where, the elimination of the BDR (Battery Discharge Regulator) is seen as an immediate gain (typical mass reduction of 6 to 7 Kg /kW of load power).

However, this simplistic mass saving has to be paid for somewhere and in this case, as illustrated in Fig. 16, it is the Service equipment and payload converters which have to accept the unregulated interface in eclipse, which imposes a requirement to use more complex and heavier PWM type converters

5.13

(mass penalty 8 to 10 Kg /kW of load power, with a factor for equipment redundancy!). In addition, the decentralised converters, employing PWM techniques, operate at lower powerjunit and hence, are lower in efficiency, resulting in an increase of the total power demand and therefore, increasing the solar array and battery mass.

configuration of batteries is possible by changing the number of series cells, use of multiple batteries and variable battery capacities. Simllarly, to achieve higher eclipse power the maximum battery capacity wiD limit the expansion of the Sunlight Regulated scheme, whereas, for the Regulated Bus concept, either more batteries can be added per bus, more or less ceDs used and of course alternative capacity values selected. Last, but not least, the Regulated bus topology allows growth to power levels in the range 5 to 10 kW, when a Battery-above-Bus concept is adopted! The Regulated Power Bus (Fag. 9) concept has the following advantages, which are unique: # Peak Power in excess of array capability can be delivered by the BDR function, hence, minimising the size of the solar array.

r.". 16.. PWM Regu.lIJtion - Location A further advantage of the regulated power system is that battery optimisation is easier. Battery optimisation depends on the following parameters: DOD Pt! tt! 'lit

Vcmd N,

N"

e"

= Battery Depth-of-Discharge = eclipse power requirement = eclipse duration (1.2 Hours) - Discharge path efficiency = Cell mean discharge voltage = number of series battery cells = number of batteries = battery Ampere-Hour capacity

The optimum battery design can then be defmed as follows:

Since, for a Sunlight Regulated (Fig- 14 & 16) scheme, where the number of batteries is faxed at one per bus, the number of ceDs is fixed and dictated by the bus voltage specification and hence, the only method of battery optimisation is by custom building battery ceDs to a specific cell capacity. The mass penalty here can be anywhere between 5 and 20 KgjkW of bus power! In the case of a regulated bus, the battery charge and discharge regulators buffer the battery from the bus voltage, hence, almost any

5.14

# Energy matching of batteries to eclipse load demand is simplified by choosing: • the number of cells used (typica1ly 20 to 30 series connected cells) • the choice of a standard range of cell capacities • the number of batteries used (any number of batteries can be connected to a bus, since, BCRjBDR provides individual charge/discharge management) . # User Power Converters can be of the simple DC/DC Converter type (lighter and absence of over-voltage failure modes).

9.

CONTROL PRINCIPLES FOR SPACE POWER REGULATORS

As indicated by the principle shown in Fig. 1, modularity is a key feature of space power control schemes. This modularity requires that power is equally shared between modules and is linearly controlled by a centralised control (error) amplifier.

Since, these modules are involved also in the regulation of the overall power bus, the control characteristics are also of major importance. Modem power regulators employ the principle of Conductance control (Ref3 & 7), where,

D. O'SuIlivan: S~ Power Ekctnmics - Design J:JtitIen

by incorporating an inner current or conductance control loop in all power module designs (see Ftg. 17), power modules can be paralleled like building blocks and in addition, the control loop has a -first-order" rather than a "second-order" nature, as opposed to that of conventional PWM techniques, resulting in simple control loop design and improved dynamic response. Conductance control: • Vutually eliminates the Inductor from the outer loop, transforming a 2nd Order control function to a 1st Order one. • Allows paralleling of PWM Power stages, with inherent power sharing. • The inner loop allows limitation of peak currents in semiconductors The equivalent circuit of the control principle is illustrated in F.. 18.

Vi

Conductance Control offers the following advantages: • 1st Order Response => Simple Stability • • Paralleling of Power Stages is inherent to the concept of "Conductance", as power is equally split between modules. • Current limitation and semiconductor protection is achieved by setting a limit to the conductance control input

10.

SPACE REGULATOR TOPOWGIES

For each of the regulator types defined in section 7:2, optimal topologies have been developed to meet the essential requirements

identified and discussed in section 1.

Vo

Vx

shown, together with the well controlled output impedance achieved.

~~~~~~~~~

10.1 Solar Array Regulators Interfacing with solar array characteristics as illustrated in Ftp.4a and b, the array regulation scheme.

-.,--....-----0-.Rs. . . . . . . .

, Added Integrator

Vc

1

f=~

...._ _............IJr--....... frequency Conductance i

I-GV e

=G v.Z

v

0

, C

c

Power Sou~

R

Unity Gain

zo/p1

KAG

e Open LoopGllin=

+:: Vo

K A G Z

F'IJIIC 18.. The resultant control characteristics are shown in Ftg. 19, where the "First order" effect is

Jii,JI.ft 19.

Condut:ttmce Control Ouuvcteristic

5.15

can either be operated in linear or digital mode. Since, clearly for high power applications, it is preferable to use non-dissipative techniques, most high power array regulators employ digital techniques. The principle approaches that can thus be considered are either series or shunt switching of array sections (where the number of sections is usually greater than 10). At low bus voltages, either series or shunt switching approaches may be used, however the two approaches have different advantages and disadvantages:

The series concept has its main advantage in having a low resistance in series with the array and can achieve a lower voltage drop and hence, has a thermal dissipation than shunt switches, where, when the shunt is open, the array current passes through a diode. However, the driving circuit for the shunt switch is simpler (ground referenced), whereas the series switch requires a floating high reliability supply, which must never rall. Another factor effecting the choice of the switching technique is the bus vokage, since, due to the sensitivity of the array voltage to temperature, voltage excursions of almost 3 to 1 (w.r.t. bus voltage) can occur. Thus when bus

voltages are in the range of 50 to 15OV, the choice of switches becomes difficult, especially considering the 65% voltage derating required for space applications. Thus for the majority of space applications, shunt switching concepts are employed. A typical type of shunt regulator used for high power (SOOw to 10kW) is illustrated in Fig. 20.

This configuration (ESA Patent, Refs.4 & 5) employs the principle of sequential switching of solar array sections as a function of the control signal emanating from the main error amplifier. Each switch is operated by a comparator, whose hysteresis levels are arranged so that sequential thresholds are achieved from one section to the next. This sequential switching thus allows all sections, except one, which oscillates digitally, to be in a digital state. For further details refer to references 4 and 5. 10.2 Battery Charge RepIaton In the majority of applications, the battery end-of-charge voltage is less than the bus voltage, hence linear series or PWM buck-type (step-down) regulators are used to control the battery charge current and the termination of charge. In addition,

Comparitor Controller

s~ - Sequential Switching Shunt Regulator FitJue 2fl ~R - Sequentiol Switching Shunt Regu/tltor 5.16

D. O'SuJ/ivan: SptII:e Power Electronics - Design Driven

in a 3-domain controlled system the regulator must allow input current control to achieve conductance control with respect to the central bus regulation loop.

SIC Protection

V.Bus o - - - - + s

For higher capacity batteries, PWM techniques are necessary to reduce dissipation. The conventional choice of a classical Buck type regulator (see Fig. 7) is ruled out for this application due to the need to provide the conductance control feature on the input of the BCR. The selected topology is a "Zero" ripple topology, which has the property that the current is continuous on each power node and conductance control can be applied to any of the 3 nodes (See Fig. 21t V-Bus, V-Batt or ground). "Zero" Ripple ~ Charge Regulator

Thus in the application for battery chargiDg, the conductance control is applied to control the current diverted from the power bus to charge the batteries. In addition to this conductance. contro~ additional battery management circuitry intervene in this controller to assure corred end-of-charge conditions for each battery.

(Used for 3-Domain

Jripw 21.

"Z6o" Ripple Regllllltor

tnpltCumM"C

Fig. 21 illustrates the additional switch added to protect against failures of the primary PWM switch. which would otherwise lead to excessive current to be drawn from the power bus.

limiter needed

eus Voltage RegulatiOn) CIuItgt!

Btlttety

~o:= (BOOST) ~~

High otipuf currenl ripple

10.3 Battery Discharge Regulators

The battery discharge task, which is essentially a Boost function (to interface the lower voltage battery to the bus) could, in principle, be handled by a conventional Boost regulator as indicated in Fig. 22 Despite the primary advantages of simplicity and high efficiency, the disadvantages of this topology are however numerous, as shown in the ftgure (22). In addition. it cannot be controlled using conventional conductance control methods.

Boost BDR Disadvantages

Conventional Regu/Iltor

Boost

DisduI1ge

The preferred topology used in the majority of space applications is the SMART (Standard Multiple Application Topology) Battery Discharge Regulator.

5.17

D. O'SuIlivtm: Space PD'MT EI«ttonics - Design Drivers

This conftgW'ation, shown in Fig. 23 below, has all the properties required for such an

- bandwidth from DC to greater than 500KHz - high common mode rejection

application: - continuous output current to the bus, required to reduce electromagnetic interference and provide the ideal conductance control property, with respect to the bus voltage controller. - series switch provides regulation and current

limiting - push-pull converter provides optimal voltage conversion and over voltage protection (e.g. stopping con\'Crter, inhibits boost function)

:::::==pIic:IIIon ~

11.

~TapoIogyl

21 SMART &ttety DisclultJe Regu/IItor

DESIGN FEATURES FOR STANDARD REGULATORS

Many of the protective features of power regulators used in space applications, require the use of sophisticated control and measurement devices. One of the most desirable features for power controllers to possess is high galvanic isolation between the power circuit and the low level control circuits.

The principle (Ref. 6) employed is based on the ideal transformer (see F'tg.25), where the Hall Meet produces an output as a function of the mapetising current, to which is then added, a signal derMd from a shorted secondary winding. The resultant measurement thus encompasses the "DC" and ·AC" components of the measured current. Since the Hall Effect measures only the magnetising curreDt, its baadwidth may be low, siacc, the shorted

secoadary proYides the high frequeDC)' response. 1-

aatlon .

"~

.

..:

~"

.... "'--"""". . .

lin

t

An example of these applications is shown

in F'tg. 24. In this example of a Buck regulator, the current measurement is performed by means of a

high performance Hall Effect Current sensor and the power switch is driven by an isolated drive circuit.

11.1 Hall meet Curreal Sensors The Hall Effect sensor has been developed to meet the following requirements: - Galvanic isolation between high power/voltage line and control circuit - good linearity of output voltage as a function of measured current

5.'18

c

Transformer Ideill Cin:uit

F'JgUfC 26 gi\'Cs a schematic illustration of how a practical current sensor, based on the above principle, has been implemented.

~

. I

•

Lu I

J.CIP 'T

I MlIKILI

I

II<

••

n,rn 26. Hall Effect Cunent SDUOI" FET Pulae DrIVe Waveforme -1r----FE-T-ON----~1 Oy

11.2 Galvanically Isolated DrIve ClreuifB Many power regulator and converter schemes require the use of series switching devices, operating with high voltages (relative to the control circuits, e.g.. 50 to 150V DC) and it was necessary to develop drive circuits that were capable of operatina at any voltage, with a control capable of modulating the switch from the permanently OFF condition, through variable duty cycles to the permanently ON condition. The circuitry had also to capable of driving parallel sets of high capacity power mosfets, with switching times in the order of a lOOnS, in order to achieve low commutation loss. The technique used in high power space applications is illustrated in Ftg.27. The principle used is based on the gating of high frequency pulses (IpS) of current to peak charge the mosfet capacity and maintaining this state by refreshing the pulses at a more moderate clock rate (100KHz). Transitions from ON to OFF, involve, reversing the process, with a second oscillator and a synchronous rectification process on the transformer secondaries. Due to the high frequency of operation, the resultant drive circuit is highly compact, but achieves the objective of a high frequency mosfet driver with galvanic isolation. 11.3 SoUd State SwItches Whilst most of the development effort of space power electronics has been concentrated in the development of control techniques, special regulator topologies and circuits, recently, with the advent of high power manned applications (HERMES, COLUMBUS) the need to develop "intelligent" switch gear has received a boost, since, there are few

. n

D

~,"-------------D ,oy

components which can operate reliably with high voltage DC busses. Thus there has been major efforts to develop sophisticated solid state switches, which provide: • active current limiting (delay less than 5pS), - intrinsic temperature protection, - latching functions, • noise disaimination on command interfaces, • galvanic isolation • analogue telemetry of throughput current The principle employed for these switches is illustrated in FIg. 28 below, where only the essential properties are shown. The essential principle applied is to control a power mosfet by means of a low power, high frequency converter (300 to 500KHz), such that the device is in the ON state when the secondary voltage is achieved. Similarly, turn OFF is achieved by removing the primary low power supply.

5.19

However, at this stage, the author is aware that many topics have only been briefly touched upoa but hopes that the curious reader may be teapted to read further literature in this specialist

fielcL ~

1

2 Solid Stote Switch - Ptindple

The protection features (all of which are located on the secondary circuit) of the switch use 3 essential ideas. The first being to operate the mosfet in a continuously linear mode, by introducing a control loop to force the drain-source voltage to be a constant ratio with respect to the voltage across the current measurement resistor. This loop thus ensures that the gate-source voltage (Vgs) tracks the channel current and thus, when an output overload occurs, the first level of limitation is achieved by the inherent limitation associated with the existing Vgs and the conductance value for the mosfet. Thus during a' rapid overload, the Vgs slowly increases to supply the increased output current, however, once the increased current reaches the second design feature, the current limit loop, the Vgs is limited to maintain constant current. Should the overload be permanent, a timer whose duration is a function of the substrate temperature (designed to limit the maximum allowable junction temperature (11SOC), will shut down the power stage, which can only be reset by turning the switch OFF and ON again.

12.

CONCLUSION

This paper has attempted to cover a wide range of topics, from system level aspects of space power system design, their basic requirements, detailed design concepts, down to the basic philosophy of how to achieve effective reliability through design.

5.20

3

4.

5.

6. 7.

Lacore B., Aulysll orp.,.... BIll Topol. . ued In TeIecommuakatioa SateWtes, European Space Power Conference, Madrid 1989 Capel A, O'Sullivan D., laflueuce of Bas Relulatloa OD TelecommuDlcatioD Spacecraft Power System ud DlstrilJUtloa, Power Electronics Specialist Conference 1985, ESA SP-230 Weinberg A., O'Sullivan D., W: AppUeatioD to Voltap Replatioa, Spacecraft Power Conditioning Seminar, 1m, ESA SP-126 O'Sullivan D., Weinberg A., Dlsposltlf de Replatloa du Courant. Partir de Ploslers Sources d'Energie. Belgian Patent: No. 0/1763 27, 31 March 1m. O'Sullivan D., Weinberg A., 1be SequeDtiaI Switching ShUDt Regulator (S.lR) , Spacecraft Power Conditioning Seminar, 1m, ESA SP-126 Ghislanzoni L. , ESA Patent Hall meet Curreat Seasor Patent O'Sullivan D., Spruyt H., erausaz A., PWM Coaduetaaee Control, IEEE-PESC-1988

Betekenis van Robotica in de Ruimtevaart

. . :w.iieln:;.de:: Peuter:.:[/ .[.:[. HetUi:::of::[·l.lObOnt::·Sect~on :.:::<::,:,

.: ..: .

'.

n' 1$SA _.:q§TEq "lipord\Vij,k; the Netherlands ' ~{{ .

;r:!,·Wih~m;: d.¢.:: :Beuler; 26 mei 1953 eti he~ff~e ~el~he ~tion~teit.

:, , :·: :Wj[;·studee(4~·:ar aan de l(athol~¢ke 'UQiversiteh

Lcuven'alSoBurgerHjk Ingenieur'

::.::
. '.

Electronika..Regelteehniek.

.: Head:':(jf:; Robotic:SectiQri~:..: : : ::::::::::::::::;::::::))/ .........

.

... .. ...

......... :.:.:.;.:.;.::-:-:-:.,.

::::.

.: : .

. . .. .

"::(:.:

'<;> :;::::.::.. :-:--, ......... ......... :::::: .;.:.:.:-:::::: .... . ..... ..................... .:........ :.::::::.-:.:-:-: ...:::-:-:::-:.:-:::-:::::-::::::..... ::.:::-:-:::::::::::;:::::::::::::-:-:.:.

\%,[:::::;':;.::

:.... .

:::::;/

6.1

W. de Peuter: De Betdcmis van Robotica in de RIlimttvalln

"Oe betekenis 'Ian Robotika vocr 1e

~uimtevaart"

frEE: :2·..1·93)

w. De Peuter Head of Robotics Section ESAiESTEC Noordwijk

1. Inleiding No een gOlf van aufomotisenng in de lancbouw. comlnistrotie, industrie, vliegwezen eTC. 'NorCT nu oak de ruimtevaort gerroffen 'Nereen door sen toenemende interesse 'Ioor (oOOTiko en oncere vormen von ouromoiisering. De redenen hiervoor zijn in ~ootazaok crievoudig : ( 1) (2) (3)

bemonde ruimtevoart is erg duur. ',ocral voor !ongdurige missies; de oongescnerpte veiligheidsnormen no net Challenger ongeval moken het mOeilijk om nieuwe technologieen ie georuiken op een bemonde vlucht: met onbemonde satellieten en rCOCT-sondes zijn veel meer bonen en monoeuvers mogelijk, en de ianceercatum kan beter gegorandeerd worden.

De vroog is don oak niet zozeer OF we moeren cUTomatiseren, dan wei HOE we dct moeten doen ! In deze overweging gOaT ~eT dan vooral om de vroag: specifieke oplosslngen, offlexibele. universele automctisering? Een olgemeen on1Woord hierop bestaat niet, en moet voor elke missie CCCrT besTudeerd worden. Zoals dikwijls zal oak hier een gulden middenweg de beste resultaren opleveren. woarbij zowel robots 015 specitieke mecnonismen en inSirumenten (en astronaut en) mekoor aonvuUen om te komen tot een optimaol werkena systeem.

I

II !'

S I' II ! I

6.2

2. Waarom robotika ? RObotisenng is voor de rUlmtevaart een ccmreKkehjke aUTomctisenngs tec:1nolcg:e om verschillende redenen. die in wezen nler z:) sterl< ofv.Jljken van ae argumenten aie gelden voor de keuze tot roootisenng in gewone oordse toepassingen: a) Robots kunnen een veelheid van taken ccn. terwijl ae komplexiteit van de machines (in de rUlmre) nagenoeg onvercnderd blijft: meer taken betekent in eerste instantie langere cOOlicctieprogramma's. en nlet extra hardware. Dus in plaats van verschillenae mechanlsmen op de satelliet te installeren voor diverse taken. kon een robot met een beperkt set end-effectors de taken uitvoeren. wot kan leiden tot minder massa en VOlume, en vooral een minder komplexe uitrusting. b) Robots kunnen geherprogrammeerd worden vanop atstand. en kunnen dus nieuwe taken en opdrachten krijgen zander dot we fysiek toegang moeten hebben tot de apparatuur. Dit is voornamelijk belangrijk in wetenschappelijke missies waarbij de taakdefinitle haast nooit voor 100% op voorhand kan gedaan worden, en natuurtijk oak ingeval van onvoorziene gebeurtenissen zoalS technische mankementen. c) RObots zijn universele machines. en kunnen dus reeas ontwikkeld worden (of Zijn) voordot de fuimtemissie in detail is gedefinteerd. Dit leidt niet aileen tot een verkorting van het ontwikkeltraject. maar laot ook toe dezelfde of soortgelijke machines te gebruiken voor verschillende projecten waardoor ervaring wordt opgedaan en de "kinderziektes· eruit gehaald. Dus: minder problemen die typisch zijn voor prototypes. Dit is een van de hoofdargumenten voor het gebruik van robots in bv. de auto assemblage, en is uiteroord een sterke troet voor het inzetten van robots in de ruimtevaort. woar een technlsch detekt meestol hard wordt afgestroft.

3. Toepassingen van ruimte robots De huidige toepassingen van ruimterobOTS situeren zich -vreemd genoeghoofdZokelijk in prOjecten van bemande ~uimtevaort. omdat dit doorgoans activiteiten zijn waor grete structuren moeten opgebouwd en onderhouden worden, en waarin sproke is van een redelijk grote heeveelheid logistieke operaties met een echte materiaol stroom. Echter. het overgrote deel von de bestoande sotellieten bevinden zich in de geostotionclre boon die onbereikboor is voor de mens met de bestoonde tronsoorttoestellen. 30vendien situeren wetenschappelijke missies zich dikwljls op grote afstanden van ae oorde, waorbij de inzet van astronauten a priori is uitgesloten. Het toepcssingsgebied van ruimterobots kon opgedeeld worden in 4 groepen: a) Loge oordse banen (LEO). in de vrije rUlmre (EVA): Dit zljn meestal grote robots. cie in hoofdZaak 5 belangrijke taken hebben: PiCk & Place van welgedefinieerde objeKten eCRU's), meestol tussen een logiStiek tronspontoestel en het rUlmrestcrtion voor het vervongen van wetenschcppelijke insrrumenren, IJltwl$Seten van cfgewerkte en nieuwe monsters. t>randstof reservOirs. koetvloeistof. t>artenjen, warmtewisselaors etc. Het ondersteunen van astronauten gedurende een ruimrewandeling, aoor re fungeren OIS ·':':ersenplukker'. of ceor her oonrelken van oOJecten en gereeaschap.

6.3

Het -ofmeren- von ruimte tronsporttoestetlen, of het oDpikken van zwevende objecten, vorierend von defelCte satellieren. Josgerookte onderdeten of gereedschOp. Het routinemotlg inspecteren von het ruimteschip, lokoliseren von meteolieten SChrootinslag. reinigen en afdekken von optische instrumenten. Het assembleren von grote strukturen in de (uimte, zools vakwerken, antennes, energie generatoren, reftecktoren. Voorbeelden hiervan zijn de robot arm von het omerikaanse ruimteveer, de conodese robot voor Space Station Freedom, en het europese ERA projekt (zje flg.2). De voomoamste technische uitdagingen situeren zich rond het garonderen von een voldoende bedrijtszekerheid van een dergetijke machine die over een lange periode (10 jaor) btootgesteld is can de physische en chemische condities von de vrije ruimte.

..

~

..

~---

Fig. 2 b) Loge aordse banen (LEO), binnenin ruimteschepen (IVA) Dit betrett kleine (max. 1.5 meter) tot zeer kleine robots die blnnenin (uimtestatlons werken of zelfs binnenin een wetenschoppetijk instrument. Zij moeten doorgoons een grote verscheidenheid aon token kunnen uitvoeren, en zjjn ook meer in gebruik. De voomaomste reden woorom deze robots zuUen aongewend worden is het onttosten van astronauten rOUTine token. en omdot de ruimtestations niet altljd bemand zijn. Vooral tijdens deze onoemonde pelicoes zullen robots onmisbaar zijn, niet aileen voor het uitvoeren von experiment token. moor ook voor het in bedrijf houden van het station. Zetts zeer eenvoudige token zoots routine inspekties (zoeken noor bv. schimmels. desinfektle> en close-up camera observaties zullen onmisbaar zijn voor een goed beheer von het kostbare station. aepaalde noodrepar01ies kunnen ook met zo een robot uitgevoerd worden. en noormate de robot technologie zich verder ontwikkeld. ZOI het pokket taken dat oon de robot toevertrouwd wordt ongetwijfeld uitgebreid worden. 70t dusver is Europa vooroo in dit aoplikotiegebied. moor de idee wlnt sterk veld in de USA en voorct in Jaoan. ESA wend' nu aon de ontwikkeling von de AMTS (Automated ManipUlation end ironsoortotlon Svstem), zools otgebeetd in fig. 3. De betongrijkste technologlSCr"le uitdogingen bij dit soort

6.4

W. de Ptllltr: De Betekenis van RoboriCQ in de RuimtIVtllUt

rocots zijn een maximum aan unlversee4heid (veelZljOigheld). komoakthetd (vOlume), minImum massa, stcnisc~e en ayncmlsc:"Ie nCU',NKeurlgneld. veliignelo & betrouwboorneid. en kompotibiiitert met de mlKrogrovltert (mlnlmale reactieverstoringen ncar oe omgevlngsstruKtuur),

o

.I Fig. 3

c) Aile andere banen (GEO, Polaire. interplanetaire) Tot dusver bestoon er geen konkrete robot projekten voor onderhoudstoken in daze banen. maar aile ruimteogentschappen bestuderen de mogelijkheden en het is duidelijk dot dit soort robot toepossingen vroeg of loot zullen uitgevoerd worden. De reden doorvoor is evident: deze bonen zijn de facto onbereikboar voor de mens omdot de geschikte vervoersmiddelen niet bestaon. of omdot deze bonen ernstige medische risiko'S inhouden voor astronauten. Van zodra er sproke zot zijn von een behoefte of interesse om onderhoudstaken Ult te voeren in aeze banen. zat dit met robots moeten uitgevoerd worden.

'./ "~

1

'"

~.

.

~'

/

I

/

/

,./

I

Fig. 4

I

/~

i

I

~I

6.5

Voorot de geostationoire boon is interessont. om dot ar nu reeds soroke is von overoezetting op aeze unieke ring. :SA oesruceert momenreel ae reciiseerocorneld von een gerobotiseerd service vOerTuig Oct sotellieten kon '.vegsleoen naor aen kerkhof-boon, mecnanlsche oss/srenrie Kan gaven bij nlet :::;f hOlf uitgaklcote zonneponelen of antennes. ot zelts Oranosrof kon bijvuilen (zie fig. 4). d) Verkenning von planeten en andere hemeUichamen Reizen naarverre hemellicncmen binnen ons zonnestelsel (planeten, manen, kometen. asteroiOen) Zijn bJJna altijd een enkele reis: de enorme ofstanden en het grote oandeet brondstof in het totoolgewicht loren eigenlijk niet onders toe. Bovendien is het weer losgeroken von een ploneet meestoJ niet meer mogelijk met de bestoande propulSie technOlogie. Omwille van de vele onbekenden die zO'n missie eigen zijn. is het gebruik von progrommeerbare en universele machines. robots dus, een evidentie. De verdere exploratie en mogelijk exploitatie van het zonnestefsef zat door gerobOtiseerde onbemonde sondes moeten gebeuren. Mobiliteit is hierbij von cruciool belong, zooat mobiele platforms. uitgerust met instrumenten en later ook met monipulatoren het beeld zullen bepolen von deze toekomstlge missies (Zie flg.S),

Rg.5 Noost gewone verkenningsfoken zullen deze mobiele rObots ook meer ingewikkelde opdrochten krijgen zools het installeren en operationeel maken von wetenschoppelijke instrumenren (seismology. :Jrmosphenscn onderzoek. astronOmie). infrOSTruktuur (antennes, zonneoonelen. return capsule). en bemonsterings coporotuur (grondboor, :n-SlfU aiagnose instrument en). Ook bij eventuele bemande missies in de verre Toekomst zal de oenodigde infrastruktuur (0.0. habitats) op voorhond door eeze (ooots moeten cpgeoouwd worden. De grate technologische uitdoglngen nier zljn ce behoefte can een veel grotere outonomie (typiseh 24 uur) in een niet-gestruktureeroe omgeving. extreem loge masso en volume, optimole energie benutting. en hoge bedrijiszekernelO in een ::eer ogressief milieu (extreme koude en hitte. stot).

6.6

W. de Peuter: De Btttunis van Robotica in tU Ruimtevaart

4. Design Drtvers De specifieke eiSen van een toepassing in de ruimte leiden vook tot oe verkeerde konktusie dot ruimterobotiko jers totoOIS aoort is en weinig te moken heett meT zljn broertjes op oorde. Niers is minder woor: het is inderdoad zo dot het ontwerp von een ruimterobot sterk versc!1ilt von bv. een industriele robot voor auto assemblage of booglossen, maar hun tecnnologische basis kan best gemeenschoppeJijk gehouden worden. In f<e is dit ook noodZakelijk: ruimte robotika is geen voldoend grote markt die toelact de benOdige exoertise en capaciteit (broinwore) uit te bouwen en operctioneel te houcen ! Niettemin moeten bepaalde technologieen uitgediept ot aangepast worden. en dot is 01' zich al een he use tOOk. De accenten bij de ontwikkeling van ruimte robots liggen op: a) Fysische & chemische condities von de (vrije) ruimte: dlt betrett sterke termische cycli (zonnestraling versus koude ochtergrond straling). vacuum (ontgossing von materialen en smeermiddelen). otomaire zuurstof radlkalen (corrosie. degeneratie von smeermiddelen en kunststoffen), en kosmische straling (hoge energie deeltjes). b) Optimisotie van masso. energie verbruik en voorat volume: afgezien van het fait von elke kilo en liter in de ruimte veel geld kosten. bestaan er meestal drempef waarden waarbOven de missie technisch of financieel niet meer haalbaar is. In zulke gevallen kon een signifikante verlaging von gewicht, volume of energieverbruik een sleutelrol vervullen bij de realisotie von het projekt in zijn geheel (miSSion enabling). c) Micro-dynamika: robOts die token uitvoeren aan werkende experimenten in de mikrogroviteit. zullen soms extreme waarden moeten halen in bewegingszuiverheid am het stoorspektrum naar de omgevingsstruktuur binnen toelaatbare perken te houden. Dtt vertaattzich niet aileen naar speciale efectromotoren, transmissies. lagenngen etc. maar ook naar specifieke rege... en planningsalgorttmen en robot kinematlsche koncepten. Bv. redundante vrijheidsgraden kunnen zeer effectief gebruikt worden am het massa-middelpunt van de robot minimaal te laten verschuiven. d) Betrouwbaarheid, robuustheid, veiligheid en onderhoudsvrijheid zijn ospekten die direct te maken hebben met het feit dot een ruimterobot erg ens in een boon een zeer ontoegankelijk iets is, aie bovendien een centrale rol vervult in een (dure) ruimtemiSSie. e) Kompotibiliteit met het ruimteschip en de nuttige last: hoewel emstig werk worat gemaakt van DFA (design for automation) en DFS (design for servicing), ishet tocn dikwijls zo dot de rObot omgeving (satelliet en instrumenten) een vast gegeven is. Daarom kunnen de princ:pes van DFA en DFS dikwijls slechts beperkt toegepast worden en wordt de robot gedeeltelijk ·rond de applikotle- gebouwd. f) Simplifikatie van het rUlmte-segment: door de beperkte beschikbaarheid van

ruimTe gekwclificeerde electronika. vooral computerchios, is het een noodZaak zoveel mogelijk kompJexe regel- en stuurfunkties uit voeren in het grand-segment. Om die reden gebeuren planning. programming, diagnostiek enz. via het grondstation en worot het ruimte--segment uitsluttend belast met de took executie. Dit leidt don revens tot een hoge graad van (korte termijn) autonomie. Echter hoe verder de miSSie van aarde gebaurt, des ta longer moet deze autonomie kunnen duren, en hoe maar complexe token moaten kunnen uitgevoerd worden.

6.7

W. tU PtIllU: De Btttlceni.s van RoboticlJ in dt Rui.InmIIJtut

5. Technotogle behoeften De technologische voorbereiding door het canpassen. uitdiepen van tecnniscne disciplines concentreert lich op de volgence gebleden: a) Concept ontwikketing & kinematische studies: net zoals bij off-shore onderwater rObots. en robots voor nucleaire onderhoudstaken is het van cruciaal belong dot vooraleer het koncept bevroren wordt. goed wordt geanalyseerd of de robot werkeUjk aile situaties aankan, en of het ontwerp geoptimaJiseerd is naar criteria die van direct belong Zljn voor de missie. b) Operationele technieken: door zijn onbereikbaarheid. Zijn veelZijdigheid en zijn centrale rol in een ruimtemissie, is het van groot belong dot de robot kan bestuurd worden in een veetheid von operationele scenarios. Twee extremen hierin zijn: TelemanipuJatie: hierbij genereert een grond operator -via een master armdirect de servo commando's voor de robot. Deze techniek is zeer gebruikelijk in de off-shore en de nucleaire industrie, maar is athankelijk von een snelle communi katie (minder dan O. 1 sek. vertraging). In ruimtevaart is dat niet haalbaar, en de bestaande tijdskompensotie technieken zijn niet robuust genoeg voor een veilig gebruik. Off-line programmering: dit is een zeer goed werkende en veilige techniek, waarbij de applikatie programma's op aarde worden ontwikkeld. en getest en geoptimaliseerd met simulatoren en testbeds. Een probleem hlerbij is dot soms -en vooral gecturende onvoorziene situoties- de took niet geheeJ apriori kan beschreven worden. zooat real-time interaktie slechts beperkt mogelijk is. Om die reden wordt ar gezocht naar nieuwe ofstandsbesturingstechnieken. Ook het gepast integreren van de commando's van wetenschappers. robot operators en astronauten met hun respectievelijk prioriteiten is een belangrijk onderwerp van ontwikkeling en test. Hiertoe behoren ook de verschillende man-machine interface technieken voor zoweJ grond stotions als astronauten. c) Controllers: de mogelijkheden von een robot worden sterk bepaald door zijn controller. Het uitbreiden von het pokket mogelijkheden blijft don ook een continue zorg. Dit leidt ook tot geovanceerder regel- en stuuralgoritmen die geooseerd zjjn op sensor informa1ie. Een onder belangrjjk aspect zijn controller architekturen en ontwikkelmethodieken. Dit is van cruciaal beiang voor het managen von prOjekten met een internationale en zelfs interkontinentale betrokkenheid. d} Callbratte techniekan: omwille von de acceleraties en trillingen tijdens de lancering. het wegvallen van luchtdruk. gravitatie en door de thermiSche cycll die de sotelliet ondergaat, moet de werkomgeving von de robot regetmatig gecalibreerd wordan. Bovendien is een gOadS absolute nauwkeurigheid noadZakelijk am vailig te kunnen werken met off-line programming. Anderzijds is de instrumentatie aan boord gelimlteerd en moeten aus calibratie technieken en strotegien ontwikkeld worden die kunnen volstaan met een beperkt set sensoren. d} Micrcrsensoren: deze zijn nodig voor verschillende taken. Niet aileen vereiSen bepaalde regetalgorithmen feeoback informaTie van exteroceptieve sensoren. maor oak calibratieroutines, telemanipulatie. diagnostiek. velligheidsregelOars (watCh-dogs) en een goede Health & Fauit Management kunnen niet zonder gedegen informatieverzomelaars. Uiteindelijk zal een beperkt set sensoren de nooige informatie moeten verscnatfen voor 01 deze funkties, en moeTen ze Uiterst betroUVtlooar en robuust zijn. Omoat deze sensoren Oikwijls in ot nabij de rooot

6.8

W. de Peute,: De Bettlcenil van Robotica in de RIlimtevaart

grijper moeten geinstalleerd worden. en ze ook efektronische circuits moeten hebben om te kunnen Intertocen met de oestacnoe infrOSiructuur, moeten ze sterl< geminiaturiseerd worden en tevens welnlg vermogen dissiperen wegens de mlncer effecktieve koeling in O-G (geen konvel
g) Design for Automation (DFA) & Design for Servicing (DFS): dft zijn relotief nieuwe begrir'?en in de ruimtevaart. en vragen doorom extra oondocht voor de komende joren. Het uitvoeren von operaties oan wetenschappelijke instrumenten, en het onderhoud plegen aan opparatuur in de ruimte is niet eenvoudig en vergt een gedegen teoretische en praktische onderbouwing die Zich uiteindelijk lOt meeten vertolen ncar nieuwe criteria en richtlijnen bij het ontwerpen van instrumenten en seteillaten. h) Grond gereedschappen voor modellering, simulatie, diognostlek etc. die de robot operator moeren ondersreunen in zijn took. Gelukkig kan hierbij sterk acngeleund worden bij de lopende ontvJikkelingen voor de reeds bestaonde robot appticaties.

5. Trends Naarmate de robot technologie zich verder ontwikkelt, zot de rol von robotlko in de ruimtevoart verder toenemen. En dit zol gebeuren, omdat (1) robotika een poputaire onderwerp van research is. en (2) omdot de toekomstige robot applikaties soortgelijke technologische benoetten hebben 01$ ruimterobots. Dit betrett 0.0. behoefte aan diverse vormen van moolHtelt (geleide, vrije), Iichtgewicht monlouloror arm en (materiOlen. geintegreerae lichtgewicht essen), minimaol volume en massa ook voor de controller. beperkre cutonomle, verscheidenheid aan besturingswijzen, en laog energieverbruik. Door ae synergie tussen ruimrerobots en de (toekomstige generatie aan) dienstverlenenae rObOTS is een hoog ontvJikkelingsniveau & -tempO gegarandeerd. De brace technologlsche Oasis cie den zoi ontstaat, zal de kosten drukken tot een nlveau waorbi; het haalbaor wordt een mini·rooot eon boord van kostbore satellieten te hebben. voor het assisteren tijdens de diagnose van een defeld. en het verhetpen ervan, zetts indlen de nomtnale operaties zulks niet vereisen.

6.9

Space-qualified Optical Memory for the Columbus Pressurized Module

.;:::;;:;

:

-::::

AJgn" '. ilj~~~~l~esFafEitgineer bij het NationatAerospace Labo~toryin . ·· · Tae~e·· f!'Jt~) ·

; :~ !E:i~~!IIr:£~e:~~~::~:5:!:~E!ti~~:i3 T stH(JJ~' venichie D~~:t : r:
:.:

I!ij::\'99¢~etde::::::fu:,:::ide:

penode

.

. ., en ·

198~1984 ele.ctroni~ . tel~co~unicatie

sign~n: Pf9~~p:tg;::~p. :::teeuwarden ~olytechnic ' en .aan .~e Um~~r$itei~'

van .-

~ ::::~~~1!~~;~:lo':~ 1~!rwa:rJ~~ · i;:~!~::r= ::~:: . (p~~{ :§p#.Gi·i:.: J: ~ppJitati8ris Concerriing Optical Me~ia ·

" technologiCis):('en.'(Hs .. hlj I

Laborat~ry.

.. _

and multimedia:

project manager' bij bet National Aero~pace

..

. .': ,:

.

,

7.1

T. Algra: Space..quaJified Optical Memory for the Columbus Pressurized Module

Space-qualified Optical Memory for the Columbus Pressurized Module T. Algra National Aerospace Laboratory NLR, the Netherlands Informatics Division, Electronics Dept. P.O.Box 90502 1006 BM Amsterdam

ABSTRACT This paper describes a space-qualified optical memory device for application in the Columbus Attached Laboratory. The unit is based on a MIL-standard CD-drive leading to a cost-effective design with full compatibility to commercial CD-ROM. The system is extendable to CD-MO (Rewriteable) and CD-I (CD Interactive).

1. INTRODUCTION The Columbus Programme [1] represents Europe's contribution, in cooperation with the U.S.A., Japan and Canada to the International Space Station Freedom. From the end of the century, Freedom, comprising a pennanently manned base for the execution of scientific and technical research tasks, will be orbiting the earth at 400 lan. The Columbus Attached Laboratory (CAL) is a pressurized cylindrical laboratory module, which will be permanently attached to the International Space Station. It has a diameter of about 4.5 m, and a length of 12.8 m and will be used primarily for Material Sciences, Fluid Physics and compatible Life Science Missions. The module will be launched on a dedicated Shuttle flight. It is equipped with one docking port for docking to the Space Station node interface. The CAL as well as the other three pressurized modules of the Space Station-the two U.S. modules and the Japanese Experiment Module (JEM)-are mounted along the flight path of the Space Station, near to the centre of gravity, to preserve an adequate microgravity environment. The CAL is dependent on the Space Station manned base for all primary technical resources like power and communication links, heat rejection, environmental and life support and crew habitation. The equivalent volume of about 50 single racks is dedicated to the accommodation of European and internationally provided payloads, with a maximum payload mass in operational conditions of 10,000 kg. The intended operational lifetime is 30 years and the crew will be relieved every three months. An important element of the CAL will be the Data Management System (OMS). The Columbus mission is primarily to support experiments and operational instruments. Therefore the OMS is designed to provide all necessary resources for an efficient operability of the payloads. In addition, the OMS will contribute to internal communications, control, and monitoring functions. Further, it will provide support to mission management, services, fault isolation and recovery, and access to external com-

7.2

T. Algra: Space-quolijied Optical Memory for the Columbus Pressurized Module

munications. The DMS includes a work station and portable terminals to enable the crew to execute these various system and payload support tasks [2]. The OMS will be equipped with two Non Volatile Memory units (NVM). Based on optical disk technology, these units will provide DMS system software storage (backup) and storage of documentation including text, graphics, and still video pictures. The requirements include [3]: Storage capacity, access time,and data rate: equal to standard CD-ROM drive SCSI interface to the Data Storage Controller Monitoring and control interface Power interface: 123 V DC Cooling method: cold plate Mechanical housing: rack drawer concept Ambient temperature, operating: +7 ... + 370C Pressure, EMI, shock & vibration, radiation, microgravity environment compatibility according applicable CAL equipment requirements. 2. OPTICAL MEMORY FOR SPACE APPLICATIONS Signaal Special Products (SSP) together with NLR developed a design and implementation approach for the NVM which is relatively cost-effective by employing a MIL-standard CD-ROM drive. The SSP/NLR design offers extendability to a multitude of additional capabilities like Rewriteable Optical Storage, CD-Interactive (multimedia), CDROM with audio, and video storage and (de)compression. The NVM will consist of two basic elements: the MIL-CD product of SSP and a drawer including all interfacing with the CAL environment. The MIL-CD is a ruggedized CD-ROM drive meeting all the requirements to withstand the severe environment of military applications [4]. The MILCD consists of an electronic box, containing the SCSI interface, encoding/decoding electronics, servo-electronics and the control processor and, secondly, of a hermetically sealed cartridge which includes the drive, optical head assembly, and the CD disk. The czruidge is exchangeable and removable, and interfaces· to the electronic box through an electrical connector. Fig. 1 gives an impression of the MIL-CD. Fig. 2 depicts a block diagram of the system. The MIL-CD design philosophy involves a high grade of modularity. No major redesign is necessary to adapt the system to future generations of CD devices. Higher disk density and data rates may be sustained by only changing the optical head and the servo processor software. This MIL-CD will be integrated into the NVM drawer as 'CAM-based' equipment. Within the Columbus project the utilization of ~ommercial, Aviation and Military (CAM) sourced and available products is adopted due to the nonavailability of newly developed and built night products. This concerns products (H/W & S/W) newly developed to a customer-controlled specification outside the Columbus programme. The item for which a CAM product is intended to be used is in general of safety criticality category III and/or reliability criticality 2 or 3 (refer to table I). Obviously, the use of CAM products results in a shorter development cycle and hence in a more cost-effective solution. In addition, the cartridge concept applied offers a number of striking advantages as compared to other options. Since all the moving parts are integrated into the cartridge, a maximal level of system reliability can be achieved. The electronic box can be upgraded to rewriteable technology with compatibility between CD-ROM and CD-MO at cartridge level.

7.3

T. Algra: Space-qualijkd Optical Memory for the Columbus Pressurized Module

The drawer contains a DC/DC power converter, mechanical and thermal provisions, a Measurement and Control Interface (MC!). The latter function relays to the OMS essential system parameters like temperature, health, power status, secondary power status. In addition, this interface allows remote operations like onioff switching and reset. Fig. 3 shows the drawer with the various sub-systems inside. As can be observed there is enough space reserved for e.g., cartridge stowage or the accommodation of a second MIL-CD. 3. QUALIFICATION The following NVM models will be produced: Two Development Models (DM), functionally equivalent to the NVM unit except for the remote monitoring and control interface. The DM is used by the DMS integrator to support software development and hardlsoftware integration. Two Engineering Models (EM). The EM is mechanically and electrically equivalent to the NVM flight model, except for the use of space qualified hi-reI components. Its function is to validate the NVM design, to be proved by a number of prequalification tests. These tests include full functional test, performance test, electromagnetic compatibility (EMC), and thermal cycling. A Qualification Model (QM). This model, equivalent to the eventual flight models, will undergo a full qualification programme (EMC, thermal cycling, mechanical shock and vibration, audible noise, pressure, and microgravity). Two Flight Models (FM). To support testing of the various models throughout the project, a computerized unit-tester will be developed. The tester simulates the electrical, mechanical, and thermal CAL environment of the NVM. Dedicated electrical interfaces are included with respect to power, SCSI bus, remote monitoring and control interface. The test software allows the definition of fully automated test sequences. The cartridge contains rotating and translating parts that cause inertia forces during operation. The effects of these forces on the microgravity environment in the CAL will be considered and verified during the design and verification phases of the NVM development. Preliminary calculations based on worst cases indicated that the proposed design is able to meet the specified micro-gravity requirements. However, in the design phase analyses on this subject will be performed. In the analyses the inertia forces will be calculated based on the time-histories of the translational and rotational movements and the corresponding masses and inertia moments. From the analysis results it will be concluded if additional measures are required. to reduce the unit's disturbing forces. In relation to the adopted CAM approach, the following measures are possible: reduction of optical head acceleration andlor of the disk acceleration (both measures influence the access time). Further, the QM will undergo a microgravity environmental compatibility test at a dedicated test facility. For the production of the electronics, space-qualified components will be used as defined in the Columbus Preferred Parts List. Hence for these subsystems no· radiation problems are foreseen. However two full custom CMOS micro-circuits of the MIL-CD are only available with extended temperature range, and in plastic packages. No military equivalents exist. These i.e. 's, manufactured in 1 p. CMOS technology, perform the red-

7.4

T. Algra: Space-quolijied Optical Memory for the Columbus Pressurized Module

book and yellow-book decoding functions. Therefore radiation test programs are foreseen to predict the in-orbit single event upset (SEU) rates. The basic mechanism of SEU is the deposition of charge by the passage of a heavy ion through the sensitive region of a device. If the ion has enough stopping power the charge may be sufficient to change the state of an electrical node or a cell within a device. The best known example of this is the change in state of a memory cell which can be subsequently rewritten. This is known as a soft error which may be defined as a erroneous but correctable logic state. A more dangerous form of upset may occur which is heavy ion induced latch-up. Latch-up is permanently and potentially destructive affecting source/drain diffusions, isolating well and substrate. Latch-up is caused by the p-n-p-n junction path in a CMOS device of which the two parasitic bipolar transistors can be brought in a low-impedance state by a deposited charge [5]. For the circuits concerned, latch-Up protection circuits will be incorporated into the design. 4. CONCLUSIONS The upgrading of a MIL-standard CD-drive to a space-qualified unit is a costeffective and low-risk approach to realize the NVM for the CAL. Moreover, this solution involves full compatibility with standard CD-ROM. This is advantageous for hardware and software development, and for the production and distribution of application data as well. In addition, the concept is extendable to CD-MO (rewriteable), digital video, and CD-I. This opens up possibilities to use multimedia documentation on-board for maintenance and service, refresh training, operational procedures, references, entertainment, and multimedia tele-support [2,6]. Not only the DMS, but also payload facilities may be equipped with such an optical storage device for data-recording and documentation purposes. Although designed for application in the CAL, the unit can be adapted rather easily for accommodation in other flight opportunities. REFERENCES [1] Altmann, G., "Columbus Programme Overiiew with Emphasis on Space Segment Activities", Space Technology, Vol.ll, No.2, pp.65-76, 1991 [2] Sijmonsma, R.M.M., T.Algra, "Multimedia Documentation System for Columbus", Study perfonned by NLR (NIVR contract), NLR CR 90332L, 1991 [3] Non Volatile Memory Specification, Toulouse: Matta Marconi Space, COL-MASPE-2463, 1992 [4] T. Algra, "Optical Media Applications in Aerospace and Specific Military Environments", Study performed by NLR (NIVR contract), NLR CR 92306 L (in Dutch), 1992 [5] Adams, L., "Cosmic ray effects in microelectronics", Microelectronics Journal, Vol. 16, No.2, 1985 [6] Bruyne, H.C.D. de, "Multimedia/Teletraining Experiment", Study performed by CAT Benelux for ESA, ESA Study contract report, ESA-CR(P)-3315, August 1991

7.5

T. Algra: Space-qr«JliJied Optical Memory for the Columbus Pressurized Module

Table 1 Criticality categories as defilU!d for Columbus equipment Safety Category

Description

Cat. I

catastrophic: loss of life, loss of launcher, loss of servicing vehicle

Cat. II

critical: loss of tlight coDfiguration, major injury/illness, major damage

Cat. III

mimr injury/illness, minor damage

Reliability Category

Description

Cat. I

failures of function/item resulting in the loss of the flight configuration

Cat. II

failures of function/item resulting in major degradation of the operational capability

Cat.

m

failures of function/item resulting in minor degradation of the operational capability

Fig. 1 The MIL-CD cartridge and electronic box

7.6

T. Algra: Space..qualijied Optical Memory for the Columbus Pressurized Module

Sealed cartridge ~--------------------------------------------------------~

Disk

;1

'VJf./?

OPtical head ..l...-_ _ _ _ _ _-I' t-...,r-______"--_____

Position mechanism

Pre-amp

servo control

_ _ _ _ _ _ _ It

,- ----------t------------ -----~----I--- ---------

:

:I ; i

'f

I

Servo electronics

I

~i E~:~C.

i ATIP Proc.

EFM data

.....--L--_

Decoding

Circuit

I (optional) I

Man Machine~> interface;

Processor

CO·ROM (yellow book) decoder (Encoder optional)

SCSI interface

To host computer

Fig. 2 Block diagram of the MIL-CD

7.7

T. Algra: Space-qualiJied Optical Memory for the Columbus Pressurized Module

TopVIew'

SCSI 123VOC

I

STAU

FrontVIew'

19

rack

• Cold Ptate Front

Fig. 3 The Columbus Non Volatile Memory (NVM)

7.8

aa..Pfate

A Modular Instrumentation Concept for experiments under microgravity

.

: .:.. :,.: .. . : ..... . : ... .; ... , ... ;.,

.

, .. ;:., •....

.. ... .

: .::.;: .:.. ;.: .

, ..... : ... , ... : ......•.:,. : ........ :,.: :. :.... . ;: .. ; ........

.

.•.• ·cm~ud.Hij-ls )~d~de~;~Sb-u1nen~onJ • • . .•.•... ..•..'." .• '. . .Test ·Society;,·.:: ..

..

.

. . . " ..

8.1

L.J. AarDntm: A modular instnImenIation concept for ttXJ¥rImenIs under mkrogravily

A Modular Instrumentation Concept for Experiments under Microgravity ing. L.J. Aartman For autonomous in-flight experiment control and data handling, a so-called Service Electronics System has been developed, for applications in microgravity research in the field of life science, fluid physics and thermal control. The on-board electronics system is based on a control concept that heavily relies on software for its functions: data acquisition, actuator control, data storage, experiment scenario execution and communication. A personal computer (PC) system is part of the ground-based setup for testing and check-out. System (re)configuration takes place by means of tables: a description of the Most Recent Parameter Base (MRPB) for data handling and the Experiment Action Control Table (BACT) for the execution of a preprogrammed experiment scenario. This basic system is used for several missions: the ESA sounding rocket modules for life science experiments CIS-J (1989), CIS-2 (1990) and CIS-3 (1992), the ESA Wet Satellite Model for fluid physics experiments and the 7\vo Phase Experiment (TFX), a thermal control experiment for ESA for flight in a Get Away Special on the NASA Space Shuttle. To adapt the size of experiment facilities without increasing the size of the Service Electronics Systems in general, and of the on-board hardware in particular, preliminary investigations have been carried out to increase the modularity. The introduction of Sman Experiment Subsystems with a simple power and communication interface was recogniZed as a way that leads towards scalable and better testable facilities. The CIS-4 module (under development for launch late 1993) will be constructed in a very modular way and will contain a Service Electronics System with provisions for distributed control. Single-chip microcontrollers and the software-based control concept mentioned above are the core of these Smart Experiment Subsystems. The flight configuration of CIS-4 will contain 25 of these subsystems: 9 experiment boxes, 9 Sman Thermal Inteiface Plates, 5 in-flight lxg centrifuges for reference experiments and 2 videorecorders. They are physically distributed over three work modules, each with its own RS-485 based local network for serial data communication. The subsystems will fulfill distributed control functions in a complex experiment facility: a hierarchy through a facility controller unit allows operator interaction from a connected computer into the facility and its subsystems or to an individual subsystem. .

8.2

L.J. Aanman: A modular inslrU1rlel1tifdon concept for experiments under microgravity

The product assurance requirements for sounding rocket electronics are less stringent than deep-space qualified electronics: less paperwork, and industrial or military qualified electronics parts are sufficient. Therefore, these products may utilize advanced technologies and require only short development times (several months up to two years)

8.3

I.M.W.M.lanssen: Nawoord

Postface J.M.W.M. Janssen

With pride, the IEEE Sudent Branch Eindhoven presented its third symposium, this time on ''Electrical Engineering in Space Applications". With this title we hoped to achieve two goals that were at least paradoxical: we wished to invite lecturers who could supply the visitors with sufficient and relevant information whereas on the other hand we wished to have a subject in which many disciplines of the broad area of electrical engineering are represented, since the IEEE basically is, as its name implies, an organisation for all electrical and electronics engineers. We were well aware of the fact that this was an ambitious goal, but still we hope that we achieved it. Furthermore, we hope that we organised the symposium in a way that matches the policy of the Student Branch board: with clarity and from a rather idealistic point of view, since the title that was chosen is not the most attractive one from a commercial point of view. The IEEE Student Branch Eindhoven has proved to be a very active group of student members during the last couple of years. This year, however, has been a year of recovery from these events. There were problems concerning the activities that had to be solved and therefore we restricted ourselves to two major events: this symposium and a study tour to Paris. As Chairman of the Student Branch board, I would like to thank you for your attendance at this symposium and I hope that you had a day of facts and pleasure. J.M.W.M. Janssen Chairman IEEE Student Branch Eindhoven

9.1

Bestuur IEEE Symco '93

Bestuur IEEE Symco'93 Voorzitter:

Chris Lennartz

Secretaris:

Marcel Breuers

Penningmeester:

Wessel Weijenberg Jeroen Rutten

Public Relations:

Suzanne Ofner

Beleidsmedewerkers: Joop Janssen Jeroen Sluyter

Bestuur IEEE Symco'93. V.l.n.r. Wessel Weyenberg, Suzanne Ofner, Chris Lennanz, Jeroen Rutten, Joop Janssen, Marcel Breuers.

10.1

Lijst van adverteerders

Lijst van adverteerders bIz.

2.0

Alcatel

3.0

ECN

4.0

P1T Research

6.0

Philips

8.0

SEP

11.1

Commissie van aanbeveling

Commissie van aanbeveling • prof.dr.ir. G. Bmssaard

Hoogleraar Telecommunicatie Technische Universiteit Eindhoven

• prof.dr. J.H. van Lint

Rector Magnificus Technische Universiteit Eindhoven

• prof.ir. M. Stevens

Hoogleraar Digitale Systemen Technische Universiteit Eindhoven

• prof. dr. L.H.T. Rietjens

Decaan Faculteit Elektrotechniek Technische Universiteit Eindhoven

• prof.ir. M. Antal

Algemeen directeur PIT Research

• prof. C.W. Turner

IEEE Regional Director Region 8

12.1

Woord van dank

Woord van dank De IEEE Symposiumcommissie '93 wil haar dank betuigen aan de volgende personen: - faculteit Elektrotechniek - dr.ir. J.M. Wetzer - ire R. Suttels - ir. B. Ambrosius - ire R. Hekmat - D. O'Sullivan, B.E. - ir. W. de Peuter - ire T. Algra - ing. L. Aartman - prof. C.W. Turner - prof.dr.ir. G. Brussaard - prof.ir. M. Stevens - prof.dr. J.H. van Lint - prof. dr. L.H.T. Rietjens - prof.ir. M. Antal - ire D. Poortvliet - prof. D.M. van Dommelen - prof.dr.ir. P. Eykboff - prof.dr.ing. J. Jess - dr. M"E.J. Jeuken - prof.dr.ir. P.C.T. van der Laan - Congrescentrum Orange, i.e. dhr. en mvr. G, Klinckenberg - Drukkerij Gemert B.V., i.e. R. van Helvoort Verder bedanken wij: Prof.ir. K.F. Wakker, ire I. Braga, ire C.AF.J. Wijffels, ire F.M. Buijs, J.M. van der Kamp, dhr. Kok, drs. J. Heijn, ire C.J. Nelemans, drs. R.M ten Cate-Dhont, mevr. G.M. Broere, drs. J. Huiskamp, bestuur IEEE Student Branch Eindhoven, de oud-voorzitters IEEE Student Branch Eindhoven en tenslotte iedereen die een positieve bijdrage geleverd hebben aan dit symposium.

13.1

Appendix to Lecture 8

A MODULAR INSTRUMENTATION CONCEPT FOR EXPERIMENTS UNDER MICROGRAVITY ing. L.J. Aartman Informatics Division, Electronics Department National Aerospace Laboratory NLR

1. Introduction A multi-user facility for microgravity research in the field of biology and biotechnology was developed for the European Space Agency (ESA). This system, built around the sounding rocket module Cells In Space (CIS-I), carried biological experiments into 7 minutes of microgravity on the Swedish MASER-3 in April 1989. [1] The experiments are contained in Experiment Boxes with several units for sample and liquid management (e.g. mixing, fixing). Each box is provided with experiment-dedicated provisions and local electronics based on a single chip microcontroller [2]. The CIS-1 facility was designed to accommodate four experiment boxes, i. e to provide power and provide for thermal conditioning by means of Peltier-based Thermal Interface Plates (TIPs). Local sensor signals, command and monitor lines are interfaced to an onboard controller that is part of the Service Electronics System. Late insert and early recovery of experiment samples are among the requirements of biological experiments that can be met with sounding rockets. System reconfigurability was taken into account to allow for future flight opportunities with new experiment types. A simple and clear experiment box interface was defmed in an early stage of the development. Hardware has been kept simple and programmable where applicable. The hardware of the on-board facility controller could be based on industrial parts, which are capable of surviving the environmental stress in the sounding rocket. Sounding rocket developments are characterized by a minimum of paperwork and by the possibility of using advanced (electronic) technologies.

AS.1

L.J. Aartman: A modular instrumentation concept for experiments under microgravity - Appendix

2. Service electronics system concept The basic control concept has been developed for the CIS-l Service Electronics System and contains four levels of control, viz: - one (or more) experiment subsystems, - an on-board controller, - EOSE with a data-link to the on-board controller, - an operator. Besides a data-link:, an external power link is connected to the module, but during flights power is supplied by a set of batteries. To achieve a high degree of flexibility, the control concept heavily relies on software and control tables for its functionality. A table-oriented approach was chosen both for data handling (input from sensors and output to actuators) and for execution of experiment scenarios. To allow the operator to change the configuration, it was decided that the preparation of the control tables for the on-board system would take place at the operators' PC. After the information in the tables has been verified they are downloaded to the on-board controller to become the current control configuration. From then on the experiment is ready for autonomous execution that will start when the connection is broken (during launch). Measurement data is collected in on-board memory during experiment execution and after the payload has returned, the data connection is re-established and the results are retrieved. Data can be replayed or converted into a format for further processing.

3. Service Electronics Box SEBCIS A facility controller 'Service Electronics Box for Cells In Space (SEBCIS)' was introduced for the following major functions: - experiment box interface, power and data handling provisions for one experiment box; thermal control of the experiment boxes, with both heating and cooling capabilities; data acquisition, (housekeeping)sensors; data storage; interface to the payload telemetry system; execution of a preprogrammed scenario; power management. The hardware of the SEBCIS is contained in a small box and consists of several printed circuit board assemblies. All these assemblies in SEBCIS are controlled over an internal bus by a microprocessor based System Processor Unit (SPU) , that provides for serial data communication (RS-422) with a PC. Data is stored on- board in a Promcorder, a solid-state recorder equipped with 128 kbyte EEPROM as a backup for telemetry.

AB.2

L.l. Aartman: A modular instrumentation concept for experiments under microgravity Appendix

4. CIS-J embedded software overview The actual internal status is present in the Most Recent Parameter Base (MRPB) , a buffer that contains all relevant actual system parameters. The embedded software, in which the MRPB is a key element, is executed by the microprocessor on the SPU in SEBCIS. The following functions are present: Acquisition. Sensors and registers are scanned and stored in the MRPB. Control. Control algorithms are implemented a.o. for thermal controL As a system philosophy, control only operates on the MRPB. Output. Commands are sent from the MRPB to the hardware(registers). Storage. A special output process collects data marked for storage. Planning accesses the EACT and transfers new settings to the MRPB to change the control flow. Communication. Bidirectional interface to a 'higher' leveL Commands and configuration tables are received while MRPB data is sent 'up'. As long as the (umbilical) connection with the PC is present a direct interface is available to the operator. The internal status of the controller is copied to the PC. After interpretation, the operator may want to influence the on-board proces and submits a command to the controller. The result of this command (including the command itself) is presented to the operator when the next copy of the MRPB is received. Since most phenomena are slowly changing (e.g. temperatures) the controller uses a one-second internal timebase, which results in an update rate of one MRPB copy per second. This MRPB transfer philosophy has proven to be very adequate during system integration.

5. Data handling with a Most Recent Parameter Base (MRPB) The MRPB can be considered as a dynamic information set: each entry contains a parameter. The MRPB is described by an MRPB Descriptor Table (MDT), a static part that is used to control the on-board data handling. Each parameter is described with respect to its: - Headertype, analog or digital parameter; - Source, in the case of an analog parameter an analog-to-digital Conversion Message has to be specified; - Gain and Offset to convert an analog parameter linearly into physical units; - Destination(s) , each process that outputs data has a column entry in this table, e.g. experiment commands (general output), telemetry and on-board storage. Part of the CIS-l MDT is given in table 1 as an example. Labels are used to present system locations: Mxxxx are MRPB locations and Exxxx labels represent hardware 110addresses. A8.3


6. Scenario execution from an Experiment Action Control Table (EACT) The pre-programmed course of experiments is described in the Experiment Action Control Table (EACT). The EACT contains the scenario as a function of time and flight phase and is accessed once per second. Flight phases can have different properties. The duration of some of them are controlled by a time counter while other phase transitions are event-driven. The EACT is used to change setpoints, phase durations, etc. After initialization, default values are transferred from the EACT into the MRPB. Part of the CIS-l EACT is given in table 2 as an example.

7. Electrical Ground Support Equipment (EGSE): functions and architecture Since many parameters were not known until a short time before launch, the on-board control hardware was kept simple and easily accessible to an on-board processor for autonomous operation and for interaction with an operator at the Electrical Ground Support Equipment (EGSE). Testability by means of full interaction is of importance during subsystem test, system integration and prelaunch operations. PC's in the EGSE are used for mission preparation and utilization as long as the connection to the on-board facility controller is present. In the mission preparation stage, both the MDT and the EACT are prepared at the PC with an ASCII text editor. Parser programs are available to verify the contents. After format conversion both tables can be downloaded to the on-board controller. In the utilization stage, the MRPB is received after (part of the) on-board MRPB contents has changed. Various ways of presentation are available (graphics, text and alphanumeric). Commands can be entered at the keyboard and checked before they are transmitted. Basic communication commands are available for: - reading MRPB locations; - writing to MRPB locations; - transferring the whole on-board MRPB to the PC; transferring (downloading) the contents of the PC-tables into the on-board Electrically Erasable Programmable Read-Only Memory (EEPROM); - transferring the contents of the recorded data to a file on the PC-disk. In addition local commands are available for the selection of display pages, for control of AS.4 j

j

i

/

i

I

i


data logging on the PC disk and to replay recorded data. Another function of the EGSE software is to provide for quick look data presentation and analysis.

8. Applications The hardware and software concepts described have been used for the first time in CIS-l and appeared very useful in all phases of the campaign. After some minor modifications and improvements the system had a re-flight as CIS-2 on the MASER-4 [1]. Using existing CIS hardware and software elements, an experiment control system was composed for a fluid physics experiment, performed during parabolic flights in an aircraft. The instrumentation of this Wet Satellite Model (WSM) served as a breadboard for a more complex instrument in a sounding rocket that flew on MASER-5 in 1992. Linear accelerations were measured and stored in free-flying conditions. After flight the satellite model was connected to a PC for data retrievaL A replay of the data gives a quick-look presentation of the behaviour. Practical re-use has .been successfully demonstrated during two flight campaigns. The control concept is also applied in the Two Phase eXperiment (TPX) , a thermal experiment currently under development for ESA [3]. A series of experiments will be executed in a Get Away Special on a Shuttle Flight (scheduled for november '93). Again, autonomous execution is required after 'power on' and data measurements are stored in on-board memory for retrieval after the flight. Although the controller hardware is different (VME-based 68020 microprocessor, different types of sensors and actuators), the basic concept is the same. Full interaction during extensive tests is possible via the datalink. The EACT features are enhanced to allow jumps and tests on conditions. All systems described so far are based on on-board controller hardware that interfaces directly to the actuators and sensors. Only one level of data communication is involved viz. between the ground-based PC and the on-board controller. On the sounding rocket module CIS-3, a new type of experiment subsystem was introduced: the Electropulser Facility (EPF). To enhance its testability, the EPF has been equipped with a serial data-link (for interfacing with a PC), which is also used in-flight to interface with the facility controller. Besides a data link:, switched 28 VDC power is provided. To accommodate the EPF in CIS, the control software concept has been modified to cope with subsystem interfacing through a data-link. These new requirements have resulted in the following changes: multiple-paged MRPB, keep related parameters together in a page; MRPB page handling (e.g. communication) with Page Update Flags (PUFs). Dedicated EPF interface hardware was introduced, equipped with a single-chip microcontroller for data and power handling. The SPU has been upgraded to a 32-bit microprocessor (MC68020) and the embedded software is re-implemented in Pascal and uses a real-time operating kernel (pSOS). CIS-3 flew on MASER-S in April 1992. A8.S


9. CIS-4 Service Electronics System For the CIS-4 sounding rocket module (scheduled for launch November 1993), new facility requirements have been defined: more experiment boxes, in-flight lxg centrifuges and video observation. Additional requirements like improved testability, simplified calibration procedures and mass/volume constraints have resulted in the continuation of the modular approach set for CIS-3 and further elaborated during related (nationally funded) technology studies. Modularity is a design driver to manage the increasing complexity of a module like CIS4. The flight system will consist of one core module and three work modules. Each work module contains three experiment boxes (some containing centrifuges or video cameras) with individual thennal control. The CIS-4 experiment subsystems are implemented as so-called smart subsystems, with a clean data and power interface. The CIS-4 flight system will contain 25 and the 1xg set-up will contain 10 of these subsystems.

10. General Smart subsystem concept Smart (microprocessor-based) subsystems can autonomously perform a local function in an experiment facility. Communication with a controller higher in the hierarchy is an important feature. A reduction in interface load and cabling is a clear advantage and a facility can easily be customized by adding or changing subsystems, without the need to physically modify the facility controller. Power requirements are met by converting a single 28 V facility power with local DC-DC converters. Other features are: - testability of the subsystems: a PC can be connected to the subsystem and full interaction is possible, in a similar manner as described for the facility, - simulation of the subsystem in the facility, a computer program can replace a physical subsystem to simulate the interaction with the facility (e.g. for demonstration purposes or feasibility studies). The first realization of a smart subsystem is the Smart Thermal Interface Plate (STIP) , a thermal conditioning unit for experiment boxes as applied in CIS. The STIP is equipped with a set of temperature sensors and Peltier elements and is built around a single-chip microcontroller (MC68HC11). Serial data communication is according to RS-485 and is used to monitor the behaviour of the STIP and to allow for operator or facility controller intervention. The local control concept of the STIP is similar to the central facility control concept and contains a local MRPB. The STIP can be equipped with several advanced thermal control algorithms. Sensor corrections are performed locally based on previous performed subsystem calibrations.

A8.6


11. Overview distributed control concept To improve the modularity, testability and accessibility of information, smart subsystems fulfill distributed control functions in a complex experiment facility like CIS-4: a hierarchy through a central control unit allows complete interaction from a connected computer into the facility and its subsystems. Each subsystem is responsible for the execution of its local task and is monitored by the on-board facility controller. Observation of its behavior is possible and if required, an operator can intervene by sending commands to the controller. If only local action is required, the commands directly related to a subsystem are passed down through the controller. Three types of MRPBs are now distinguished: - local MRPBs in the subsystems; - the MRPB in the on-board facility controller; - a copy of the facility MRPB at the operators side. A change in a local MRPB of a subsystem (e. g. change of internal status) is flagged and can be transferred to the MRPB of the facility. If connected, this information is sent to the operators PC, where a copy of the on-board MRPB is maintained. This information flow forms also a feedback for commands entered at the highest level. The receipt of the command cannot be observed until the local MRPB contents is available to the operator.

12. SEBCIS-4 Compared to the previous CIS missions, the SEBCIS for CIS-4 is reduced in size, thanks to the introduction of serial experiment interfaces for communication (RS-485) and (+ 5V and + 28V) power distribution with the experiment subsystems. These interfaces are equipped with a local communication controller that maintains the communication protocol between the experiment subsystems and the facility controller.

13. Concluding remarks During several campaigns the CIS control concept has proven to be very useful. The table-oriented approach introduced sufficient flexibility during integration and operations; software changes have not been necessary after acceptance of the subsystems. Using the CIS concept as a reference model, it has been possible in some cases to build an experiment control system in a short time using existing hardware and software elements. In other cases, different environmental requirements have resulted in a different hardware implementation. AB.7

LJ. Aartman: A nwdular instrumentation concept for experiments under microgravity ~ Appendix

The distributed control approach, presently implemented into CIS-4, is very promising and leads towards scalable facilities. A study to evaluate the application of this technology for flight on the Space Shuttle or the MIR is scheduled for 1993. The control concept described is not limited to the field of experiment control in microgravity applications, but could be baseline for other applications (e.g. in the process industry) as well.

14. Acknowledgements The Cells In Space and Wet Satellite Model sounding rocket facilities are prepared and operated by Fokker Space & Systems (FSS) for the European Space Agency (ESA) under contracts with the Swedish Space Corporation (SSC). The Center for construction and Mechatronics (CCM) is subcontractor for CIS. The Electropulser on CIS-3 was designed and built by the Central Research Workshops of the Ersamus University Rotterdam, as a subcontractor to FSS. Within the framework of the Technology Demonstration Program (TDP-l) the Two Phase Experiment (TPX) is prepared by the National Aerospace Laboratory (NLR) under contract with ESA. Technology development for the CIS facility is partially financed by the Netherlands Agency for Aerospace Programs (NIVR).

15. References [1]

R. Huijser, L. Aartman and H. Willemsen, "Cells In Space: Sounding rocket facilities for cell biology and biotechnology in microgravity" in Proceedings of the Fourth European Symposium on Life Sciences Research in Space, Trieste, Italy, May 28 - June 1 1990, pp. 455-466.

[2]

Aartman, L.J., Huijser, R.H., "Towards distributed control systems for experiments under microgravity" , NLR TP 92027, IEEE Instrumentation/Measurement Technology Conference, New York, May 1214, 1992.

[3]

A.A.M. Delil, J.F. Heemskerk and W. Supper, "TPX: Two-Phase Experiment for Get Away Special G-557", SAE 911521, Thermal Control Technology Session of the 21st International Conference on Environmental Systems, San Fransisco, USA, July 15-18, 1991.

[4]

J.P.B. Vreeburg, "Investigations preparatory to the Wet Satellite Model Experiment" , International Symposium on Hydromechanics and Heat/Mass Transfer in Microgravity, Perm-Moscow, 6-12 july, 1991.

A8.8

A Modular Instrumentation Concept for Experiments under Microgravity

'A MODULAR INSTRUMENTATION CONCEPT FOR EXPERIMENTS UNDER MICROGRAVITY

L.J. "Aartman National Aerospace Laboratory NLR Informatics Division/ Electronics Department

Mod-ins. sht 22 April 1993

1

A Modular Instrumentation Concept for Experiments under M icrogravity

Contents Introduction CIS-1 Sounding Rocket Module Basic Service Electronics System architecture CIS-1 hardware interfaces Service Electronics Box SEBelS CIS-1 embedded software overview Data handling with an MRPB Scenario execution from an EACT EGSE functions and architecture Applications

CIS-4 Service Electronics System General Smart subsystem concept Overview distributed control concept CIS-4 Interfaces Concluding remarks Mod-ins.sht 22 April 1993

2


INTRODUCTION

Objective Provide a multi-user sounding rocket facility for cell biology experiments

System requirements Automated experiment execution; Several experiments (multi-user); Active temperature control; Ground reference 1 xg set-up;

Development features Short development time, 'Low-budget' w.r.t. large (deep)space programs, Limited paper work, Application of advanced technologies.

Result The CIS-1 module made a successful flight on MASER 3 on April 1 0, 1989.

Mod-ins.sht 22 April 1993

3


~

e

,)t'

CIS-1 SOUNDING ROCKET MODULE EXPERIMENT BOX ON THERMAL INTERFACE PLATE

DECK

SKIN - - - - - t o....

BATTERY BOX (1 OF 2)

~. ~


Fokker Space & Systems

4


BASIC SERVICE ELECTRONICS SYSTEM ARCHITECTURE

EGSE

on-board

control tables

control tables

-

c::

data link operator

~


_____________

.-E

ACT

-

SNS

c.. x

Q)

onboard data storage

measurement data ~~

Q)

PWR

______________

Q)

~

~m

5

~@\

s:o ~

:r !" VI

~

"-> "->

»

»

~ IQ IQ

W

.J..

.J-

EXPT BOX #1

EXPT BOX #2

EXPT BOX #3

EXPT BOX #4

TIP #1 TABLE

TIP #2

TIP #3

TIP #4

I

J. BATTERY BOX #1

SEBCIS ..-----

T M

FEEDTROUGH HARNESS '--~-

II

s

E

~

e ~

0

(')

en I -Jr.

I

» ::D

••

0

~

II

DECK

0)

...

I~.l

~

FLIGHT DIRECTION

» ::D m

J.

Z -I

BATTERY BOX #2

m

::D

UMBILICAL CONNECTOR

» " (') m

~

;i;

on

a;

en

a.

c

~


SERVICE ELECTRONICS BOX SEBCIS

IHK2

I SEBCIS

HOUSEKEEPING HK1

I J

I I I I RS422 SERIAL I OATALINK I WITH EGSE PCI

SEBCIS PROCESSOR UNIT

I

I

I I I I I I I 28VDC FROM I INTERNAL PWR BATIERIES OR I EXTERNAL I POWERSUPPLY I MODULE I POWER ~I

!coL

IEXIF4 I EXIF3 IEXIF2 EXPERIMENT INTERFACE EXIF1

SUBSYSTEM

en

w

z

ITIPIF1 TIP INTERFACE TIPIF2

l I ~

I I

p= r-+---

rr-

l-

l-

0

z

SERVICE SMIF

u

MODULE INTERFACE

0

en :::>

CD

I

I I

l-

as co

128 KByte EEPROM

_ _ _ _ _ _ _ _ _ _ _ _ _ _ ..J


LIFT 0 FF ZERO- G

36 DIGITAl

32ANALOO

'PROMCORDER'

PT1 000 SENSORS

I I I

M

+ en :::> r::c 0

4ANALOGIN 4 DIGITALiN 4 DIGITALOUT 4POWER 1 ANALOGIN 1 DIGI TAL IN 1 DIGITALOUT 1 POWER PELTI ERCONTROL

I

....J

cr:

1

,I

~

I-

POWER VOLTAGES CURRENTS FACILI TV TEMPERATURES

TMSE TELE· METRY SUB

ENCODER

7


CIS-1 EMBEDDED SOFTWARE OVERVIEW Basic tasks: • data acquisition; • control; • actuator output; • scenarion execution/planning; • communication; • storage. actuators

sensors

RS422

EEPROM data EACT


MRPB CM = A = P = CL = ·0 = R =

Communication Acquistion Planning Control Output Recording

8


DATA HANDLING WITH AN MRPB Most Recent Parameter Base

=

dynamic

MRPB description table = static

Example

Table J

Part a/the CIS-J MDT

##-- MRPB table ____ +description ___ +________ + ____ (file). +__________ +__________ +________ +________ +______ _

I

# LAB # EL

0

___

I I

I I

I I I ~:~~i I

lHEAI MESSA 10_ I GAIN OFFSET 10_ PROM TMSE DER GE SOURCE DEST CORDER #------+---+--------+--------+----- ----+----------+--------+--------+-------

:~;TS

I I

I

0 I ESRSTS #------+---+--------+--------+----------+----------+--------+--------+------# Housekeeping Current/Voltages .. #------+---+--------+--------+----------+----------+--------+--------+------MI28H A EAHK01 EAHHK1 +001.250 +000.000 EPCYT MI28L EPCYT MV28H A EAHK02 EAHHK1 +003.492 +000.000 EPCYT MV28l EPCWT MIsH A EAHK03 EAHHK1 +001.250 +000.000 EPCWT MIsL EPCWT MVsH A EAHK04 EAHHK1 +000.624 +000.000 EPCWT MVsl EPCWT #------+---+--------+--------+----------+----------+--------+--------+-------

CM = A= p= CL = 0= R=

Communication Acquistion Planning Control Output Recording


9


.,."

9~

SCENARIO EXECUTION FROM AN EACT Automation for Autonomous operation

Example

TalJle 2

ParI (~f'lhe

CIS-J EAC1'

# EACT description table (file). #-------+-------+----------+------------+-----------------------------------# PHASE I TIME I FUNCTION I EXPERIMENT , DATA #-------+-------+----------+------------+-----------------------------------PRE 00000 TMIN 1 +016.000 # minimum temp Algae A PRE 00000 TMAX 1 +016.500 # maximum temp Algae A PRE 00000 TMIN 2 +036.500 # mininun tefll> l~ocytes PRE 00000 TMAX 2 +037.500 # maximum tefll> l~ocytes PRE 00000 COMMAND 1 O#lamps off/start off/mode TEST PRE 00000 COMMAND 4 0 #-------+-------+----------+------------+-----------------------------------LAUNCH 00000 DURATION 64 # launch phase duration LAUNCH 00000 COMMAND 1 4 # l amp on, start off ,mode -TEST LAUNCH 00000 COMMAND 3 4 #-------+-------+----------+------------+-----------------------------------HICROG 00000 DURATION 425 # microg phase duration MICROG 00000 COMMAND 1 4 MtCROG 00000 COMMAND 2 a #-------+-------+----------+-------- ---+-----------------------------------DESCENT 00000 DURATION 106 # descent phase duration DESCENT 00000 THIN 3 +005.000 # mininun tefll> Algae B DESCENT 00000 TMAX 3 +020.000 # maximum tefll> Algae B #-------+-------+----------+------------+------------------------------------

I I

I I

I I

I

Communication for Interactive operation

Read from MRPB Write into MRPB Transfer MRPB pages

Mod-jns.sht 22 April 1993

10


EGSE FUNCTIONS AND ARCHITECTURE

,

I

~

I

~s

C\I I ~

·w

...J

1I00f.DOC

(/')

Zw a:...J

a: ~.Q.. I

(!)

~8 wo

W

en

W C)

«

X

o

~ ...J

« (.)

o::~ wa.. o 3: a.. x 0::::> o a.. Cf.)

::i

as

:e ::l

~~J

I

I "I:t

J

W

F-

0

C? ~ N

a::::J WI

Z wa..

tu

1= o tl:i ~O::(J)

....

Mod-ins. sht 22 April 1993

0:>w"'" a.. o 3: a.. ....xO:::J a.. (J)

w

en

J

Cf.)

0

oal xW

.,...Cf.)

~1

(!)

w C)

X

"

or-

11


APPLICATIONS CIS-2 Reflight of CIS-1 (1990) CIS-3 Upgrade from CIS-2 with new experiments (1992) WSM-BB Wet Satellite Model (bread-board) Re-use of CIS-1 technology for data handling of an instrumented free-floating liquid tank WSM-EB Wet Satellite Model (Ejectable Ballistometer) Re-use of CIS-3 technology for data handling of an instrumented small satellite. TPX Two Phase Experiment • VME-based data handling system • thermal experiment • GAS-container on the Shuttle • ESA Technology Demonstration Program


12


CIS-4 MODULE LAYOUT

CIS-4 Work Module 3

CIS-4 Work Module 2

CIS-4 Work Module 1

SEBCIS4

CIS-4 Core Module

Fokker Space & Systems


13


CIS-4 SERVICE ELECTRONICS SYSTEM

•

Modularity Easy reconfiguration Step-by-step upgrading

•

Clean interfaces Power (28 and 5 V) Data according to RS-485

•

Distributed control Enhanced testability No end-to-end calibrations required

I I I

I

I

I I

I I

I ,1--

i

----::;. :. r I

.:

r.f.~i--T

i l

I I

BATT3

r--------------------------------, : CIS WORKING MODULE 3 :

-1

I: I

I

I I.L:::::::::::::::::!..lt!:_'.'_1 I

I I

I I

I I I

I I I I I I I

I -----------~EAI~T~~------------l

I I l ________________________________ ~

~----~--iJTU

I

I I Il _______________________________ I

J

TO MASER-6 SERVICE MOOUI..E TEl-fIETAY LIFT-OFF ZEAOGE£


14


l1'"

e ~

GENERAL SMART SUBSYSTEM CONCEPT • • •

MRPB-based data handling Local microcontroller access through communication

STIP ELECTRONICS CONTROL UNIT ~----------------------T------------------' I I

3 EXTERNAL TEMP. SENSOR

~

I I

I

: I

I I

TEMP 1

,

TEMP 2

MICRO CONTROL

1

IT1 I, II

~: 1....-: , I I I

I

...------,:

68HC11

PELTIER I ELEMENT : ARRAY I I

, I I

I I I

I I I

I

'----~

PELTIER POWER CONVERTER

I

I I I

t

I I

: I

:

:

I ,

I I

,

I L______________________ ~I __________________ • I

I

I

I

I,

I,

"'I"

!! ,

, I

I

: , I I I I

I

n

I

I

f

, I I

t I

I

:

STIP

:

,

IL~_~~~_~~!~~~~~_J

eNTAL

,

II.

:

il

:

en

~,..

=~ x

....

~~

I I I I

,

: I

ISTIP SOFTWARE ARCHITECTURE : ________________ J ~-------------------


15

tq)\

s:

o

9-

S'

!Ii

III

;r 1'0.) 1'0.)

>

"CI

CM A CL

3: CD

CD

c.:>

0

----------------

= = = =

communication acquisition control output

1- 1

- - -

- - -

- -

- - -

Isensor I

-

- - -

- - -,

Iactuator

II

I I I , 1 '" I I

.9 as

•• _._-

I

,

O~

,--------------------, I I i I

me

1

!Subsystem L _ _ _ _ _ _#= _ _1 ____________ ,

Q;

o

0 Oen

_I

c.. I

I I

I

I sensor

sensor

actuator

I I

I I I I

~__

ground based personal computer

....),

Om -u~ ~m

z-;

_ ________________ _

facility controller

local MPRB #= 2 ,subsystem #= 2 L__ _________ _ I I

subsystems

en

Z:D

0 0 0

1

I I I I L ___________________ 1 I I

< m :JJ < m ~

---~ ~--------------------------~

!

0

, I I ~

,~ _ _ _ _ 'iij

::IJ

0

r


SEBCIS-4 r

I PRt5

-

~I ~I ~I

I PAt 1

POWER OISTAIB. UNIT PR AS

P-

I I I I

MPR CONTROL.

-.

I I I I

I

i

~ ~

11 PR13 JI PR141 ~

en t\I

[/ PRtel

a::

0

'"lit

I I

HOUSE KEEPING

I I I

fS

f

SYSTEM PAOCESSOf 0 UNIT SPU-2

r--

.....

f

I--

0

~

0

M

"'"-

I I I I I I I I

PR CS

~ PR PI

I I I I

P

HOUSE KEEPING '2

,

P

I

I

I I

,.-

It-

SERVICE MODULE INTERFACE ~ SMIF-2

~

~

I I I

PR Gl

I I I

PR 02

I I

SEXIFt 0 P

0

r-

P

~ r0-

t I I I

I I

PR Ht

I

I

1 I

SEXIF2

I I I

0

0

r-

t I t

P

FI

~

TO CIS WORKING MODULE 2

r-

I I

I

PR H2

I I I I

I

I I

SEXIF3

I I I I I I I I


PROM CORDER 512KB MARK-2

1

0

0

.r--

p

P

~


L.-

I

0

PR H3

I

I I

~

1

SEXIF .. 0

0

p

p

PR Nt

r~.

r-

PR H4

I I I I I

-----------------------------~


17

A Modular Instrumentation Concept for Experiments under Microgravitv

~

e ~

CONCLUDING REMARKS

•

the CIS control concept has proven to be very useful

•

re-use of hardware and software elements has been demonstrated

•

the CIS-4 distributed control system is candidate for future payloads

•

the applicability is not limited to space ACKNOWLEDGEMENTS

The Cells In Space and Wet Satellite Model sounding rocket facilities are prepared and operated by Fokker Space & Systems (FSS) for the European Space . Agency (ESA) under contracts with the Swedish Space Corporation (SSC). The Center for Construction and Mechatronics (CCM) is subcontractor for CIS. Technology development for the facilities is partially funded by the Netherlands Agency for Aeropsace Programs (NIVR).


18

ELEKTROTECHNIEK RUIMTEV AART. IEEE Student Branch Eindhoven

Recommend Documents