WHY VIKRAM FAILED

THE MISHAP TOOK PLACE DUR­ING A PRO­GRAMMED CHANGEOVER PHASE OF THE LAN­DER’S DE­SCENT ON THE MOON—A MA­NOEU­VRE HITH­ERTO UNTESTED BY ISRO AND FRAUGHT WITH COUNT­LESS POS­SI­BLE VARIATIONS

India Today - - CONTENT - By RAJ CHEN­GAPPA in Ben­galuru

Chan­drayaan 2: The in­side story of what went wrong with the pre­ci­sion moon land­ing, planned to a nicety

There was plenty rid­ing on Vikram, In­dia’s squat moon lan­der, be­fore its pre­ci­sion land­ing on the lu­nar sur­face on Septem­ber 7 went mys­te­ri­ously awry. There was Pragyan, the com­pact rover with Ashoka em­blems em­bossed on its wheels, to leave a per­ma­nent foot­print of In­dia’s pres­ence on the moon. If the lan­der, which was named af­ter the fa­ther of the coun­try’s space pro­gramme, Vikram Sarab­hai, had suc­ceeded, it would have been a fit­ting fi­nale to the 50th an­niver­sary cel­e­bra­tions of the In­dian Space Re­search Or­gan­i­sa­tion (ISRO). It would also have sig­nalled to the world that In­dia had come of age in space ex­plo­ration and made the coun­try only the fourth na­tion to achieve a soft land­ing on the moon af­ter the US, the erst­while USSR and China. Coin­ci­den­tally, it would also have marked 100 days of the Naren­dra Modi-led gov­ern­ment’s sec­ond term in of­fice. More than any­one else, ISRO chair­man Kailasavad­i­voo Si­van was acutely aware of the risks in­volved in load­ing this par­tic­u­lar space event with na­tional as­pi­ra­tions. “In rocket

sci­ence,” he told in­dia to­day, “there are al­ways un­known un­knowns” (see in­ter­view). Of the 109 lu­nar mis­sions since 1958, only 61—or a lit­tle more than half—had been suc­cess­ful. Of the 46 mis­sions that, like Chan­drayaan 2, had planned a soft land­ing, only 21, slightly less than half, were suc­cess­ful. ISRO had suc­ceeded in its first at­tempt at send­ing an or­biter, Chan­drayaan 1, to cir­cle the moon in 2008 and also had a Moon Im­pact Probe loaded with in­stru­ments crash on the lu­nar sur­face. Be­fore it fell apart, it re­layed vi­tal in­for­ma­tion about the pres­ence of wa­ter mol­e­cules on the moon. ISRO then stunned the world by send­ing the or­biter Man­galyaan to cir­cuit Mars in 2014—again in its very first at­tempt. Man­galyaan will com­plete five years in the red planet’s or­bit on Septem­ber 24 and con­tin­ues to beam data back. But, de­spite all these suc­cesses, soft-land­ing a rover on the moon to ex­plore its sur­face in­volved new chal­lenges and com­plex tech­nol­ogy that ISRO had to mas­ter. So, even as ex­cite­ment over the moon land­ing built across the coun­try, Si­van, a vet­eran of space launches, had fa­mously con­fessed that the phase of Vikram’s de­scent to the moon’s sur­face from its or­bital path would be “15 min­utes of ter­ror” for space sci­en­tists.

Un­for­tu­nately for Si­van and ISRO, his ap­pre­hen­sions turned out to be pre­scient. With the prime min­is­ter look­ing on in the Mis­sion Op­er­a­tions Com­plex in Ben­galuru, those 15 min­utes ended in dis­ap­point­ment and de­spair. Space sci­en­tists lost com­mu­ni­ca­tion with Vikram in the 12th minute while it was mak­ing some crit­i­cal ma­noeu­vres 2.1 km above the lu­nar sur­face and have no idea what hap­pened to it af­ter that. Though ISRO an­nounced on Septem­ber 10 that the Chan­drayaan 2 or­biter had pho­tographed Vikram on the moon’s sur­face, it has since played down such re­ports. Asked about the im­port of Vikram being found on the moon’s sur­face, Si­van told in­dia to­day, “It means noth­ing. We have not been able to es­tab­lish any form of com­mu­ni­ca­tion with it and, till we do so, it is of no sig­nif­i­cance.” More im­por­tantly, Si­van also said that sci­en­tists are still analysing the de­tails and have no an­swers yet as to what went wrong with the lan­der.

So, why did Vikram be­have so er­rat­i­cally in the fi­nal three min­utes of the con­clud­ing stages of its de­scent, re­sult­ing in the abrupt ter­mi­na­tion of the mis­sion? Even as an of­fi­cial fail­ure-anal­y­sis com­mit­tee ex­am­ines the rea­sons, in­dia to­day

spoke to a host of space sci­en­tists and ex­perts to piece to­gether what could have pos­si­bly gone wrong with In­dia’s pres­ti­gious lan­der.

A CHE­QUERED BE­GIN­NING

A lit­tle his­tory about Vikram at this point can help one un­der­stand why a lu­nar lan­der is com­pli­cated busi­ness and why one out of two such mis­sions ends in fail­ure. In­dia had not planned to make its maiden at­tempt at a soft land­ing on the lu­nar sur­face on its own. Even be­fore the in­dige­nously-built Chan­drayaan 1 or­biter was launched, ISRO had de­cided that it could use the help and ex­pe­ri­ence of Rus­sia’s Fed­eral Space Agency (Roscos­mos) for the Chan­drayaan 2 mis­sion and signed an agree­ment with it in Novem­ber 2007. For the joint In­doRus­sian lu­nar mis­sion, ISRO would have the prime re­spon­si­bil­ity for the or­biter; Rus­sia for the lan­der and rover. The launch was planned for 2012. Though ISRO was ready with the or­biter on sched­ule, Roscos­mos pulled out of the agree­ment in De­cem­ber 2011. This was af­ter its Pho­bos-Grunt mis­sion to put a lan­der and rover on a Mar­tian moon in col­lab­o­ra­tion with the Chi­nese space agency failed. ISRO then de­cided it would build a lan­der and rover on its own and sched­uled a launch for 2016. Mean­while, the or­gan­i­sa­tion re­pur­posed its or­biter for the Man­galyaan, ac­com­plish­ing it in record time.

De­spite ISRO’s vast ex­pe­ri­ence in build­ing launch­ers and satel­lites, it soon found de­sign­ing and de­vel­op­ing a lan­der and rover a com­plex and up­hill task. Ac­cord­ing to M. An­nadu­rai, a for­mer direc­tor of ISRO’s U.R. Rao Satel­lite Cen­tre and till last year the in­charge of plan­e­tary mis­sions, “It is one thing to send an or­biter [to the Moon] as we did with Chan­drayaan 1, fire an im­pactor probe to the Moon or send an or­biter to Mars. But to bring down an or­bit­ing space­craft and make it land softly on the lu­nar sur­face is vastly more com­plex and chal­leng­ing.” The key tech­nolo­gies ISRO needed to mas­ter were a flex­i­ble propul­sion sys­tem that would reg­u­late the lan­der’s de­scent, and a con­trol sys­tem that would guide and nav­i­gate the space­craft to a pre-des­ig­nated spot on the lu­nar sur­face. Both

these tech­nolo­gies have been de­vel­oped in the past five years and are now the prime sus­pects in the pre­ma­ture ter­mi­na­tion of Vikram’s mis­sion.

THE LAND­ING PLAN

Af­ter its sep­a­ra­tion from the Chan­drayaan 2 or­biter on Septem­ber 4, Vikram was or­bit­ing the moon at a speed of 1,680 me­tres per sec­ond (or 6,048 km per hour, six times the speed of a com­mer­cial jet) and at a height of 30 km above the lu­nar sur­face. That ve­loc­ity, along with the height, had to be brought down in a con­trolled man­ner to al­most zero within 13 min­utes of the de­scent phase. The lan­der would do so us­ing the ar­ray of five rocket en­gines and eight tiny at­ti­tude con­trol thrusters fit­ted on its base, which ISRO had de­vel­oped for the mis­sion. The en­gines were de­signed as throt­tle-able ones, their thrust vary­ing with the reg­u­lat­ing of the fuel flow, just like an ac­cel­er­a­tor in a car.

The con­trol and guid­ance sys­tem was also de­vel­oped to meet the com­plex­i­ties of a moon land­ing. With the dis­tance be­tween the earth and moon being 3.84 lakh km, there is a time lag of more than a sec­ond be­fore com­mands sent from mis­sion con­trol reach the craft, and of an­other sec­ond when data about its im­ple­men­ta­tion is re­layed back. As de­ci­sions had to be taken in mil­lisec­onds dur­ing Vikram’s rapid de­scent to the lu­nar sur­face, ISRO de­vel­oped a fully au­tonomous guid­ance and con­trol sys­tem that would take care of all the ex­i­gen­cies and anom­alies that may arise on the 15-minute flight. The craft was also equipped with highly pre­cise mea­sur­ing in­stru­ments to mon­i­tor its ve­loc­ity, height, at­ti­tude, di­rec­tion and po­si­tion with re­la­tion to the moon’s sur­face, en­abling Vikram’s on­board com­puter to take de­ci­sions in real time. The craft’s con­trol and propul­sion sys­tems were also de­signed keep­ing in mind that the moon’s grav­ity is one-sixth of the earth’s. Both these sys­tems were sub­jected to rig­or­ous tests, sim­u­lat­ing con­di­tions cor­re­spond­ing to the moon’s er­ratic grav­ity pro­file.

THE BIG CHANGE

What was also un­der test was ISRO’s new plan for pow­ered de­scent that was put in place just two years ago. When de­sign­ing the lan­der, ISRO sci­en­tists had ini­tially de­cided to work with only four en­gines in­stead of five. In this con­fig­u­ra­tion, the en­gines and the guid­ance con­trol sys­tem would grad­u­ally bring the speed and al­ti­tude down to around 10 me­tres above the moon’s sur­face. But then the con­cern arose that the en­gine thrusters, at this dis­tance, would kick up a mini lu­nar dust storm that would en­velop the craft and dam­age its vi­tal equip­ment. ISRO then planned to shut all the four en­gines and in­stead strengthen the four legs of the craft to with­stand the free fall from that height with­out dam­ag­ing ei­ther the lan­der or the rover. A launch was sched­uled for Jan­uary 2018.

Mean­while, a fierce de­bate had bro­ken out among space sci­en­tists over the dan­gers of hav­ing a four en­gine-con­trolled de­scent for a moon lan­der. ISRO de­cided to cir­cum­vent the free fall by in­tro­duc­ing a fifth en­gine at the cen­tre of the lan­der’s base. This would have two ad­van­tages. The fifth en­gine would be fired only af­ter all the other four en­gines were shut down at 10 me­tres and en­sure a pow­ered de­scent till touch­down. And since the en­gine was lo­cated at the cen­tre of the craft, the plume of dust it would kick up would be pushed away from it. That de­ci­sion would add more weight to the space­craft: along with other changes in the con­fig­u­ra­tion, the com­pos­ite Chan­drayaan 2 with the or­biter, lan­der and rover would now weigh 3.8 tonnes. This meant that Geosyn­chronous Satel­lite Launch Ve­hi­cle, or GSLV MarkII was no longer suit­able as a launch ve­hi­cle, as it was ca­pa­ble of car­ry­ing a pay­load of only 2-3 tonnes. So the Chan­drayaan 2 pro­ject team had to wait for GSLV MarkIII, ISRO’s heav­i­est rocket, then un­der devel­op­ment, to be val­i­dated. Rather than wait for the full range of trial flights, ISRO de­cided to take a risk by launch­ing Chan­drayaan 2 on

THE KEY TECH­NOLO­GIES ISRO NEEDED TO MAS­TER WERE FLEX­I­BLE PROPUL­SION AND CON­TROL SYS­TEMS

GSLV MarkIII’s first op­er­a­tional flight. As it turned out, af­ter an ini­tial scare, the GSLV MarkIII fired beau­ti­fully on July 22, 2019, launch­ing Chan­drayaan 2 on its lu­nar jour­ney.

THE WOB­BLE

For Vikram’s de­scent to the moon, ISRO homed in on a par­a­bolic pow­ered de­scent tra­jec­tory di­vided into four dis­tinct phases. The process would be­gin when the space­craft was at a height of around 30 km above the lu­nar sur­face and 650 km away from the land­ing site. In the first phase, known as the Rough Brak­ing Phase and last­ing for 10 min­utes 20 sec­onds, Vikram would use the brute force of its en­gines to brake its hor­i­zon­tal speed of 1,648 me­tres per sec­ond down to around 150 me­tres per sec­ond. In this phase, it would come down from 30 km to 7.4 km. While de­tach­ing from the or­bit and in­de­pen­dently re­volv­ing around the moon, Vikram was ejected with the ex­haust fun­nels of its five en­gines fac­ing the di­rec­tion of its revo­lu­tion in­stead of on the op­po­site side. At the be­gin­ning of the de­scent phase, its on­board com­put­ers ig­nited four of the five en­gines to steadily kill its ve­loc­ity. To en­sure that both the craft’s hor­i­zon­tal and ver­ti­cal ve­loc­i­ties were within pa­ram­e­ters, all four en­gines had to fire with per­fect syn­chronic­ity. If one of the en­gines de­vi­ated, the com­puter was pre-pro­grammed to use the other en­gines to pro­vide dif­fer­en­tial thrusts to cor­rect the anom­aly.

The live tele­cast by Do­or­dar­shan showed sci­en­tists clap­ping at the com­ple­tion of the Rough Brak­ing Phase, in­di­cat­ing it was suc­cess­ful. But some ex­perts be­lieve that there are in­di­ca­tions that er­rors may have been build­ing up in this phase. For while the hor­i­zon­tal ve­loc­ity (the speed the craft was mov­ing at) was to be around 150 me­tres per sec­ond at the end of the phase, the read­ings on the large con­sole in the mis­sion op­er­a­tions com­plex showed that it was around 200 me­tres per sec­ond, faster than what it should have been. On the other hand, the ver­ti­cal ve­loc­ity or the speed with which the lan­der was de­scend­ing, hov­ered be­tween 70 me­tres and 68 me­tres per sec­ond for sev­eral sec­onds.

It was at this point that the sec­ond phase, termed the Ab­so­lute Nav­i­ga­tion Phase (ANP) and last­ing around 40 sec­onds, kicked in. In this phase, Vikram should have cor­rected any er­rors in cal­cu­la­tions of the key nav­i­ga­tion pa­ram­e­ters such as its height and ve­loc­ity dur­ing the Rough Brak­ing Phase. It did this by dou­ble-check­ing the read­ings of its on-board mea­sur­ing in­stru­ments, in­clud­ing cam­eras pho­tograph­ing the lu­nar ter­rain, to mea­sure Vikram’s ve­loc­ity and height. Variations in the ve­loc­ity, al­ti­tude or in­cli­na­tion of the space­craft were to be cor­rected by the au­tonomous con­trol sys­tems, which ar­rive at their own log­i­cal de­ci­sions on the ad­just­ments that need to be made. As a se­nior sci­en­tists put it, “The num­ber of ex­i­gen­cies and er­rors you can cal­cu­late and feed into the com­puter is only lim­ited by your imag­i­na­tion. The best con­trol sys­tems are the ones where sci­en­tists let their imag­i­na­tions run free and plan for as many con­tin­gen­cies as pos­si­ble.”

It was at the brief ANP phase that the anom­alies in Vikram’s pow­ered de­scent be­gan to mount. In the con­trol room, the large con­sole sim­u­lat­ing Vikram’s de­scent showed the lan­der de­vi­at­ing from its 45-de­gree in­cli­na­tion. It in­ex­pli­ca­bly ex­e­cuted a som­er­sault, mak­ing the en­gines face up­wards in­stead of down­wards (see graphic: 15 Min­utes to De­spair). One ex­pla­na­tion is that the on­board com­puter was

cor­rect­ing the space­craft’s at­ti­tude to en­able the cam­eras to po­si­tion it prop­erly for tak­ing the pic­tures it needed to cal­i­brate vi­tal pa­ram­e­ters. But that ma­noeu­vre went hay­wire and re­sulted in in­creas­ing the ver­ti­cal de­scent ve­loc­ity rather than de­creas­ing it. The other ex­pla­na­tion is that the con­trol sys­tem no­ticed a drop in the ve­loc­ity and cor­rected it even though it was still within the thresh­old. In do­ing so, it first er­ro­neously ro­tated the craft by 140 de­grees to boost the ve­loc­ity, then re­versed it to the orig­i­nal po­si­tion. By then, the space­craft had lost its ori­en­ta­tion and con­trol.

LOSS OF CON­TROL

It was at this point that the third phase, the Fine Brak­ing Speed Phase last­ing 90 sec­onds, be­gan. To bring down Vikram’s hor­i­zon­tal and ver­ti­cal speeds to near-zero and the craft to an al­ti­tude of 400 me­tres, two of the four en­gines were to shut down. There is ev­i­dence to show that the space­craft was des­per­ately try­ing to re­gain its ori­en­ta­tion and was pitch­ing from side to side. The con­sole showed that the ver­ti­cal speed had in­creased; it was also at this junc­ture that all com­mu­ni­ca­tion with the con­trol room snapped. There was no ev­i­dence to show that the two en­gines had shut down as per plan. All the con­sole showed was that the hor­i­zon­tal ve­loc­ity was still a high 48 me­tres per sec­ond and the ver­ti­cal ve­loc­ity 59 me­tres per sec­ond. Both these key pa­ram­e­ters should have been con­sid­er­ably lower for the lan­der to go into its ter­mi­nal de­scent phase. Its speed at this point should have been near-zero and it should have been hov­er­ing over the lu­nar sur­face at a height of 400 me­tres. Its on­board cam­eras were then to take pic­tures for its con­trol sys­tem to check whether the land­ing site was suit­able.

ISRO had de­cided that Vikram would land near the colder South Pole where wa­ter mol­e­cules were ex­pected to be found in greater abun­dance. This was the first time a lan­der was do­ing so—for good rea­son, as there are more craters on the lu­nar poles than its equa­to­rial belt. Vikram’s con­trol sys­tem, us­ing its in­stru­ments in­clud­ing the cam­eras, was to en­sure that the craft would land on a flat

sur­face. If it landed on any sur­face that had an in­cline be­yond 12 de­grees, it would top­ple over. Vikram was to then descend to 10 me­tres be­fore its on-board con­trol sys­tem would switch off two en­gines. The fifth en­gine lo­cated at its base would then be switched on for a con­trolled de­scent. All this was to hap­pen if ev­ery­thing had gone well in the ear­lier phases. But, since com­mu­ni­ca­tion snapped at a height of 2.1 km, there is no ev­i­dence so far to show whether the ter­mi­nal de­scent phase was ac­ti­vated or not.

Ac­cord­ing to ex­perts, Vikram’s abrupt end can be at­trib­uted to three ma­jor rea­sons, but they do not quite agree which one was the main cul­prit. Some be­lieve that the propul­sion sys­tems mal­func­tioned dur­ing the tran­si­tion from the Rough Brake Phase to the Ab­so­lute Nav­i­ga­tion Phase, when the en­gines were to fire syn­chronously to re­duce the lan­der’s speed. Since the throt­tle-able en­gines were based on a new tech­nol­ogy, there is sus­pi­cion that one of them could have mis­fired, caus­ing un­sta­ble con­di­tions be­yond the sys­tem’s tol­er­ance, and con­fused the com­mand and con­trol sys­tem. Oth­ers be­lieve the er­ror lay in the con­trol sys­tem it­self, with an im­proper logic built in, that made the lan­der do a com­plete turn dur­ing the tran­si­tion be­tween the ab­so­lute nav­i­ga­tion and fine brak­ing phases. Yet an­other sec­tion of opin­ion ar­gues that it was a com­bi­na­tion of er­rors in both propul­sion and con­trol sys­tems that led to the set­back. Mean­while, ISRO sci­en­tists are gath­er­ing ev­ery bit of data the lan­der trans­mit­ted be­fore its sig­nal was lost. They are us­ing such data to sim­u­late all pos­si­ble sce­nar­ios and ex­plain Vikram’s aborted land­ing.

Space is harsh and un­for­giv­ing of er­rors. But ISRO sci­en­tists need not feel dis­cour­aged with the out­come. As a to­tal mis­sion, Chan­drayaan 2 has notched up many suc­cesses. It has val­i­dated ISRO’s big­gest launcher, the GSLV MarkIII. It has proved ISRO’s ca­pa­bil­ity to suc­cess­fully send an or­biter to the moon and ex­e­cute a com­plex ma­noeu­vre of un­dock­ing the lan­der from it. More im­por­tantly, the or­biter it­self is per­form­ing out­stand­ingly and has eight ma­jor in­stru­ments that are trans­mit­ting reams of data about the moon, in­clud­ing in­di­ca­tors of the avail­abil­ity of wa­ter on its sur­face. This could help space­far­ing na­tions de­ter­mine whether the moon can be colonised and used as a base for deep space ex­plo­rations by us­ing the wa­ter not just for hu­man needs but also as fuel to power rock­ets. The bonus is that the or­biter has enough fuel to go around the moon for seven and a half years in­stead of one year. The ma­jor tech­nolo­gies Chan­drayaan 2 couldn’t val­i­date re­late to the thwarted soft land­ing on the moon and the op­er­a­tion of the rover. These can be ad­dressed only af­ter ISRO is able to de­ci­pher what went wrong with Vikram.

Sci­en­tists should take heart from the ex­pe­ri­ence of the late Dr A.P.J. Ab­dul Kalam when he was pro­ject direc­tor of ISRO’s first satel­lite launch ve­hi­cle, SLVIII. When the first launch failed in 1979, Kalam ten­dered his res­ig­na­tion to his boss, who tore it up and asked him to carry on. Years later, Kalam grate­fully re­called that the fail­ure taught him more about mak­ing bet­ter space sys­tems than a suc­cess could have. Prime Min­is­ter Modi was per­haps speak­ing for the en­tire coun­try when he said that de­spite the set­back, the na­tion is solidly be­hind ISRO and remains proud of its many achieve­ments. ■

THE MA­NOEU­VRE TO RO­TATE THE CRAFT FOR NAV­I­GA­TIONAL PUR­POSES SEEMS TO HAVE GONE HAY­WIRE

ANI

PHASES OF MOON LAND­ING (Clock­wise from left) ISRO of­fi­cials watch live tele­cast of Vikram’s Fine Brak­ing Phase; Si­van mon­i­tors fi­nal de­scent of the lan­der; the PM con­sol­ing Si­van

WORD OF EN­COUR­AGE­MENT The PM in­ter­acts with ISRO sci­en­tists in Ben­galuru on Septem­ber 7

Newspapers in English

Newspapers from India

© PressReader. All rights reserved.