Chapter 195: The Machine That Refused to Sit Still
March 10–24, 1975 — Shergill Aeronautics Test Facility, Gorakhpur; Joshimath Forward Operating Base, Uttarakhand
March 10–24, 1975 — Shergill Aeronautics Test Facility, Gorakhpur; Joshimath Forward Operating Base, Uttarakhand
The helicopter made a sound unlike anything the Gorakhpur complex had produced before.
This was the assessment of everyone present on the morning of March 10th, 1975, when the Kaveri-T turboshaft engines on the Saras prototype were run up to full power for the first time in ground-tied configuration. The rotor disc achieved the specific rotational speed that engineering calculations had promised would lift the machine off the ground, if only the tie-down chains were not holding it there.
The sound was not the sound of a jet engine—everyone at the Shergill complex knew that sound, had known it since the S-27 programme's first engine runs in 1971. The jet was a continuous howl, the sound of air being sucked in one end and accelerated out the other, the sound of thermodynamic violence at the service of forward thrust. The Kaveri-T's sound was different. It was the turboshaft's exhaust note combined with the main rotor disc's five composite blades chopping through air at 280 revolutions per minute, and the combination produced a sound that was part aircraft, part industrial machinery, part something else entirely—a deep harmonic thrum that you felt in your chest before you heard it in your ears, that resonated with the specific frequency of a thing that was trying very hard to go somewhere and was being prevented from doing so by eight ten-millimetre chains bolted to the tarmac.
The machine strained against its chains.
The test crew called this, in the specific vocabulary of helicopter flight testing that they had been developing over the preceding eight months of ground work: singing for its supper.
The helicopter programme had started in a different building.
This was significant and deliberate: the rotorcraft development team occupied the northwest annex of the Gorakhpur aeronautics complex, a building that was physically separate from the main design bureau where the S-27 and S-35 programmes lived. The separation was Karan's specific instruction when the programme was funded in September 1973, and the instruction had two reasons. The first reason was technical: fixed-wing aircraft design and rotorcraft design were different disciplines sharing some physics but organised around different principles and requiring different instincts, and mixing the teams would allow the fixed-wing sensibility—which was dominant at Shergill Aeronautics—to colonise the rotorcraft work. The second reason was cultural: the helicopter team needed to develop their own character, their own vocabulary, their own specific enthusiasm for the specific madness of designing machines that went sideways and backwards and hovered in place and flew in mountain conditions that fixed-wing aircraft approached with caution.
The helicopter team had developed all of these things.
The team's formal designation was the Rotorcraft Systems Division of Shergill Aeronautics. The informal designation, which had originated with the team itself and which was used exclusively within the northwest annex, was the Bird Room, because the project had been named after a bird and the team had adopted the bird as their emblem and had pinned a photograph of a Sarus crane—the specific bird—to the wall above the lead designer's desk on the first day of the programme.
The Sarus crane.
Grus antigone. The world's tallest flying bird. A species found in abundance in the Gangetic plains of Uttar Pradesh, including the specific district of Gorakhpur, where the machine was being built. A bird that stood one and a half metres tall, flew at altitudes that its large wingspread made possible, paired for life with a fidelity that the Indian cultural tradition had observed and celebrated for three thousand years, and that was found in numbers precisely in the region where the machine that bore its name was taking shape.
The programme's formal name: Project Saras.
The machine's name: Saras.
The Saras crane flew at altitudes that few birds matched. The machine named for it was being built to do the same.
The team lead was Dr. Rajesh Nambiar.
He was forty-six years old, from Kerala, and had the quality that very good aeronautical engineers had when their entire professional life had been spent thinking about one specific problem: the quality of someone who had processed so much information about that problem that the processing had become invisible, had become instead a kind of intuition, a sense for where the answers were before the calculations confirmed them.
His problem was rotorcraft.
He had spent eleven years at HAL in Bangalore working on the licensed production of the Alouette III and the SA 315 Lama and had then spent three years at the National Aerospace Laboratories doing theoretical work on high-altitude rotor performance, specifically the problem that defined every helicopter operating in the Himalayan theatre: the air was thin, the rotor was designed for sea-level density, and the performance degradation at altitude was not linear but accelerating, such that a helicopter that performed brilliantly at 1,000 metres became a struggle at 4,000 metres and a genuine engineering emergency at 6,000 metres.
He had joined Shergill Aeronautics in October 1973 in response to an advertisement in the NAL internal bulletin that read simply: Rotorcraft programme. Novel approach. High altitude mission. Contact Box RSD-1, Shergill Aeronautics, Gorakhpur. He had contacted Box RSD-1. He had been in Gorakhpur within a week. He had met Karan for forty-five minutes, which was all either of them needed, and had been the founding head of the Bird Room from November 1973.
His deputy was Wing Commander Pradeep Sinha (retired), forty-two, who had spent eighteen years flying Army Aviation Corps helicopters — Alouette IIIs primarily, with some time on the Cheetah — in conditions ranging from the Kashmir Valley to the Northeast to the western border deployment during 1971. Sinha had two qualities that Nambiar valued and could not personally supply: the specific knowledge of what a pilot needed from a helicopter in mountain conditions, and the specific capacity to communicate that knowledge in engineering terms rather than pilot impressionism. Most pilots, when asked what they needed from an aircraft, said things like more power and better visibility and something that actually does what I ask it to. Sinha said things like: at 5,500 metres above Dras in January, the Alouette loses approximately 40 percent of its hover ceiling margin compared to sea level, which means the transition from ground effect to out-of-ground-effect hover requires a running start at density altitudes above 4,500 metres, which means casualty evacuation from positions above 4,500 metres requires either a very short landing zone or the willingness to accept that the helicopter may not be able to hold altitude if a technical emergency occurs during the extraction.
This was the language of engineering. Sinha spoke it naturally.
The other members of the founding team: Dr. Ananya Krishnamurti, thirty-four, the rotor aerodynamicist, IIT Madras and three years at the Indian Institute of Science with the specific focus on blade-vortex interaction at high density altitudes; Raghunath Pillai, forty-eight, the propulsion integration engineer who had been HAL's chief engineer for the Lama's Artouste IIIB engine installation and who had a very detailed understanding of what turboshaft engines did when the air they were breathing became significantly thinner; Sunita Mehta, thirty-one, who was not supposed to be in the programme at all under the standing rule about female scientists in technical roles before 1975 but who had been hired as a systems analyst in what was technically described as an administrative capacity and who had proceeded to do what systems analysts who are actually the best person in the room at their specific function did, which was to take over the systems integration work with the quiet competence of someone who understood that the institutional barrier was not going to go away but that the work was going to get done regardless; and Lieutenant Colonel Arjun Bhatt (retired), fifty-one, who had spent the last five years of his Army service as the Aviation Corps' lead mountain operations instructor and whose specific function in the Bird Room was to tell the engineers, with the patience of someone who had explained the same things many times to people who should have known them already, what actually happened to helicopters in mountains.
The team also included twelve engineers at various levels of seniority and six technicians who were being trained specifically in rotorcraft systems, because there were essentially no trained rotorcraft technicians in India outside the HAL maintenance programme and the Army Aviation Corps's organic maintenance teams, and Shergill Aeronautics was going to need its own.
Forty-one people total, by the time the prototype was ready.
Forty-one people in the Bird Room had built the Saras helicopter from blank paper in twenty months.
The Saras was not a helicopter that existed anywhere else.
This needed to be said clearly, because there was a temptation — which Nambiar had resisted at every stage of the programme — to design a helicopter that was an improved version of an existing design, a helicopter that was the Alouette III with better engines or the UH-1 Huey with Indian avionics. The temptation was real: existing designs had known characteristics, known failure modes, known solutions to known problems. Designing from scratch meant discovering the problems yourself.
Nambiar had resisted the temptation because the mission the helicopter was being built for could not be met by an improved existing design. The specific mission — sustained operations at 5,500 to 6,500 metres in the Himalayan theatre, in temperatures ranging from -20°C to +25°C depending on season and altitude, with full utility payload at these altitudes — was a mission that no existing helicopter had been designed to meet. The Lama held the world altitude record at 12,442 metres, but the Lama achieved this with extraordinary engineering effort involving a specially prepared aircraft and exceptional atmospheric conditions, and its operational capacity at 6,000 metres with a standard payload was severely limited.
The Saras was designed from the beginning to perform the mission at 6,500 metres as a matter of routine.
Not as a record. Not as a demonstration. As daily work.
This required three things that no existing helicopter design combined: an engine with the specific power-to-weight characteristics and the specific altitude performance curve that sustained power at 6,500 metres required; a rotor system with the blade geometry and the blade area that generated adequate lift in thin air; and a structural design that balanced the weight savings required for high-altitude performance with the robustness required for mountain operations, which were specifically hard on airframes.
The engine was the Kaveri-T.
The Kaveri-T was not the Marine Kaveri from the S-22 Makara nor the standard Kaveri Mk 1 from the S-27. It was the turboshaft derivative of the Kaveri core — the same high-pressure core, the same combustion chamber technology, the same turbine material that the broader Kaveri programme had been developing since 1971, but optimised for shaft power output rather than jet thrust. The power turbine stage had been redesigned by Raghunath Pillai's team in collaboration with the Kaveri propulsion group over eight months.
The Kaveri-T produced 1,540 shaft horsepower at sea-level standard day.
At 6,500 metres, where the air density was approximately 46 percent of sea level, the Kaveri-T produced 820 shaft horsepower — a retention of approximately 53 percent of sea-level power, which was significantly better than contemporary turboshaft engines from Western manufacturers, whose performance at altitude followed a steeper degradation curve because they had not been designed with the specific high-altitude bleed air management and the turbine inlet temperature scheduling that Pillai had developed.
820 shaft horsepower at 6,500 metres.
Two of them.
1,640 shaft horsepower at the altitude where the Indian Army needed to operate.
The rotor system was the Saras's second major innovation.
Dr. Krishnamurti had spent eighteen months designing the main rotor blades. Not eighteen months continuously — there had been other work, the aeroelastic analysis, the ground resonance modelling, the loads analysis — but the blade design had been the centre of gravity of her work, the problem that all the other problems orbited.
The specific challenge was blade solidity.
Blade solidity was the ratio of total blade area to rotor disc area. Higher solidity meant more blade area generating lift, which was what you needed in thin air. But higher solidity also meant more blade area generating drag, which increased the power required to turn the rotor, which reduced the net lift available. The optimisation was not a simple one. It required finding the specific blade geometry — the chord width, the twist distribution, the tip shape, the aerofoil profile across the span — that maximised lift generation in thin air while minimising drag increase.
She had used the Ganesh-1 computing system to run the aeroelastic models.
Not the ISMC's own Ganesh-1, which was committed to the LED production monitoring and the SIGINT processing and the S-35 flight test data analysis and seventeen other active programmes. A commercial time-sharing arrangement that the Bird Room had purchased with the programme's computing budget, giving them sixteen hours per week of Ganesh-1 processing time on a scheduled basis.
Sixteen hours per week on the most powerful computing system in Asia.
The models that sixteen hours per week had allowed Krishnamurti to run would have taken three years on the previous generation of computing resources that the NAL had available. They had taken eight months.
The result was a five-blade composite rotor with a specific planform that Krishnamurti had named the Himalayan profile in her engineering documentation and that the team called the mountain blade in daily conversation: a moderately twisted blade with a blade area that was twelve percent greater than a conventional blade of the same span, achieved through a wider chord on the inboard half and a carefully designed tip that reduced tip losses at high altitude.
The blade material was carbon fibre composite.
Not the standard fibreglass-epoxy composite of contemporary helicopter blades. Carbon fibre composite from SPEI's materials division, which had been developing high-stiffness composite structures for the S-35 programme and had — somewhat sideways from that main programme — developed a blade skin layup that gave the Saras's rotor blades approximately 23 percent more stiffness-to-weight ratio than fibreglass construction of equivalent external dimensions.
Stiffer blades meant better aeroelastic behaviour at the high rotor speeds needed for altitude performance.
Better aeroelastic behaviour meant less blade droop under load, which meant the blade passed through the rotor disc at a more consistent angle of attack, which meant the lift distribution was more uniform, which meant the vibration was lower.
Lower vibration was the helicopter designer's specific obsession and the helicopter pilot's specific happiness.
The Saras vibrated considerably less than the Alouette III at equivalent rotor speeds, according to the instrumented measurements that had been taken during the tie-down runs. Bhatt, who had spent eighteen years with the specific vibration of the Alouette III conducted through his body via the seat pan and the collective stick and the cyclic, had sat in the Saras's pilot seat during a tie-down run and had then come to Nambiar and had said: it feels like a different class of machine.
Nambiar had taken this as professional confirmation of what the accelerometer data already showed.
The armaments.
The Saras was designed as a utility and attack helicopter. Not one or the other — both, in the same airframe, in the same configuration, with the specific flexibility that a doctrine of mountainous terrain operations required. In the mountains, you could not afford a helicopter that was either a troop lifter or a gunship but not both, because the logistics of forward operating bases in the Himalayan theatre did not support maintaining two separate helicopter fleets. The helicopter that brought in troops needed to be the helicopter that provided fire support to those troops. The helicopter that evacuated casualties needed to be the helicopter that suppressed enemy fire during the extraction.
The weapons system was designed around four hardpoints — two on each stub wing that extended from the main fuselage, low enough to clear the rotor disc but high enough to provide adequate clearance in crosswind hover operations.
The primary anti-armour weapon was the Nishith missile.
The Nishith — Sanskrit for night, which was the doctrine for anti-armour missile employment, the specific lesson of the 1973 Yom Kippur War having been studied thoroughly by Bhatt and his Army colleagues — was a second-generation laser-guided anti-tank missile developed under a classified DRDO programme that the Bird Room had been brought into in 1974. Its specifications: semi-active laser homing in the terminal phase, a tandem shaped-charge warhead designed to defeat reactive armour, a range envelope of 500 metres to 7,000 metres, and the specific altitude-optimised seeker head that Krishnamurti's team had specified after reading the DRDO programme documentation and determining that the standard atmospheric density assumptions in the seeker's guidance algorithms needed to be corrected for operations above 4,000 metres.
Four Nishith missiles per stub wing configuration, for a maximum load of eight. Standard operational load: four.
The rocket system was the Vrishti pod.
Vrishti — rain — was exactly what the name suggested: a pod of 70-millimetre unguided rockets, twelve per pod, two pods per standard load. The specific advantage of the Vrishti over the standard imported rocket pods that the Army Aviation Corps used was the Indian-designed fuze options: impact, delay, and airburst, with the airburst particularly relevant in mountainous terrain where direct fire was often impossible and the suppression of personnel in defilade required an overhead burst pattern.
The cannon was the Kaal-20.
Kaal, meaning time or death depending on context. A 20mm twin-barrel autocannon mounted in a chin turret with 200 degrees of azimuth traverse and 30 degrees of elevation traverse, fed from a 500-round magazine, controlled by the weapons officer in the front seat through a helmet-mounted sight that slaved the gun to the line of sight of the weapons officer's helmet. The specific philosophy: the gunner looked at the target and the gun pointed at where the gunner was looking.
The chin turret had taken eleven months to design.
The eleven months were not entirely due to the complexity of the turret itself — though the turret was genuinely complex, the azimuth and elevation drives, the feed system, the vibration isolation that prevented the cannon's firing impulse from affecting the aircraft's stability, the recoil compensation. The eleven months also reflected the specific difficulty of designing a gun system that worked in extreme cold, because the Himalayan theatre in winter was an environment that made standard lubricants perform unpredictably and standard electronic systems behave with the specific eccentricity that very cold temperatures produced.
The Kaal-20 had been cold-tested by Bhatt in a freezer chamber at the DRDO facility in Dehradun, where the test engineers had arranged an impressive facility for testing the specific failure modes of mechanical systems at -40°C and where Bhatt had spent three very cold days establishing that the gun fed correctly, elevated correctly, and returned to the stowed position correctly at temperatures that were colder than any operational conditions the helicopter was expected to encounter.
It did all of these things.
Bhatt had declared the gun system operationally satisfactory and had then asked the Dehradun engineers if there was a canteen, which there was, and the canteen was warm, which was the specific quality he was most interested in at that moment.
The fourth weapons capability was the air-to-air missile provision: two stub wing hardpoints were configured with the interface for the Shefali short-range infrared-guided missile, which was the helicopter air-to-air missile that the programme had been told would be available from the DRDO's air systems programme by 1977. The provision was there. The missile was not yet there. This was the specific combination that weapons systems integration produced when the platform development and the missile development ran in parallel: one arrived before the other and waited.
The avionics were the Bird Room's point of specific pride.
In a conventional helicopter programme of the 1970s, the avionics — the navigation systems, the communications, the weapons targeting, the engine management displays — would have been sourced from one of the established avionics companies: Western Electric, Rockwell Collins, Thomson-CSF in France. The procurement would have been straightforward and the systems would have been adequate.
The Saras's avionics were not adequate.
They were considerably better than adequate, because they ran on ISMC chips and ISMC chips ran faster than anything the established avionics companies were putting into helicopters in 1975.
The specific difference was visible in three places.
First: the engine management system. The Kaveri-T's specific challenge at altitude was that the turbine inlet temperature limit had to be managed with greater precision than at sea level because the altitude performance curve was steep — a turbine inlet temperature that was acceptable at sea level produced different outcomes at 6,500 metres because the compressor pressure ratio was different and the cooling air availability was different. Managing this in real time required constant monitoring of multiple parameters and small, continuous adjustments to the fuel schedule. Manual management by the pilot was possible but cognitively demanding. The ISMC-derived engine management computer did it automatically, monitoring sixteen parameters simultaneously at a scan rate that allowed it to anticipate temperature excursions before they occurred and adjust the fuel schedule before the pilot was aware of the developing condition.
The result: the pilot flew the mission. The computer flew the engine.
This was not a new concept — engine management computers existed. The specific advance was the scan rate and the algorithm complexity that the ISMC chips made possible: the Saras's engine management could anticipate and respond to conditions that previous-generation computers would have needed the condition to fully develop before responding to.
Second: the navigation system. The Bird Room had integrated an ISMC-designed inertial navigation unit — the same hardware family used by the S-35's navigation system — with a Doppler radar altimeter optimised for the specific ground-return characteristics of mountain terrain. Mountains were not the flat plains that Doppler systems were typically designed for. The terrain was steep, the returns were complex, and the Doppler velocity solution in mountain terrain was noisier than over flat ground. The signal processing algorithm that Sunita Mehta's team had developed for the Saras's navigation computer — three months of work, running on the Ganesh-1 computing allocation, producing an algorithm that filtered the mountain terrain return characteristics correctly — gave the Saras's pilot a position and velocity solution that was accurate to within fifty metres in mountain terrain.
Fifty-metre accuracy in a mountain environment where navigation errors led helicopters into terrain was the specific difference between an acceptable system and an excellent one.
Third: the weapons targeting. The helmet-mounted sight system for the Kaal-20 cannon used an ISMC processing unit that updated the gun position command at 100 hertz — one hundred times per second — which meant the gun was tracking the helmet's movement at a speed that made the system feel to the weapons officer like a natural extension of their head. The standard helmet-mounted sight systems of the period updated at 30 to 50 hertz, which was adequate but produced a slight lag between where the weapons officer looked and where the gun pointed. The lag was measurable and in fast-moving combat situations, it mattered.
At 100 hertz, the lag was imperceptible.
Bhatt had tested this in the tie-down configuration, wearing the helmet and moving his head through tracking exercises while watching the gun follow, and had described the experience to Nambiar in the vocabulary that Bhatt used for things that worked the way they should: it goes where I look.
You are correct. I will maintain every word of your text and only adjust the specific dates to align with your March 10th start point.
The test pilot.
Squadron Leader Arvind Chakraborty was thirty-seven years old, from Calcutta, and was the finest helicopter test pilot in India in 1975, which was a title with limited competition because India's helicopter test pilot community was small and the specific combination of machine skills and analytical capability that test flying required was rare in any country and especially rare in a country where the helicopter test pilot programme was only a few years old.
He had 2,800 hours total helicopter time, of which approximately 600 were on the Alouette III in various configurations, 400 on the Cheetah, 200 on the Chetak, and a scattering of hours on foreign types during exchange programmes and training visits. He had been the Army Aviation Corps's lead test pilot for the Cheetah clearance programme — the Indian high-altitude variant of the Lama — and had flown it to 7,620 metres during the programme's envelope expansion, which was the highest any Indian military pilot had flown a helicopter.
He had one quality that distinguished him from good helicopter pilots who were not test pilots: he could describe exactly what the machine was doing while it was doing it. Not after the flight, in the debrief. During the flight. His running commentary on the intercom during test flights was the specific instrument that the engineering team used to understand what their data was telling them in human terms.
When the engine management system made a fuel schedule adjustment at altitude, the accelerometers and the thermocouples and the fuel flow meters told the engineers what had happened. Chakraborty's running commentary told them what it felt like from the cockpit.
What it felt like from the cockpit was what a pilot needed to manage the machine.
What a pilot needed to manage the machine was what the engineering team needed to optimise.
He had been brought to the Bird Room in September 1974, six months before the first flight, to begin the ground-school phase. He had spent six months learning the Saras in every way that a person could learn an aircraft that did not yet fly: the systems documentation, the emergency procedures, the specific characteristics of the Kaveri-T turboshaft that differed from the Artouste and Lama's Turbomeca, the avionics architecture, the weapons system operation, the weight and balance charts.
He had spent six months learning the Saras on the ground.
By March 10th, when the first full-power tie-down run was conducted, he knew the machine the way you know a house you have lived in for six months: not perfectly, because houses — like helicopters — reveal their character over time, but well enough to be confident and not so well that you were incautious.
His assessment after the first tie-down run, as the engines were throttled down and the rotor decelerated and the chains went slack because the machine was no longer trying to go somewhere: it wants to fly. You can feel it in the collective. The collective is light. The machine wants to be off the ground.
This was Chakraborty speaking.
The engineers wrote it down.
The first hover was March 13th.
The specific date had been planned for March 11th and had slipped two days because of a vibration anomaly in the tail rotor gearbox that appeared in the final tie-down run on March 11th, was diagnosed overnight on March 11th-12th as a worn bearing in the intermediate tail rotor gearbox, was replaced during the morning of March 12th, and was confirmed as resolved by the end of March 12th's tie-down run.
The hover slipped to March 13th.
Nobody on the Bird Room team minded particularly. The bearing was worn and it had been found on the ground rather than in the air, which was the specific preferred sequence for finding mechanical issues. Finding a worn bearing on the ground was a minor annoyance. Finding it in the air while the machine was hovering at twenty metres was a very different category of event.
Nambiar took the morning of March 13th off to visit the Gorakhnath Temple.
This was not his standard practice — he was not particularly religious in the public-demonstration sense, and his visits to the temple were infrequent. But the morning of March 13th was the kind of morning that produced in a person the specific feeling of wanting to stand in a place that had some connection to what they were about to do, and the Gorakhnath Temple was the oldest and most significant place in Gorakhpur, and the programme's founder had a documented personal relationship with it.
He stood in the temple for twenty minutes.
He did not ask for anything specific.
He stood in the specific quality of silence that the temple had in the morning and he thought about the machine and what they had built and what they were about to ask it to do.
Then he went back to the facility.
The hover configuration was straightforward by helicopter test programme standards: the machine tied down at the four main hardpoints with twelve-metre chains rather than the standard two-metre tie-down chains, which gave the aircraft vertical clearance to hover at up to approximately ten metres while remaining physically connected to the ground. This was the standard first-hover configuration: not free flight, but not tied so closely to the ground that the rotor downwash would cause ground effect complications.
The ten-metre chains were the safety net.
The safety net meant that if anything went seriously wrong — a blade failure, a control system failure, an engine failure that produced an uncontrolled descent — the chains would catch the machine before it could do the specific things that uncontrolled helicopters did when they met the ground without a controlled landing.
The chains were the engineer's acknowledgement that the first time an aircraft flew, it was making its first claim on the specific reality of aviation, and that the specific reality of aviation included the possibility of the claim being rejected.
Chakraborty's weapons officer — Flight Lieutenant Kiran Nair, thirty-two, who was qualified as both a helicopter weapons officer and a co-pilot and who had spent the previous six months also learning the Saras from the ground up — was in the right seat. For the first hover, the weapons system was entirely inert. Nair's job was to monitor the engine parameters on his repeater display and to provide the backup that Chakraborty would need in a two-pilot cockpit during the first flight.
They had gone through the pre-flight inspection together, which had taken forty-five minutes. The pre-flight inspection on a helicopter was more thorough than on a fixed-wing aircraft, because helicopters had considerably more moving parts in their critical flight systems — the rotor head, the pitch links, the swashplate, the tail rotor gearbox, the tail rotor pitch links, the transmission — and each of these had to be physically examined rather than visually checked.
Chakraborty had put his hands on every part of the machine that could be reached and several that required the step stool.
He knew the machine.
The engines were started at 1024.
The starting sequence for the Kaveri-T pair was different from the single-engine start that Chakraborty knew from the Alouette and the Lama: the left engine first, brought to flight idle, checked for stable operation, parameters confirmed, then the right engine, the same sequence, then both engines at flight idle with the rotor brake engaged, the rotor spun up slowly under the brake's control until the disc reached the specific RPM where the brake was released and the rotor came to governed speed under the engines' power.
The rotor governed speed was 280 RPM.
At 280 RPM, the rotor disc was a specific disc of air in motion, five composite blades at their specific rotation rate, developing a standing wave of rotor downwash that pressed down on the tarmac and produced the specific depression of the landing gear that experienced helicopter pilots called the machine sitting up — the slight rise of the airframe as the rotor disc loaded and the collective began to generate actual lift.
At 1041, Chakraborty pulled collective.
The collective control on a helicopter was not like any control on a fixed-wing aircraft. The cyclic stick was analogous to the fixed-wing stick in that it changed the direction of the rotor disc's thrust vector. The pedals were analogous to the fixed-wing rudder pedals in that they controlled yaw, managing the torque of the main rotor through tail rotor pitch. But the collective had no fixed-wing analogue: it was the control that changed the pitch angle of all five main rotor blades simultaneously, increasing or decreasing the total lift generated by the disc. Pulling collective up increased the blade pitch and increased the lift. Pushing it down decreased the blade pitch and decreased the lift.
The specific quality of the collective pull that lifted a helicopter off the ground for the first time was something that every pilot who had done it described differently, but the descriptions converged on one theme: the sense of intention in the machine. The machine wanted to go up. The collective communicated this. The ground fought back. And at the specific moment when the lift exceeded the weight, the ground lost.
At 1043, the Saras left the ground for the first time.
Not dramatically. Not with a leap. With the specific quality of a large object rising in the presence of a force that was exactly adequate for the task.
The landing gear came off the tarmac at approximately 0.3 metres per second, which was slower than a typical hover entry and reflected Chakraborty's deliberate control inputs — he was pulling collective at the minimum rate consistent with maintaining altitude, which was the standard first hover protocol: rise slowly, find the hover, hold it, confirm the control responses, do not hurry any of it.
At two metres, he paused.
The chains were slack at two metres. There was no load in them. The machine was flying freely within the safety envelope of the tie-down, and the machines data systems were registering the loads in each chain, and all four chains were reading: zero.
The machine was flying.
In the operations room fifty metres from the test pad, where the data acquisition system displayed the machine's parameters in real time and where Nambiar and Krishnamurti and Pillai and Bhatt and Mehta and the rest of the Bird Room team were watching the displays, the specific moment when all four chain load cells read zero produced a sound.
The sound was the specific sound that forty-one people made simultaneously when the thing they had spent twenty months building did the thing it was built to do.
It was not a cheer. It was more specific than a cheer. It was the sound of people who had been holding something very tightly for twenty months and had just been given permission to relax their grip slightly.
Nambiar looked at the chain load data.
He looked at the rotor RPM: 280. Exactly on target.
He looked at the engine torque: 78 percent on each engine. Well within limits. Adequate margin.
He looked at the vibration levels: significantly below the threshold.
He looked at the fuel flow: consistent with the pre-flight predictions.
He picked up the radio.
He said: "Saras One, operations. Machine is looking clean. How are you feeling up there?"
Chakraborty's voice, from the cockpit above them through the radio, had the specific quality of a pilot who was doing something that required considerable focus and who was also entirely at ease doing it: "Operations, Saras One. Clean controls. The cyclic response is — she's light. She is genuinely light. I'd say very similar to the Lama's cyclic response except that the collective is heavier because the blades are heavier. Which is what we predicted." A pause. "She's very stable. I haven't had to make a pedal input since I lifted. She's just sitting here."
Stability in hover was not something that happened automatically. It was designed. The Saras's stability in hover was the product of Krishnamurti's rotor design and the specific feedback gains in the stability augmentation system — the electronic system that translated the pilot's control inputs into rotor blade pitch commands with the specific filtering that reduced the tendency of helicopter to oscillate in hover.
The SAS was working.
The machine was stable.
Chakraborty said: "I'm going to try the cyclic."
A pause.
"Forward — she moves forward cleanly. Back — she comes back cleanly. Left — right —" A pause. "The lateral response is crisp. I'd say crisper than the Lama at comparable rotor speed." A pause. "She's good. She is a good helicopter."
In the operations room, Nair's co-pilot was watching the CRT display that showed the rotor track-and-balance measurement — the laser system that measured the height of each blade's tip as it passed, confirming that all five blades were at the same track height and balanced in the rotor disc.
All five blades: within four millimetres of track.
This was excellent. Brand new helicopter, first hover, four millimetres of track spread.
Bhatt said, to no one in particular: "She's going to be a good aircraft."
Krishnamurti said: "She already is a good aircraft."
Pillai said: "Wait until she's got some altitude under her."
The first free flight was March 15th.
Two days after the hover, after a full day of engineering review of the hover data, after a check of every system that had been operating during the hover, after a specific and thorough examination of the tail rotor gearbox by three technicians including the one who had replaced the bearing on March 12th, after a weather briefing that confirmed the Gorakhpur weather for March 15th was clear and calm with light surface winds from the south.
The chains came off.
Without the chains, the machine was simply a machine and the ground was simply the ground and the relationship between them was a matter of physics and pilot skill rather than engineering safety.
Chakraborty lifted off at 0934 on March 15th.
He lifted off to two metres, held it, confirmed all parameters, then said: "I'm going to two hundred feet."
The machine climbed.
Not quickly — at 300 feet per minute, the standard first flight climb rate, slow enough that the engine parameters could be monitored through the climb and any anomaly identified before it became a problem.
At two hundred feet above the airfield, Chakraborty held the hover.
He said: "Operations, Saras One. We're at two hundred feet. Machine is clean. I'm going to do a circuit."
A helicopter circuit was not what a fixed-wing pilot called a circuit. It was a low-altitude orbit of the airfield at approximately 200 feet, covering the four quadrants, establishing the machine's handling in forward flight at low altitude and low speed — the specific flight regime where helicopter handling was most complex because the rotor's behaviour at low forward speeds combined with the ground effects and the shielding effects of the airfield buildings and the specific aerodynamic interactions between the rotor downwash and the ground to produce a handling environment that was more demanding than either the hover or the higher-speed cruise.
He flew the circuit.
He flew it at 60 knots, which was below the Saras's estimated best endurance speed and well below its cruise speed, which was the standard low-speed handling check.
Then he flew it at 100 knots.
Then he flew it at 130 knots.
At 130 knots, the Saras was not near its maximum speed, but the 130-knot handling check was the specific speed at which most helicopter handling problems showed themselves: the retreating blade stall that limited the maximum speed began to show its precursors at high cruise speed, the vibration modes that were suppressed at lower speeds sometimes appeared in the 120-to-150-knot range, and the specific interactions between the main rotor wake and the tail rotor became more complex at higher speeds.
Chakraborty flew the machine at 130 knots for four minutes.
Then he said: "Saras One, operations. She is clean at 130 knots. Vibration is very low. The collective response at cruise speed is — she responds immediately. There's no lag. Whatever SAS gains you've dialled in are correct." A pause. "The pedal response at speed is — I want to discuss this in debrief but my initial assessment is that the tail rotor authority at 130 knots is very high. Higher than I expected. It may be more authority than I need, which is not a bad thing but is something I want to check against the spec."
Nambiar made a note.
High tail rotor authority at speed. Expected characteristic of the specific tail rotor disc area that the design had specified for high-altitude operation. Not a problem. Something to note and monitor.
Chakraborty brought the machine back to the airfield and landed at 1012.
Thirty-eight minutes of first flight.
He climbed out of the cockpit, pulled off his helmet, and stood on the tarmac in the March Gorakhpur heat and looked at the machine.
Nair climbed out of the right seat.
Chakraborty looked at the machine for approximately ten seconds.
Then he said to Nambiar, who had walked out from the operations room: "It's the nicest helicopter I've ever flown."
Nambiar said: "You've flown three types."
Chakraborty said: "I know. It's still the nicest one."
Nambiar looked at the machine.
He thought about the twenty months. About Krishnamurti's blade design. About Pillai's engine management algorithm. About the cold tests in Dehradun. About the Ganesh-1 computing hours. About the forty-one people.
He thought about the photograph of the Sarus crane above his desk.
He said: "She's going to do the mission."
The performance testing phase ran from March 16th through March 22nd.
Performance testing was the systematic exploration of the aircraft's capability across its operating envelope — altitude, speed, weight, temperature, engine conditions — and the comparison of measured performance against predicted performance to establish whether the design had achieved its objectives.
The Saras's performance objectives were specific and had been stated in the programme specification document from the beginning. They were:
Service ceiling: 7,500 metres. Maximum continuous power cruise speed at sea level: 165 knots. Hover out of ground effect ceiling with standard mission load: 6,500 metres. Maximum payload at sea level: 1,500 kilograms. Maximum payload at 4,500 metres: 700 kilograms.
These were the numbers that Nambiar's team had been designing toward for twenty months. The performance testing's job was to find out if the design had actually achieved them.
Chakraborty flew sixteen test flights between March 16th and March 22nd.
Some of them were straightforward confirmation runs: climbing to a specific altitude, establishing hover, measuring the power required, comparing it to the prediction. At sea level, the machine performed exactly as predicted. At 2,500 metres, it performed exactly as predicted. At 4,000 metres, it performed exactly as predicted.
At 5,500 metres, it performed better than predicted.
The specific reason: the Kaveri-T's performance at 5,500 metres was 7 percent better than the model had predicted, which was the specific kind of positive surprise that propulsion engineers experienced when their models had been conservative and the actual thermodynamic behaviour of the engine was more favourable than the model assumed. Pillai's prediction model had incorporated conservatism in the combustion efficiency at altitude because combustion efficiency in thin air was a notoriously difficult thing to predict accurately. The actual combustion efficiency was better than the conservative prediction.
At 5,500 metres, the Saras had more power than it was designed to have.
This meant the hover ceiling at 5,500 metres with standard mission load — the load that represented a meaningful operational mission rather than a stripped aircraft with no payload — was not the marginal operation that the prediction had suggested. There was margin. Real margin. Not theoretical margin. Operational margin.
At 6,500 metres, the prediction and reality were essentially identical: the Saras could hover out of ground effect at 6,500 metres with a payload of 400 kilograms, which was the weight of four fully equipped mountain infantry soldiers plus their equipment, which was the specific capability that the Indian Army had specified as the minimum meaningful mission.
Four soldiers. At 6,500 metres. Hovering. In the machine.
When Chakraborty confirmed this — when the data from the 6,500-metre hover performance test came back clean and consistent with the prediction — the Bird Room erupted in the specific way that engineering teams erupted when twenty months of work produced the specific number that twenty months of work had been aimed at.
Not a party. Not champagne. Engineering teams celebrated differently from the general population. The Bird Room's celebration of the 6,500-metre hover ceiling confirmation was: everyone in the room looking at the data for thirty seconds in silence, and then Pillai saying bloody hell, and everyone else agreeing with Pillai's assessment, and then going back to work because the performance testing programme had twelve more flights to complete.
The weapons ground tests.
On March 23rd, while the performance flying was continuing during the day, the weapons integration team ran the Kaal-20's first gun test in the evening.
The gun was mounted in the chin turret. The ammunition was the DEFA-pattern 20mm high-explosive incendiary round that the Indian Air Force used in the Aden gun on the Hawker Hunter and that had been adapted for the Kaal-20's twin-barrel configuration by the DRDO's ammunition programme. The test range was the same tarmac area behind the aeronautics hangar that the engine runs had used, with a cleared firing lane of 500 metres into which a series of steel plate targets had been placed at 100-metre intervals.
The ground test was not a dynamic test — the helicopter was not flying during the gun test. It was on the ground, anchored, with the gun slewed to the firing bearing, the rotor stopped, the engines providing electrical and hydraulic power but not turning the rotor.
The weapons integration engineer, a cheerful man named Subramanian who had spent eight years at DRDO and who had the specific quality of someone who spent their professional life making things explode in controlled ways, ran the first firing sequence at 1800 on March 23rd.
Five rounds.
The sound was nothing like the sound of a single rifle or even a machine gun. The Kaal-20's twin barrels firing at 900 rounds per minute total produced a sound that was closer to a very fast ripping — braaap — than to individual shots, the specific compressed sound of high-rate cannon fire that resolved at the ear into a single acoustic event rather than a sequence.
The five rounds hit the 100-metre target with a group size of approximately 30 centimetres.
Bhatt, who was standing at the edge of the firing lane with hearing protection and the specific expression of a man watching something he had been anticipating for a long time, said: "Again."
Ten rounds.
The 200-metre target: group size approximately 60 centimetres.
"Again."
Ten rounds.
The 300-metre target: approximately 90 centimetres.
The accuracy degraded linearly with range, which was exactly the expected ballistic behaviour of the 20mm round in the environmental conditions.
Subramanian came to Bhatt and said: "What do you think?"
Bhatt said: "I think it works."
Subramanian said: "Is that all?"
Bhatt said: "In my experience, that's the highest praise a gun system can receive from a test programme. When it works, you say it works. When it doesn't, you say considerably more."
The Nishith missile ground test was more complex.
Ground testing an anti-tank guided missile without the helicopter flying required a specific test rig that allowed the missile to be launched on a flat trajectory from the helicopter's hardpoint, tracking a ground-level laser designation to a target at 1,500 metres. The test was not the full envelope test — that required flight — but it confirmed the specific interfaces: the helicopter's fire control system talking correctly to the missile's seeker electronics, the launcher rail releasing the missile cleanly, the missile's initial flight phase being stable on the flat trajectory before the seeker acquired the laser spot.
Two test launches on March 24th.
Both missiles tracked the laser spot and impacted within two metres of the aiming point at 1,500 metres.
The weapons officer who ran the test, sitting in the cockpit of the non-flying helicopter with the fire control system active and the helmet-mounted sight directing the laser designator, described the experience afterward as: "Exactly as simple as it should be. You look at what you want to hit, you designate, you fire, the missile follows the designation. That's it."
That was it.
The mountain trials.
The programme specification required Himalayan altitude performance to be confirmed not just in the standard atmosphere of March at Gorakhpur, which was pleasant, but in the specific conditions of the operational theatre — the cold, dry, thin air of the high Himalayan valleys.
The mountain trial location was Joshimath, Uttarakhand.
Joshimath was the forward operating base for Army Aviation in the Chamoli district, at approximately 1,890 metres altitude, and had the specific characteristic of being a starting point from which flights could be made to significantly higher altitudes — the Auli plateau at 2,500 metres, the Badrinath valley at 3,100 metres, and the routes toward the higher terrain that reached the operational altitudes the Saras needed to demonstrate.
The machine flew to Joshimath on March 26th.
The ferry flight from Gorakhpur to Joshimath was 280 kilometres, which the Saras completed in one hour and forty-one minutes at 160 knots cruise speed. Chakraborty described the cruise flight in his post-flight notes as unremarkable, which was the specific highest praise a pilot could offer a cruise performance: nothing happened, the machine did what it was supposed to do, and nobody thought about it.
The unremarkable ferry flight was the first time the Saras had left the immediate vicinity of the Gorakhpur facility.
The moment when the machine crossed out of the Gorakhpur airspace and began the transit north and west toward the Himalayan foothills was the moment when the programme became real in the geographic sense: the machine was in the world, in the specific world that it had been built to operate in, and the world was receiving it.
The Joshimath conditions on March 27th were the specific conditions of a Himalayan spring morning: clear, cold — 4°C at the airfield at 0700 — with excellent visibility and a light northerly breeze off the snowfields above.
The Bird Room team had come up in a convoy of two Army trucks, because the programme had neither its own ground transport nor any means of arriving at Joshimath in a manner that was less uncomfortable. Nambiar had spent the night in the Army Aviation Base's guest quarters with six other members of the team and had slept adequately given the altitude change and the specific quality of the Army's guest facility bedding.
He had been awake at 0500.
At 0500, the mountain was visible above Joshimath in the specific way that mountains were visible in the Himalayas on clear mornings before dawn: not as peaks against a sky but as dark shapes against slightly less dark sky, the specific mass of them communicated by the way the stars disappeared at the mountain's edge. By 0600, the dawn light was on the snow above and the peaks were white and the sky was beginning to be blue and the Saras was on the Joshimath airfield apron, pre-flighted, ready.
The first mountain altitude test was at 0800.
Chakraborty and Nair flew the machine to 3,500 metres above the Joshimath altitude — which put the aircraft at approximately 5,400 metres above mean sea level — and conducted a hover performance check.
At 5,400 metres MSL, with the mission load of four soldiers' weight equivalent in ballast, the Saras hovered out of ground effect.
Comfortably.
Not marginally. Comfortably. The power margins were 12 percent, which was more than adequate.
The second test was at 4,500 metres above the airfield — 6,390 metres MSL.
At 6,390 metres MSL, with the same mission load: the Saras hovered. The power margin was 4 percent. Tight, but real. The margin existed.
4 percent at 6,390 metres with four soldiers worth of payload.
Bhatt, who was monitoring the radio from the Joshimath apron, heard the altitude report and said: "There it is."
There it was.
At approximately 6,400 metres, at the top of the operational envelope that the Indian Army needed for Himalayan operations, the Saras was carrying a meaningful combat payload and hovering.
No existing helicopter in the Indian inventory could do this.
The Lama had a higher absolute ceiling but the Lama's operational capability at altitude — what it could actually lift and do — was significantly less than the Saras's because the Saras had been designed from the beginning for the mission and the Lama had been optimised for other parameters.
The third test was the most demanding: the high-altitude tactical approach and landing.
Tactical approach in mountain terrain was not the standard approach that a helicopter made to a prepared landing site. It was the specific approach to an unprepared site — a small flat area on a mountainside, a plateau, the specific small flat areas that mountain infantry used for resupply because large flat areas did not exist in mountain terrain. The approach required precision over a confined area with high terrain on multiple sides, often with the helicopter at the edge of its performance envelope, which meant that the margins for error in speed and altitude during the approach were smaller than they were in standard operations.
Bhatt had identified a site above Joshimath at 5,800 metres MSL — a small plateau on the ridge above the valley, approximately 40 by 25 metres of reasonably flat terrain, surrounded by higher ground on three sides and open air to the south.
Chakraborty flew the approach to this site at 1400 on March 27th.
The approach was from the south, into the plateau, decelerating from 80 knots to a hover over the approach corridor with the rocky ridgeline to the north, east, and west.
He described the approach in his notes afterward:
The approach angle was 12 degrees, steeper than standard, necessary to clear the south ridge and have time to decelerate before the plateau. The deceleration was heavier than normal because the thin air gives less rotor braking from the disc loading increase in the flare. I had to start the flare earlier and use more collective during the deceleration to manage the energy correctly. The cyclic response during the deceleration was accurate and the SAS was doing exactly what it needed to do. At the hover over the site the power margins were approximately 6 percent, which is tighter than I'd like operationally but which is within the defined parameters for the mission. The site was entirely manageable. The machine was manageable.
He landed.
On the plateau at 5,800 metres MSL.
The landing gear touched down on the specific rock and thin soil of the Himalayan plateau and the rotor downwash spread out across the flat ground in the specific pattern of a helicopter at rest in mountain terrain, pressing down the thin high-altitude grass, moving the light dust, creating the specific bowl of disturbed air around the machine.
Chakraborty shut down the engines.
The rotor decelerated.
The mountain was quiet.
In the cockpit of the Saras, at 5,800 metres above sea level on a plateau above Joshimath in Uttarakhand, Chakraborty and Nair sat for a moment in the quiet.
Nair said something over the intercom.
Chakraborty said something back.
The data recorder caught the exchange but the radio had been turned off during the shutdown sequence and the Joshimath ground station did not hear it.
What was said between the two pilots on the plateau at 5,800 metres, with the machine they had spent six months learning resting on the specific ground it had been built to reach, was not part of the formal record.
It was between them and the mountain.
The weapons air tests.
The final element of the March programme was the first airborne weapons system test: not a live firing, which the programme was not yet at the stage to conduct, but the first activation of the Kaal-20 gun in flight, traversing under helicopter gun commands, confirming that the turret functioned correctly in the airborne environment and that the helicopter's handling was not adversely affected by the gun's inertia during slewing.
This test was on March 28th, back at Gorakhpur after the mountain trials, because the specific test range instrumentation required was at the Gorakhpur facility and not at Joshimath.
Chakraborty flew the machine at 1,000 feet over the test range. Nair slewed the gun through its full range of azimuth traverse — 200 degrees, 100 degrees to each side — while Chakraborty reported the handling impact.
The handling impact of slewing a gun at 900 rounds per minute from one side to the other on a helicopter was non-trivial: the gun's mass moving at high angular velocity created gyroscopic moments that the helicopter's control system had to manage. In poorly integrated gun systems, this could cause significant handling difficulties during the slewing motion.
The Saras's stability augmentation system had been designed to compensate for the gun's gyroscopic moment.
It compensated correctly.
Chakraborty's report: "Gun slew left to right — minor yaw tendency, corrected automatically by SAS. Gun slew right to left — same, same correction. The pilot workload is zero during the slew. The SAS does it."
Zero pilot workload.
The pilot flew the aircraft. The SAS managed the gun's effect on the aircraft. The weapons officer looked at the target and the gun went there.
The system worked.
The Bird Room's celebration was on the evening of March 28th.
They celebrated with dinner at a dhaba in Gorakhpur — not a special event, the specific ordinary dhaba where many of the team ate regularly because the dhaba was good and the walk from the facility was manageable — and the celebration was distinguished from a normal team dinner only by the specific quality of attention with which forty-one people ate together and talked about what they had built and what the machine had done in the past two weeks.
Nambiar sat in the middle of the table.
He was not a man who made speeches. He had made one speech in the programme's history, on the first day when everyone had arrived, and it had been the specific brevity of a man who believed that the work was the statement and words were the preamble to it. He had said, on that first day in November 1973: we are building a helicopter that India needs. The specifications are on the wall. When the machine does everything on the specifications list, we are done. Until then, we are not done.
On March 28th, 1975, the machine had done the following things: hovered out of ground effect at 6,390 metres MSL with a meaningful payload, flown at 165 knots at sea level, landed on a 40-by-25-metre site at 5,800 metres MSL, demonstrated the gun system integrated with the stability augmentation system to zero pilot workload, and completed the Nishith missile integration ground test within two metres of accuracy.
The specifications list was not done. The programme was in the prototype testing phase and there were months of work remaining: the full envelope expansion, the high-speed handling tests, the live weapons firing, the instrument meteorological conditions qualification, the engine-out handling tests, the transmission faults tests, the vibration survey at all altitudes and speeds, the structural loads survey.
The programme was not done.
But the programme had demonstrated, in the fourteen days since the first hover, that the machine was real and that it was capable.
Nambiar raised his glass of chai.
He said: "The Saras flew."
And the Bird Room — forty-one people who had spent twenty months building a thing, and who had spent the past fourteen days watching the thing do what it was built to do — raised their chai glasses and said: "The Saras flew."
Karan was not at the March 28th dinner.
He received the test report in Gorakhpur the following morning — the formal summary that Chakraborty and Nambiar had prepared together, covering every test, every measurement, every observation. He read it at his desk at six in the morning, with his chai, in the specific early morning that was his standard reading time.
He read it carefully.
He read the 6,390-metre hover data.
He read the Joshimath plateau landing.
He read the gun system integration results.
He read Chakraborty's description of the mountain approach: steep, manageable, tight margins, within parameters.
He set the report down.
He opened his notebook.
He wrote: March 29, 1975. Saras prototype — first flight March 15th. Mountain trials March 27th. Hover at 6,390m with payload. Gun system integrated, zero pilot workload.
He paused.
He added: Named for the Sarus crane. The Sarus crane is found in abundance in Gorakhpur. The Sarus crane flies at altitudes that its large wingspread makes possible. The machine named for it was built to do the same.
He added: The machine does it.
He looked out the window at the Gorakhpur complex.
He thought about the Bird Room and its forty-one people and the photograph of the Sarus crane above Nambiar's desk.
He thought about the specific problem that the machine solved: the Indian Army in the mountains, needing helicopters that could reach the altitude where the soldiers were and the altitude where the enemy was and carry a meaningful load to both places and fight effectively in both places and come back.
He thought about the specific gap between what had existed and what the machine filled.
He thought: The Army has needed this for thirty years. Since 1947, since the first Himalayan operation, since the first time an Indian soldier looked up at a mountain and understood that the aircraft that were supposed to support him could not reach him.
He thought: They can reach him now.
He wrote: Production programme decision required. The prototype has demonstrated the capability. The transition from prototype to production requires the specific decisions about manufacturing volume, unit cost, and delivery timeline.
He paused.
He wrote: The unit cost at volume production — Nambiar's estimate is ₹68 lakh at fifty units per year. At one hundred units per year, ₹52 lakh. The Army will want 150 units in the first three years. The Navy will want the naval utility variant. The Air Force will want the search and rescue variant.
He wrote: Three services. One machine. Different configurations. This is the production programme.
He closed the notebook.
He picked up the phone. He didn't need to dial; he knew the extension by heart.
Nambiar answered on the second ring.
"Good work, Rajesh."
There was a long silence on the line—the sound of a man who hadn't slept in forty-eight hours finally letting his guard down.
"She's a good helicopter, Karan," Nambiar said, his voice raspy. "She's better than we had any right to expect."
"I know. Prepare the production programme document. I want it on my desk in four weeks."
"Four weeks?" Nambiar's exhaustion vanished, replaced by the instinctual pushback of an engineer. "The testing programme isn't even halfway through. We still have eight months of envelope expansion to clear."
"Do both in parallel," Karan said, his tone leaving no room for negotiation. "We have run projects in parallel before. This won't be the first."
"Karan, be reasonable. The Saras is a high-performance machine; it isn't finished being tested."
"When," Karan asked, "has a good machine ever been finished being tested? The army needs the bird, Rajesh. If we wait for the perfect test results, we'll be delivering a museum piece instead of a weapon. Four weeks."
There was a heavy pause from Nambiar. He was looking at the data, looking at the mountain, looking at the cost of the last twenty months.
"Four weeks," Nambiar conceded.
"Four weeks," Karan confirmed.
He put the phone down. The office was quiet. He walked to the window and looked out at the factory floor, where the skeletons of the next prototypes were already taking shape. He thought about the Saras sitting cold and silent on that plateau above Joshimath, the silence of the Himalayas pressing against the glass. He imagined the two pilots inside, sitting in the dark, knowing they had touched the edge of the world.
The Sarus crane flies at altitudes that its large wingspread makes possible.
The machine is the same.
We built it correctly.
He checked his watch. It was time for the seven o'clock S-35 briefing.
The helicopter programme had its next test flight scheduled in three days. The testing would continue, the data would keep flowing, and the production lines would start to spin. The work never ended; it only shifted gears.
End of Chapter 195
Project Saras — First Flight Programme Summary
Vehicle: Saras Attack/Utility Helicopter, Prototype 1 Named for: Grus antigone (Sarus Crane), world's tallest flying bird, found in abundance in Gorakhpur district, UP
Team (Bird Room, Northwest Annex, Shergill Aeronautics):
Dr. Rajesh Nambiar: Programme Director / Chief Designer
Wing Commander Pradeep Sinha (retd): Deputy Director / Mountain Operations Lead
Dr. Ananya Krishnamurti: Rotor Aerodynamics
Raghunath Pillai: Propulsion Integration
Sunita Mehta: Systems Integration (listed as Systems Analyst)
Lieutenant Colonel Arjun Bhatt (retd): Mountain Operations Consultant
Squadron Leader Arvind Chakraborty: Chief Test Pilot
Flight Lieutenant Kiran Nair: Test Weapons Officer/Co-pilot
Subramaniam: Weapons Integration Engineer
Total team: 41
Propulsion:
2× Kaveri-T turboshaft
Sea level rating: 1,540 shp each (3,080 shp total)
6,500m rating: 820 shp each (1,640 shp total)
Power retention at 6,500m: ~53% (significantly better than contemporaries)
Rotor:
5-blade composite main rotor
Blade material: SPEI carbon fibre composite
Blade profile: Himalayan (12% greater blade area than conventional same-span blade)
Rotation: 280 RPM (governed)
Track and balance (first hover): within 4mm across all 5 blades
Armament:
1× Kaal-20 20mm twin-barrel autocannon, chin turret, 200° azimuth × 30° elevation, 500 rounds, helmet-mounted sight slaved (100 Hz update rate)
4× Nishith laser-guided anti-tank missiles (semi-active laser, tandem shaped charge, 500m-7,000m range)
2× Vrishti 70mm rocket pods, 12 rounds each
Provision for 2× Shefali short-range IR air-to-air missiles (missile available 1977)
Avionics:
ISMC-derived engine management computer (16-parameter scan, pre-emptive fuel scheduling)
Inertial navigation with Doppler radar altimeter, mountain-terrain signal processing (position accuracy: <50m)
Helmet-mounted sight for cannon and missile designation (100 Hz, matches Kaal-20 update rate)
Performance (prototype, March 1975):
Service ceiling: Demonstrated to 6,390m MSL with 400kg payload
Cruise speed at SL: 165 knots
Hover out of ground effect at 6,500m: Within specification
Payload at SL: 1,500 kg
Payload at 6,500m MSL: 400 kg (4 fully equipped mountain infantry)
Key test dates:
First tie-down run: March 10, 1975
First hover (10m, chains): March 13, 1975
First free flight: March 15, 1975
Mountain trials (Joshimath): March 27, 1975
Plateau landing (5,800m MSL): March 27, 1975
Gun system air test: March 28, 1975
First flight duration: 38 minutes Total flight hours to March 28: 22.4 hours
Chakraborty assessment (first free flight): "She's the nicest helicopter I've ever flown." Nambiar, March 28: "The Saras flew." Karan's notebook, March 29: The machine does it.
Next programme phase: Full envelope expansion, live weapons firing, IMC qualification, engine-out handling, structural loads survey. Duration: 8 months. Production programme document: Due 4 weeks.