Cryogenic setback

WHAT went wrong with the GSLV-D3? This question will haunt the Indian Space Research Organisations (ISRO) rocket engineers, especially the cryogenic engine technologists, for some time to come. On April 15, the three-stage Geosynchronous Satellite Launch Vehicle-Development flight 3 (GSLV-D3) lifted off from the spaceport at Sriharikota in Andhra Pradesh at 4-27 p.m., with an indigenous cryogenic engine for the first time in the third, upper stage. The engine, which was in the making for 19 years, was to inject a 2,220-kg satellite, GSAT-4, into a geosynchronous transfer orbit (GTO) with a perigee of 170 km and an apogee of 36,000 km. However, the cryogenic engine, and with it the mission, failed.

Cryogenic engines are a class apart in terms of complexity and performance. Of all types of rocket propulsion, cryogenic technology is the most complex to develop, says S. Ramakrishnan, Director (Projects), Vikram Sarabhai Space Centre (VSSC), Thiruvananthapuram. Cryogenics is the science of dealing with liquids at very low temperatures. In launch vehicles, it involves the use of liquid hydrogen as fuel at -253 Celsius and liquid oxygen as oxidiser at -183 Celsius. A cryogenic engine is essential to put heavy communication satellites, weighing more than 2,000 kg, into a GTO.

The technology is highly guarded. The United States, Russia, Europe (the European Space Agency, or ESA), Japan and China are the only ones to have acquired it so far. In fact, the U.S. even put pressure on Russia to renege on its agreement with India to transfer the cryogenic know-how. Five GSLV flights between 2001 and 2007 were powered by Russian cryogenic engines. A successful GSLV-D3 flight would have made India independent in all aspects of launch-vehicle technology, giving it the capability to put any satellite in any type of orbit low-earth, polar sun-synchronous or geosynchronous transfer orbit.

Mission Director G. Ravindranath told visiting journalists in Sriharikota on April 9 that the mission was crucial because we are testing our own cryogenic engine in the flight. It had come about despite technology denial regimes, he added. He called the 49-metre-tall GSLV-D3, weighing 420 tonnes, the most reviewed vehicle. The flight was scheduled to take place in December 2009, but it had to be postponed to allow an in-depth review of the vehicle by a national committee consisting of veterans in rocket technology, former ISRO chairmen and academics.

Mohammed Muslim, Project Director, Cryogenic Upper Stage Project (CUSP), said ISRO took no chances with the flight and reviewed the vehicle point by point.

Prior to the launch, everything went as planned. The GSLV-D3 was to lift off at 4-27 p.m. There was no hold at all in the 29-hour countdown that began at 11-27 a.m. on April 14. The entire flight was to last 1,022 seconds, during which the cryogenic engine was to fire for 720 seconds and inject GSAT-4 into orbit at a velocity of 10.2 km a second.

At 4-27 p.m., the vehicle rose from the launch pad and sped into the sky, riding on plumes of flame and smoke. The four liquid strap-on booster motors surrounding the core first stage ignited on time. The first stage, fired by solid propellants, then came into play even as the strap-on booster motors fell away. The first stage jettisoned 151.6 seconds after lift-off. The second stage, powered by liquid propellants, ignited at 151.7 seconds. The payload fairing, which protects the satellite and the vehicles electronics from intense heat during the vehicles ascent into the atmosphere, separated down its seams and fell into the Bay of Bengal 228.8 seconds from blast-off. The second stage shut down at 293 seconds. The cryogenic engine ought to have ignited at 304.9 seconds and burned for the next 720 seconds to provide the necessary velocity to inject GSAT-4 into the intended GTO.

As the engineers scanned the plot-boards in the Mission Control Centre (MCC), their faces turned ashen. It was clear that the cryogenic engine had not ignited. The rocket was losing velocity and altitude, and soon it veered off its path and plunged into the sea. As the vehicles trajectory fell within the safety corridor, the Range Safety Officer did not press the destruct button to destroy the rocket.

Within minutes, ISRO Chairman K. Radhakrishnan addressed those present at the MCC. The performance of the vehicle was normal up to the burnout of the second stage, and the vehicle was travelling at 4.9 km a second. The cryogenic engine ignition occurred as planned, he said, but added: This needs to be confirmed after a detailed analysis of the data. According to Radhakrishnan, the vehicle was seen tumbling and losing control, most probably because the two vernier engines [in the cryogenic engine] could not ignite. (The vernier engines, also called steering engines, control the roll, pitch and yaw of the vehicle in flight.)

A detailed analysis of the flight data would be done to find out why the vernier engines did not ignite, he said. We will put all our efforts to ensure that the next GSLV flight with an indigenous cryogenic engine takes place in a year from now. We need to go a long way. We will do that. But it will be tough, Radhakrishnan said.

About an hour later, when he addressed a press conference at Sriharikota, he blamed the non-ignition of the cryogenic engine for the missions failure. This assessment was made on the basis of telemetry data received for about 30 minutes after the flight. Radhakrishnan said: We are not sure whether the main cryogenic engine did ignite. We have to confirm this after looking at various parameters that were monitored during the flight. The vehicle was tumbling. It means it lost control and altitude. Finally, it splashed into the sea.

P.S. Veeraraghavan, Director, VSSC, blamed a problem in the start-up cycle of the cryogenic engine for the missions failure. ISROs rocket engineers are perplexed why the cryogenic engine failed to ignite despite successful testing for about 7,776 seconds at the Liquid Propulsion Systems Centre (LPSC) at Mahendragiri near Nagercoil in Tamil Nadu.

On November 15, 2007, the cryogenic stage, which includes the engine, fuel tanks, propellants and electronics, had passed the qualification test when it fired for the full duration of 720 seconds.

Of five GSLV flights so far, the first ended in failure. The computer aborted the launch, on March 28, 2001, one second before the lift-off because one of the four liquid strap-on motors surrounding the core first stage, did not develop enough thrust. The vehicle was saved. On April 18, 2001, the Russian upper cryogenic engine underperformed, resulting in the satellite going into a lower orbit than planned. The fourth GSLV, on July 10, 2006, deviated from its path after one of its liquid strap-on motors failed to ignite, and the vehicle was destroyed.

ISROs hopes were riding high on this flight achieving for the GSLV the level of maturity and reliability that the Polar Satellite Launch Vehicle (PSLV) commands. The last 15 PSLV flights have been successful.

We are definitely disappointed, said Ramakrishnan, who has been appointed chairman of the Failure Analysis Committee (FAC). He, however, argued that the failure was nothing unusual because several countries, including France and the U.S., had gone through such failures.

Cryogenics is a very complex technology which cannot be fully tested on the ground, he said. The cryogenic propellants, that is, liquid hydrogen and liquid oxygen, exist in liquid form only at extremely low temperatures. To test the cryogenic propellants on the ground, simulating the vacuum of space and the cold conditions there [space], is a complex job. We can test certain things only in flight. For example, the ignition of the cryogenic engine in the vacuum [of space]. Radhakrishnan asserted that the ignition of the cryogenic engine in vacuum could not be simulated on the ground.

S. Satish, Director of Publications and Public Relations, ISRO, explained: The cryogenic engine has to be ignited only in the vacuum of space. The vacuum cannot be simulated because of the extremely low temperatures of liquid hydrogen and liquid oxygen on the ground, which, when put on fire, rise to about 3,000 Celsius.

There is another angle to the ignition of the cryogenic propellants. While the solid propellants in the first stage and the liquid propellants in the second stage are hypergolic, that is, when they come together they burn, the cryogenic propellants are anergolic, that is, they need an external ignition source. Ravindranath explained that in a cryogenic engine a small rocket motor generated a flame, which, in turn, produced heat and gave combustion to the liquid hydrogen and liquid oxygen. The flame was provided by a pyro technique with a specific time-delay. It is very tricky. That is why the technology is so guarded, Ravindranath said. The combustion process is also tricky. It should be sustained for 720 seconds.

ISRO specialists were sceptical about news reports that claimed the cryogenic engine had ignited for a second but that the turbo-pump that supplied fuel to it had stopped working. The reports quoting an unnamed source also claimed that the rockets acceleration had increased for a second before it veered off its path.

A top ISRO engineer was emphatic that there was no ignition at all of the main cryogenic engine. He added: It is a matter of timing.

It hardly matters whether the engine ignited for one second or more, said Satish. What was important was that the ignition was not sustained. Prima facie, it looked as if the cryogenic engine had not ignited. Ramakrishnan declined to comment on the claim that the engine had ignited for one second. These are all observations, he said.

ISRO began work on the cryogenic engine in the late 1980s, when it tested a single element injector generating a 60 kg thrust. In January 1991, ISRO signed a Rs.235-crore contract with Glavkosmos of Russia for buying cryogenic engines and the transfer of cryogenic technology to India. But the U.S. pressured Russia in 1992 and 1993 not to transfer the technology, claiming that it would violate the Missile Technology Control Regime. As Professor U.R. Rao, former ISRO Chairman, asserted, First, nobody uses cryogenic engines for missiles. Anyway commercial motives are behind all these. (Space India, an ISRO publication, October 1993-March 1994).

India took up the challenge to develop a cryogenic stage. It was a multi-disciplinary task. Handling, storing and pumping liquid hydrogen at -253 Celsius and liquid oxygen at -183 Celsius demanded advanced technology because they were volatile. It was not easy to develop fuel tanks that could withstand such low temperatures. Metals become brittle at such freezing temperatures. So new alloys, welding techniques, lubricants and special insulation techniques were developed.

Besides, liquid hydrogen and liquid oxygen should be pumped into the engine by a turbo-pump, which should run at a very high speed of 40,000 revolutions a minute. The project took a long time because we had to develop a lot of special materials. That is how we got delayed, explained Ravindranath. According to Asir Packiaraj, deputy general manager, LPSC, as many as 1,500 intricate welds were done in the cryogenic stage, which was 8.50 metres long and had a diameter of 2.90 m.

The FAC may take some weeks to submit its report. As Radhakrishnan says, ISRO indeed has a long way to go before it launches a GSLV with its own cryogenic engine in a year from now.