Inside India’s quest to 3D-print parts for space missions

Pankaj Mishra October 5, 2017 10 min

India’s space agency Indian Space Research Organisation, sent its maiden 3D-printed satellite part, a radio antenna, in the outer orbit in June this year, and is now planning to leverage the next-gen manufacturing for adding more muscle against rivals in the global space war.

For ISRO, which has been looking to break free from its legacy, cumbersome manufacturing processes, 3D printing means not just lower costs because of reduced weight of the components, but more importantly, a much faster turnaround time. In the ongoing global space war, which saw Isro launch 104 satellites in one go in February this year, the most in history, 3D printed components can potentially add more advantage beyond just lower costs.

To be sure, other space agencies such as NASA have long been experimenting and sending 3D-printed parts in the space. In September this year, for instance, NASA tested its first 3-D printed rocket engine part made with two different alloys. In December last year, The American space agency also tested parts of the combustion engine of a rocket, including fuel turbopump, fuel injector and valves–all 3D printed.

Last month, I travelled to Ahmedabad to sit down with Sukhjeet Singh Gill, a scientist and engineer at ISRO. He’s been leading the organisation’s next-generation manufacturing experiments including 3D printing for the last six years or so. Gill, who’s been working with ISRO for over two decades in Ahmedabad, has also published a paper titled “On the development of Antenna feed array for space applications by additive manufacturing technique” earlier this month in the Journal of Additive Manufacturing, Elsevier, Science Direct.

Sukhjeet Singh Gill, Scientist-Engineer, Space Application Center, ISRO

Isro’s baby steps in 3D

Unlike conventional manufacturing, where bulk iron and other metal parts have to be cut and designed as per the requirement, additive manufacturing uses a metallic powder that gets accumulated on a tray and eventually gets solidified to give a single, ready to use component. In traditional manufacturing, different parts are designed and cut separately, and then joined together using nuts and bolts, or welding techniques.

Around six years ago, when Gill first heard that complex metals were being 3D printed, he got excited about the prospects. Later, he identified a partner in Germany and started experimenting with different metallic shapes, helping ISRO take baby steps in the additive manufacturing.

“My department is responsible for all mechanical requirements of the centre, all the manufacturing and (related) processes are run through my department. I keep on looking for new technologies that come into the field, not just 3D printing,” he says. The German partner shipped some 3D printed components to ISRO but nothing functional yet.

“They gave us some samples to be evaluated by our quality assurance department, mostly mechanical prototypes. We found it to be promising and realised this (3D-printed, simple machine parts) could be used.”

The next and the most difficult question was about the surface finish of the printed parts, which didn’t look good at that time. ISRO was looking for a Roughness Average (RA), of at least about 2 or 3, but the printed parts were offering 12. RA is a measure of the surface roughness of a metal and determines how smooth or rough the texture is.

“But we still decided to keep pursuing hoping that with time, it will shape, and that’s what happened eventually,” he says.

Why 3D print at all?

While experimenting with new technologies was an overall driver for Gill to explore 3D printing satellite parts, it wasn’t the reason enough. For additive manufacturing or 3D printing to make real business sense, there were other questions that triggered its adoption at ISRO.

“There are few components that cannot be manufactured in a single piece using traditional machines. Some parts have their tops and bottom machined separately, then joined together to form a single unit using nuts and bolts. Now, these are microwave components…RF components….when there are joints, there are leakages too,” says Gill.

They asked themselves: Can we reduce the number of joints? Can we make such parts in a single unit?

“Another thing was when a plumb line (through which the microwave signals travel) for a satellite goes, there are many components. These parts had to be joined not just horizontally, but also vertically, increasing the number of joints overall. Joints are the cause of leakage,” he says. Joints add to the weight. And each gram of additional weight costs dearly.

Also see: This Bengaluru startup has developed a satellite propulsion technology that can make space missions cheaper

“We decided to go for 3D printing so that we can print few parts, clubbed together without joints, and also reduce the weight,” says Gill.

Getting these jobs done from Germany involved several regulatory clearances each time, and it was overall cumbersome. We then learnt that Wipro had the machine in India that could potentially print out the parts we wanted.

“It was an EOS machine…I then contacted Wipro.” EOS is one of the world’s largest makers of industrial 3D printers.

“We started with a big one (3D printed part).”

“Back then (4 years ago), I was very sure that this will not give the surface finish we were looking for, apart from the dimensions. So, if i made smaller parts, the percentage of errors will be more too. That’s why we went for “the X-band twist”. When frequency increases, your job gets smaller, it’s inverse.”

“The part we made is called ‘a twist.’ It changes the polarisation of the radio waves.”

In simple terms, the component is an antenna that enables the exchange of radio waves.

“The first one we printed wasn’t as good.”

“Our chairman said we should do jobs where machining is not required…..once you 3D print a part, it shouldn’t need to be worked upon again because then it defeats the purpose by adding efforts and costs.”

“I took the job of printing horns, which are difficult to manufacture through conventional machines. On a given tray, there are four horns and another four on top of each of them.”

Can they fly?

Notwithstanding the process of making them, each of the components meant to fly in the space needs to go through long, ruthless tests. Before any component can be inducted into any space mission, it has to go through layers of strenuous tests. These tests include testing of a component’s metallic strength, ability to work with other parts and above all, their space-readiness.

Vibration test is just one among several such hurdles to be passed.

“The sceptics said 3D printed parts can break apart when put through the long flight in the space because it’s “a layer by layer” technology,” recalls Gill.

“All these RF components need to be silver-plated for better conductivity. I had to get it qualified for that, post which the component went for microwave test. Then, it was showcased in different exhibitions, but it was still a non-functional experiment.”

Gill had to wait until June this year, to finally see a 3D printed antenna in the outer orbit, when Indian communication satellite GSAT 19 was launched in the space. But the preparations started over a year ago.

“When GSAT 19 came, the requirement was for a 2/2 horn and no 4/4, which we had. It was bigger. I threw the challenge at Wipro, they said it’s not possible because the size of their 3D machine was smaller. They offered to make two separate units, but that would again mean joining them together much the same way as done in conventional manufacturing.”

“They modified their machine to accommodate the height of the machine, which was only around 300 millimetres.”

“Also, we had to change the design from the start because it was earlier meant to be for conventional manufacturing. You have to design for 3D separately, which we did along with Wipro folks.”

For Wipro3D, the additive manufacturing division of the multi-billion dollar oil to software conglomerate, the assignment meant a chance to prove its capabilities on the national stage.

Also see: India needs manned missions to space. Or we’ll lose out to China in the race for resources: U R Rao

The component with a height of approx 320 mm and wall thickness of 2 mm posed significant challenges when it comes to realization through current standard metal AM (additive manufacturing) systems,” says Pratik Kumar, CEO of Wipro’s infrastructure engineering division. 

Isro engineers had to redesign the antenna from start keeping the 3D manufacturing process in the mind.

“It took 3-4 months because, with every change, our design engineers had to recheck the component for its frequency. This process started sometime in May last year.”

“After printing, it was again tested for all the things–for vibrations, microwave and so on.”

“Then, the director said, it can fly.”

“Now it’s working in the space and the performance is excellent I am told. It’s no different from how a conventionally built part would have performed.”

It is the antenna that receives and transmits the radio signals, linking space to the earth.

“First thing is to have our designers to start thinking from 3D point of view. We are teaching, explaining limitations.”

“When the jobs are simple, we should not attempt to 3D print because the costs are high. Also, it helps if the job is very intricate because such parts are difficult to build through conventional manufacturing.”

Another important aspect is time saved because it equals real savings. Amid all the design changes, time becomes even more crucial because just fabricating these parts can take 2-3 months if done the traditional way.

“If I do the same component through the conventional process, then I will have to make four separate horns, make the bottom and top plates and assemble them, inspect each part separately. Here, in 3D, it all happens in a single setup. Also, the parts that were not required, were merged together in the 3D printed part to save weight. That can be only done in 3D.”

In this case (the antenna for GSAT 19), at least 112 hours of machining were required for assembly alone. “Here, we’ve done this in only 50 hours, and that too because this is the first time. We will crunch this even further from the next time,” says Gill

What else can be 3D printed for space?

Most of the support structures needed in large volumes, such as brackets are future candidates (for 3D printing), says Gill. In 3D printing, you can have any features printed, anything you can design on a paper or your screen. “And that too, with around 30-40% weight reduction. Brackets are the structures on which antenna and other stuff are mounted, so we are trying to tackle that next,” he said.

“Quick delivery, intricate shapes and weight savings. These three put together will give real benefits.”

Also see: ISRO, it is time you outsourced to get to the next orbit

Some intricacies are almost impossible to make traditionally. Take, for instance, a wall wherein you need to insert a hexagon or a pentagon shape.

Already, according to Wipro3D, private space enterprises, state-run-space organizations, and other members of space industrial ecosystems are replacing existing and conventional geometries, with creative and complex designs using 3D printing.

“Components such as antennae, waveguides, brackets, thrusters, main oxidizer valves, combustion chamber liners, and propellant injectors, additively manufactured, are either in the prototyping stage or are actually flying,” says Ajay Parikh, business head of Wipro3D. 


Disclosure: FactorDaily is owned by SourceCode Media, which counts Accel Partners, Blume Ventures and Vijay Shekhar Sharma among its investors. Accel Partners is an early investor in Flipkart. Vijay Shekhar Sharma is the founder of Paytm. None of FactorDaily’s investors have any influence on its reporting about India’s technology and startup ecosystem.