Artemis 1 – what happened?

Artemis 1 Artemis-sed its launch window yesterday.

Proudly standing erect on Launch Pad 39B at the Kennedy Space Centre, ready for its 42-day spin around the moon and back, Artemis 1’s launch was scrubbed, one minute into the 2-hour launch window.

Depending on when you tuned in to NASA’s live feed of the event, you might have heard a whole lot of fancy-sounding stuff about bleed lines, cracks in flanges, slow and fast fill of LOX and LH2 and lightning storms.

Let’s back track and break it down.

Lightning: It’s not great when you’re playing golf; it’s even worse when you want to fill your rocket tanks with some 890,000-odd million litres of liquid oxygen and almost 2.5 million litres of liquid hydrogen.

The first hiccup for Artemis 1’s launch occurred in the wee hours of the morning before the launch (about 7 hours before the launch window opened), when a small weather cell, replete with lightning, developed within 5 nautical miles off the Florida coast. NASA’s launch protocol stipulates that the chance of lightning must be 20% or less in the first hour of tanking – that is, filling the fuel tanks with propellant. So, the fuelling team had to delay.

Lightning strikes a tower, missing artemis 1 at launch pad 39b
Lightning strikes a tower, missing Artemis 1 at Launch Pad 39B. Credit: NASA Kennedy Space Centre

Lightning strikes do happen.

In fact, there were three strikes just two days prior to this first launch attempt. Luckily, Tower 1 and Tower 2, part of the three-tower 181 metre-tall lightning protection system at Launch Pad 39B, did their job magnificently, funnelling the electric current diagonally down and away from the launch pad perimeter.

It’s not just strikes on the launch pad which worry NASA – just like aeroplanes, rockets can be struck by lightning while in the air. A rocket launch won’t go ahead if there’s a probability of lightning within 10 miles of the flight path, which includes the cloud producing the lightning.

Lightning Strikes at NASA’s Kennedy Space Centre, Launch Pad 39B. Credit: NASA

Luckily, the bad weather passed – allowing the launch sequence to continue.

A too-full purge can

Fuelling teams began to fill the rocket’s core stage tanks. Liquid oxygen – or LOX – cooled to -183°C was added in “slow fill” phase to the smaller tank sitting at the top of the core stage, while the hydrogen fuel lines for the larger, lower tank were cooled to the frigid -253°C needed for liquid hydrogen filling. After chilling, teams began filling the tank with hydrogen propellant on “slow fill” while transitioning the LOX fuelling to “fast fill”.

Artemis 1 - core stage expanded
[Click to enlarge]
Diagram of Artemis 1 with core stage expanded. Credit: NASA

Read More: How do rockets and boosters work

The LOX tank filled merrily, taking on more and more liquid oxygen, with some boiling away to gas and venting out the sides as designed, but the hydrogen tank proved to be more fickle. When they transitioned to fast fill, engineers noticed excess hydrogen leaking out into the purge can – which is designed to cover the spot on the rocket the fuel line enters (these connectors are also known as umbilicals).

Engineers eventually solved the problem by manually cooling down the liquid hydrogen, so they could resume fast fill operations.

The little engine that wouldn’t bleed

Artemis 1’s Space Launch System (SLS) gets its grunt from four RS-25 liquid-fuel cryogenic engines sitting at the base of the rocket. Otherwise known as the Aerojet Rocketdyne RS-25, these are tried and tested rockets, having been used throughout NASA’s Space Shuttle program. For the four RS-25 engines on Artemis 1, this will be their final service, as the core stage will break apart on re-entry, falling to the bottom of the Ocean, making recovery too difficult.

Artemis 1's four rs-25 engines
The four RS-25 engines on Artemis 1. Engine position 3 is back right. Credit: NASA/Cory Huston

These four engines combine to produce around 4 million kilograms of thrust, making the SLS the most powerful rocket so far – although it won’t hold this title for all that long as future configurations will deliver even more power.

These engines need to be cooled from the ambient temperature – around 27°C in Florida at this time of the year – to a working temperature close to that of the cryogenic fuels, so they can be started up. This is done through a step in the launch process called an engine ‘bleed’, through which pressure is increased on the core stage tanks (which are now being constantly topped up as they boil off some of their fuel) to enable them to run some of their cryogenic liquid hydrogen propellant through the engines.

Engine bleeds to engines in positions 1, 2 and 4 functioned as anticipated, but the bleed to position 3 – to an engine known as Engine 2058, which has been part of six previous shuttle missions – did not work properly, meaning the engine was not being cooled enough.

This would have been caught in testing when the rocket was last on the launch pad in June for what’s known as a “wet dress rehearsal”, but a leak was detected in one of the lines related to this engine bleed, stopping the dress rehearsal before the bleed tests began.

Engineers tried a couple of ideas, eventually attempting to force the bleed by shutting down engines 1, 2 and 4, but to no avail. They couldn’t fix the issue in time, so the mission was scrubbed one minute into the launch window.

What next?

Engineers will process the data coming from the launch sequence, looking to come up with and implement a fix before the next two-hour launch window, which will be in on September 2. They will also be looking to gather information on and confirm what looks like a crack in the thermal protective foam on a flange on the core stage. These are connective joints that enable the connection of the core stage’s fuel tanks to the body of the rocket.

The report for Friday, however, suggests only a 40% chance of fine weather. The next launch option after that is a 1.5-hour window on September 5. If this is also missed, Artemis 1 will need to wait for one of the launch windows between Sept 19 – Oct 4 or October 17-31.

There are launch dates after this, but let’s hope we don’t have to wait that long!

Please login to favourite this article.