The first one failed to reach orbit. The second died soon after getting to space, when its helium coolant was accidentally dumped. The third one lasted for 37 days before its spacecraft broke apart in a fatal spin.

The Japan Aerospace Exploration Agency (JAXA) is hoping the fourth time is the charm for a revolutionary x-ray instrument that will give astronomers unprecedented views of the hot gases around supernovae and black holes and within galaxy clusters. On 26 August, the agency plans to launch the X-Ray Imaging and Spectroscopy Mission (XRISM), a telescope fitted with a NASA-developed device to do something that has long been a challenge for x-ray telescopes: Tease apart the x-rays’ wavelengths, the way a prism splits visible light. Detailed x-ray spectroscopy will allow researchers to not only see the hot gases, but also discover what they’re made of and how they’re moving. “It’s a completely new kind of detector,” says astrophysicist Poshak Gandhi of the University of Southampton.

Because Earth’s atmosphere blocks x-rays, astronomers must go to space to see them. Even there, making x-ray images is a challenge. Because x-rays go straight through conventional mirrors, the photons must be gathered and focused by glancing reflections off nested cylindrical mirrors. X-ray telescopes using that approach can image the hot gases, which make up more than half the visible matter in the universe. But astronomers want more, Gandhi says. “We really need to be able to distinguish the different colors of x-ray light.”

Regular spectrometers struggle with xrays, especially high energy photons from extended sources. But in the 1990s, engineers at NASA’s Goddard Space Flight Center developed a chip-based sensor called a microcalorimeter, which can measure the energy—closely related to wavelength—of individual x-ray photons. When an x-ray strikes one of the mercury telluride pixels in the calorimeter, it knocks loose an electron and transfers all its energy to it. The electron bounces around the pixel, raising its temperature by a tiny fraction of a degree and warming an adjacent temperature sensor. To register these tiny quantities of heat, which indicate the energy of the original photon, the whole device must be cooled to 1/20 of a degree above absolute zero.

Astronomers got a taste of a microcalorimeter’s capability when one flew aboard the ill-fated Hitomi telescope, JAXA’s third try. Before the 2016 mission broke up in the fatal spin, it made groundbreaking observations of the Perseus galaxy cluster and a handful of other objects. “We glimpsed the promised land, but could not go in,” says Brian Williams, NASA’s project scientist for XRISM. In those few weeks, Hitomi did “transformational science with one single pointing,” says Elisa Costantini of the Netherlands Institute for Space Research, principal investigator for the Swiss-Dutch contributions to XRISM. “It proved how necessary it was.”

The Perseus cluster is one of the most massive objects in the universe, a conglomeration of thousands of galaxies swimming in a sea of gas heated to 50 million K. With Hitomi’s microcalorimeter, researchers were able to see unprecedented details in the gas’ x-ray glow. They detected spikes in its spectrum that revealed specific elements, such as iron, indicating what types of supernovae had spewed their heavy elements into space. To their surprise, the chemical recipe was very similar to that of the Sun. It looked “strangely familiar,” Costantini says.

Researchers also saw some of the spikes were smeared out by the motions of the gas. But not by much: The cluster’s gas was surprisingly quiescent, not the maelstrom theorists had predicted. One of XRISM’s first tasks will be to look at other clusters to see whether Perseus is an oddity or the norm. “That’s the homework left by Hitomi,” says Makoto Tashiro of Saitama University, XRISM’s principal investigator.

In addition to galaxy clusters, XRISM will study the hot gases swirling around supernovae remnants and black holes, both the supermassive ones in galactic centers and stellar mass ones that suck material from companion stars. “It’s dead easy to find black holes in the universe with an x-ray telescope,” Gandhi says.

What isn’t so easy is knowing how swirling matter moves and whether some of it is blasted outward into the surrounding galaxy, affecting star formation and the galaxy’s evolution. Some of the outflows are known to travel at hundreds of kilometers per second, but there are hints of winds blowing hundreds of times faster, which would have a profound effect on the host galaxy. “There are many things we don’t know,” Costantini says.

JAXA has worked hard to ensure the $190 million XRISM doesn’t succumb to the fates of its predecessors. It has a revamped attitude control system, redesigned coolant pipework, and a backup mechanical cooler that could allow it to operate after its helium runs out in 3 years. Although XRISM has fewer other instruments than Hitomi, “spectroscopy is the strongest point, [and] that can extend the science,” Tashiro says.

Hopes for XRISM are brightening an otherwise difficult time for x-ray astronomy. The field relies heavily on two flagship missions—NASA’s Chandra X-ray Observatory and XMM-Newton from the European Space Agency (ESA)—that are both more than 20 years old, long past their design lifetimes. “Chandra’s health is very good overall,” says Pat Slane, director of the Chandra X-ray Center. But, he says, deposits on its filters are reducing sensitivity and its shiny thermal shielding is degrading, leading to overheating. As for XMM-Newton, its detectors are aging, says Norbert Schartel, ESA’s principal investigator for the mission, but it “can still do science.” The ESA team hopes to eke out enough coolant to last until 2030.

If either telescope dies early, “it will be a loss to the field,” says Jiachen Jiang of the University of Cambridge. Replacements won’t come until the mid-2030s at the earliest.

That raises the stakes for a successful launch even higher. XRISM isn’t a general-purpose machine like Chandra and XMM-Newton. But, Williams says, it “will be the predominant x-ray mission of the 2020s.”