The Perseus Black Hole Note Was Never Measured

Chapters

1. 0:00 A black hole that plays a B flat
2. 0:44 The Perseus cluster and its hot gas
3. 1:13 Jets, bubbles and holes
4. 2:28 What Chandra recorded
5. 3:25 Finding the ripples
6. 4:06 From an angle to a length
7. 4:47 From a temperature to a speed
8. 5:14 The if that carries the note
9. 5:56 The note, played as arithmetic
10. 6:35 Three inputs, three notes
11. 7:21 The 2022 sonification
12. 7:55 Sound or turbulence?
13. 8:34 Hitomi measures the gas
14. 10:51 XRISM, with its door shut
15. 11:36 The number that was measured

Transcript

This is the Perseus galaxy cluster, seen in X rays. In two thousand three, NASA said the black hole at its centre was playing a note. A B flat, fifty seven octaves below middle C.

The lead astronomer put it this way: now we have detected the sound. But no instrument heard anything, and the paper behind the announcement never names a note at all. The B flat first appears in the press release. What the telescope recorded were single X ray photons, each one a position, an arrival time and an energy. This episode follows them to the note, and ends on the one motion of that gas an instrument actually measured.

Perseus is a cluster of thousands of galaxies, about two hundred fifty million light years away. Most of its ordinary matter is not in the galaxies. It is in the gas between them. That gas sits at tens of millions of degrees. It is too thin to see, and too hot to glow in visible light. It shines in X rays, and Perseus is the brightest galaxy cluster in the X ray sky.

The clearest picture of what comes next is in another galaxy, M eighty seven, shown here. Its black hole fires a jet that can be followed all the way in. Perseus has the same machinery. In the middle sits the galaxy NGC twelve seventy five and a supermassive black hole, and its jets inflate bubbles. In Perseus the jets show up in radio. In X rays, what shows up is where they pushed the hot gas aside.

Bubbles of radio plasma, tens of thousands of light years across, show up in the X ray picture as holes. The gas around them is pushed out. That push is where the sound story begins.

Why would anyone care? Gas in the core radiates its heat away. Left alone, the inner fifty kiloparsecs would cool in about two billion years and rain inwards, several hundred Suns of gas a year. It does not happen. The central galaxy is not forming stars at anything like that rate. Something keeps the gas warm, and the black hole is the suspect. The question is how its energy gets out into the gas.

In August two thousand two, NASA's Chandra X ray Observatory pointed at Perseus twice, for about ninety five thousand seconds each time. Almost two hundred thousand seconds in all, roughly fifty three hours.

An X ray camera does not take a picture the way a phone does. It counts. Each photon lands on a pixel, and the detector writes down where, when, and with how much energy. The image is a tally. More photons from a patch of sky means brighter, which means denser gas there. Every structure in the picture is a difference in how many photons arrived.

Chandra's mirrors are the sharpest ever flown for X rays, about half an arcsecond. That matters here, because the structure this story is about is about ten per cent in brightness, and a few dozen arcseconds wide.

The ripples are not visible in the plain image. To find them, the team blurred the picture twice, once at about one arcsecond and once at about ten, and subtracted one from the other. What survives is structure on scales between the two blurs. Around the bubbles it showed curved bands, roughly concentric, like rings on a pond. The paper calls them ripples.

How strong are they? At the sharp front to the northeast, the brightness drops by twenty to thirty per cent. The first ripple to the northwest is about ten per cent. The counting noise is one to two.

That is the first measurement in this story. A brightness ripple of about ten per cent, five or more times the noise. It is real, and it is a pattern in a count of photons, not a sound. The second measurement is the spacing. On the sky, neighbouring ripples sit about thirty arcseconds apart. To make that a length, you need the distance to the cluster.

The paper takes it from the cluster's redshift and a Hubble constant of seventy, so one kiloparsec is about two point seven arcseconds. The spacing becomes about eleven kiloparsecs, thirty six thousand light years.

The third input is temperature. The same photons carry energies, and their spectrum gives about five kilo electron volts, near sixty million kelvin, good to about ten per cent if the spectral model is right. Temperature becomes a speed of sound through the textbook formula for a hot ideal gas. At five kilo electron volts that is about eleven hundred seventy kilometres per second.

Then comes the sentence that carries the whole note. If the ripples are sound waves moving at constant speed, eleven kiloparsecs at that speed is one wave every nine point six million years.

The period had a second use. One bubble every ten million years, kept up for two and a half billion years, is about two hundred seventy bubbles, roughly what it takes to replace the heat the core radiates. That is why the ripples mattered to astronomers. Not as music, but as a way to move energy from the black hole into the gas, spread out in every direction, as long as the sound is absorbed before it leaves the core.

One cycle every three hundred trillion seconds. Raise that fifty seven octaves, a factor of two to the fifty seventh, and it lands near four hundred seventy six hertz. Here it is, then the same sum done with fifteen kiloparsecs.

Four hundred seventy six hertz is a third of a semitone above the B flat just above middle C. So the honest wording is fifty seven octaves below that B flat, or about fifty six below middle C itself.

And the inputs are all approximate. Divide the paper's own rounded numbers, eleven by eleven seventy, and you get nine point two million years. Raised the same way, that is a B, a semitone higher.

In two thousand seventeen, a paper led by the same first author described the same ripples with a wavelength of about fifteen kiloparsecs. Put fifteen in, and the note falls by more than four semitones, near F sharp. So the note is a chain. An angle, times a distance, divided by a speed that comes from a temperature, under an if. Each link moves the pitch, and the last link is the one nobody measured.

NASA turns many telescope images into sound. This one is the Mice galaxies, read from left to right. In two thousand twenty two it did the same to Perseus, reading the waves outwards from the centre and raising them fifty seven and fifty eight octaves, one hundred forty four and two hundred eighty eight quadrillion times. Those factors are exactly two to the fifty seventh and two to the fifty eighth. The mapping is declared, and that is the right way to do it. But what you hear is a choice made on Earth.

Sound waves leave a second fingerprint. Where the gas is compressed it should be slightly hotter. In the deeper Chandra data those temperature changes were at the limit of detection, and were not seen.

In two thousand fourteen, another team read the brightness fluctuations in Perseus as turbulence, gas churning rather than ringing. Same photons, different physics.

The way to decide is to measure how the gas moves. A camera that counts photons cannot do that. A spectrometer sharp enough to see one emission line spread by motion can.

In February two thousand sixteen, Japan launched Hitomi. Its successor, XRISM, shown here, carries nearly the same instrument: a calorimeter, thirty six pixels cooled to one twentieth of a degree above absolute zero, each measuring the energy of single X rays.

Hitomi looked at Perseus for two hundred thirty thousand seconds. It caught twenty one thousand seven hundred photons in one line of highly ionised iron, near six point seven kilo electron volts. Gas moving toward us shifts those photons up in energy, gas moving away shifts them down. A mix of motions widens the line. Remove the width that heat alone gives, and what is left is a speed.

The answer was one hundred sixty four kilometres per second, plus or minus ten. That is a broadening of under four electron volts, narrower than the detector's own five. It was measured anyway, because the response is calibrated. Before Hitomi, the best a grating spectrometer could say about Perseus was less than six hundred twenty five kilometres per second. One short-lived satellite cut that to one hundred sixty four, with ten either side. Five weeks after launch, on the twenty sixth of March, Hitomi lost contact and broke apart. Perseus was one of the few targets it ever observed.

One hundred sixty four is slow. Gas churning at that speed carries energy only about thirteen kiloparsecs in the eighty million years before it must be replaced. Turbulence alone seemed too weak to spread the black hole's energy.

It also means the turbulent push in the gas is four per cent or less of its thermal pressure. A core stirred by a black hole and by passing galaxies turned out to be surprisingly calm.

Sound waves get around that. Their energy travels at the speed of sound, about six times faster than the gas itself moves. If all of Hitomi's width were sound, the gas in those waves would swing at about three hundred kilometres per second.

XRISM launched in two thousand twenty three. The door in front of its detector, meant to open in orbit, is still stuck shut, so it observes through a thin window of beryllium. In January two thousand twenty four it mapped Perseus outwards, three hundred seventy nine thousand seconds in five observations. Near the centre the gas moves at one hundred sixty to two hundred kilometres per second.

Farther out, mergers stir the gas. Near the middle the black hole does, and the XRISM team says that inner stirring could help heat the core, the reading Hitomi's numbers had seemed to rule out. Around seventy kiloparsecs, the gas slows to about seventy.

So whether the ripples are sound, turbulence, or both, is still being argued, with better instruments every decade. The note sits at the end of that argument, not at the start. Which leaves the number this episode ends on. Not a pitch. The width of one iron line, the only motion of that gas an instrument measured directly. One hundred sixty four kilometres per second.

Plus or minus ten, at ninety per cent confidence. No sound wave was heard. A line was measured. One hundred sixty four, plus or minus ten.

Next time, the expansion of the universe, measured with the same space telescope by two teams. Seventy point four, and seventy two point six, and what the first rung of each ladder actually recorded.

Description and sources

In August 2002 Chandra spent almost 200,000 seconds on the Perseus galaxy cluster. An X-ray camera counts photons: position, arrival time and energy. Blurring the image at 1 and 10 arcseconds and subtracting revealed curved ripples around the bubbles inflated by the central black hole -- about 10 per cent in brightness, against 1 to 2 per cent of counting noise, and about 30 arcseconds apart. Everything after that is conversion. The angle becomes about 11 kiloparsecs through an adopted distance; the spectrum gives about 5 keV, which an ideal-gas formula turns into a sound speed near 1,170 km/s; and the paper's own 'if' -- if the ripples are sound waves moving at constant speed -- gives one wave every 9.6 million years. Raised 57 octaves that is 476 Hz, a third of a semitone above the B flat above middle C. The paper's rounded inputs give a B; the 15 kpc spacing of a 2017 paper gives something near F sharp. The temperature changes sound waves should leave were not seen, and another team read the fluctuations as turbulence. In 2016 Hitomi's calorimeter measured the width of one iron line: 164 plus or minus 10 km/s. XRISM mapped 160 to 200 km/s near the centre in 2024. The note was never measured; the line width was.

PRINT-READY, FROM THIS CHANNEL

The Conversion Chain, Vol. 01 — the method, and four numbers taken apart with it

https://therepository.gumroad.com/l/the-conversion-chain

PRIMARY SOURCES

• Fabian, A. C. et al. (2003). A deep Chandra observation of the Perseus cluster: shocks and ripples. MNRAS 344, L43; arXiv:astro-ph/0306036 -- ACIS-S, 8 and 10 Aug 2002, 95.8 + 95.4 ks; 1 kpc ~ 2.7 arcsec; ripples ~10%, Poisson noise 1-2%; ~1170 km/s at 5 keV; 'about 11 kpc', 'about 9.6 x 10^6 yr'. No note is named.
• Chandra X-ray Center press release, 9 September 2003: 'Chandra Hears A Black Hole For The First Time' -- 'B flat, 57 octaves lower than middle-C'. Chandra Chronicles, 16 September 2003.
• Chandra X-ray Center, 4 May 2022: sonification of the Perseus cluster, raised 57 and 58 octaves.
• Fabian, A. C. et al. (2017). Do sound waves transport the AGN energy in the Perseus cluster? MNRAS 464, L1; arXiv:1608.07088 -- ripples ~15 kpc; expected temperature changes not seen; sound energy at ~1000 km/s.
• Hitomi Collaboration (2016). The quiescent intracluster medium in the core of the Perseus cluster. Nature 535, 117; arXiv:1607.04487 -- SXS 4.9 eV, 230 ks, 21,726 counts in Fe He-alpha, 164 +/- 10 km/s (90%), turbulent pressure 4% or less.
• XRISM Collaboration (2026). Disentangling multiple gas kinematic drivers in the Perseus galaxy cluster. Nature 650, 309; arXiv:2509.04421 -- Resolve ~4.5 eV, Jan 2024, 379 ks; ~160-200 km/s near the centre.
• Freedman, W. L. et al. (2025), ApJ 985, 203; Riess, A. G. et al. (2024), ApJ 977 -- for the next episode.

ARCHIVE PLATES — public-domain documents, with author, date, archive and licence.

• XRISM Mission Captures Unmatched Data With Just 36 Pixels | NASA's Scientific Visualization Studio | public-domain
• Gigantic Wave Discovered in Perseus Galaxy Cluster | NASA's Scientific Visualization Studio | public-domain
• Zoom In on Galaxy M87 | NASA's Scientific Visualization Studio | public-domain
• Artist’s impression of the Black Hole at the heart of M87 | ESO | cc-by-4.0
• Zooming in on the black hole and jet of Messier 87 | ESO | cc-by-4.0
• Sonification of NGC 1300 | NASA's Scientific Visualization Studio | public-domain
• Sonification of the Mice Galaxies | NASA's Scientific Visualization Studio | public-domain

... and 5 more plates, each credited on screen in the video and listed in this project's manifest.

Absolute Magnitude takes one number the internet repeats and asks what an instrument actually returned. Every episode ends on a measured figure with its source on screen, and says plainly which part of it was converted afterwards.

No AI-generated imagery is used on this channel. Every frame of archive footage is real, publicly released material, credited on screen with its licence; every diagram is drawn in code from the numbers in the cited papers. The tones in this episode are synthesised from the published period, and the description of the mapping is on screen.

Corrections are pinned, not quietly edited.

#Perseus #BlackHole #Chandra #Astronomy #AbsoluteMagnitude