Three hours to save Integral spacecraft | Hacker News Reader

Three hours to save Integral spacecraft

esa.int

376 pointsjensgk41 comments

The article briefly mentions dumping energy at regular periods without using thrusters. Does anyone know what mechanism they're using to do this? At first glance I'd assume it would look like rotating/configuring the satellite so the solar wind imparts angular momentum.

"using its highly sensitive reaction wheels"

Yes, the article uses those words. But an understanding of elementary physics tells us that momentum is conserved in an isolated system. Spending the reaction wheels can cause the outer frame of the satellite to rotate in a certain fashion but it cannot possibly alter the overall momentum or angular momentum of the entire satellite unless it somehow interacts with an external force.

I guess the motors are lossy and some of it gets converted to heat if they speed the wheels up and down

That won't solve the conservation of momentum problems though, it'll just use more energy to get to the same state. If you could demonstrate controlling the rotation of a spacecraft just by using more inefficient motors you would upend a large portion of modern physics and could probably go and collect your Nobel prize within the week.

Ok, read TFA now. It sounds like they need to dump energy from the reaction wheels. Maybe they can do this by inducing a spin, rotating 180 on a different axis then cancelling the spin?

You can't cheat conservation of momentum, no matter how cleverly you move your spacecraft. That's why all the solutions involve expelling reaction mass. (Or, in the case of solar wind, interacting with an external force.)

Right, the reaction wheels help if you want to maintain orientation of the satellite under external forces that want to (gently!) rotate it. And eventually your reaction wheel saturates and you can't get more useful rotation out of it.

This is where thrusters are useful, because you can use the thrusters to give you a "source of force" while you spin the reaction wheels back down. I believe this is where the original parent's question is coming from - if you have no thrusters, how do you avoid saturation?

Presumably, they've worked out a way to accurately predict the external forces and gently operate against them while spinning the wheels down in a way that doesn't make the satellite spin faster.

https://en.m.wikipedia.org/wiki/Control_moment_gyroscope

Perhaps? The ISS uses these.

Excellent question and it turns out there's a whole article about the solution! https://www.esa.int/Enabling_Support/Operations/Rescuing_Int...

> By using a specially designed sequence of manoeuvres, the control team realised they could redistribute the angular momentum stored onboard the satellite using two different reaction wheels spinning in opposing directions, causing the spacecraft to flip.

> "So at this point we knew we could control the build up of energy absorbed from the Sun, and christened this new manoeuvre the z-flip".

There’s actually several small to medium effects on any spacecraft in orbit around earth. There will be tiny perturbations from the interaction of earths electromagnetic field as it moves through it slowly, there’s atmospheric drag on solar panels and the spacecraft body that can create “aerodynamic” torques related to what the normal orientation is relative to the earth. There’s even smaller perturbation from the solar wind… some spacecraft have a thing (or can fudge one using some part of the electrical subsystems) called a “magneto-torquer” which is basically a solenoid/electromagnet designed to tug on the earth’s magnetic field so they can trade electrical power for physical force.

It’s a pretty cool thing that seems to be happening more and more frequently with spacecraft flight dynamics teams developing such a detailed model of all these forces that they can invent new “stable modes” like the one they used in this article, or the Kepler missions K2 mode where they found another useful mode instead of just ending the mission. It’s not as dramatic as something like a solar sailing mission, but I think it’s a pretty awesome technical achievement, like watching a trained stuntman drive a car on two wheels through a slalom course, like a form of highly skilled, kinetic, robot based gymnastics.

https://en.wikipedia.org/wiki/Tennis_racket_theorem

Just turn it 180 every few days this way, so the torque from external influences is cancelled out.

The NASA video on that page is mesmerizing.

I think what they are referring to is "angular momentum dumps". The reaction wheels have a fixed maximum amount of angular momentum. This is a function of the speed of the wheels themselves since the mass/moment of inertia is fixed.

Over time external forces continue to add momentum to the system (from drag/solar radiation pressure). In order to maintain a desired pointing direction this means the wheels must increase in velocity to "absorb" this momentum.

Eventually you need to "dump" the momentum using another external force, such as thrusters, or in this case the same solar radiation pressure.

Goal is to orient spacecraft such that external forces can be used to spin down the various reaction wheels, thereby "dumping" momentum out of the wheels.

This is pretty common on all spacecraft, but gets complicated/innovative when you have to do it in an emergency or if you don't have a full set of reaction wheels.

Just adding it because I think it's a cool use of the word. When reaction wheels are operating at maximum speed and can no longer exchange momentum with the spacecraft in both directions, they are considered 'saturated'.

'Saturated' is used in control systems engineering for any value (sensor input, drive output etc.) that has reached maximum or minimum value and can't be further increased/decreased. It's the same sense of the word as used in the HSV colour space in computer graphics.

One of the interesting things about the telescope on Integral [1] is that it uses a coded aperture mask [2] to image the sky in high energy X-rays to gamma rays, rather than a focusing optic.

[1] https://en.wikipedia.org/wiki/INTEGRAL [2] https://en.wikipedia.org/wiki/Coded_aperture

Wow, thank you for this! I had no idea such thing was possible, now I'm reading a lot on it.

AFAIK thats because we dont have any usable material for X-ray lenses, right ?

Used to be, they would loft a whole bunch of concentric cylinders, and rely on total internal reflection off their surfaces, which works even with X-rays.

If I understand correctly what they do now, they treat the diffraction pattern from X-rays going through a mask as a sort of 2-D Fourier transform, and transform it to an image with the inverse transform based on the known mask. Which is very clever. I don't doubt that this description glosses over the actually interesting bits.

Total internal reflection gets harder at higher energies since the angle of incidence has to be smaller, meaning a the effective area of the mirrors gets smaller. Once you get up to gamma ray energies, mirrors are not really feasible, and straight line optics like coded masks are the way to go.

The coded mask technology is really cool. The mask pattern is chosen in such a way that the inverse transform generates a unique solution (along with a whole lot of background noise that gets distributed more evenly and so can be subtracted out). Not to downplay that subtraction - modelling the noise in a coded mask system is one of the hardest parts of analysing the data. Worked on data from a coded mask mission once - 2 years to model the noise and correct for it, 6 months to do the science on the results.

The cylinder approach (actually paraboloid-hyperboloid or Wolter-I optics) is the main design used today and for future telescopes for lower energy X-rays. Focusing optics have a big advantage of having a lower level of background compared to a coded mask. Future Wolter-I optics use clever designs for the telescope, such as stacking silicon wafers into modules planned for the ESA Athena observatory [1] or modular lightweight single crystal silicon mirrors planned for future NASA missions [2].

[1] https://www.esa.int/ESA_Multimedia/Images/2019/02/Athena_mir...

[2] https://www.nasa.gov/feature/goddard/2019/nasa-s-new-lightwe...

Cool story! I wonder if the mission engineers do a lot of operational training for incidents like this: for example the SREs at Google do "tabletop" role-playing exercise to practice incident response. Would be cool to hear similarities/differences with ESA folks.

There are many simulations like that when sadistic engineers (sorry, simulation engineers) will inject errors in the simulator and see how the operators react. Then the operators have to find a way out. This is also used to make sure that monitorings and procedures cover all the errors.

Did anyone notice the amazing cookie banner?

1. It offers a 'accept essential cookies' button 2. It doesn't take up as much space as many other cookie banners do.

Re #2: Are you on a desktop/laptop? It took up a third of my screen on my phone. That’s not much better than others.

I like the accept essential cookie button, but it doesn't much smaller than most banners I come across.

I thought "essential" cookies didn't even need consent? Either way, cookie consent nonsense and the banners caused by it are a blight whose benefits ("solving" the absolute non-problem of advertisement tracking) pale in comparison to the enormous collective inconvenience and wasted developer hours they've caused us all. It's sad that we look at a banner that's not as bad as it can be and appreciate it, like someone saying their spouse only hit them three times today instead of ten.

Yeah, it's unfortunate that the bar is so low

After checking uBlock's debug log, yes, I did notice it! It is ##.cc-cookies from the rulesets "Fanboy's Annoyance" and "AdGuard Annoyances".

(To anyone unfamiliar: these "extra" rulesets ship with uBlock Origin, are disabled by default, and can be toggled in the "Filter lists" tab of Settings).

https://github.com/gorhill/uBlock/wiki/Dashboard:-Filter-lis...

I wonder why did they call it so. Either they did not know of prior art[1], or knew it all too well and embraced it.

1. https://en.wikipedia.org/wiki/We_(novel)

This one regained spin control too late,

https://en.wikipedia.org/wiki/Wide_Field_Infrared_Explorer#W... (1999)

“Flash, I Love You! But We Only Have Fourteen Hours Left to Save the Earth!”

I am surprised to read that Safe Mode did not automatically disable all non essential systems. That seems to be very risky behavior.

These reaction wheels have such a high failure rate. Is it a fundamental Tribology problem? edit I see they think it was a software issue but I still wonder about reaction wheel failure rates.

One failure mode is static electricity build up leading to arc discharge and pitting the bearing races. Also, the vacuum tends to make the bearing grease evaporate. Back in the 60's they used spermaceti, maybe we should go back to that?

https://est-aegis.info/2018/08/in-space-no-one-can-hear-your...

I'd wager that modern synthetic oils could beat whale oil, but now I'm curious what the state of the art is.

High failure rate? This particular spacecraft is 19 year old.

There was a string of reaction wheel failures in the past that were tracked to a single supplier (forgot exactly what the issue was, but I think it was some contaminant).

Regardless, it's a device that has to continuously spin at high speeds, in space. It's actually remarkable that they survive for as long as they do.

Did the reaction wheel recover from this or is it dead? Article started by saying they used 3 so now only 2 are operational? Can they sustain the satellite with 2?

From what I gather it was just a radiation caused software glitch that momentarily shut it down.

> Integral has since remained under control, and from 27 September all systems are back online. Since 1 October, after an extended checkout, its instruments are back observing the high energy Universe.

Seems like they're back in shape.

As a comparison the Kepler telescope had 4 wheels, of which 2 eventually completely failed and they managed to get it into a neutrally stable orientation that could be managed with just the two and some thruster assistance. Until RCS ran out that is. 3 are generally the minimum you'd need to orient yourself in 3 axis, as one might expect.

> “Most of the Control Team were working from home at this point – I was following operations from the train!”

The future is here!

Reply on news.ycombinator.com