NASA's Swift Observatory May Have Suffered An Attitude Control Failure (engadget.com) 35
After 17 years of relatively smooth sailing, NASA's Neil Gehrels Swift Observatory has entered safe mode after detecting a "possible failure" in one of the six reaction wheels used to change attitude. Engadget reports: While it's not clear exactly what (if anything) went wrong, NASA has halted direction-based scientific observations until it can either give the all-clear or continue operations with five wheels. This is the first potential reaction wheel problem since the Swift Observatory began operations in February 2005, NASA said. The rest of the vehicle is otherwise working properly.
The Swift Observatory has played an important role in astronomy over the past two decades. It was primarily built to detect gamma-ray bursts and detects roughly 70 per day. However, it has increasingly been used as a catch-all observer across multiple wavelengths, spotting solar flares and hard-to-find stars. NASA won't necessarily run into serious trouble if Swift has a lasting problem, but it would clearly benefit from keeping the spacecraft running as smoothly as possible.
The Swift Observatory has played an important role in astronomy over the past two decades. It was primarily built to detect gamma-ray bursts and detects roughly 70 per day. However, it has increasingly been used as a catch-all observer across multiple wavelengths, spotting solar flares and hard-to-find stars. NASA won't necessarily run into serious trouble if Swift has a lasting problem, but it would clearly benefit from keeping the spacecraft running as smoothly as possible.
Re: (Score:2)
Reaction wheel = gyroscopic flywheel (Score:5, Informative)
From Wikipedia:
A reaction wheel (RW) is a type of flywheel used primarily by spacecraft for three-axis attitude control, which does not require rockets or external applicators of torque. They provide a high pointing accuracy, and are particularly useful when the spacecraft must be rotated by very small amounts, such as keeping a telescope pointed at a star. Link [wikipedia.org]
Just in case anyone wondered why a satellite needed wheels.
--
Re: (Score:2)
I can't think about reaction wheels without remembering this thing:
https://youtu.be/n_6p-1J551Y?t... [youtu.be]
broke: spaceflight
woke: walking cubes
Re: (Score:2)
Just in case anyone wondered why a satellite needed wheels.
Gotta admit - I saw "Swift" and "wheels" and thought this was an Apple story. Fortunately the satellite doesn't look like a cheese grater.
Re: (Score:2)
And you only need three to stabilise the attitude, although satellites usually have at least four to provide some redundancy.
So I don't think this is a big deal, yet.
Another Ithaco reaction wheel failure? (Score:2)
Re: (Score:2)
The questions that then need to be asked are:
1. What alternatives to reaction wheels exist that could cope with Swift's need to turn quickly?
2. Are there (better?) ways to keep electrical fields relatively uniform?
3. Since charged particles don't strike satellites from all directions evenly, would it be practical to have backup reaction wheels?
Re:Another Ithaco reaction wheel failure? (Score:5, Informative)
I think it's safe to say that NASA considered these factors in great detail when designing SWIFT, they hire what I'd call 'pretty good' engineers and scientists. I would guess that six reaction wheels was a compromise between anticipated service life and weight. To give you an idea how that worked out, the mission lifetime of the SWIFT was 2 years, meaning it was planned to conduct all the science that justified its launch within 2 years. It's now just over 17 years old.
Honestly, it shouldn't last forever: there's such a thing as over-engineering! And even with one reaction wheel down, the news is that the smart folk at NASA will probably find a way of making it usable with five.
Re: (Score:2)
Re: (Score:2)
To be fair to the person that I was responding to [notes 4-digit user ID], I guess the point was more along the lines that if you need 3-redundant reaction wheels, what alternatives are there?
Reaction wheels are kind of great, in that they can have exceptionally long service life and (unlike a thruster, which could do a similar job) require no fuel apart from sunlight.
The real question seems to be - does Ithaco (the space tech company making a some of the failed reaction wheel on Kepler, Dawn and Hayabusa)
Re: (Score:3)
I think they have 2 on each axis, rotating in opposite directions.
So they can slow down one or speed up one to change.
Re: (Score:3)
The reason for this is because if you only have 1 per axis, eventually you will reach a situation where one wheel is going as fast as it can go, or as slow as it can go, and you lose attitude control in that axis. With 2, you can transfer the stored momentum between them.
Re: (Score:2)
"Having 6 is a highly redundant configuration"
No it isn't "highly redundant". You must have three on x, y and z axes. You can have the other three on the same axes (which means you have simple redundancy, one "main" and one "spare"), or on differently oriented axes (in which case one failed wheel could be replaced by rotation of all the "redundant" three wheels.
So basically you have "simple redundancy" - like having a motorcycle with differently sized front and rear wheels, and one spare wheel of each.
Re: (Score:2)
Generally, 4 is good enough for redundancy.
Reaction wheels suffer from saturation (eventually, you can't slow it down any more, or can't speed it up anymore). You then need to use non-reaction wheel methods to transfer momentum to something that can take it. Another set of reaction wheels works great for this, as well as also being a redundancy mechanism.
Either way, any long-lived craft will have need another set of attitude adjustments to relieve the saturati
Re: (Score:2)
Honestly, it shouldn't last forever: there's such a thing as over-engineering!
While there is such a thing as overengineering, making something last as long as possible means wasting less resources, and we've seen before that even outdated instruments can provide valuable data.
I wonder if it's practical to use maglev bearings on reaction wheels in spacecraft, or if not, why not.
Re: (Score:2)
An interesting question, although I suppose it's possible that would worsen charge buildup. It might be made to work, though.
Re: (Score:2)
I'm wondering if - on Earth orbit at least - you could rotate your craft using large electric coils that interact to Earth's magnetic field (like an electric motor).
Re: (Score:2)
You can: https://hal.archives-ouvertes.... [archives-ouvertes.fr]
The problem is that the direction and strength of torque you can create depends on your orientation and where you are in your orbit. Not so good for maintaining orientation in a telescope. You can use reaction wheels to store angular momentum, and magnetorquers to dump it, though.
Re: (Score:2)
They do:
https://spinoff.nasa.gov/spino... [nasa.gov]
https://asmedigitalcollection.... [asme.org]
Other solutions are ceramic bearings and various grounding schemes. It wasn't until fairly recently that people really nailed down the causes of the failures though. E.g.:
https://esmats.eu/esmatspapers... [esmats.eu]
IIRC twenty years ago they knew there was a problem and were trying various fixes, which is one of the reasons why Swift has lasted so much longer than expected.
Re: (Score:2)
Remarkably, Tesla had a similar problem in their early Model S. Owners from back then have good stories about multiple replacements of drive units due to motor bearing failures - caused by stray induced currents in the rotor that would arc through the bearings to the case, causing bearing pitting and failure. Eventual solution (after many different attempts)? Ceramic bearings.
Re: (Score:2)
they hire what I'd call 'pretty good' engineers and scientists
I know, I was one.
A lot changes in 17 years, though, and what was a perfectly sound engineering choice back then needn't be now. Since my questions are framed in the present tense and refer explicitly to design choices that could be made today, the choices made back then probably aren't what I'm after. So, to repeat, what could NASA do differently now?
Re: (Score:3)
So, to repeat, what could NASA do differently now?
They could crew their satellites with cats, who are well known for being able to reorient themselves in a freefall without needing flywheels.
Re: (Score:2)
I've heard of this being the cause of premature failures, but in this case we're talking year 17 of a 2 year mission.
Re: (Score:2)
Reaction wheels are a frequent cause of failures. This paper (PDF) [esmats.eu] discusses a possible cause as "electrical discharge across the bearings due to rapid spacecraft charging in a volatile space plasma environment". I'm not claiming that this is the root cause for the Swift's problem, but it seemed like a reasonable explanation.
It is also a common problem on Earth with sensitive equipment that includes bearings. You can sometimes find mechanical designs which include a way to drain accumulated charge across a bearing.
Does Nasa need an attitude adjustment? (Score:2)
Attitude? (Score:2)
Needs more discipline.
Re: (Score:2)
Re: (Score:2)
> Needs more discipline.
If the satellite has an attitude problem, NASA should send up a LART ( https://en.wiktionary.org/wiki... [wiktionary.org] ) , a.k.a. clue-by-four ( https://idioms.thefreedictiona... [thefreedictionary.com] ).
After 17 years of relatively smooth sailing... (Score:2)
Can't they just use the spinnaker?
Some context (Score:4, Informative)
For some more info and context, Scott Manley on youtube produced a video a few years back [youtube.com] explaining the purpose and function of reaction wheels. He also has another video about the reaction wheel failures that took the Kepler telescope out of commission. That video promulgated the theory that electrostatic discharge through the bearing components resulted in micro-pitting, accelerating wear, and leading to failure.
ITHACO, which makes a lot of the reaction wheels and gyros for spacecraft, have since switched to ceramic bearing designs, which are non-conductive and should not experience this failure mechanism.
Not entirely unexpected (Score:2)