Joe, what you are saying is your balls are steel and impotent. Im sure thats not true, but just couldnt resist making a joke. No hard feelings brother.
Ceramic bearings would probably have fewer problems because of the surfaces. Would probably handle heating/cooling cycles better than steel and electrical discharge. But, then again, if the grease/lubricant being used is causing undue friction for some reason you'd still have failures.
So I will begin with "I am not a rocket scientist" however as an industrial HVAC and commercial equipment service technician I replace bearings all the time and as a matter of fact I am currently covered in grease from a 15 hp electric motor that blew a rear bearing and cost my customer over $2000 to fix. There is this extreamly common thing that happens to electric motors that are being driven off what we call a VFD (variable frequency drive) it allows us to control the speed of a motor but because of how a vfd works and converts the voltage it creates voltage spikes in the rotor that hava no where to go but through the bearings and into the housing, this pits the balls, races and destroys the bearing over just a few months. The thing to keep in mind is that there is a very small amount of contact area from the ball to the race so it doesn't take much to destroy the bearing in a very short period of time. Something as simple as a rotor grounding shaft would have eliminated the bearing failure.
This is exactly what my mind went to when Scott mentioned solar flares. Searching for "bearing common mode current" provides lots of relevant results for anyone else looking for more info on the VFD/earthly sort of failures.
I dunno grounding across bearings can be hilarious when a trainee does it. Flames (from the lubricant), a "bearing" that is now rock solid, and mickey taking material for the next month at least.
Not so funny for the welder that does it across multiple bearings in a turbine engine, We had one the was welding an exhaust nozzle crack, but grounded to the front fan assembly on the engine, welded every bearing in the engine.
James Collier it's a wheel that you spin to spin your spacecraft the opposite way (hence reaction) without wasting propellant. Used to orient and stabilize. Every action has a equal and opposite reaction. In order to spin up something in space you need something else gaining the same angular momentum in the opposite direction
You can get same failure mechanism with cheap frequency convert drives on electric motor bearings. The ceramic bearing is good but expensive fix the other solution is install grounding carbon brush but that's needs probably some maintenance over long periods of time so using just ceramic bearings on stuff like that is probably good idea, especially when they are already stupidly expensive :D
I worked on commercial space communication satellites, and my dad worked on Hubble Space Telescope. Reaction wheels were always acting up. On Hubble some of the reaction wheels have been replaced on servicing missions, the only spacecraft (other than the ISS) that such replacement in orbit is possible. On our commercial satellites, there were usually more than 4 wheels, so there was some redundancy. I always wondered why such a (seemingly) simple device was so troublesome. Your video provides some interesting insight into this issue.
An easy fix, if someone would have guessed the issue, would be to add some radioactive material near the bearings, because ionizing radiation dissipates away static electrical charge as it accumulates. Or make the lubricant electrically conductive. Or, as they did by accident, switch to ceramic bearings.
I was surprised that they were using BBs and not magnetic bearings. It's possible for metals to cold weld together in a vacuum if they've lost enough of their oxide layer so that could be happening if the lubrication film breaks down enough to allow bare metal to contact in the bearing.
Meme Master Lubricants that work long-term in a vacuum are used. Everything was and is lubricated very well and coated in a lubricant layer in any of these mechanical bearings out in space.
ed p anything using bearings will always fail at some point, and something spinning incredibly fast (some at the limit of their tensile strength) will fail more quickly.
"...the only spacecraft (other than the ISS) that such replacement in orbit is possible." WAS possible. Since the retirement of the space shuttle fleet, there is no spacecraft currently able to rendezvous with Hubble and secure it sufficiently to perform any sort of maintenance on it. Its days are now numbered. HST might well last well into the 2020s. Even if the JWST is completely successful, HST is capable of "seeing" in wavelengths that JWST cannot. A 6th servicing mission may well be approved some day, using a spacecraft yet to be deployed, in order to save one of the most beloved spacecraft ever launched into orbit.... at least, for a time.
I remember when Bill gave that presentation to all of us former Ithaco Space Systems employees on the fifth anniversary of it's closing. He set up a projector in his barn and after our barn party reunion we all listened very intently. I saw the picture of a CME on the screen, looked back at Jeff Golden, then to Eric Stromswold, and my jaw dropped as they let me know that they agreed. The ball bearing is basically the gateway for static potential to discharge from a relatively massive rotor (hidden under a relatively thin cover) to a massive dish. Yes, of course it would be subject to ionic buildup, regardless of any Faraday cage of the rotor cover. Remember, the outer bearing cover is attached to the outside of the main cover housing. Essentially, it's all right there waiting to pick up these ionic discharges. I so miss working at Ithaco. Everyone there worked so hard at making those things work right... Laurie spent so many hours in the inner cleanroom taking so many measurements fir Bill and Eric WHILE ALSO building so many other RWAs. I stick to Torqrods so I didn't know much but I just remember how heartroken we we're when we lost the Iridium contract because of the failures. That was the death of a great company.
Wow, any more stories? Such a small world. I was at that barn party too. I was the small, meek fellow in the back corner, nervously chewing on sunflower seeds expecting someone to step on me at any moment.
@@SpenserRoger You worked there, or family? I was only st Ithaco for the short remainder of its existence as a UTAS company, October 2011-March 2013. I built a prototype motor driver card for GOES-R and a lot of test cables. Also wound one SMAP Torqrod
Now, if we could just figure out a massive photon gatherer and grab enough photon mass equal just 1/10 th the mass of what is being driven. Wow, ranks right up there with the wormhole thing.
I still don't understand why he said reaction control propellant.. when clearly their motorized wheels actually I don't know what kind of thing is propelling the wheels
I work for a large organisation and look at equipment used at our facilities. One of our hardware vendors insisted that nothing was wrong after i directly showed them what i believed was a problem. They later admitted the issue but by that time they'd rolled out a new model, and guess what? The new model had exactly addressed the issue i pointed out, meaning that at least someone in their organisation knew about it for a long time. Corporations can be as dishonest with each other as they want if they feel they can get away with it. I wouldn't be shocked if it's true that they moved to ceramics here because they secretly knew about the defect.
I mean hell, my MS explorer trackball has had ceramic bearings in it for years since the metal ones wore flat forever ago. You’d think they’d want ceramic just for lifespan if that holds true for this application.
@@gags730 I’m sorry but I hate cranberry juice now some apple punch? That’s my jam. Also no fidget spinners sadly I broke them all by using them too much. And instead of a glowing one I did have one that flew so there’s that 😉.
- 🥺 Bombarded with audio all day I like to read the closed captions you provide in your always very informative video presentations. I really apreciate your thoughtful consideration. Keep up the good work. 👍
A friend who is building large CNC machines said it is a huge problem to ground these machines properly because ESD (when machining some matertials) and the currents of the motors are causing these problems with the bearings
Add some ambient ionizing radiation, this will dissipate electrostatic buildup, and it will give you cancer, but for grounding moving parts across an air-gap, it really works, by knocking electrons out of surfaces.
A good experiment for the X-37A would be to put some reaction wheels in it and run them in space for a year or two then bring them back and take a look at them.
I absolutely love it that you insist on using the satellite girl figurine every time you talk about the Hayabusa. Maybe next time show a picture of the motorcycle :^)
Most fascinating. I work for a major satellite fleet operator. We've been seeing this exact type of wheel failure in some ~15 year-old spacecraft which were expected to last even longer. I'm going to go back and check the data for correlations. Glad that there seems to be a fix in the newer wheels, even if accidental. Thanks for this one, Scott!!
Damn Scott.. you always pick the most interesting things to talk about.. Plus you put them to us in a way that this 66 year old fart can actually understand.. or maybe it is just because of my Scottish heritage (Lamont clan) helps me to understand... Well done brother.. carry on...
And he seems to provide further, conclusive evidence that Scots will be the first Space Superpower. Because everyone else's stuff will be breaking down all the time.
This has to be one of my favourite videos on this channel yet. A nice inside on the engineering problems involved in rockets and space probes but easy enough to be understood by laymans.
They should make a craft full of reaction wheels, send it to orbit, Wait for like 50% to fail, then have it rendevous with the ISS for inspection and analysis.
It would need to get out past the Van Allen belts to really soak up the rads. Might as well put it inside a heatshield, send it on a hyperbolic orbit out to L2 and then let it reenter.
Unless ceramic bearings aren't the only needed fix, it could be something like the lubricant not having the same conductivity across it leading to the exact same problems, where parts of the fluid will conduct the electricity and cause micro pits while others won't. What if the pits aren't the problem, but the sudden heat produced is burning off the lubricant on that spot and over time it just burns away enough to cause high friction/wear?
People often fail to realize that there is no such thing as a perfect Faraday cage. You can't just wrap something in micron thick foil and expect infinite attenuation. Seems to me like you'd need something more than just a standard metal enclosure to keep voltages across the bearings below 6V during a decent size solar flare.
It's not the thickness, it's how good the connection is along all the joints. You use things like this: www.tech-etch.com/shield/images-shield/assortment-of-emi-shielding-products.jpg But ionizing radiation can also cause a voltage spike and flares give off some quite energetic particles. If that's the culprit, good luck shielding from that.
Mining - you're confusing grounding with earthing. Neither have anything to do with effectiveness of a Faraday shield. That's determined by the conductivity/admittance of the cage.
My first thought to solve the problem was to use heat resistant ceramics to replace bearings; you can do a home experiment with anyone who has had an electric train set, to demonstrate what static discharge event being discussed looks like in real life: All you need is an outlet adapter for a race track or a train set. The racetrack or train set adapter will usually come with lead wires that travel to the racetrack or track connections. This is the wire I used as a child to conduct this experiment. With the adapter charged to any degree above zero, you will see a blue spark as the wire contacts the quarter and grounds out the current through the quarter in a short circuit. This would be the same effect as having the charge jump from one bearing to the next, as another aspect, the bearings themselves may have different charges from each other during a CME (Coronal Mass Ejection). Remember, these Coronal Mass Ejections were recorded to make sparks on telegraph wires within our own atmosphere during the 1800's when electricity was not as wide-spread and in use as a part of major industry as it is now. That is surely enough to make the same kind of spot scarring you see on the surface of the quarter you've been creating sparks on. You can feel the part of the metal where parts of the wire got fused to the quarter during the exchange of electrons, also used as a practical application in welding, called arc welding. These small imperfections would be catastrophic to bearings needing any degree of precision in a very highly machined quality product of microscopically precise application. Good article!
I think things like this is why we should try to plan for a potential recovery of some probes. I'm no rocket scientist or astrophysicist, but surely it would be worth the extra trouble to actually get a good ol' fashioned 'MK-01 Eyeball' looking at things, right?
Maybe if some of them are in low Earth orbit, we can take a look. Otherwise, it's simply not practical. Spacecraft are only given enough fuel for their mission, and not enough fuel to return to Earth.
I think most of these craft are in solar orbits, or very high Earth orbit, so we don't yet have anything capable of reaching them and returning to Earth. They are also mostly outside of Earth's magnetosphere, and thus more exposed to solar radiation, which contributed to the problem.
Pfft, you're clearly not thinking the real kerbal way. Just build another Saturn V and send up a little robot arm + capsule to grab and return one of these probes.
Oh that's what I do. I just....misinterpret the numbers on occasion (read: all the bloody time), resulting in recovery vessels needing recovery vessels....Which need recovering in true Kerbal fashion.
This makes perfect sense to me. I remember reading a story many years ago about how an electric motor bearing was found to be failing due to a short between the rotor winding and the rotor itself. The small electric current through the bearing caused increased bearing wear, and eventually, early bearing failure.
This is so awesome. In one video I learned that satellites use reaction wheels, that they malfunction, why they probably malfunctioned and how it likely has been solved.
@Scott Manley - Thank you for helping to reveal how and why these mysterious space borne failures took out many of my favorite missions. Not surprised and it makes very good sense. Appreciate the effort. Astrophysics and astronomy are my first loves... I too went into tech.. as a hardware engineer in multimedia telecom and network products. Retired now.. JK in Berkeley
I know I'm late but this was intriguing but more so surprising. I would have figured they considered this and implemented some capacitive grounding rail with some discharging mechanism to mitigate the charge. I always assumed this would be done with any system launched into space/upper atmospheres.
I'm sure we'll see a conductive lubricant on the bearings in the future or more ceramic bearings. I'm really surprised ceramics were not used in the first place. With a much lower mass, smoother operation and longer life they seem like the go to choice. I'm sure there was more to it though. Great video!
Great video buddy as always Scott. I'm a mechanical engineer , and it makes alot if sense. There 's a mechanica proces called "electric sprak erosion" , that is done in an oil bath . The work piece is connected to the + pole and a graphite electrode - When currency is switches on , the elektrode cavitates a hole in the workpiece. From the moment you mentioned static charge , i already knew what the study would say 😉😁... Grtz huge fan johny geerts
Not sure about the ceramic balls solution. Static charges can build up on the rotor before jumping out within a dramatically thousand volts spark. Saw this failure on high perfomance machine-tool spindles.
What exactly would be the problem if the sparks go around the bearings instead of damaging them? Also, Einstein got his Nobel Prize not for relativity, but for the photoelectric effect. Add a bit of radioactive material, exposing the whole thing to some low level of ionizing radiation, electrons will get knocked out of surfaces and will travel to reduce any electrostatic charge. At high levels, this is poison to any electrical devices, this is why all robots fail inside the fukushima nuclear ruins. At low levels, this is poison only to any electrostatic buildup of charge. This works in air or in a vacuum. No further design considerations as long as you have conducting surfaces seperated by nothing other than air or vacuum, add some ambient radiation to drop voltages and electrostatic charges across the system.
@ Conductive lubricants tend to have higher viscosities and particulate content. Perfectly fine for an industrial motor, but completely unsuitable for high speed high precision applications like a reaction wheel.
Very interesting failure mode. Easy enough to remedy using a simple spark gap though. A few closely gapped sharp points in the right locations should allow any discharge to occur at non-critical areas of the spindle assembly. Or, as kurtilein3 suggested, use radioactive material to effectively decrease the work function, the results would be similar.
As someone who uses K2 data, its a pain that the reaction wheels broke because since it makes the data messy. But on the other hand its great because we could use Kepler to hunt for exploding things in K2. So for me its a good thing that Kepler didn't have ceramic reaction wheels!
The more you learn, the more you realize you still have discovered nothing, and that you are nothing in the grand scheme of things. That is frightening and fascinating at the same time.
Theres a small pebble of attractive conglomerates in my batroom as an ornament. The story it could tell is millions of years old. It was as it is millions of years before our ancestors descended from the trees. I take a bath, many baths as I grow ancient, soon to die, but this pebble will remain as it is for millions of years to come.
I had noticed this and had a strong curiosity as to why the reliability had seemingly not improved over quite a span of years. Thank you for speaking about this potential failure mode as it is both interesting and satisfying explanation.
Would it be feasible to create a zero-contact reaction wheel with magnetic rails like a maglev train? Even though there's no contact, angular momentum would still be transferred by pushing off the magnetic fields. As an added bonus, the system's magnetic field may offer some protection against solar winds if it's strong enough.
@@deanrensberger631 Oil is a dielectric. It doesn't conduct electricity. In fact, all those large power transformers you see on poles and on the ground are filled with oil.
@@stargazer7644 you dont understand what we're talking about. Oil isnt a perfect insulator like the vacuum. Its conductivity is what allowed these mcgs to fail.
This idea is old enough that lubricants for use in space are tested for this in nuclear research facilities. You can just go and buy high-performance lubricants that are rated to withstand both a vacuum and intense levels of radiation. It is too obvious.
Very interesting, as I’m a retired marine machinery mechanic & the Navy back in the late 1990’s experienced a similar problem after a week of conducting a heat run test/monitoring of a newly installed bearing. It was determined that the manufacturers were to blame.
This reminds me of a known problem with bearings in movable bridges. When a bridge deck is not properly grounded, lightning strikes can actually weld the bearings together, much like the described effect in this video.
I was aware they were built by a different source but as long as they used metal parts similar to Ithaco instead of the newer ceramic they would still be able to use it to isolate the variable of who made it to narrow it down to possible common design flaws.
Deric Anthony Didn't Hubble servicing mission 3B bring back the gyros and the reaction wheel? I think the gyros and reaction wheels were designed in the early 80's. Someone should find out and correlate to any solar events.
@@scottmanley Even if these were made by Ithaco, they still shouldn't be susceptible to this particular failure mode since Hubble is within the Earth's magnetosphere.
Not sure this is true. While they are shielded partially there are solar flares that do damage to electronics on earth so I think it is safe to say Hubble is still able to be damaged by solar flares.
I have seen this exact problem in a welding fixture that was mounted on bearings. It worked great for months but eventually the bearings would be so damaged from arcing that they had to be replaced. This just became regular maintenance but that is a little tougher to replace in a spacecraft.
Here's a question: Is the manufacturer at fault? Assuming the best intentions, they did their best to make sure it worked. It took many people years of studying the data to figure out a possible explanation, not to mention a crtain one. On the other hand, it hardly seems like an unknown phenomenon. The manfacturer also seems to be the one that made it cheapest so maybe one with more money invested in each, would have been more resistant. If you look at the end result only, you might find that you would have been better off buying twice as expensive units from someone else even if all the available data said that these were just as good, but those maybe did not end up failing.
It's just the manufacturers fault if someone figured out the problem beforehand. But the reason for those failures was only discovered after many years. Also those spacecraft ALL functioned well beyond their intended lifetime.
Any other manufacturer that used metallic bearings would have had the same problem. This isn't an Ithaco problem, it's a we-don't-understand-spaceflight problem. And I'm surprised NASA didn't attempt a scientific analysis of this earlier. Reaction wheels failing in space but not on Earth is exactly the type of thing a space agency should be trying to solve.
No, it's a design flaw that only could've been seen coming with either a multi-year test in space or them getting lucky and getting a solar flare very soon after launching said test. With NASA's current budget it's a bit hard to justify the cost of testing things like this sadly.
Interesting! Certainly seems like a plausible mechanism for the failures. Also interesting (and good) that no ceramic reaction wheels have failed (so far).
The problem with the metal bearings is that the lubricant is a very good insulator, otherwise there would not be a problem, a conductive lubricant would be an elegant solution for metal bearings. Arcs are not an issue if they do not damage any surface needed for the bearing rolling smoothly. The trouble is only that the arcs hit precisely the contact points where the balls touch the tracks they are rolling in. For this problem, solutions are numerous and trivial once the problem is known.
There was a 1200 horsepower motor at a cement plant where I frequently work that had repeated bearing failures from electrical discharge across the bearings. They ended up replacing the bearings of the tail end of the motor with ceramic bearings to reduce the electrical current problem, and had a ground brush connected to the shaft to drain the voltage away without it crossing the bearings.
An odd question related to reaction wheels: Has a Ferrofluid-Induction Field combination ever been used? And would the loss of energy in the system to Brownian motion/fluid friction still conserve the total system momentum? I guess I'm wondering if it's possible to dissipate momentum into chaotic systems as thermal energy without Newton rising from the grave...
There's a lot of unanswered questions regarding this. I'm guessing that these bearings are in a sealed unit? Has anyone checked to make sure that whatever lubrication is being used will actually work long term exposed to vacuum? Has anyone tested to make sure some kind of vacuum welding isn't taking place? Generally the most inane things are studied when it comes to aerospace engineering, so I would be surprised if there wasn't a lot of documentation on it... But the problem is how much LONG TERM study is being done? What happens when this lubricant/grease is subjected to vast temperature changes (because going from being in the sun's light to behind a planet tends to cause a drastic heating/cooling phase) cycles? That's prime territory for metal fatigue from constant and continuous thermal expansion and cooling. What happens if something causes this grease/lubricant to dry up or harden? It might take months to do it. And, of course, you can run into issues with non-insulated metal and high voltage - we have have issues with melting, spontaneous welds and electrical discharge damage here on earth with some things (although admittedly probably not much in the way of ball bearings).
It takes very little current to ruin a ball bearing, and this has been a known problem for a very long time. There's even a special term for this damage in electric motors: en.wikipedia.org/wiki/Shaft_voltage Bearing companies themselves have studied this to death: www.waukbearing.com/en/technical-resources/bearing-damage-index/bearing-damage:-electrical-pitting/ This is something any experienced mechanic or welder is aware of, I have seen multiple failed bearings from poor grounds.
Which makes you wonder if they ever consulted a bearing manufacturer(that would immediately point out shaft voltage) or if the design was fully home brew.
This was known to happen in caged roller bearings on cars and trucks. They would get ESD damage (pitting and spalling) over time if certain ground straps were broken. It was my first thought when I heard about reaction wheel bearing failures. Makes sense.
An aluminum housing would shield RF fields but not magnetic. Induced currents from magnetic fluctuations would take out the bearings. Good detective work.
John Drachenberg It's easy enough to test that electrical discharge will damage the bearings. Testing to see if solar flares will cause sufficient voltage across the bearing is a different story.
It's not just that it's the type of radiation - protons and muons for starters as well as everything from about microwaves through to gamma. Even though the sun is 93 million miles away the amount of crud and energy that hits us per square meter has to be seen to be believed. 1kw/m2 is just the optical light. The amount of power the sun kicks out is mind boggling and it's not a big star.
Some cars are prone to having this sort of problem -- if the ground strap comes loose from the engine it forces the current to find another route. Sometimes this can go through a bearing or bushing, causing it to eventually fail.
EXCELLENT !!!!!!!!!!!!!!!! This video shows so wonderfully the degree to which no matter how much effort and intelligence you put into a design, you simply don't know what you don't know!!! But we do learn!
Why is there a minus symbol between Newton and meter in the table at 3:00 ? That doesn't comply with any standard notation I'm aware of. If anything, there should be a multiplication symbol.
One of the sad consequences of not having some replacement program for the former US Shuttle program is that we can't retrieve a sat to do a post mortem on it. I remember so many missions where the shuttle crew was just replacing parts on the Hubble.
Thanks for that. It has relevance to less critical things down here too. Not the solar flare part but the spark erosion mechanism through the lubricant aspect.
My dad was an electrical engineer, worked 40 yrs for a company that made solenoids for the aerospace industry. Many times he was asked to bid on making parts that he suspected would not work properly in space. He would ask how the part would be used, under what conditions, temperature extremes, etc. then he would run it past his engineering team. It wasn't unusual for them to suggest that maybe they should redesign the parts before putting them out for bid.
This is an interesting study of why some spacecraft are experiencing problems. It is nice to see some science being applied when space craft have problems. This is an interesting item.
Oh snap! I work in one of the buildings linked with Salesforce park and walk around it all the time. Crazy to see you standing there! Love your channel.
I used to work in industry building 1, 2 and 3 axis positioning systems for satellite and telemetry tracking as well as radar and EW systems. These were usually outside in the environment. We always had a hard cable from one axis to the other, when possible, for grounding in case of lightning strikes because the discharge would go right through the bearings and fuse them. It was like hitting the balls with an arc welder. If we had slip rings for continuous motion we would reserve a number of circuits just for ground in case of lightning. It was easier and cheaper to replace a slip ring than a bearing.
Grounding paths are very important. Back in the seventies the VW Rabbit had a missing grounding strap that attached to the engine. The alternative grounding path was through the front left axel and the wheel bearings. Arcing in the bering race would destroy the bearing. It did a great job of destruction, speaking from personal experience with a 1976 VW Rabbit.
Scott, I don’t understand why the designers of these extremely expensive space craft don’t add some form of locking points on the external parts of the vessel. This would make several things much simpler. 1: maintenance robots could be flown out to the craft and have a solid locking point to use to secure themselves to the craft. 2: if a guidance system like the reaction wheels failed a supplemental guidance system could simply be flown out to the craft lock itself onto the craft and be used to extend the mission of the craft. 3: let’s say a second or third or forth guidance system was needed, with the correct design all the systems could be added by a modular system. I’ve never read anything that would suggest that any of our spacecraft are related to each other in any modular system. I understand each new spacecraft is developed under very narrow and specific guidelines, but a back up guidance system set up as a modular system does make a lot of sense.
Going out on a limb here, but in my youth I use to be a tesla coin fanatic and one thing that became very clear to me after a while is that when you are applying a large amount of energy to a high mass object is that eddy currents inside of the object, even if it is what you would consider shielded can easily reach into the hundreds of volts for an instant.
Thank you for podcastposting this one Scott, I have indeed wondered. In a ball bearing there's very little surface on the top of the ball, so yes comparably little amps is needed to weld it in place or leave a residue as shown on your video. Ceramics seem to be the way to go, or at least in some part of the ball bearing to stop it from being conductive, especially on spacecraft that use Ion engines - or any space application that use the kilopower units.
Funny you brought up this topic. My bathroom fan is failing, but only the bearing that has metal connected to each side is affected. The problem seems deeper than anybody thought.
Surprising this wasn't addressed in design of the bearings. EM discharges through bearings causing spalling and failure is a known issue with inverter driven motors and is addressed by additional bonding of the rotating shaft.
I did a similar study on Relays for my MSc. That was about identifying failure mechanisms in advance but the mechanisms were the same. I imagine that there might be a condition where the arc causes metal to vaporise into the lubricant and reduces the breakdown voltage, resulting in more arcs. That’s a big issue in immersed electrical gear.
Anybody know what that movement is starting at 3:52? It's on the far left edge of the sun. (Sort of looks like a "tornado" shaped thing). Just wondering.
All hail ceramic bearings
When having balls of steel isn't enough
haha good one buddy!
Joe, what you are saying is your balls are steel and impotent. Im sure thats not true, but just couldnt resist making a joke. No hard feelings brother.
Ceramic bearings would probably have fewer problems because of the surfaces. Would probably handle heating/cooling cycles better than steel and electrical discharge. But, then again, if the grease/lubricant being used is causing undue friction for some reason you'd still have failures.
Joe I really didnt mean it as anything but a cheap joke, but thank you for taking it sportively.
matchesburn Use no lube then!
So I will begin with "I am not a rocket scientist" however as an industrial HVAC and commercial equipment service technician I replace bearings all the time and as a matter of fact I am currently covered in grease from a 15 hp electric motor that blew a rear bearing and cost my customer over $2000 to fix. There is this extreamly common thing that happens to electric motors that are being driven off what we call a VFD (variable frequency drive) it allows us to control the speed of a motor but because of how a vfd works and converts the voltage it creates voltage spikes in the rotor that hava no where to go but through the bearings and into the housing, this pits the balls, races and destroys the bearing over just a few months. The thing to keep in mind is that there is a very small amount of contact area from the ball to the race so it doesn't take much to destroy the bearing in a very short period of time. Something as simple as a rotor grounding shaft would have eliminated the bearing failure.
Conductive lubricants will also eliminate these failures, Polyrex EM or Retinax EPX/M aerospace.
Thanks for taking the time to post a valuable comment. Respect to HVAC!
Voltage spikes are easily suppressed just use a snubber a RCD snubber will do
But what do I know , just a amateur hobbyist speaking
This is exactly what my mind went to when Scott mentioned solar flares. Searching for "bearing common mode current" provides lots of relevant results for anyone else looking for more info on the VFD/earthly sort of failures.
wheelitzr2 - ok, next time I’m gonna use a VFD, I’m definitely going to ground the rotor! 🤔
Ah yes electric arc across a bearing is why you have to be careful to not ground across a bearing when welding.
Yep, what amazes me is that because of the tiny gaps involved in the bearings even a few volts can create arcs with enough energy to melt steel.
I dunno grounding across bearings can be hilarious when a trainee does it. Flames (from the lubricant), a "bearing" that is now rock solid, and mickey taking material for the next month at least.
Well, there are no complete solids at atomic level anyway. Some particles sometimes get out of their places in grid.
It's just a matter of amount.
Not so funny for the welder that does it across multiple bearings in a turbine engine, We had one the was welding an exhaust nozzle crack, but grounded to the front fan assembly on the engine, welded every bearing in the engine.
James Collier it's a wheel that you spin to spin your spacecraft the opposite way (hence reaction) without wasting propellant. Used to orient and stabilize.
Every action has a equal and opposite reaction. In order to spin up something in space you need something else gaining the same angular momentum in the opposite direction
You can get same failure mechanism with cheap frequency convert drives on electric motor bearings. The ceramic bearing is good but expensive fix the other solution is install grounding carbon brush but that's needs probably some maintenance over long periods of time so using just ceramic bearings on stuff like that is probably good idea, especially when they are already stupidly expensive :D
I see you are still doing research in to rocket science.
how is going you rocket program?
His wünderwaffe rockets are nearly ready to be deployed for the next step in the secret world domination plan.
I have my big rocket with 1000 N of thurst still under work :D I have been to busy but I think I should get back to it
@@HydraulicPressChannel = Elon Musk's alter ego confirmed 0.o
ceramic bearings aren't even that much more expensive when you consider the cost of the spacecraft
Missed opportunity: "Bearings fail when the friction on them become unbearable."
Are you saying he should just roll with it?
Rofl
Genius comment and replies.
Rolling on the Races Laughing - RORL
You three will make great dads if you aren't already.
I worked on commercial space communication satellites, and my dad worked on Hubble Space Telescope. Reaction wheels were always acting up. On Hubble some of the reaction wheels have been replaced on servicing missions, the only spacecraft (other than the ISS) that such replacement in orbit is possible. On our commercial satellites, there were usually more than 4 wheels, so there was some redundancy. I always wondered why such a (seemingly) simple device was so troublesome. Your video provides some interesting insight into this issue.
An easy fix, if someone would have guessed the issue, would be to add some radioactive material near the bearings, because ionizing radiation dissipates away static electrical charge as it accumulates. Or make the lubricant electrically conductive. Or, as they did by accident, switch to ceramic bearings.
I was surprised that they were using BBs and not magnetic bearings. It's possible for metals to cold weld together in a vacuum if they've lost enough of their oxide layer so that could be happening if the lubrication film breaks down enough to allow bare metal to contact in the bearing.
Meme Master Lubricants that work long-term in a vacuum are used. Everything was and is lubricated very well and coated in a lubricant layer in any of these mechanical bearings out in space.
ed p anything using bearings will always fail at some point, and something spinning incredibly fast (some at the limit of their tensile strength) will fail more quickly.
"...the only spacecraft (other than the ISS) that such replacement in orbit is possible." WAS possible. Since the retirement of the space shuttle fleet, there is no spacecraft currently able to rendezvous with Hubble and secure it sufficiently to perform any sort of maintenance on it. Its days are now numbered. HST might well last well into the 2020s. Even if the JWST is completely successful, HST is capable of "seeing" in wavelengths that JWST cannot. A 6th servicing mission may well be approved some day, using a spacecraft yet to be deployed, in order to save one of the most beloved spacecraft ever launched into orbit.... at least, for a time.
I remember when Bill gave that presentation to all of us former Ithaco Space Systems employees on the fifth anniversary of it's closing. He set up a projector in his barn and after our barn party reunion we all listened very intently. I saw the picture of a CME on the screen, looked back at Jeff Golden, then to Eric Stromswold, and my jaw dropped as they let me know that they agreed.
The ball bearing is basically the gateway for static potential to discharge from a relatively massive rotor (hidden under a relatively thin cover) to a massive dish. Yes, of course it would be subject to ionic buildup, regardless of any Faraday cage of the rotor cover. Remember, the outer bearing cover is attached to the outside of the main cover housing. Essentially, it's all right there waiting to pick up these ionic discharges.
I so miss working at Ithaco. Everyone there worked so hard at making those things work right... Laurie spent so many hours in the inner cleanroom taking so many measurements fir Bill and Eric WHILE ALSO building so many other RWAs. I stick to Torqrods so I didn't know much but I just remember how heartroken we we're when we lost the Iridium contract because of the failures. That was the death of a great company.
That's a cool personal story.
@therealnightwriter Are you a bot?
Wow, any more stories? Such a small world. I was at that barn party too. I was the small, meek fellow in the back corner, nervously chewing on sunflower seeds expecting someone to step on me at any moment.
@@SpenserRoger You worked there, or family? I was only st Ithaco for the short remainder of its existence as a UTAS company, October 2011-March 2013. I built a prototype motor driver card for GOES-R and a lot of test cables. Also wound one SMAP Torqrod
lol sorry I was just pretending to be a mouse in a corner. I did want to hear any other stories though. Thanks.
"the pressure of the sun's light" now that's some awesome physics
Now, if we could just figure out a massive photon gatherer and grab enough photon mass
equal just 1/10 th the mass of what is being driven. Wow, ranks right up there with the wormhole
thing.
Michael A. yes it’s called solar sail and the japanese has tested some in space already
Damn spoilers
Airbus have been using solar sailing to stabilise communication satellites since the 1980s. Old technology.
I still don't understand why he said reaction control propellant.. when clearly their motorized wheels actually I don't know what kind of thing is propelling the wheels
you know its also possible they realized this about 10 years ago said "yikes!" and quietly switched to ceramics
While this is absolutely possible, I rather think they switched because of weight saving.
Or durability. Ceramic bearings should wear a lot slower.
that is exactly what i though, also, if theses reaction wheels are "high quality" how bad is the competion????
Probably the same quality, just more expensive.
I work for a large organisation and look at equipment used at our facilities. One of our hardware vendors insisted that nothing was wrong after i directly showed them what i believed was a problem. They later admitted the issue but by that time they'd rolled out a new model, and guess what? The new model had exactly addressed the issue i pointed out, meaning that at least someone in their organisation knew about it for a long time. Corporations can be as dishonest with each other as they want if they feel they can get away with it. I wouldn't be shocked if it's true that they moved to ceramics here because they secretly knew about the defect.
Finally, the aerospace industry is catching up to fidget spinner technology.
I mean hell, my MS explorer trackball has had ceramic bearings in it for years since the metal ones wore flat forever ago. You’d think they’d want ceramic just for lifespan if that holds true for this application.
@@gags730 haha so funny.... not
@@gags730 I’m sorry but I hate cranberry juice now some apple punch? That’s my jam. Also no fidget spinners sadly I broke them all by using them too much. And instead of a glowing one I did have one that flew so there’s that 😉.
@@gags730 too bad 😂 lemonade ain’t half bad either
Made me lol
Fascinating! Space is such an extreme environment and there's still so much we don't know yet.
- 🥺 Bombarded with audio all day I like to read the closed captions you provide in your always very informative video presentations. I really apreciate your thoughtful consideration. Keep up the good work. 👍
A friend who is building large CNC machines said it is a huge problem to ground these machines properly because ESD (when machining some matertials) and the currents of the motors are causing these problems with the bearings
Add some ambient ionizing radiation, this will dissipate electrostatic buildup, and it will give you cancer, but for grounding moving parts across an air-gap, it really works, by knocking electrons out of surfaces.
Balanced Ion showers are the only way to work on instrument grade CCDs.
@@kurtilein3 An alpha emitting radioisotope would accomplish this.
@@taraswertelecki3786 Actually it is Beta radiation that is composed of electrons. (Alpha is essentially a Helium nucleus).
A good experiment for the X-37A would be to put some reaction wheels in it and run them in space for a year or two then bring them back and take a look at them.
I absolutely love it that you insist on using the satellite girl figurine every time you talk about the Hayabusa. Maybe next time show a picture of the motorcycle :^)
Which other times has he done that? I wanna see now.
... or a picture of the Nakajima Ki-43 Hayabusa (WW2 fighter)
ruclips.net/video/Pwjvg_-7Ayw/видео.html
It's right to the end of the video.
moe makes the world go round.
Most fascinating. I work for a major satellite fleet operator. We've been seeing this exact type of wheel failure in some ~15 year-old spacecraft which were expected to last even longer. I'm going to go back and check the data for correlations. Glad that there seems to be a fix in the newer wheels, even if accidental. Thanks for this one, Scott!!
And here I thought it was Sun Rust caused by damp and salty photons. Boy, do I feel silly now.
Could be stowaway tin worms.
fogllama you are so weevil sir.
Special humor. I love it.
OMG 😂
Reminds me how the FIA band the uses of ceramic connecting rods in the mid 90's highly evolved and insanely expensive for most team's to develop.
I put ceramic bearings on my fidget spinner and it changed my life.
Damn Scott.. you always pick the most interesting things to talk about.. Plus you put them to us in a way that this 66 year old fart can actually understand.. or maybe it is just because of my Scottish heritage (Lamont clan) helps me to understand... Well done brother.. carry on...
And he seems to provide further, conclusive evidence that Scots will be the first Space Superpower. Because everyone else's stuff will be breaking down all the time.
@Cosmo Genesis Sheesh... Now there's a thought for ya... Woof!
This has to be one of my favourite videos on this channel yet. A nice inside on the engineering problems involved in rockets and space probes but easy enough to be understood by laymans.
They should make a craft full of reaction wheels, send it to orbit, Wait for like 50% to fail, then have it rendevous with the ISS for inspection and analysis.
It would need to get out past the Van Allen belts to really soak up the rads. Might as well put it inside a heatshield, send it on a hyperbolic orbit out to L2 and then let it reenter.
Meh, a cube sat, stiffed with fairings on different axis, all spinning and sensors or maybe a mechanism to open and photograph the bearings.
Unless ceramic bearings aren't the only needed fix, it could be something like the lubricant not having the same conductivity across it leading to the exact same problems, where parts of the fluid will conduct the electricity and cause micro pits while others won't.
What if the pits aren't the problem, but the sudden heat produced is burning off the lubricant on that spot and over time it just burns away enough to cause high friction/wear?
Have funding will travel...
@@jamestheotherone742 No bucks, No Buck Rogers
New mod for KSP : Carbon ceramic reaction wheels
New mod for KSP: failing reaction wheels
People often fail to realize that there is no such thing as a perfect Faraday cage. You can't just wrap something in micron thick foil and expect infinite attenuation.
Seems to me like you'd need something more than just a standard metal enclosure to keep voltages across the bearings below 6V during a decent size solar flare.
Good point.
It's not the thickness, it's how good the connection is along all the joints. You use things like this: www.tech-etch.com/shield/images-shield/assortment-of-emi-shielding-products.jpg
But ionizing radiation can also cause a voltage spike and flares give off some quite energetic particles. If that's the culprit, good luck shielding from that.
So, the grounding is to blame, as usual
lets attach a long wire from the satellite to the ground, that should fix the problem.
Mining - you're confusing grounding with earthing. Neither have anything to do with effectiveness of a Faraday shield. That's determined by the conductivity/admittance of the cage.
"I'm Scott Manley, fly ceramic!"
I love videos like this, answering questions I never even knew I wanted the answer to.
My first thought to solve the problem was to use heat resistant ceramics to replace bearings; you can do a home experiment with anyone who has had an electric train set, to demonstrate what static discharge event being discussed looks like in real life:
All you need is an outlet adapter for a race track or a train set. The racetrack or train set adapter will usually come with lead wires that travel to the racetrack or track connections. This is the wire I used as a child to conduct this experiment. With the adapter charged to any degree above zero, you will see a blue spark as the wire contacts the quarter and grounds out the current through the quarter in a short circuit. This would be the same effect as having the charge jump from one bearing to the next, as another aspect, the bearings themselves may have different charges from each other during a CME (Coronal Mass Ejection).
Remember, these Coronal Mass Ejections were recorded to make sparks on telegraph wires within our own atmosphere during the 1800's when electricity was not as wide-spread and in use as a part of major industry as it is now. That is surely enough to make the same kind of spot scarring you see on the surface of the quarter you've been creating sparks on. You can feel the part of the metal where parts of the wire got fused to the quarter during the exchange of electrons, also used as a practical application in welding, called arc welding.
These small imperfections would be catastrophic to bearings needing any degree of precision in a very highly machined quality product of microscopically precise application.
Good article!
I think things like this is why we should try to plan for a potential recovery of some probes. I'm no rocket scientist or astrophysicist, but surely it would be worth the extra trouble to actually get a good ol' fashioned 'MK-01 Eyeball' looking at things, right?
Maybe if some of them are in low Earth orbit, we can take a look. Otherwise, it's simply not practical. Spacecraft are only given enough fuel for their mission, and not enough fuel to return to Earth.
I think most of these craft are in solar orbits, or very high Earth orbit, so we don't yet have anything capable of reaching them and returning to Earth. They are also mostly outside of Earth's magnetosphere, and thus more exposed to solar radiation, which contributed to the problem.
Play some KSP and try to recover a probe you deployed into interplanetary space and I think you'll answer your own question.
Pfft, you're clearly not thinking the real kerbal way. Just build another Saturn V and send up a little robot arm + capsule to grab and return one of these probes.
Oh that's what I do. I just....misinterpret the numbers on occasion (read: all the bloody time), resulting in recovery vessels needing recovery vessels....Which need recovering in true Kerbal fashion.
I spent an extra 20 minutes on the toilet pondering this, Scott. I'm fascinated and completely evacuated as a result.
Somebody once told me, the Sun is gonna roll me, and even r-wheels would fail.
Electrostatic shocks, caused our ball bearings to block and left the Rocket Engineers looking pale
It was looking kinda bleak, till one paper found a cheet with the shape of the so-lar pan-els
She was looking kind of smart with her paper and her chart in the top shared news on Nature
(Ah, damn, I'm late)
Temp Name I had the same realization with an on's post, I liked yours before you removed it though ^^
This makes perfect sense to me. I remember reading a story many years ago about how an electric motor bearing was found to be failing due to a short between the rotor winding and the rotor itself. The small electric current through the bearing caused increased bearing wear, and eventually, early bearing failure.
This is so awesome. In one video I learned that satellites use reaction wheels, that they malfunction, why they probably malfunctioned and how it likely has been solved.
Thanks, Scott. This deep-dive into an obscure facet of deep space exploration is why I keep tuning in.
I just love the picture you choose to represent Hayabusa.
Im sure this hypothesis can be put to the test in lab to provide more definitive results right? Great episode by the way
@Scott Manley - Thank you for helping to reveal how and why these mysterious space borne failures took out many of my favorite missions. Not surprised and it makes very good sense.
Appreciate the effort. Astrophysics and astronomy are my first loves... I too went into tech.. as a hardware engineer in multimedia telecom and network products. Retired now.. JK in Berkeley
I know I'm late but this was intriguing but more so surprising. I would have figured they considered this and implemented some capacitive grounding rail with some discharging mechanism to mitigate the charge. I always assumed this would be done with any system launched into space/upper atmospheres.
I'm sure we'll see a conductive lubricant on the bearings in the future or more ceramic bearings. I'm really surprised ceramics were not used in the first place. With a much lower mass, smoother operation and longer life they seem like the go to choice. I'm sure there was more to it though. Great video!
Great video buddy as always Scott.
I'm a mechanical engineer , and it makes alot if sense.
There 's a mechanica proces called "electric sprak erosion" , that is done in an oil bath .
The work piece is connected to the + pole and a graphite electrode -
When currency is switches on , the elektrode cavitates a hole in the workpiece.
From the moment you mentioned static charge , i already knew what the study would say 😉😁...
Grtz huge fan johny geerts
Not sure about the ceramic balls solution. Static charges can build up on the rotor before jumping out within a dramatically thousand volts spark. Saw this failure on high perfomance machine-tool spindles.
I thought they would go in the direction of conductive lubricants.
What exactly would be the problem if the sparks go around the bearings instead of damaging them?
Also, Einstein got his Nobel Prize not for relativity, but for the photoelectric effect. Add a bit of radioactive material, exposing the whole thing to some low level of ionizing radiation, electrons will get knocked out of surfaces and will travel to reduce any electrostatic charge. At high levels, this is poison to any electrical devices, this is why all robots fail inside the fukushima nuclear ruins. At low levels, this is poison only to any electrostatic buildup of charge. This works in air or in a vacuum. No further design considerations as long as you have conducting surfaces seperated by nothing other than air or vacuum, add some ambient radiation to drop voltages and electrostatic charges across the system.
@ Conductive lubricants tend to have higher viscosities and particulate content. Perfectly fine for an industrial motor, but completely unsuitable for high speed high precision applications like a reaction wheel.
Very interesting failure mode. Easy enough to remedy using a simple spark gap though. A few closely gapped sharp points in the right locations should allow any discharge to occur at non-critical areas of the spindle assembly. Or, as kurtilein3 suggested, use radioactive material to effectively decrease the work function, the results would be similar.
Thank you for putting the link to the paper in the description
Solar Freakin' Welding!
As someone who uses K2 data, its a pain that the reaction wheels broke because since it makes the data messy. But on the other hand its great because we could use Kepler to hunt for exploding things in K2. So for me its a good thing that Kepler didn't have ceramic reaction wheels!
The more we study the more we find out.. human knowledge is increasing rapidly.
The more you learn, the more you realize you still have discovered nothing, and that you are nothing in the grand scheme of things. That is frightening and fascinating at the same time.
Theres a small pebble of attractive conglomerates in my batroom as an ornament. The story it could tell is millions of years old. It was as it is millions of years before our ancestors descended from the trees. I take a bath, many baths as I grow ancient, soon to die, but this pebble will remain as it is for millions of years to come.
And then a flat Earther appears
@@chris-hayes Well, of course, stupidity is increasing as well. Don't worry, stupidity is deadly, but that's why it will die off soon.
OP and his edgewad suck buddies think they're deep because.... the universe.
I had noticed this and had a strong curiosity as to why the reliability had seemingly not improved over quite a span of years. Thank you for speaking about this potential failure mode as it is both interesting and satisfying explanation.
Would it be feasible to create a zero-contact reaction wheel with magnetic rails like a maglev train? Even though there's no contact, angular momentum would still be transferred by pushing off the magnetic fields. As an added bonus, the system's magnetic field may offer some protection against solar winds if it's strong enough.
Assuming you use superconductors then sure but spacecraft already have issues radiating heat
Conventional bearings are zero contact - they ride on a film of oil. Sparks jump the gap. Sparks can still jump the gap in your magnetic bearing..
@@stargazer7644 except that oil can conduct electricity, whereas the vacuum of tinyfoxtoms mag bearing couldnt
@@deanrensberger631 Oil is a dielectric. It doesn't conduct electricity. In fact, all those large power transformers you see on poles and on the ground are filled with oil.
@@stargazer7644 you dont understand what we're talking about. Oil isnt a perfect insulator like the vacuum. Its conductivity is what allowed these mcgs to fail.
This video is the kind of high quality content that makes me watch your channel. Great stuff Scott!
It could also be polymerization in the lubricant- secondary radiation can turn lube into glue pretty easily.
This idea is old enough that lubricants for use in space are tested for this in nuclear research facilities. You can just go and buy high-performance lubricants that are rated to withstand both a vacuum and intense levels of radiation. It is too obvious.
Very interesting, as I’m a retired marine machinery mechanic & the Navy back in the late 1990’s experienced a similar problem after a week of conducting a heat run test/monitoring of a newly installed bearing. It was determined that the manufacturers were to blame.
Sounds like a solid theory. Nothing worse than a good theory that can’t be tested.
Great food for thought, I've actually been wondering about this! Thanks Scott!!
Is the Parker Solar Probe using the new or old wheels?
This video is a perfect example of how many different ways there are for something to go wrong, and very few ways for something to go right.
Does this mean there could be a higher risk of failure for the Parker Solar Probe?
doubt is uses reaction wheels prone to failure. thats if it does use reaction wheels
This reminds me of a known problem with bearings in movable bridges. When a bridge deck is not properly grounded, lightning strikes can actually weld the bearings together, much like the described effect in this video.
Didn't we return reaction wheels from Hubble? I assume they were able to measure the changes in material properties.
The Hubble reaction wheels weren't built by Ithaco, so they weren't involved.
I was aware they were built by a different source but as long as they used metal parts similar to Ithaco instead of the newer ceramic they would still be able to use it to isolate the variable of who made it to narrow it down to possible common design flaws.
Deric Anthony Didn't Hubble servicing mission 3B bring back the gyros and the reaction wheel? I think the gyros and reaction wheels were designed in the early 80's. Someone should find out and correlate to any solar events.
@@scottmanley Even if these were made by Ithaco, they still shouldn't be susceptible to this particular failure mode since Hubble is within the Earth's magnetosphere.
Not sure this is true. While they are shielded partially there are solar flares that do damage to electronics on earth so I think it is safe to say Hubble is still able to be damaged by solar flares.
Metallic vs Ceramic bearings in those harsh conditions... that is such a no-brainer.
Thank you Scott, we enjoy your amazing videos alot.
That’s an interesting problem, funny how some things we regard as non-significant can actually be in some ways!
That is fascinating, who knew we couldn't use the same materials on spacecraft because space is so harsh! Amazing video Scott!
It’s called bearing fluting, a ceramic coated outer race is the cure.
I have seen this exact problem in a welding fixture that was mounted on bearings. It worked great for months but eventually the bearings would be so damaged from arcing that they had to be replaced. This just became regular maintenance but that is a little tougher to replace in a spacecraft.
Here's a question: Is the manufacturer at fault?
Assuming the best intentions, they did their best to make sure it worked. It took many people years of studying the data to figure out a possible explanation, not to mention a crtain one.
On the other hand, it hardly seems like an unknown phenomenon. The manfacturer also seems to be the one that made it cheapest so maybe one with more money invested in each, would have been more resistant.
If you look at the end result only, you might find that you would have been better off buying twice as expensive units from someone else even if all the available data said that these were just as good, but those maybe did not end up failing.
It's just the manufacturers fault if someone figured out the problem beforehand. But the reason for those failures was only discovered after many years. Also those spacecraft ALL functioned well beyond their intended lifetime.
Any other manufacturer that used metallic bearings would have had the same problem. This isn't an Ithaco problem, it's a we-don't-understand-spaceflight problem.
And I'm surprised NASA didn't attempt a scientific analysis of this earlier. Reaction wheels failing in space but not on Earth is exactly the type of thing a space agency should be trying to solve.
If only *someone* gave them funding...
If it is designed to all of NASA's requested specifications, that is all they need to fulfill
No, it's a design flaw that only could've been seen coming with either a multi-year test in space or them getting lucky and getting a solar flare very soon after launching said test. With NASA's current budget it's a bit hard to justify the cost of testing things like this sadly.
A life without Scott is a life not worth living. Thanks for keeping my brain busy on my downtime, otherwise I'd be trying to take over the world.
try welding on an engine lathe see how well it works after. I learned this in high school.
Interesting! Certainly seems like a plausible mechanism for the failures.
Also interesting (and good) that no ceramic reaction wheels have failed (so far).
"Oh come on guys! It's so simple, maybe you need a refresher course. It's all ball bearings now days."
Fletch!
I really like seeing Cody'sLab commenting on these videos. But as always a seemingly random video with a lot of interesting material to view.
first thought was to 'go ceramic'. but then if the lubricant is conductive, a coated ceramic bearing will still arc.
The problem with the metal bearings is that the lubricant is a very good insulator, otherwise there would not be a problem, a conductive lubricant would be an elegant solution for metal bearings. Arcs are not an issue if they do not damage any surface needed for the bearing rolling smoothly. The trouble is only that the arcs hit precisely the contact points where the balls touch the tracks they are rolling in. For this problem, solutions are numerous and trivial once the problem is known.
There was a 1200 horsepower motor at a cement plant where I frequently work that had repeated bearing failures from electrical discharge across the bearings. They ended up replacing the bearings of the tail end of the motor with ceramic bearings to reduce the electrical current problem, and had a ground brush connected to the shaft to drain the voltage away without it crossing the bearings.
I was just about to suggest ceramics and... I'm some ten years late. XD
An odd question related to reaction wheels: Has a Ferrofluid-Induction Field combination ever been used? And would the loss of energy in the system to Brownian motion/fluid friction still conserve the total system momentum? I guess I'm wondering if it's possible to dissipate momentum into chaotic systems as thermal energy without Newton rising from the grave...
I'm sure it has but the biggest problem is that space is a vacuum and so fluids tend to boil away
There's a lot of unanswered questions regarding this. I'm guessing that these bearings are in a sealed unit? Has anyone checked to make sure that whatever lubrication is being used will actually work long term exposed to vacuum? Has anyone tested to make sure some kind of vacuum welding isn't taking place? Generally the most inane things are studied when it comes to aerospace engineering, so I would be surprised if there wasn't a lot of documentation on it... But the problem is how much LONG TERM study is being done? What happens when this lubricant/grease is subjected to vast temperature changes (because going from being in the sun's light to behind a planet tends to cause a drastic heating/cooling phase) cycles? That's prime territory for metal fatigue from constant and continuous thermal expansion and cooling. What happens if something causes this grease/lubricant to dry up or harden? It might take months to do it. And, of course, you can run into issues with non-insulated metal and high voltage - we have have issues with melting, spontaneous welds and electrical discharge damage here on earth with some things (although admittedly probably not much in the way of ball bearings).
This is why so many people want these spacecraft recovered - so we can crack them open and see exactly what happened.
Is it possible to create a conducting grease to dissipate the charge? Perhaps with graphite?
It takes very little current to ruin a ball bearing, and this has been a known problem for a very long time. There's even a special term for this damage in electric motors: en.wikipedia.org/wiki/Shaft_voltage
Bearing companies themselves have studied this to death:
www.waukbearing.com/en/technical-resources/bearing-damage-index/bearing-damage:-electrical-pitting/
This is something any experienced mechanic or welder is aware of, I have seen multiple failed bearings from poor grounds.
Well, the obvious suspects are the most likely ones to have been adressed.
Which makes you wonder if they ever consulted a bearing manufacturer(that would immediately point out shaft voltage) or if the design was fully home brew.
This was known to happen in caged roller bearings on cars and trucks. They would get ESD damage (pitting and spalling) over time if certain ground straps were broken. It was my first thought when I heard about reaction wheel bearing failures. Makes sense.
You get what you pay for, even in rocket science.
An aluminum housing would shield RF fields but not magnetic. Induced currents from magnetic fluctuations would take out the bearings. Good detective work.
Why couldn't we test this theory on the ground? Surely we could simulate a sufficiently large enough plasma discharge...
John Drachenberg It's easy enough to test that electrical discharge will damage the bearings. Testing to see if solar flares will cause sufficient voltage across the bearing is a different story.
It's not just that it's the type of radiation - protons and muons for starters as well as everything from about microwaves through to gamma. Even though the sun is 93 million miles away the amount of crud and energy that hits us per square meter has to be seen to be believed. 1kw/m2 is just the optical light. The amount of power the sun kicks out is mind boggling and it's not a big star.
We can and they did. The issue is figuring out whether that is actually what caused the reaction wheels in space to fail.
That sounds like exactly the kind of thing we don't want to build on earth. Therefore I'm all for it.
Jukelo we can replicate the type but nothing close to the intensity experienced after a CME.
Some cars are prone to having this sort of problem -- if the ground strap comes loose from the engine it forces the current to find another route. Sometimes this can go through a bearing or bushing, causing it to eventually fail.
Great explanation. And I thought it was just more "space cooties" bogging up the works.
EXCELLENT !!!!!!!!!!!!!!!! This video shows so wonderfully the degree to which no matter how much effort and intelligence you put into a design, you simply don't know what you don't know!!! But we do learn!
"Hwy do Reaction hweels fhail?"
~Scott Manley.
Why is there a minus symbol between Newton and meter in the table at 3:00 ? That doesn't comply with any standard notation I'm aware of. If anything, there should be a multiplication symbol.
One of the sad consequences of not having some replacement program for the former US Shuttle program is that we can't retrieve a sat to do a post mortem on it. I remember so many missions where the shuttle crew was just replacing parts on the Hubble.
The Kepler spacecraft is as far away as the Sun. Humans can only reach low Earth orbit.
Thanks for that. It has relevance to less critical things down here too. Not the solar flare part but the spark erosion mechanism through the lubricant aspect.
In before the electric universe nutjobs
ive never pay any attention to electric universe, but just keep in mind. 2000 years ago they would have said "in before the round earth nutjobs"
@@aPOTg They believed the earth was a sphere even way back then.
Wat
Yep. Greek guy figured it out looking at shadows in wells.
@@NikovK And an Egyptian guy figured it out looking at an obelisk sinking below the horizon.
My dad was an electrical engineer, worked 40 yrs for a company that made solenoids for the aerospace industry. Many times he was asked to bid on making parts that he suspected would not work properly in space. He would ask how the part would be used, under what conditions, temperature extremes, etc. then he would run it past his engineering team. It wasn't unusual for them to suggest that maybe they should redesign the parts before putting them out for bid.
This is an interesting study of why some spacecraft are experiencing problems. It is nice to see some science being applied when space craft have problems. This is an interesting item.
Oh snap! I work in one of the buildings linked with Salesforce park and walk around it all the time. Crazy to see you standing there! Love your channel.
I used to work in industry building 1, 2 and 3 axis positioning systems for satellite and telemetry tracking as well as radar and EW systems. These were usually outside in the environment. We always had a hard cable from one axis to the other, when possible, for grounding in case of lightning strikes because the discharge would go right through the bearings and fuse them. It was like hitting the balls with an arc welder. If we had slip rings for continuous motion we would reserve a number of circuits just for ground in case of lightning. It was easier and cheaper to replace a slip ring than a bearing.
Grounding paths are very important. Back in the seventies the VW Rabbit had a missing grounding strap that attached to the engine. The alternative grounding path was through the front left axel and the wheel bearings. Arcing in the bering race would destroy the bearing. It did a great job of destruction, speaking from personal experience with a 1976 VW Rabbit.
Scott, I don’t understand why the designers of these extremely expensive space craft don’t add some form of locking points on the external parts of the vessel.
This would make several things much simpler.
1: maintenance robots could be flown out to the craft and have a solid locking point to use to secure themselves to the craft.
2: if a guidance system like the reaction wheels failed a supplemental guidance system could simply be flown out to the craft lock itself onto the craft and be used to extend the mission of the craft.
3: let’s say a second or third or forth guidance system was needed, with the correct design all the systems could be added by a modular system.
I’ve never read anything that would suggest that any of our spacecraft are related to each other in any modular system.
I understand each new spacecraft is developed under very narrow and specific guidelines, but a back up guidance system set up as a modular system does make a lot of sense.
Going out on a limb here, but in my youth I use to be a tesla coin fanatic and one thing that became very clear to me after a while is that when you are applying a large amount of energy to a high mass object is that eddy currents inside of the object, even if it is what you would consider shielded can easily reach into the hundreds of volts for an instant.
Good information as usual. Thanks for your work..
Thank you for podcastposting this one Scott, I have indeed wondered.
In a ball bearing there's very little surface on the top of the ball, so yes comparably little amps is needed to weld it in place or leave a residue as shown on your video.
Ceramics seem to be the way to go, or at least in some part of the ball bearing to stop it from being conductive, especially on spacecraft that use Ion engines - or any space application that use the kilopower units.
“I’m Scott Manley, fly safe” how did I become addicted to this frase? Love it! 😁💫
hweel, hwy. I love that there's a genuine Stewie out there, roaming the world like it's nobody's business. Keep on rockin!
I wonder if anyone has ever just stood near Scott while hes filming a video like this just for a lesson about space.
Funny you brought up this topic. My bathroom fan is failing, but only the bearing that has metal connected to each side is affected. The problem seems deeper than anybody thought.
Surprising this wasn't addressed in design of the bearings. EM discharges through bearings causing spalling and failure is a known issue with inverter driven motors and is addressed by additional bonding of the rotating shaft.
I did a similar study on Relays for my MSc. That was about identifying failure mechanisms in advance but the mechanisms were the same. I imagine that there might be a condition where the arc causes metal to vaporise into the lubricant and reduces the breakdown voltage, resulting in more arcs. That’s a big issue in immersed electrical gear.
I'm starting to like this channel more and more. Interesting stuff
Anybody know what that movement is starting at 3:52? It's on the far left edge of the sun. (Sort of looks like a "tornado" shaped thing). Just wondering.