Note this is only non-american carriers. The Nimitz class in the video is only stabilized by just being fucking giant and heavy. The issue is that the model is too light, plus waves aren't to scale. On the real deal the deck is about 60 feet above the water. That's right under the max size of any recorded rouge wave, so the likelyhood of a wave cresting over the deck if a Nimitz is near zero.
largish recreational boats have the stabilizers and I have seen a gyro based one on boats as small as 27' What this video is showing is basically like the Seakeeper.
Internal solution: Two reaction wheels, each one only spinning in one direction. Because they have to be smaller, you can have two and improve the response by them not having to stop and speed up the other way but only accelerate. They would have an idle speed instead of full stop and just speed up to create the stabilization effect in the direction required.
contra-rotating ones still get fully saturated and have to de-spin. and if they're not mounted next to each other, they'll exert a yaw torque along the length of the ship
@@recurvestickerdragon no, they dont have to "de-spin". You mount them inline, they have a certain idle rpm and there will be no yaw induced for that exact reason.
@@GaniscolOnce you've sped one wheel up to resist roll in one direction, it will have to be slowed down as you can't keep accelerating it indefinitely, and when you slow it down it generates a torque.
This brought back memories long lost. I once worked on the design of a reaction wheel for an FPSO (~450,000t tanker conversion to FPSO) almost 20 years ago, to control roll for improved seakeeping. The wheel was massive and to be installed inside the vessel.
@@NoNameAtAll2 Floating Production Storage and Offloading unit. It's a huge vessel permanently moored on top of oil production wells in deep waters, which is capable of receiving and processing the oil and gas, separate one from the other, store the oil and then offload it to shuttle tankers which then transport the oil directly to the national or international customer.
For your electronics, in wet environments, use solar panel bits. Like the roof pass-through boxes and bulkhead fittings. They're perfectly sized for small electronics and waterproof. $ is ok too, not too much.
Stability on surface ships does not normally rely on the centre of gravity being below he centre of buoyancy. The stabilizing effect comes from the centre of buoyancy moving faster outwards than the centre of gravity when the ship heels over.
@@BartJBols That stability is mostly created by the hull shape, not by a big honking heavy keel dangling under it. With the consequence @nerdyengineer7943 mentions: such a ship won't right iself once capsized. Simplest example: a catamaran: very stable around the neutral, but once you tip it over, it will remain upside down.
You should set the reaction wheel to spin at a relatively high speed and then adjust that speed up or down to control for roll axis, while the high speed acts as a passive gyro which would stabilize the pitch axis.
Great addition,thats a definite improvment. I'm making a reaction wheel Odrive cubli just now so wonderful to see the relative sizes of your parts and the effect.
@@smalltime0 well, there's possible problems from that, I just haven't seen it tried. Let's say that the net roll torque (because waves aren't symmetrical and you spend more time on one side of the wave). Then you have a net roll torque on the vessel, which because of the gyro, causes a net pitch on the vessel. This pitch will be resisted by the buoyancy and also by the gyro, creating another roll, I think in the opposite direction. So I think this will set up a very definite oscillation that just looks like precession, except it will be a factor of 3 to 10x larger oscillation in roll. I mean, it is SIMPLE, yeah. but the stability is questionable. I was mainly wanting to be entertained.
With a long ship like this, maybe a smoother option would be two or three smaller wheels, one towards front, one mid and one back. Could probably also better control pitch/twist correction that way.
I wonder also if you had them mounted closer to your center of gravity what kind of change it would do. And is the change better with a large flywheel were multiple small flywheels.
ODrive makes some awesome stuff, been following them for several years now, I have a ton of ideas, but have never gotten to actually using them in a project yet.
When I saw how stable it was, I thought it would be cool to try and land something on it. Then, it looked like you want to land a heli on it. I'm now subbed. Doesn't matter if you do or dont land something on it man,... I just like the way you think.
Use a fast spinning flywheel parallel with the boat, and then rock it back and forth to counteract roll. Seakeeper makes these for fishing boats and yachts
On big boats they mount a flywheel at 90 degrees but have it free to rotate on both axis. So when the boat rolls the gyro tries to yaw and similar for pitch. I think this results in dampening both roll and pitch Then by using actuators to move the flywheel in yaw they can create a roll reaction. I think the advantage of this is
Great video. The wheel is a little impractical for large vessels. You could try using a push pull weight on a slide rod, attached to either a cable or belt inside the vessel. You can use one for pitch as well as roll. It would mean two sets of electronics, but it would stabilize the ship so you can land the Apache! lol
What about mounting a spinning disc horizontally, to give passive roll and pitch stabilisation? Or maybe two smaller ones, rotating in opposite directions.
roughly mathematically you can make your gyroscope half the size if you can either quadruple the speed with a half size wheel the same weight as yours here. or by quadrupling the weight of the wheel and keeping the speed the same (half size wheel as well). maintaining the angular momentum in different form factors.
I think several smaller wheels could go on the inside, closer to the axis of rotation could be more effective. They would also help preserve the look of the carrier.
For the section at 2:10, it would've been cool to see a 2x2 of the footage we see now on the left, and the same two bits but stabilized on the horizon instead on the right
Marine engineer student here. At 2:13 you say it has some natural stability because the center of gravity is below the center of buoyancy, but that actually not necesarry to be stable. What´s important is the "metacenter height", which is a result of how the center of buoyancy shifts to the side when the ship leans over. Nearly no ships has a center of gravity below the center of buoyancy, both because it´s geometrically tricky to get such a low center of gravity, but also because the ship will get very twitchy if the distance between the metacenter height and the center of gravity is too long.
This makes me wonder if you would have a better effect with a pair of counter rotating gyro wheels? The counter rotation would cancel out any yawing effect caused by them that could slowly translate in the water while still allowing the boat to turn freely. They could also be smaller, lighter and more easily mounted (one in bow, one in stern) yet still provide massive amounts of roll and pitch stability.
@@nixer65 A single for the same effect would need to be larger and be more difficult to mount, plus it would create a rotational force (think of a helicopter) that would create a slow spin in the water thile the boat is stationary.
Looks impressive. Why do you place the wheel so high above the mass center and length axis? Bottom side of the ring working opposite direction you want to, but if the wheel will completely fit under deck -- both sides (top and bottom) will stabilize the ship, do not it?
We had a seakeeper gyro in one of our boats, and it uses a fast spinning weight spinning horizontally (like the old fashion spinning top) and then hydraulic piston to offset the roll to rotate the spinning top fore and aft. That way you still have all the torque available at anytime and it doesn't start and stop the flywheel. Although it will take almost an hour to spin up and reach full speed. Great job... 🤠👍
I'm sure a gyro would be much much better than a reaction wheel, that would be interesting too! Just spin it up, and use a servo to yaw it left and right, which will result in a roll torque.
You could spin it up to some higher baseline speed, adjust around that point for roll damping. The spinning inertia will then help stabilize pitching motions (although precession will impart a yaw moment onto the ship). Would be interesting to see how that works out.
I wonder if you could put some little bow thruster jets under the water line to help do the same thing but keep the runway open :) This is so cool though. Looks like a fun hobby.
You mentioned the roll axis bias. As you already know, that is from the bias in the roll axis angulare rate sensor. I am assuming you are not using a expensive rate sensor and they typically have a bias. To fix this issue, run the angular rate sensor value through a high pass filter. This will eliminate the bias in the sensor. I would expect 2-4 second filter would work sufficiently. I have not mess with the microcontroller you are using but I am fairly sure, you can add the filter. I would start with a 2 second filter, 2s/(2s+1). If you do not know how to implement this, I can create the code to use as long I I know the frequency of the microcontroller. Nice thinking on how to stabilize the platform.
Slimming down the wheel and mount it internal could help. Closer to center of gravity and lowering the center of gravity. You might not even loose reaction wheel force by slimming down the wheel if the motor is fast enough to compensate for the smaller wheel.
Was it difficult to keep it straight when going forward? The pitch up and down of the bow should have caused gyroscopic procession at 90 degrees on the gyro meaning it should try to turn you left/right each time a wave hit. just me pondering about it.. great idea for an experiment by the way! A company called Seakeeper actually make a vey expensive similar system for large yachts. thanks for the video!
Maybe adding some "ailerons" would help too. Surely they would be less weight than the gyro and extra ballast. Maybe the lighter weight would make up for the extra drag?
How much of a difference does it make that the fly wheel is offset from the rotational axis of the boat? Would it be more efficient for the flywheel to be centered on the ships rotational axis or does it get some mechanical advantage like a lever arm?
It needs the offset offset by a level sensor. A pendulum won't work for the same reason quads have problems with hanging loads. Centrifugal force is not going to be always down.
I am curious, is you method better than mounting the wheel 90 degrees around a Z axis? This is kind of what makes it easier to stay upright on a moving vs stationary bicycle. Might be simpler, but would use more energy because the fly wheel would need to be moving at a constant speed. It should not need a controller though. Thoughts?
If you add another reaction wheel or a pendulum at 90° to that one, you could control pitch too. Even better, two sliding weights inside the hull, at 90° to each other would act like pendulums and stabilise both axes whilst leaving the deck clear and the centre of gravity low.
I have a 16 foot fishing boat that can get sketchy when leaning over the gunnel for a fish. I wonder if a person could come up with something scaleable to a larger boat without being too obtrusive.
What impact may running balanced (so a pair, or pairs of) counter-rotating gyroscopes without any of the disturbance detection? It seems like the real carriers have some stabilisation tech on them, maybe a review of their stabilisation is a quick way to figure out if there are other Daniel-scale ideas to implement.
Reminds me of tuning enthalpy loops. Customers always asking why they cant get the exact enthalpy they want. You can hang steadily at a very small offset from setpoint, or have it see saw a bit above and below setpoint forever. They get brain cramp when explaining why.
would a simple PID loop trying to maintain "zero tilt"? So long as it has an integral factor it should try to correct over time. Though you might wind up with the wheel building up speed over time and going bananas
What happens when you turn via the rudder, phase lag? Why not lay the wheel flat inside the ship spinning at constant velocity to contain roll movement?
Just thinking out loud here. Would it help to have additional fans/engines on the side, front and end, to counteract this? Mounted at a 90 degree angle, so side to side, iso the normal front to back direction. Much more complicated structure, extra maintenance etc of course.... And on the topic of the helicopter. auto landing based on a mark on the ship?
Interesting. It reminded me of the gyro-wheel based roll damping devices installed on old aircraft carriers and small ships . The stabilizer with reaction wheels was also an interesting video, as it showed that it was effective enough. If it worked well with a large gyro, it could work well with multiple smaller installations.
Great work guys. OK I did some visual scaling and for a full size carrier it would be approx 85m above the deck - about the same height as the superstrucure. It would look pretty awesome out there in space meeting up with the Starship Enterprise. 🙂
That's excellent, thanks for showing that! Don't you just love Science and Engineering! You should be able to achieve the same effect with a smaller diameter flywheel hidden inside the hull, using a gearbox to get an equivalent referred inertia. You might also add some fore and aft stabiliser vanes underwater to help with the for and aft stabilisation which could also sort out the long term drift issue. I wonder how big the flywheel would need to be to achieve noticeable for and aft stabilisation?
This was supposed to be our thesis in minimizing heeling factor in racing catamarans, our prof bonked us at that topic. Glad to see someone put it to test. Edit: typo *faction* -> factor
Was there any windup noticed?? - forcing the algorithm to drive the velocity back to zero - in relatively steady state conditions avoids maximum speed limits. ( offset mass balance - SS offsets sorted..)
- high speed gimballed gyro is more suited to transient balance upsets (sure you know that) with the mass' rotational inertia allowing a small mass to affect a rather heavy vessel.
Adding an I coefficient to the PID control will eliminate the offset, as you suggested. Seems like it's time to add a reaction wheel in the longitudinal axis, too! Or maybe a reaction mass on a linear track within the hull?
I was thinking that too, but I actually don't think it would since he said he's using roll rates as the set point and feedback, not the angle of the ship. An I term would compensate for a steady state error in roll rate, but that's not really a thing that this system can have, since that would mean the ship was constantly rolling over.
The fly "wheel" doesn't need to be a wheel it could be a long rod with high mass at the point of buoyancy or a pendulum that hangs below the center of buoyancy. For pitch or even roll you could also use a tube with bird shot in it that tilts to counter act rotation on ither access by tilting the tube. The tube would need to be slightly curved up so that the shot would want to go to the center of it but should work just fine. It would be on a stepper motor with limiters so it would tilt the tube and because the motor is fixed solid to the hull it would force it to remain upright
I don't think it's possible to cancel out steady-state error with a reaction wheel alone, the wheel would gradually spin up to maximum speed trying to counter whatever natural list the ship has. Reaction wheels can't handle constant long-term forces in a single direction (like using one to take a ship that naturally lists and hold it upright), as they have a maximum amount of force they can apply in one direction by accelerating the wheel before the wheel hits maximum speed and can't accelerate in that direction anymore. The forces they're applying need to be balanced long-term, or you need a way of momentarily applying force to the vehicle in a specific direction so the wheel can unload the momentum it's accumulated. Something like a control moment gyroscope would be able to counter natural steady-state list, as they rely on pushing around a spinning gyroscope at right angles to it's spin, instead of accelerating it faster or slower.
You should have placed it on the deck facing the other way and just had it continuously Spinning at a higher RPM to create a Gyro stabilization effect I think that would work way better
After i watched this video this morning i can't stop thinking about it. I got to see if more is possible with this. Like installing 1 in the X axis and 1 in the Y axis in order to HYPER Stabilize a Ship. (Maybe start with a mid size ship. Something that rescue workers might use when battling Violent ocean waves?)
Try building the gryo stabiliser from the self balancing train and mount that! (The Brennan monorail) Should produce some really interesting results mounted on a boat🤔 not sure if that's ever been tried.
If the wheel were made of lead, it could be proportionally smaller for the same weight as the aluminum one. This would also lower the CG (possibly allowing below deck mounting?).
I have been watching a ton of your videos and you come up with all kinds of RC applications for things. Let’s say you wanted to drop a 1,000 gram payload out in the ocean with a retrieval line attacked to it what vehicle would you use (drone, boat, plane, helicopter)?
what if you have two smaller flywheels where you mount them horizontal (so axis is along Z-axis) and have them counterrotate, and just spin them up and use them as gyros to physically stabilize the ship? i only mention two because only having one might add yaw in the opposite direction of the spin
Why didn’t you use a gyroscopic stabiliser? Can maintain your angular momentum by increasing speed and decreasing weight, mount it lower in the boat, and have it spin about the vertical axis so it dampens pitch and roll.
Interesting. Have you ever tested kfm profiles on bottom of boat? Pontoons have step but what happens if you put three or more steps to bottom of a boat?
when I was aboard the Carl Vinson, a machinist mate buddy of mine was showing a new guy around, and when they passed once of the spinning prop shafts, he told him "and this is the ship's gyroscope. if you ever see this stopped, alert everyone immediately, cuz this top-heavy ship will flip right over without it"
dude, I do lettering at a boat yard in Maine and they have huge gyroscopes inside them. Private ships not government. Crazy tech. On of the gyros weighs in at 14000 lbs. and it's the smaller one. Someone said they thought the bigger one was like 40,000
I'd like to see you make the natural next evolution of this, with four smaller steel wheels mounted internally. One in the front, one in the back, and one on each side of the center of mass. Making them out of steel would mean they'd be much heavier so you could make them small enough to mount internally, but I'm sure there are cheaper metals you could use instead. Maybe brass? I used to do stuff like this in Besiege a lot when building my flying machines. The difference between no stabilization and four stabilization points was craaaazy.
"You could land any plane on that thing", as long as you can dodge the massive spinning wheel on the aft deck :D
You just gotta time your descent profile to come in between the spokes, it's something any military pilot picks up on naturally over time
@@verdiss7487 Yeah ... If they are any good, of course .....
It’s the kind of thing you only get wrong once
😂😂
Have you hear the word "INNOVATION"
As soon as I saw you set that flywheel on top of it. I just immediately pictured everything at scale and it gave me a really good laugh 🙂
Clearly a ferri wheel would be a good addition to the nimitz class.
I'm sure it won't impact take off and landing procedures at all.
@@CriticoolHit suuuuuuuuure
Same XD I'm imagining a wayward plane going through it and making the same noise a leaf makes when it gets sucked into an oscillating fan
This video was 4 minutes long because for once something worked exactly the way it was supposed to. Keep up the fine work good sir!
The newer aircraft carriers have wings that come out underwater to help stabilize the carrier as well as gyroscopic dampeners
They're the most stable ships in existence currently, if I know correctly
Likewise cruise ships.
Note this is only non-american carriers. The Nimitz class in the video is only stabilized by just being fucking giant and heavy. The issue is that the model is too light, plus waves aren't to scale. On the real deal the deck is about 60 feet above the water. That's right under the max size of any recorded rouge wave, so the likelyhood of a wave cresting over the deck if a Nimitz is near zero.
largish recreational boats have the stabilizers and I have seen a gyro based one on boats as small as 27' What this video is showing is basically like the Seakeeper.
Cruise ships 2
Internal solution:
Two reaction wheels, each one only spinning in one direction. Because they have to be smaller, you can have two and improve the response by them not having to stop and speed up the other way but only accelerate. They would have an idle speed instead of full stop and just speed up to create the stabilization effect in the direction required.
contra-rotating ones still get fully saturated and have to de-spin. and if they're not mounted next to each other, they'll exert a yaw torque along the length of the ship
@@recurvestickerdragon no, they dont have to "de-spin". You mount them inline, they have a certain idle rpm and there will be no yaw induced for that exact reason.
@@GaniscolOnce you've sped one wheel up to resist roll in one direction, it will have to be slowed down as you can't keep accelerating it indefinitely, and when you slow it down it generates a torque.
This brought back memories long lost. I once worked on the design of a reaction wheel for an FPSO (~450,000t tanker conversion to FPSO) almost 20 years ago, to control roll for improved seakeeping. The wheel was massive and to be installed inside the vessel.
what's fpso?
@@NoNameAtAll2 Floating Production Storage and Offloading unit. It's a huge vessel permanently moored on top of oil production wells in deep waters, which is capable of receiving and processing the oil and gas, separate one from the other, store the oil and then offload it to shuttle tankers which then transport the oil directly to the national or international customer.
"Will it capsize? No it won't because we have a reaction wheel B"
hahahah. Already thought I was the only one who heard that 😄
He's providing us that "itch" to create something 😎
@@vsci79 Nah. He gave the "B". We need to provide the "itch".🤓
CYKA
For your electronics, in wet environments, use solar panel bits. Like the roof pass-through boxes and bulkhead fittings. They're perfectly sized for small electronics and waterproof. $ is ok too, not too much.
Stability on surface ships does not normally rely on the centre of gravity being below he centre of buoyancy. The stabilizing effect comes from the centre of buoyancy moving faster outwards than the centre of gravity when the ship heels over.
What does this mean
@@BartJBols It means ships are only stable below a certain roll angle.
@@BartJBols That stability is mostly created by the hull shape, not by a big honking heavy keel dangling under it. With the consequence @nerdyengineer7943 mentions: such a ship won't right iself once capsized. Simplest example: a catamaran: very stable around the neutral, but once you tip it over, it will remain upside down.
That makes a lot of sense actually
You should set the reaction wheel to spin at a relatively high speed and then adjust that speed up or down to control for roll axis, while the high speed acts as a passive gyro which would stabilize the pitch axis.
Came here looking for this comment, well put!
That would mean it could get saturated pretty quickly though, if it built up too much speed the direction it was already turning
1:41 "This aircraft carrier is so stable, you could land any plane on that thing"
Meanwhile the Boat: *boing boing boing*
Where do I get that GPS hold Apache that he was using?
@@KandiKlover Asking the real questions here
@@KandiKlover true
Great addition,thats a definite improvment. I'm making a reaction wheel Odrive cubli just now so wonderful to see the relative sizes of your parts and the effect.
Maybe turn the reaction wheel 90 degrees, so the spin axis is vertical. Then spin it up and leave it spun up, for damping in the pitch and roll axes.
Yeah, that's what I don't get - why reaction wheel this? Logically an Anti-rolling gyro would work better.
KISS principle always
@@smalltime0 well, there's possible problems from that, I just haven't seen it tried. Let's say that the net roll torque (because waves aren't symmetrical and you spend more time on one side of the wave). Then you have a net roll torque on the vessel, which because of the gyro, causes a net pitch on the vessel. This pitch will be resisted by the buoyancy and also by the gyro, creating another roll, I think in the opposite direction. So I think this will set up a very definite oscillation that just looks like precession, except it will be a factor of 3 to 10x larger oscillation in roll. I mean, it is SIMPLE, yeah. but the stability is questionable. I was mainly wanting to be entertained.
The gyro wheels inside those RC motorcycles work great to stabilize a boat to.
Yes, they do. Used in RC race boats and scale models ships.
With a long ship like this, maybe a smoother option would be two or three smaller wheels, one towards front, one mid and one back. Could probably also better control pitch/twist correction that way.
I wonder also if you had them mounted closer to your center of gravity what kind of change it would do. And is the change better with a large flywheel were multiple small flywheels.
Smaller reaction wheels are (practically, not theoretically) less effective for the weight of the wheel. That's one downside.
@@roboman2444You'd need more speed instead, but you could mount the system under deck.
and they'll exert a torque that makes the ship yaw more easily in one direction than the other
Add another flywheel on the perpendicular axis. So it'll be locked more😂 It's a cool project thanks for sharing bro!
ODrive makes some awesome stuff, been following them for several years now, I have a ton of ideas, but have never gotten to actually using them in a project yet.
When I saw how stable it was, I thought it would be cool to try and land something on it.
Then, it looked like you want to land a heli on it. I'm now subbed.
Doesn't matter if you do or dont land something on it man,... I just like the way you think.
Use a fast spinning flywheel parallel with the boat, and then rock it back and forth to counteract roll. Seakeeper makes these for fishing boats and yachts
On big boats they mount a flywheel at 90 degrees but have it free to rotate on both axis.
So when the boat rolls the gyro tries to yaw and similar for pitch.
I think this results in dampening both roll and pitch
Then by using actuators to move the flywheel in yaw they can create a roll reaction.
I think the advantage of this is
Great video. The wheel is a little impractical for large vessels. You could try using a push pull weight on a slide rod, attached to either a cable or belt inside the vessel. You can use one for pitch as well as roll. It would mean two sets of electronics, but it would stabilize the ship so you can land the Apache! lol
What about mounting a spinning disc horizontally, to give passive roll and pitch stabilisation? Or maybe two smaller ones, rotating in opposite directions.
roughly mathematically you can make your gyroscope half the size if you can either quadruple the speed with a half size wheel the same weight as yours here. or by quadrupling the weight of the wheel and keeping the speed the same (half size wheel as well). maintaining the angular momentum in different form factors.
The breadth and depth of your Engineering skills and knowledge is very impressive, and more awesome with each video.
I think several smaller wheels could go on the inside, closer to the axis of rotation could be more effective. They would also help preserve the look of the carrier.
and to compensate for the smaller wheels' lower mass, they could be cylinders instead of flat.
For the section at 2:10, it would've been cool to see a 2x2 of the footage we see now on the left, and the same two bits but stabilized on the horizon instead on the right
A new video from your channel is like my hit of crack now
Fren!
Marine engineer student here. At 2:13 you say it has some natural stability because the center of gravity is below the center of buoyancy, but that actually not necesarry to be stable. What´s important is the "metacenter height", which is a result of how the center of buoyancy shifts to the side when the ship leans over. Nearly no ships has a center of gravity below the center of buoyancy, both because it´s geometrically tricky to get such a low center of gravity, but also because the ship will get very twitchy if the distance between the metacenter height and the center of gravity is too long.
That reaction wheel sitting so far back on the deck is actually kicking the bow left and right.
You're so smart, Daniel! This very short video is very fun to watch!
This makes me wonder if you would have a better effect with a pair of counter rotating gyro wheels? The counter rotation would cancel out any yawing effect caused by them that could slowly translate in the water while still allowing the boat to turn freely. They could also be smaller, lighter and more easily mounted (one in bow, one in stern) yet still provide massive amounts of roll and pitch stability.
Just a single would be fine
@@nixer65 A single for the same effect would need to be larger and be more difficult to mount, plus it would create a rotational force (think of a helicopter) that would create a slow spin in the water thile the boat is stationary.
Looks impressive.
Why do you place the wheel so high above the mass center and length axis?
Bottom side of the ring working opposite direction you want to, but if the wheel will completely fit under deck -- both sides (top and bottom) will stabilize the ship, do not it?
We had a seakeeper gyro in one of our boats, and it uses a fast spinning weight spinning horizontally (like the old fashion spinning top) and then hydraulic piston to offset the roll to rotate the spinning top fore and aft. That way you still have all the torque available at anytime and it doesn't start and stop the flywheel. Although it will take almost an hour to spin up and reach full speed. Great job... 🤠👍
1:55 could you lock roll to horizon like a quad copter? Horizon mode. Or maybe increase I term.
I'm sure a gyro would be much much better than a reaction wheel, that would be interesting too! Just spin it up, and use a servo to yaw it left and right, which will result in a roll torque.
have you thought about two giros spinning in opposite directions and then applying independent brakes to control the roll? It would be smoother
You could spin it up to some higher baseline speed, adjust around that point for roll damping. The spinning inertia will then help stabilize pitching motions (although precession will impart a yaw moment onto the ship). Would be interesting to see how that works out.
I wonder if you could put some little bow thruster jets under the water line to help do the same thing but keep the runway open :)
This is so cool though. Looks like a fun hobby.
You mentioned the roll axis bias. As you already know, that is from the bias in the roll axis angulare rate sensor. I am assuming you are not using a expensive rate sensor and they typically have a bias. To fix this issue, run the angular rate sensor value through a high pass filter. This will eliminate the bias in the sensor. I would expect 2-4 second filter would work sufficiently. I have not mess with the microcontroller you are using but I am fairly sure, you can add the filter. I would start with a 2 second filter, 2s/(2s+1). If you do not know how to implement this, I can create the code to use as long I I know the frequency of the microcontroller. Nice thinking on how to stabilize the platform.
This was a fun little project. Definitely cool to see the impact that wheel had
Slimming down the wheel and mount it internal could help.
Closer to center of gravity and lowering the center of gravity.
You might not even loose reaction wheel force by slimming down the wheel if the motor is fast enough to compensate for the smaller wheel.
Was it difficult to keep it straight when going forward? The pitch up and down of the bow should have caused gyroscopic procession at 90 degrees on the gyro meaning it should try to turn you left/right each time a wave hit. just me pondering about it.. great idea for an experiment by the way! A company called Seakeeper actually make a vey expensive similar system for large yachts. thanks for the video!
What about using a gyroscope? Continuously running creating a stable plane and could be placed under the landing runaway
Now make reaction wheels in both axis but using the same weight in tungsten, you might be able to fit the rig under the deck of the ship
Maybe adding some "ailerons" would help too. Surely they would be less weight than the gyro and extra ballast. Maybe the lighter weight would make up for the extra drag?
How much of a difference does it make that the fly wheel is offset from the rotational axis of the boat? Would it be more efficient for the flywheel to be centered on the ships rotational axis or does it get some mechanical advantage like a lever arm?
It needs the offset offset by a level sensor. A pendulum won't work for the same reason quads have problems with hanging loads. Centrifugal force is not going to be always down.
Neat, a diy rc seakeeper
No good having an aircraft-carrier unless you land on it !!! ❤❤ go for it 😊
I am curious, is you method better than mounting the wheel 90 degrees around a Z axis? This is kind of what makes it easier to stay upright on a moving vs stationary bicycle. Might be simpler, but would use more energy because the fly wheel would need to be moving at a constant speed. It should not need a controller though. Thoughts?
Quick sponsor vid, nice. I respect the grind
I built the interfaces for Seakeepers for years. The amount they mitigated the rocking was pretty impressive.
How hard would it be to mount the reaction wheel under the ship closer to centerline? Obviously with more waterproofing.
If you add another reaction wheel or a pendulum at 90° to that one, you could control pitch too. Even better, two sliding weights inside the hull, at 90° to each other would act like pendulums and stabilise both axes whilst leaving the deck clear and the centre of gravity low.
Was watching part one few hours ago, this is insane that u uploaded part two today.😍
I have a 16 foot fishing boat that can get sketchy when leaning over the gunnel for a fish. I wonder if a person could come up with something scaleable to a larger boat without being too obtrusive.
What impact may running balanced (so a pair, or pairs of) counter-rotating gyroscopes without any of the disturbance detection? It seems like the real carriers have some stabilisation tech on them, maybe a review of their stabilisation is a quick way to figure out if there are other Daniel-scale ideas to implement.
Reminds me of tuning enthalpy loops. Customers always asking why they cant get the exact enthalpy they want. You can hang steadily at a very small offset from setpoint, or have it see saw a bit above and below setpoint forever. They get brain cramp when explaining why.
would a simple PID loop trying to maintain "zero tilt"? So long as it has an integral factor it should try to correct over time. Though you might wind up with the wheel building up speed over time and going bananas
What happens when you turn via the rudder, phase lag?
Why not lay the wheel flat inside the ship spinning at constant velocity to contain roll movement?
Check out control momentum gyroscopes. CMGs. You can build them smaller and get higher torques.
Just thinking out loud here. Would it help to have additional fans/engines on the side, front and end, to counteract this? Mounted at a 90 degree angle, so side to side, iso the normal front to back direction. Much more complicated structure, extra maintenance etc of course.... And on the topic of the helicopter. auto landing based on a mark on the ship?
Would a simple accelerometer (or an IMU) help to maintain a level deck rather than just a stable deck?
Interesting.
It reminded me of the gyro-wheel based roll damping devices installed on old aircraft carriers and small ships .
The stabilizer with reaction wheels was also an interesting video, as it showed that it was effective enough.
If it worked well with a large gyro, it could work well with multiple smaller installations.
Very cool short video, also love longer format ones but this from time to time is awesome!!🎉
This would be a great platform to play with the ardupilot moving platform landing feature.
Great work guys.
OK I did some visual scaling and for a full size carrier it would be approx 85m above the deck - about the same height as the superstrucure.
It would look pretty awesome out there in space meeting up with the Starship Enterprise. 🙂
That's excellent, thanks for showing that! Don't you just love Science and Engineering! You should be able to achieve the same effect with a smaller diameter flywheel hidden inside the hull, using a gearbox to get an equivalent referred inertia. You might also add some fore and aft stabiliser vanes underwater to help with the for and aft stabilisation which could also sort out the long term drift issue.
I wonder how big the flywheel would need to be to achieve noticeable for and aft stabilisation?
This was supposed to be our thesis in minimizing heeling factor in racing catamarans, our prof bonked us at that topic. Glad to see someone put it to test.
Edit: typo *faction* -> factor
Wouldn't that end up having to be impractically huge
@@taliesinriver...and that's why we fumbled that thesis.
@@harveylorenzedejesus2383 Ah I see haha
Was there any windup noticed?? - forcing the algorithm to drive the velocity back to zero - in relatively steady state conditions avoids maximum speed limits. ( offset mass balance - SS offsets sorted..)
- high speed gimballed gyro is more suited to transient balance upsets (sure you know that) with the mass' rotational inertia allowing a small mass to affect a rather heavy vessel.
Adding an I coefficient to the PID control will eliminate the offset, as you suggested. Seems like it's time to add a reaction wheel in the longitudinal axis, too! Or maybe a reaction mass on a linear track within the hull?
I was thinking that too, but I actually don't think it would since he said he's using roll rates as the set point and feedback, not the angle of the ship. An I term would compensate for a steady state error in roll rate, but that's not really a thing that this system can have, since that would mean the ship was constantly rolling over.
The fly "wheel" doesn't need to be a wheel it could be a long rod with high mass at the point of buoyancy or a pendulum that hangs below the center of buoyancy.
For pitch or even roll you could also use a tube with bird shot in it that tilts to counter act rotation on ither access by tilting the tube. The tube would need to be slightly curved up so that the shot would want to go to the center of it but should work just fine. It would be on a stepper motor with limiters so it would tilt the tube and because the motor is fixed solid to the hull it would force it to remain upright
I don't think it's possible to cancel out steady-state error with a reaction wheel alone, the wheel would gradually spin up to maximum speed trying to counter whatever natural list the ship has.
Reaction wheels can't handle constant long-term forces in a single direction (like using one to take a ship that naturally lists and hold it upright), as they have a maximum amount of force they can apply in one direction by accelerating the wheel before the wheel hits maximum speed and can't accelerate in that direction anymore. The forces they're applying need to be balanced long-term, or you need a way of momentarily applying force to the vehicle in a specific direction so the wheel can unload the momentum it's accumulated.
Something like a control moment gyroscope would be able to counter natural steady-state list, as they rely on pushing around a spinning gyroscope at right angles to it's spin, instead of accelerating it faster or slower.
You should have placed it on the deck facing the other way and just had it continuously Spinning at a higher RPM to create a Gyro stabilization effect I think that would work way better
It looked like a propeller in the thumbnail but when you revealed the stabilizer I literally LOLd. Very cool!
After i watched this video this morning i can't stop thinking about it. I got to see if more is possible with this. Like installing 1 in the X axis and 1 in the Y axis in order to HYPER Stabilize a Ship. (Maybe start with a mid size ship. Something that rescue workers might use when battling Violent ocean waves?)
would it remain stable if you are a second one that is rotated 90 deg, or would they create a conflict?
Try building the gryo stabiliser from the self balancing train and mount that!
(The Brennan monorail)
Should produce some really interesting results mounted on a boat🤔 not sure if that's ever been tried.
Does the wheel need to be that big, to dampen the roll, or is that weight based?
how well does this scale up?
one of those standing kayaks seems like a perfect application for a reaction wheel, for those of us a bit less stable lol
could that reaction wheel be shaped like rollingpin, so it would be in side hull? it would bring center of gravilty lower.
If the wheel were made of lead, it could be proportionally smaller for the same weight as the aluminum one. This would also lower the CG (possibly allowing below deck mounting?).
Is there anyway you could put multiple small gyros under the deck?
Why not mount it parallel to the water line? It will resist roll and pitch when you spin it at constant velocity.
I have been watching a ton of your videos and you come up with all kinds of RC applications for things. Let’s say you wanted to drop a 1,000 gram payload out in the ocean with a retrieval line attacked to it what vehicle would you use (drone, boat, plane, helicopter)?
what if you have two smaller flywheels where you mount them horizontal (so axis is along Z-axis) and have them counterrotate, and just spin them up and use them as gyros to physically stabilize the ship? i only mention two because only having one might add yaw in the opposite direction of the spin
Why didn’t you use a gyroscopic stabiliser? Can maintain your angular momentum by increasing speed and decreasing weight, mount it lower in the boat, and have it spin about the vertical axis so it dampens pitch and roll.
Interesting.
Have you ever tested kfm profiles on bottom of boat? Pontoons have step but what happens if you put three or more steps to bottom of a boat?
Have you thought about trying a losi pro Moto gyro they sit right at a pound I used one in my recoil 2 26 inch
would love to see test with Gyroscopic effect. put it flat with max spinning speed. :D
Could the accumulated error could be canceled out by an I gain in your PID?
im actually currently building a miniature us csg so this video was pretty helpful
Try a high speed gyroscope installed so that the rotation is in line with the hull and the top of the wheel is moving aft.
Would it be useful on the hydrofoil paddle board? 👀
when I was aboard the Carl Vinson, a machinist mate buddy of mine was showing a new guy around, and when they passed once of the spinning prop shafts, he told him "and this is the ship's gyroscope. if you ever see this stopped, alert everyone immediately, cuz this top-heavy ship will flip right over without it"
I made the same rc aircraft carrier out of a bath toy when I was a kid, this video is making me nostalgic
dude, I do lettering at a boat yard in Maine and they have huge gyroscopes inside them. Private ships not government. Crazy tech. On of the gyros weighs in at 14000 lbs. and it's the smaller one. Someone said they thought the bigger one was like 40,000
I'd like to see you make the natural next evolution of this, with four smaller steel wheels mounted internally. One in the front, one in the back, and one on each side of the center of mass. Making them out of steel would mean they'd be much heavier so you could make them small enough to mount internally, but I'm sure there are cheaper metals you could use instead. Maybe brass? I used to do stuff like this in Besiege a lot when building my flying machines. The difference between no stabilization and four stabilization points was craaaazy.
What kind of programming did you have to do to get the Odrive to control roll? Is its own sensor providing the roll data?
do you use a PID-controller to control the reaction wheel?
Ive done something similar when making boats in the game Stormworks, but i used big reaction masses on sliders within the hull
Would love to see this imbedded in the hull. Maybe a keel with side-to-side actuators to counteract for roll and even pitch dampening.