Fist-sized 11M-to-1 gearbox | Hacker News Reader

Fist-sized 11M-to-1 gearbox

makezine.com

157 pointsp4bl066 comments

The Space Needle's spinning top level is powered by a 1.5hp motor[1] at a gear ratio of 360,000:1 [2]. It makes one complete rotation every 47 minutes.

[1] https://en.wikipedia.org/wiki/Space_Needle

[2] https://books.google.co.uk/books?id=Jox4CAAAQBAJ&pg=PT262&lp...

I stayed at the Westin Bonaventure in LA recently. An employee told us rotating bar at the top takes something like two hours to make a complete rotation. It was disorienting to have everything around you be constantly changing, yet so subtly.

Using the contraption in the article hooked up to a motor at 600rpm (typical for a diesel engine), it'd take about 12 and a half days to rotate something.

The point of this gearbox is the small size. The space needle gets to use a gear that is the size of the top level, and drive it with a gear that is a few inches in diameter. That has always been easy to do and is the sort of thing this compact gearbox avoids.

[edited for mistake: motor -> gearbox]

I seriously doubt they use a gear the size of the base. I doubt the facilities to manufacturer a gear that big exist anywhere. Gear machining is a very specialized operation.+

More likely it's a train of gears of much more moderate ratio...a series of 5:1 or 10:1 reductions, say.

At that size and speed, I don't think you need the gear to be particularly precisely machined.

You do if you're going to be meshing a much smaller gear to it. For the two to fit together, the teeth on the larger gear need to be the same size as the teeth on the smaller one.

Based on this photo [1], I'm guessing they use a system much like a cog railway. There would be a continuous ring of teeth along the tracks and a small motor on the platform with a few inch gear to engage the cog. Or vice versa. The cog ring would probably be made up of several smaller segments.

http://www.heraldnet.com/article/20111118/NEWS01/711189872 makes it look like they do.

As kevhito says (but (s)he forgot to include the URL), it is more like a train running on a cog railway, with the train fixed and the rail moving.

I can't find any diagrams explaining the mechanics of the restaurant (or even exactly what rotates); which upsets me. -Can someone provide more information/pictures? Even with amazing balance I don't understand how there isn't massive friction caused by whatever it's running on...

Nearly everything says it's a single 1.5hp engine (upgraded from 1hp), however there's a case study by Graybar who did a retrofit in 2013 [0] which makes it very clear that there is (and was) actually _two_ motors, both in play. -once again, if anyone has more information? I'm massively intrigued.

"The major problem was that the two existing drives were fighting to share the load – one was lagging and the other leading. The two gear boxes were constantly adjusting according to inputs from the motors, which caused both the noise and roughness during the rotation. The team determined that the best solution was to upsize the motors, drop motor revolutions per minute (RPM) and reduce the reduction ratio with the new gearboxes. So they recommended Schneider Electric drives as well as various automation and control products to the Space Needle engineering staff. The new motors were equipped with encoders feeding the drives in a master/slave configuration. Along with this, all new controls were installed, including a wireless control system. Schneider Electric also provided engineering and software support, both at the integrator and jobsite facilities."

>>> edit: I'm wondering what wattage this/these 1.5hp motors are; the Falkirk Wheel is a rotating canal lock up in scotland. It manages to simultaneously lift/lower two canal boats + water 35 meters for only 1.5kWh of energy (and has a pleasing gear system) [1] [2]

[0] http://www.graybar.com/documents/graybar-keeps-tables-turnin...

[1] https://en.wikipedia.org/wiki/Falkirk_Wheel

[2] https://www.youtube.com/watch?v=n61KUGDWz2A

The restaurant floor, a ring around the core, rotates. The observation deck, exterior and center do not.

I highly recommend going through Oskar's other puzzles and contraptions which includes tons of very very strange cubes, puzzle rings and other oddities:

https://www.youtube.com/user/OskarPuzzle

http://oskarvandeventer.nl/hanayama.html

He also has done a ton of collaborations with Bram Cohen of Bittorent fame like this one:

https://www.youtube.com/watch?v=RTcGCqz4ajs

He's definitely one of those guys who just can't stop creating puzzles and new ways of thinking about problems.

The application is obvious: DIY "Machine with Concrete" (http://makezine.com/2012/04/25/arthur-gansons-machine-with-c...) kits. I'm sure Arthur Ganson won't mind.

Really neat, thank you!

This is not too different from a harmonic drive. In fact it's basically a single-planet epicyclic drive inside a harmonic drive. Kind of nice but I don't know if it counts as a new type of gearbox.

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

Well, harmonic drives are cycloid drives that use elasticity rather than eccentricity to do the same task, so let's focus on cycloid drives.

It's not quite the same your typical cycloid drive, but it has a strong relationship. Cycloid gears are about having an inner circle of radius A rotating within an outer circle of radius B by putting the center of A's circle on its own circle of radius C = B - A.

If C is small, then a fast rotation of C produces a slow off-center rotation of B relative to A. Since it's off-center, usually a bunch of holes (theoretically also of radius C, I think?) are drilled in B and filled with "output rollers" which connect to one single shaft which is on-center relative to C. Thus C connects on-center to this asymmetric B, which connects to an on-center output shaft that otherwise rotates as B does.

In his case, these holes don't exist and the rotation is never on-center. But you can use the same mechanism to convert his amateur cycloidal drive to one that has an on-center rotation of the input shaft.

Now normally in a cycloidal drive, A is regarded as fixed, while C is rotating, and B is still. In this case his C is totally implicit, his B is rotating, and he is at two different A's which rotate relative to each other. There are therefore presumably two different Cs, too.

It's hard to really isolate the exact mechanism in terms of connected cycloid drives. It is not as simple as fixing C and watching a single cycloidal drive's A rotate as a response to a rotating B, because that would have the reverse mechanical advantage. It's also not obviously "two cycloidal drives connected together by C", which would have a mechanical advantage near 1.

So I'm not sure what it is, but I don't think you can just say "oh, it's a harmonic drive."

This "extract the difference of speeds between two parts" method is quite common to get extremely high reduction in a small amount of space; this is a ~70:1 single-stage gearbox found in a rearview mirror positioner, and it only has 6 parts:

https://en.wikipedia.org/wiki/File:Rearview_Mirror_Epicyclic...

Put 4 of those in series and you'll get 24M:1.

> We live in a world where an idea that does something that has never been done before can be built at home and shared without cost [...], all without it having any use or any place in the market. Outside of art, that is something we did not have a few years ago.

I do not agree with this idea. Many inventions have been driven by curiosity. People playing with magnets or 8 bits computers were not trying to solve any market problem. Even physics were developed by simple people sharing curious "unuseful" experiments for centuries.

If anything, individuals seem to have stopped inventing/investigating. It used to be that persons of means would use their means in the pursuit of some expansion of their knowledge. Now it's in pursuit of more means.

I don't know about this - one of the things I like to do when I travel is visit the local hackerspace. I've been to quite a few of them around the country, and they all seem to be full of wonderful people exploring and inventing for the fun of it.

This is a common historical fallacy. Like "It used to be that Greeks would sit around inventing geometry, philosophy, and democracy. Now they just ruin banks."

There were always both kinds of people, and the inventors who are driven by the love of science and learning have always been a minority. It's just that the people who only care about personal wealth rarely make it into history books, whereas the iconoclastic scientists frequently do. And the same is true today.

"It's just that the people who only care about personal wealth rarely make it into history books, whereas the iconoclastic scientists frequently do."

Arguably even this isn't true - Many people know the name of, say, Archimedes and Newton, but I doubt it's more than the people who know Julius Caesar and Genghis Khan. Might be close though.

What I'm trying to say is, what percentage of people can name a couple of great thinkers from the classical era? Now compare that to the percentage of people who can name a wealthy merchant or banker from the same period.

Every generation thinks it invented sex, rebellion, and curiosity.

That's because these things seem to have an inverse relationship with age.

(Not a flame! I feel like statistically speaking, probably true for the majority. Thankfully, there's a minority that shines even brighter with age)

You aren't wrong about the sentiment, but parsimoniously the cause is that youth is callow

The article mentions slippage, grinding, play, noise and vibration. These are generally undesirable in machinery. Slippage and grinding suggest significant problems with wear and durability. That would limit the lifespan of the gearbox.

It's still pretty cool and could be very useful. That doesn't mean there aren't useful applications. It just means there are design tradeoffs.

It looks like a pretty rudimentary prototype. Many of those issues are addressed with better manufacturing tolerances, different materials, and tooth profiles. I'm not sure those features are inherent limitations as much as drawbacks to the prototype implementation.

> Headed in a different direction, Van Deventer has a new type of gear he calls “grinder gears” that comprise an inner gear with one fewer tooth than the outer gear, causing it to wobble its way through the rotation. For the engineers, it’s a variation of a cycloidal drive.

Oh, he reinvented a Gerotor.

> A gerotor unit consists of an inner and outer rotor. The inner rotor has N teeth, and the outer rotor has N+1 teeth. The inner rotor is located off-center and both rotors rotate. https://en.m.wikipedia.org/wiki/Gerotor

It's not completely something new https://www.youtube.com/watch?v=MBWkibie_5I

The article points out issues with this exact implementation at an 11M:1 gear ratio but begs the question whether a less extreme ratio using "grinder" gears has any application or not.

Regardless, the conclusion about new innovations at the end of the article was exciting/encouraging to read.

I am most concerned about the efficiency of a design like this. People usually look for a gear reduction to increase torque, but an inefficient gearbox can make the output both slower and lower torque which is rarely desirable.

The wobbling input shaft would also be tough to work around.

Perhaps not with this ratio, but something like this with a quartz movement, with the right ratios should be able to advance a "day of year hand" around the face of a clock.

If you wanted to get really fancy, advance a DoY-hand for Venus, Mars, Jupiter, as well as the Earth's.

I want to see the inverse.

Eleven million spins with one rotation of the shaft.

Couldn't you just rotate the "output" shaft, to accomplish exactly this?

Sure, the required torque would be immense, but... ;)

Friction will prevent that, similar to the self-locking behavior of a worm gear. Even if you could get it started, it would take as much energy to spin the output shaft once as it takes to spin the input shaft 11 million times. If you tried to quickly spin the output shaft, the forces exerted on the gears would shred the materials almost instantly.

It would be fun to calculate the power input required for that. I suspect it would be in the range of megawatts or so - instantly vaporizing the gearbox (assuming you could somehow push that power into it before you destroy it mechanically).

Maybe make the gearbox out of mirrors and the "shaft" out of light. Suspend the whole thing in a magnetic field.

The mechanism is reminiscent of the one in the Newbould Indexer: http://www.imperialnewbould.com/the-newbould-indexer.html

This discussion has a link to RJ Newbould describing how he invented the indexer: http://www.chiefdelphi.com/forums/archive/index.php/t-68265.... It's a really interesting explanation for us mechanism geeks :-)

    Gune: [holding up a small device] Does this look familiar? Do you know what it is? 
          Neither do I. I made it last night in my sleep. 
          Apparently I used Gindrogac. Highly unstable.
    Preed: Gune...
    Gune: I put at button on it. Yes. I wish to press it,
          but I'm not sure what will happen if I do.

http://www.imdb.com/title/tt0120913/

Seriously, this is a cool device. And I have no idea what I could use it for...

I wonder what's the highest torque it can sustain without breaking. That will in turn severely limit applications because with that demultiplocation little torque will be huge at the other end.

Reminds me of a fun trick that was popular when I was in undergraduate mechanical engineering: A gear train that has one end spinning, powered by a small DC motor, and one end bolted/welded/glued/otherwise fixed solidly to a wall.

Increase the ratio enough and you can have an output shaft that will take years to go through even a fraction of a revolution ;)

There is one by Arthur Ganson in the MIT museum[1]. It's part of a great exhibit of fun/quirky machines and there's an interesting video about it on youtube[2].

[1] https://www.youtube.com/watch?v=5q-BH-tvxEg [2] https://www.youtube.com/watch?v=2Rov_Lriqvk

Will this also work as a 1 to 11M gearbox? If so it might allow a strong force (e.g. the weight of a large body of water) to spin a small electric generator. That will allow small scale hydroelectric storage for renewable sources.

If not it can probably still find some applications for turning something like weak wind power into something that can create great pressure.

I know nothing about this, but I would imagine that this gearbox would need to be made out of some other-worldly material in order to cope with the force necessary to drive it in reverse

I'm pretty sure it's already potentially self destructive if there is any snag in the output.

nope, cycloidal drives can't be backdriven because of the eccentric part, the friction is higher than the driving force for geometry reasons.

No, it cannot. As you can see in this video from the site where he explains the mechanism:

https://www.youtube.com/embed/vbU1afvCRwc?version=3&rel=1&fs...

When he tries to move the red ring against the green ring (at 2:39), the thing doesn't budge. The gearbox concatenates two such transmissions, therefore, it cannot be operated backwards.

Looks like a bunch of harmonic drive gears in a row. Every industrial robot contains a couple of these.

What you are describing is called "backdriving" the gearbox. Generally speaking, gear trains with high reduction ratios don't like to be backdriven. Small steel gear trains with ratios in the 1000:1 range can destroy themselves when back driven.

Could it be used for high precision movements? For example a stepper motor with very small and large movements? I know it's mentioned in the article.. but if it could be done then one could maybe create a CNC for DIY CMOS or something.

Here's a planetary drive using roughly the same concept. This is 64:1 per stage.

https://www.youtube.com/watch?v=P-Obt-9tZVo

Just a 3D printed gearbox. For me this is rather a promotional article than something really interesting. We've been able to print these kind of gears for years with reprap or reprap-like 3d printers.

If it had enough torque (which seems to be a problem given the plastic nature) I could see it breaking loose rusted or stubborn bolts. Maybe? In a scenario like that you wouldn't really care about precision or sound. Though 11 million turns still seems like a lot of cranking... maybe if it had a 9v attachment it could be a portable bolt breaker. Feels like a stretch though, would be fun to play with.

Though now that I think about it the holder of the bolt breaker would probably start slowly turning before the bolt does... weeeee

The problem with rusted or stubborn bolts I've encountered is the mechanical friction on the threads usually greatly exceeds the shear force (??) that the bolt can tolerate. I've snapped a head off more than once. That's a real bummer.

I feel like it could be used to construct mechanical calendars.

Precision could be an issue according to the faq.

I know the answer..think Tesla..but instead of a small auto a big truck or better yet a big ship...

both require a rethinking of the transmission to transfer power from the engine to the wheels or propeller in order to get bigger torque

this gear device solves that particular problem

Because of the non-back-drivability, I would use this for making winches.

Its sometimes nice to take a break from software/digital engineering and build stuff in the real physical world. Feels like a vacation for my brain almost.

Would be good for a telescope mount, maybe.

Why wouldn't one just use three nested cycloidal gears to achieve the same result?

Why they include an animated gif of someone turning another gearbox with a far lower ratio puzzles me.

But I'm going to go with "yes, this is useless, unless remade with much better materials". The current version is imprecise, high-friction, snd can't even survive one rotation of the output.

Huh. Was someone under the impression that the GIF was of the gearbox in question, or did they not RTFA enough to realize I was simply agreeing with the inventor for the same reasons he enumerates?

That is the gearbox. The output shaft is the red one.

You're right, sorry. I hadn't watched the second video.

Reply on news.ycombinator.com