
“Fluid gears” invention shows promise for improving mechanical devices.
A group of researchers at New York University has developed a new type of gear system that uses fluid motion to produce rotation. This approach could pave the way for mechanical devices that are more adaptable and resilient than traditional gears, which trace their origins back to ancient China.
The findings are detailed in the journal Physical Review Letters.
“We invented new types of gears that engage by spinning up fluid rather than interlocking teeth—and we discovered new capabilities for controlling the rotation speed and even direction,” says Jun Zhang, a professor of mathematics and physics at NYU and NYU Shanghai and the senior author of the paper.
Rethinking gears beyond solid teeth
Gears have been a fundamental part of machinery for thousands of years, with early examples dating to around 3,000 BCE in China, where they were used in two-wheeled chariots that traveled across the Gobi Desert. Over centuries, they have played roles in devices such as the Antikythera mechanism in ancient Greece, which predicted astronomical events, as well as in windmills, clocks, and modern robotics.

Despite their long history, traditional gears come with limitations. Their teeth, whether made of wood, metal, or plastic, are rigid and prone to damage, and they must align precisely to function correctly.
Motivated by these constraints, Zhang worked with colleagues Leif Ristroph, an associate professor of mathematics at NYU’s Courant Institute School of Mathematics, Computing, and Data Science, and Jesse Etan Smith, an NYU doctoral candidate, to explore whether gear-like behavior could be achieved without physical teeth or direct contact between components.
Because moving air and water already drive systems such as turbines, the researchers proposed that carefully controlled fluid flows could effectively take on the role of gear teeth.
Fluid flows replace mechanical contact
To test this concept, the team carried out detailed experiments using cylindrical rotors submerged in a liquid mixture of glycerol and water. By adjusting properties such as viscosity and density, they were able to control how the fluid behaved.
A gear mechanism created by NYU researchers relies on water to generate movement. Under some conditions, the rotors spin in the same direction like pulleys looped together with a belt (first video below); in others, the rotors spin in opposite directions like a pair of gears (second video below). Credit: Videos courtesy of NYU’s Applied Mathematics Laboratory
In each setup, one cylinder was powered to rotate while the second remained passive. The researchers predicted that motion from the active cylinder would generate fluid currents capable of turning the passive one. To visualize these flows, they introduced tiny bubbles into the liquid, allowing them to observe how the fluid transferred motion. Experiments were conducted at different distances between the cylinders and at varying rotation speeds.
The results showed that the interaction between rotating cylinders and the surrounding fluid could mimic different mechanical systems. When the cylinders were positioned close together, the fluid acted like the interlocking teeth of traditional gears, causing the passive rotor to spin in the opposite direction. When the cylinders were farther apart and the active rotor moved faster, the fluid wrapped around the passive cylinder in a way that resembled a belt and pulley system, causing both to rotate in the same direction.
New control and design possibilities
The researchers highlight several potential advantages of this fluid-based approach over conventional gear systems.
“Regular gears have to be carefully designed so their teeth mesh just right, and any defect, incorrect spacing, or bit of grit causes them to jam,” explains Ristroph. “Fluid gears are free of all these problems, and the speed and even direction can be changed in ways not possible with mechanical gears.”
Reference: “Hydrodynamic Spin-Coupling of Rotors” by Jesse Etan Smith, Leif Ristroph and Jun Zhang, 13 January 2026, Physical Review Letters.
DOI: 10.1103/m6ft-ll2c
The research was supported by a grant from the National Science Foundation (DMS-2407787).
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
11 Comments
Is this article an April Fools joke?
I would like to see the patents on this
What about the energy loss transferred to the fluid? How can you overcome this?
I wondered the same thing, but I’m also asking why this appears to be some great news since anyone who has ever been in a swimming pool knows how to create whirlpools that make other objects in different areas spin in different directions?This isn’t new at all, it’s very old tech that’s not practical to use or apply BECAUSE of one inevitable energy loss. It’s like trying to plug in a drill to an inverter that’s being powered by a battery that’s being charged by an alternator that is spun by the drill. The laws of diminishing returns kicks in. This is what our tax money is creating grants to fund?
This is moronic! Traditional gears work cause they’re not made of fluid xD what moron thinks he’s gonna power anything more than a party trick with fluid gears. I spit my drink out when I read the title! What a joke! Would be some new york yuppie that comes up with something so stupid. This article is good for nothing more than a laugh!
You mean re-invented? This was invented by Tesla. Tesla turbine.
So you invented an inefficient torque converter. Just like is in your car.
Isn’t this the same principle as the automatic transmission in your car that uses transmission fluid as the medium?
You took the words right out of my mouth. Sounds like the exact same principle.
Please explain how this research is in any way new and different from GM and Ford patents for hydro-flow transmissions. The hydro flow uses fluid dynamics as a contactless clutch for two, three and four speed transmissions, plus reverse. Fluid flow dynamics has been used for contactless gears for 100 years.
Not “gears”, maybe a fluid coupling but not a very efficient or useful one and certainly not an “invention”. Have you people ever heard of a torque converter?