Coflow Jet - a new kind of sail?

I recently read about this new form of cylindrical "sail", a modern redesign of Flettner rotors.
Cylinder sails promise up to 90% fuel consumption cut for cargo ships https://newatlas.com/transport/cylinder-sails-promise-90-reduction-cargo-ship-fuel-consumption/

and here is a bit more:
CoFlow Jet – IWSA Full Member | International Windship Association https://www.wind-ship.org/en/coflow-jet-iwsa-full-member/

I am intrigued with the thought of putting one of these on a catboat. But I have no idea whether the technology scales down to support that. I have not been able to find any discussion of utilizing this technology on recreational or cruising sailboats.

Is anyone else following this, or interested in it?

 

To answer the question - it isn't new, and it isn't a sail, either. It works better at smaller sizes than at bigger sizes.

 

Why do you say it's not new? This isn't the Flettner rotors.

Do you know any more about this technology? I have not been able to find out much.

 

The concept of boundary layer suction lift has been around since the 1920's. What they don't say is how much power that is not propulsive that is consumed and lost to generate the propulsive force. As most thermo teachers will tell you, it's all about where you draw your energy boundary. If you need more info, look it up in Heorner's Fluid Dynamic Lift.

And, if you're not that internet savvy, you can look up the plethora of papers here:
GeCheng Zha https://people.miami.edu/profile/cd10bc38896121a13e4b70152b7c41e7

 

This appears to be the best description - from (CFJ) Introduction https://acfdlab.miami.edu/CFJ_webpage/CFJ_web_dir/introduction_CFJ.html

So loosely speaking, it's a Coanda effect device that tries to also create a boundary bubble around the cylinder shaped so that the ambient breeze sees an airfoil and creates lift. You get some thrust from the breeze, and you recover power used to blow the air via the Coanda effect.

The difficulties here are many. There is no scaling law to get from small experimental devices to practical sized machines, because you can't just scale a stagnation bubble or boundary layer around a form. There is no way to predict the performance as a function of power consumption. Coanda devices work best where the peak jet velocities adjacent to the cylinder approach Mach 1. But how these flows develop across the face of the cylinder is notoriously hard to predict. Very thin jets tend to be more stabile, but won't induce a sufficient far-field circulation to generate much thrust from the wind.

And then there's this part - The airfoil always achieves a significantly higher lift due to the augmented circulation. That's simply wrong. There is no such thing as augmented circulation. A jet is not circulation. A jet in the near-field of the cylinder only produces the Coanda effect, not lift. It does not in any way change the far field pressure gradient that we relate to classical self-induced circulation, lift, and energy extraction from a breeze. The only way to get energy, lift, or thrust from a breeze is to present the breeze with a shape that causes a favorable changes to the momentum of the breeze. It's that momentum change, when we look at the details of the flow geometry and conservation laws, that we can decompose into a circulation term. It's just a book keeping convention that scales beautifully and allows the forces to be predicted at any scale and breeze condition. What we do not know how to do is accurately predict how the geometry of energy injection near the surface affects the momentum change in the breeze or the shape that the breeze sees and flows around. It doesn't scale with size or wind speed. The is no way to design the injection system to modify the flow of the breeze in the best way. It's all experimental. It's different for every machine. And there is no reason to think that the best injection geometry will morph in a smooth predictable manner from one scale to another. The entire realm is chaotic.

It looks like a perfect job for transformer AI, though. Telemeter actual wind conditions and ship motions from a ship at sea to a wind tunnel and duplicate the wind speed, orientations, and motions on the test article. Hand the blower controls off to an AI. Measure the net thrust, power consumption, and side forces and moments. Compute the effect on ship motions and apply to the test platform (I suspect vessel stabilization may help overall performance of both the masts and the primary propulsion system). Compute a measure of merit. Turn the lights off and come back in six months. At least you'll know something about how one machine will work on one ship. And perhaps the AI will identify a strange attractor in all the chaos.

 

Concur, some of this seems like the old horse handicapping physics joke...."first you assume all horses are perfectly smooth. perfect spheres"....

LOL, as has been pointed out many times CFD fares poorly in real world water. Roll, pitch, sway, surge, and heave will play havoc on any boundary layer manipulation at masthead height.

 

Thanks guys. It sounds like I can't build one of these in my garage, and even if I could there's no way to calculate the correct configuration, and then there's no way to predict performance.