There has recently been a lot of attention given to "Toroidal Propellers".
MIT has a patent. There are also expensive boat "Toroidal Propellers".
Link with links to triloop and biloop stl experimental models at printables.com and thingiverse. Also shows rails and profiles for sweeps.

https://www.youtube.com/watch?v=hby6U47oV_k

- Brian

So MIT has patented the "dounut" shape:-)
Thrust and Noise may be better or worse than more traditional designs?
The video shows an attempt to reproduce the loop designs, using rudimentary propeller design knowledge.

Interesting. I haven't heard about these before. It looks like a swept forward and swept backward blade used at the same time. It should be able to be analyzed using existing methods. The tips would meet up and be joined. For the performance calcs, you would probably want to subtract out the tip station, since it wouldn't be functioning as it normally would. The wake would be pretty unusual, but probably not an issue.

Hi Brian,

I did a quick test and it seems like this is fairly easy to do. The way I did it was use the program I created, PROP_DESIGN. I randomly used some inputs that I thought might work out. The first blade is swept back, the second swept forward. The other geometry characteristics are the same for each blade. By dumb luck, it seems to work out without any real issues. It also wouldn't have to be 'patented' like the MIT design. If they end up getting a patent for this, it's a real joke. This isn't any different than any other propeller, despite their claims. This capability has been in PROP_DESIGN for over a decade. The nice thing is no code changes are needed, since the tips don't end at the same point, like I originally thought.

Some performance items that would go against their 'grand' claims. Generally, sweep is not a good thing. Straight blades with a constant chord are usually best. Noise has nothing to do with the geometry they are changing. Noise is caused by pressure drops. The more the pressure drops, the more noise you have. Doubling the blade count will make comparisons virtually impossible. The propellers will be very different. The propeller with double the blade count will be at a much lower diameter, to meet the same power requirements. This will end up changing the rpm. So basically, it's a whole different design from the ground up. It's easy to say one prop is quieter than another. However, comparing two props that draw the same power, is the key. There are a lot of ways to deal with noise. I doubt this toroidal prop thing is the best method.

I'm really at a loss for how this is patent worthy. But the way the world is, they will probably get it.

Anthony

I made the geo in Rhino 5. My foray into MoI was a failure. The MoI step files don't work with Netgen. That killed the transition for me. However, I still like to follow the forum. It's a great community.

Edit; Updated the pictures, to reduce the file size. However, I can't seem to figure out how to get them inline.

Anthony, thank you for your aerodynamic information and prop designs.

- Brian

hi brian,

i ran the designs through the performance codes, to try and do the best comparison i could. i am seeing the swept forward / swept back config about 2 decibels quieter than the straight blades. i kept the radius and power the same for both designs. given the drastically different geometry, that made the rpm for the straight blades half of the swept blades. so a 2:1 reduction gear is the only difference. the motor would be the same for this comparison.

they stated they did this for noise and not having exposed tips for drones. i'm not really sure that it's any safer though. if your finger got caught in the rotor it would still be a problem. for marine applications i'm thinking seaweed and lily pads would still get all mangled in there. the straight blades were slightly more efficient, so they made a little more thrust. it's a pretty negligible difference though. the cad model i showed has the annulus to try and duplicate the loop they have in their design. the annulus wouldn't cause any aero modeling issues and is easier to model in cad. performance wise, ducting/annulus/tip devices don't make any difference. so i usually don't use them. i only put that in the model to try and produce an equivalent to what they were doing. that loop they have to connect the blades would be an issue though. you would have to use 3d cfd to model that. i suspect it would cause performance degradation, to some extent.

in any event, i don't know that this is patent worthy or even really worthwhile. it limits your options a lot. there are a lot of things you can do to deal with noise. i'm not really sure this is the way to go. but to each their own. what's sad to see is this is just one of many configs you can run and for some reason they think they can patent it. if i were to really build something like this, i think i would do it the way i showed in the pictures, but leave the annulus off. the blades shown have 60 degrees of sweep. that's a massive amount. pretty much the maximum allowed, otherwise the blades double back and would be nonsensical.

anthony

Update 1; Added pictures of the swept geometry that was used in the comparison

Update 2; Added the normal swept geometry. It performs exactly the same as the swept forward / swept back config. This is another reason why the patent seems so senseless to me.

Update 3; I conducted another comparison, to try and make it more fair. Previously, the rpm difference benefited the swept design. It allowed the noise to be less, since the torque was higher for the straight blade. For the updated comparison, I kept radius, shp, and rpm the same. This makes torque the same as well. So the noise calcs are on an even playing field. As expected, the straight blade performed better. It had more efficiency, thrust, and comparable noise. So I definitely don't see an advantage to the torodial propeller. In fact, it's worse in a lot of ways. You can just use a regular swept blade config and get the exact same performance as the swept forward / swept back config (toroidal propeller). However, as mentioned, you can outperform a swept blade with straight blades. I added some pics of the updated swept design. The duct / annulus is of no benefit and adds weight and cost. So I wouldn't use it. If safety is a concern, for drones, you can just put a cage around the drone or props.

Just for fun, I connected the tips on the last example I created. The prop consists of a mix of swept back and swept forward blades. It's performance would be similar to a propeller with all swept back or all swept forward blades. A few issues will hurt the performance of what is shown. The airfoils get close at the hub and tips, the hub has to be wider than a normal design, the connected tips extend past the diameter of the propeller. It's either extend them out or change the interior of the shape. Changing the interior would deviate from the calculated shape. Extending them out will also deviate from the performance calcs. Since connecting the tips is not of any use, this was just to play with the shape in cad. It would be better to use all swept back or all swept forward. Even better than that would just use straight blades or straight blades with swept airfoils (if sweep is really needed).

Edit; Updated pictures