Everything Old is new again - Flettner Rotor Ship is launched

I'm liking daiquiri's stabilizing effect better than Jeremy's prediction of deep rolls and risk of capsize in a tack or gybe. How serious is this gybe/tack stability risk?

 

Depends on the weight of the rotor, shape if the hull, ballast etc.

Remember the advice to get some engineering work done ?

 

Jeremy talked about "high roll moment when the wind is directly ahead or astern, and no lift under those conditions, just drag". Wind from 0° or 180° and no lift means that the rotor is not turning. In that condition, and depending on the wind speed, a Von Karman (V.K.) vortex shedding can trigger, which creates an oscillatory (alternating with +/- sign) heeling moment.
See this video for a visualization of the V.K. vortex street: https://www.youtube.com/watch?v=zPDyLE6UMPY
and this one to see what happens to a cylinder subject to V.K. vortex shedding: https://www.youtube.com/watch?v=_Hbbkd2d3H8

When the rotor is turning, the gyroscopic effect stabilizes the rolling motion. It doesn't mean that the V.K. vortex street is not there anymore. If the Velocity ratio (Vrot/Vwind) is sufficiently small (less than approx. 1.5) V.K. vortices are certainly there. The frequency of vortex shedding is approximately equal to 0.5*Vw/D. The problems might arise when this frequency couples with the natural roll or pitch frequency of the ship. In this case the gyroscopic effect (which transfers rolling excitation force into a pitching motion) might even contribute rather than cure the problem.

But I wouldn't worry too much about this, if the rotor is light enough and the boat has a sufficient transverse stability. V.K. vortex shedding occurs naturally on sailboats too, when they are running or very broad reaching. I believe that every sailor has experienced the unusual oscillatory rolling of their boat when navigating with wind at 180°. I still remember how it drove me crazy when I was a kid and couldn't explain to myself why was my Laser rolling so much in that condition, and why cannot I do anything to stop it. It is an annoyance rather than a danger, imo.

Cheers

 

Okay. Whew! (wipes sweat from brow) Thanks.
On occasion I have had to run in the trough beam on, in storm seas. The roughest ride I can remember wasn't in the largest seas I've encountered. Towing across the Gulf of Maine in 30 ft seas whose period was so short, they looked like teeth on a sawblade.
Speed improves ride in those conditions. Flatten the throttles on the dash and HANG ON!

 

The important thing to remember with a rotor is that the lift vector is always at 90 deg to the relative wind direction. If you don't stop the rotor when the wind is directly on the bow or stern, then you will get a very strong roll moment, as the lift vector will be acting athwartships, and the keel won't be able to translate lift into forward thrust, as thrust will be virtually zero (just the small amount of drag from the rotor)..

Also, remember that to tack (or gybe) you need to stop the rotor then reverse the rotor direction.

This isn't a hard problem to overcome, as a simple relative wind vane at the top of the rotor mast could be used to automatically detect wind shifts and change/adjust the rotor speed/direction as needed.

 

Good info, Jeremy.
The issue of stopping and reversing the rotor was addressed in the other Yobarnacle's thread about modifications to his boat. The idea about wind vane commanding the reversal of the motor is imo very good and simple.
Cheers

 

I think this is one reason why proas are particularly suited to Flettner rotors - their natural 'at rest' position is beam on to the wind and they should only really be head to wind (or running dead downwind) if something has gone somewhat wrong.

As for tacking boats I think using a rotor may require some conventions to be overturned. Most sailing boats are set up with slight weather helm so that in the absence of any other input they will round up to a head to wind position. This would be undesirable in a rotor boat which would probably be better set up with slight lee helm so that it naturally bears off to a beam or broad reach. My gut feeling is that this would give pretty poor windward performance unless a bow rudder was used

 

Lift is at 90 deg to the apparent wind because that's the definition of lift. There is significant drag associated with a rotor as well. So the maximum rolling moment will happen well before the ship is head to wind, when the total force is athwartships.

If you are trying to go to windward with a rotor ship, I don't think a little lee helm is going to make much difference to the performance.

Lift-induced drag for a rotor is the same as for a conventional sail producing the same spanwise lift distribution, and this drag is inversely proportional the the square of the height of the rig. So the rotors need to be just as tall as the mast of a normal rig to have any hope of achieving decent windward performance. If the rotors are shorter in order to reduce the heeling moment and gyroscopic torques, then they will have the induced drag of sail rig that is just as short. Plus, the rotors have a much higher profile drag because they experience separated flow on their aft face.

You never see rotors on a boat that is designed to go to windward under wind power. You always see them used as auxiliary propulsion on a motor vessel. The only advantages rotors have over conventional sails is a small deck footprint and the ability to continuously "reef" by varying the rotor rpm. If windward performance is a consideration, then a different rig choice is in order.

 

I am not sure I understand this statement. There are quite a few examples of rotors on primarily wind powered sailboats. The Flettner Rotors are reputed to perform as well as, if not better than conventional sailboats to windward. The first example in the photos below matched an identical conventionally rigged yacht in a race.

The 'only' advantages of small footprint and instant variable reefing. together with no swinging appendages or control lines seem to be very desirable indeed, even on non-commercial craft.

The reason that they are not popular on the average yacht is probably one of aesthetics and engineering complexity.

http://www.boatdesign.net/forums/projects-proposals/collapsible-flettner-rotor-project-50587-13.html

may be of interest.

 

Rwatson, you have rised the problem of aesthetics, which is indeed important for pleasure yachts. From that point of view, a Flettner rotor is a no match to a nice and elegant bermuda or latin rig. Which is a problem for its wider diffusion in the world of pleasure boats.

Last year, when I was approached by a builder to design a new line of sailboats for his brand, I have examined his SOR and have proposed a scow-type hull. The proposal was refused with the following argument: a boat has to look like a boat.
So the design efforts were switched to a more classical monohull type, which is more widely accepted as elegant and eye-pleasant.

I can foresee the same type of issue and argumentations for the case of Flettner rotors.

Cheers

 

I'm not completely happy with this statement.

There may be an analog to induced drag. But I think it would have some very different properties to the induced drag of simple fixed wings. The difficulties lie in the fact that induced drag is calculated with respect to a fixed surface. It involves a vector operation where there is a change in the flow vector (effective angle of attack), but the flow tangency condition is held fixed (parallel to the surface of the wing). In a rotor, the flow tangency condition in the separated flow behind the rotor would not be fixed, and this is the cause of 100% of the lift of a rotor, whereas the separated flow has at most a tiny effect on the lift of typical wings.

Given a one piece rotor with only diameter to work with, it is not obvious that any single lift distribution can be maintained over varying wind speeds. That makes the idea of an induced drag coefficient for a rotor a bit of a nonstarter. Whatever a rotor's appropriate drag decomposition may turn out to be, it will be different from that of a wing. I've seen nothing to suggest that an elliptical loading would be possible, or that anyone knows how to design a rotor with a particular spanwise load distribution.

<edit> of course tspeer's original comment compared the rotor to a sail, not a fixed wing. Sails also lack fixed geometry and respond to changes in wind speed. That fact contributes to their versatility. I just don't see how that versatility is going to be duplicated in a rotor.

 

I am at a loss to figure out your point. The physics of the rotor are well documented, from Flettner through countless studies since, as has the airfoil.

The 'versatility' of sails is probably best characterized as the 'complexity' of sails, with complex control system that ranges from sheets to vangs to winches struggling to get some useful shape out of an inherently less than optimum engineering solution. Sure, that's great for creating the wonderful mental challenge to get the combination right in different conditions, but from a practical point of view, hard to justify.

I regard the rotor concept more as a predictable tool to harness the wind for practical purposes, rather than the most efficient method to extract every last inch of performance from the wind.

The debate should probably be more on the economics of each method if one is discussing rotors V other wind capture devices.

 

No RWatson, it isn't. Not in terms of engineering. There is no information suitable to design a rotor "to meet expectations" as has been asserted. There is no way to guess how much lift will be produced from an arbitrary rotor because there are no scaling laws to go from one size to another. All of the data presented is only good for duplicating that particular rotor. It can not be scaled up or down. The drag calcs are all but nonexistent even on the ones tested.

The point I was making wrt tspeer's comment boils down to this. If you have an airfoil that has a uniform lift slope across the span - which happens to be every airfoil that is reasonable under thin airfoil theory that predicts a deltacl = 2*pi*deltaalpha - the minimum induced drag for a given lift requires the downwash to be uniform across the span. The trouble is that there is no such thing as a lift slope for a rotor that is a function of angle of attack. The minimum drag condition will not be one with constant downwash velocity. So the idea that the spans would have to be the same is not true. (Which isn't to say that nothing constructive can be said about it, just that the equality is not there. I suspect the steady downwash condition is the minimum for all possible configurations, and anything that optimizes differently would need a longer span to equal the induced drag of thin wing theory. So you might match the induced drag at one airspeed design point, but at any other, there would be more induced drag than on a wing of equal span.)

At any rate, I will continue to try to punch holes in any arguments that treat rotors with wing theories. Induced drag qualifies.

 

I am at a loss to understand this statement,

I have a number if sources of calculation of drag/lift that are well validated

To quote Jeremy Harris reply to you on another of your assetations a long itme ago
"I'm afraid I disagree, as the wind tunnel test data shows very clearly indeed that Cl and Cd are proportional to u/V. It's given in the NACA wind tunnel test data in NACA Technical Note 209, "Tests on rotating cylinders" by Elliott G Reid, Langley Memorial Aeronautical Laboratory (copy here: http://www.boatdesign.net/forums/att...cyclinders.pdf).

Once again, over my head, feel free to explain further,

 

Crazy alternate ship propulsion

Wow - really over the top fuel saving, including rotors

http://www.shippingscenarios.wartsila.com/Ship_Concepts_For_Shipping_Scenarios_2030.pdf