Everything Old is new again - Flettner Rotor Ship is launched

[Jeremy Harris]

Largely the potential problem can be managed in the same way as on any other sailing boat, always being aware of where the wind is coming from.

For a big ship this may not be an issue except for a couple of cases. The first is that, unlike a conventional sailing boat, a Flettner rotor boat cannot sail directly downwind. Because the lift vector is at right angles to the wind direction with the wind aft the vector just gives a rolling moment. This may present problems, as the option to run before high winds, or to keep the ship heading into waves, isn't there, the ship has to always try and keep the wind on the beam, or else turn the rotor off and try and heave to (which will only work if the centre of the total superstructure drag is located in the right place, fore and aft).

The second case is close quarters manoeuvring, where the ship may not be able to always keep the wind on the beam. I suspect a separate propulsion system might be required to allow harbour approaches etc to be navigated safely with the rotor stopped.

 

[rwatson]

Largely the potential problem can be managed in the same way as on any other sailing boat, always being aware of where the wind is coming from.

For a big ship this may not be an issue except for a couple of cases. The first is that, unlike a conventional sailing boat, a Flettner rotor boat cannot sail directly downwind. Because the lift vector is at right angles to the wind direction with the wind aft the vector just gives a rolling moment..

You might like to refer to the test Polar Graph in post #45 to see that this is not the case.

The second case is close quarters manoeuvring, where the ship may not be able to always keep the wind on the beam. I suspect a separate propulsion system might be required to allow harbour approaches etc to be navigated safely with the rotor stopped.

Absolutely, no large modern sailing vessel could survive comfortable without a good diesel and screw.

 

[Jeremy Harris]

Exactly. Look closely at the data and you'll see it was for the ship, not the rotor alone.

If you look at the way the rotor works you'll see that the lift vector is orthogonal to the wind vector, just as the lift vector for a wing is orthogonal to the direction of the airflow over it. Just as in a conventional sailing ship, the keel and hull act to convert the direction of the lift vector into forward thrust.

However, when the lift vector is dead abeam, as it is for the head to wind and stern to wind cases, then the lift vector cannot be translated into forward thrust and the force becomes just a strong rolling moment. The much smaller drag vector will be the only force acting to provide thrust.

The whole ship case (as in the data on Buckau and the smaller model) includes the drag from the ship hull, superstructure and rotor, not just the lift from the rotor. This means that for the whole ship there will be a drag element providing thrust in the forward direction with an aft wind direction, but if the rotor is turning there will still be the large rolling moment, too.

The reason I'm stressing this is that when I was looking at, and modelling, rotor performance a while ago (the spreadsheet is somewhere earlier in this thread) I exchanged emails with Stephen Thorpe, whose rotor boat is also featured somewhere here. He pointed out that the need to stop and reverse the rotor quickly was important, as he'd discovered practically by sailing his own rotor boat. It was he who alerted me to the potential problem of the large rolling moment if the rotor is spinning when you try to tack or gybe, as he'd experienced it first hand in his boat.


[Edited to add these links:

This is my reply to the post where you made the point once before about the apparent ability of a rotor to generate lift other than orthogonally to the wind direction: http://www.boatdesign.net/forums/bo...-rotor-ship-launched-24081-11.html#post446014

This post has a copy of NACA TN209 which has wind tunnel force balance data for a rotor, and shows that the lift vector is always orthogonal to the air flow vector: http://www.boatdesign.net/forums/bo...-rotor-ship-launched-24081-12.html#post446611

That last post also has a copy of a spreadsheet I put together to try and model rotor performance]

 

[rwatson]

Exactly. Look closely at the data and you'll see it was for the ship, not the rotor alone.......
However, when the lift vector is dead abeam, as it is for the head to wind and stern to wind cases, then the lift vector cannot be translated into forward thrust and the force becomes just a strong rolling moment. The much smaller drag vector will be the only force acting to provide thrust.

I dont beleive that the speed shown in that graph, almost 1/3 of the reaching speed, can be attributed to the drag from the superstructure. Just as lift on an aerofoil is not totally perpendicular, it appears that a significant forward vector applies..

Stephen Thorpe, whose rotor boat is also featured somewhere here. He pointed out that the need to stop and reverse the rotor quickly was important, as he'd discovered practically by sailing his own rotor boat. It was he who alerted me to the potential problem of the large rolling moment if the rotor is spinning when you try to tack or gybe, as he'd experienced it first hand in his boat.

Yes, but to be fair, he is using a very long rotor on a sailing dinghy. The length alone would make just the drag component significant for a small boat.

This is my reply to the post where you made the point once before about the apparent ability of a rotor to generate lift other than orthogonally to the wind direction: http://www.boatdesign.net/forums/bo...-rotor-ship-launched-24081-11.html#post446014

This post has a copy of NACA TN209 which has wind tunnel force balance data for a rotor, and shows that the lift vector is always orthogonal to the air flow vector: http://www.boatdesign.net/forums/bo...-rotor-ship-launched-24081-12.html#post446611

That last post also has a copy of a spreadsheet I put together to try and model rotor performance]

Yes, I dug those up to. The thing is, that as we both know, the theory and the actual performance can vary considerably.

Did Stephen say anything about downwind performance specifically ? I bet he wasn't becalmed, listing to one side.

 

[Jeremy Harris]

Lift from an aerofoil is always perpendicular to the direction of airflow though, and the same is true of the rotor (which is just another form of aerofoil). Much as I distrust Wikipedia (http://en.wikipedia.org/wiki/Lift_(force)) it does express this well: "Lift is the component of this force that is perpendicular to the oncoming flow direction.[1] It contrasts with the drag force, which is the component of the surface force parallel to the flow direction." The reference [1] is to a simplistic explanation on the NASA website.

Drag from the rotor and ship hull/superstructure can be substantial. The L/D of a rotor varies from around 8:1 or so down to less than 1:1, so for the worst case drag can exceed lift and even for the best case drag is around 12% or so of lift. This means there will always be a fair bit of forward thrust from drag when the wind vector is from aft.

I've just dug out the emails I exchanged with Stephen back then and these are his words on the heeling moment problem:
"There is a dangerous case peculiar to rotors though - namely that your heel increases the closer to the wind you get, but unlike a conventional sail which stalls and loses power (thus heel), the rotor keeps delivering max thrust even when pointing right into wind. As that thrust is always perpendicular to the wind, you will capsize in a canoe hull. To avoid this you need to anticipate the condition (it can occur very quickly) and have a very good rotor brake (a rapid brake will help you in other ways too)."

Clearly having a substantial keel or a hull shape that provides a strong restoring force against a heeling force will mitigate this.

 

[Paul Walbran]

I've just discovered this thread and enjoyed reading it. Like RWatson have had a long interest in the Flettner rotor concept ever since coming across it when a student in the late 1960's.

At that time the Uni had an annual raft race, so my friend and I set our sights on winning the prize for the most creative craft. So I visited the Uni's engineering library and photographed Flettner's 1925 report in the "Engineering" journal (no flash photocopiers then but I was into photography) and we built a Flettner rotor powered catamaran, each hull consisting of 6 x 44 gallon drums, with a very crude bit of "streamlining" on the forward one of each hull. Overall width as I recall was some 10-12 feet. So you could say a heavy and cumbersom hull.

We made the rotor 12 feet high, 3 feet diameter with 6 foot diameter end plates. Construction was plywood end plates and internal frames (2 of them suitably lightened with lots of holesas were the end plates) with lightweight (spruce I think) longitundinal stringers. To these, circumfrential and spiral alloy stips (held on with masking tape) provided bracing, and corrugated cardboard formed the outer coating. Oh, and polyethelyne wrapping round the outside to protect it from the marauding opposition, one of whom had a fire engine pump mounted on their raft. The whole rotor was rigid and unlike the hull light.

power was provided by a small (and temperamental) Villiers 2 stroke parked on one corner, and drive vial a roughly made rope "belt" (we did at least splice the ends together) with an old bicycle wheel (minus tyre) attached under the bottom plate acting as a pulley.

The mast was a length of 3" galvanised water pipe, no stays, just canitlevered off a suitably strong space frame under the deck line.
The bearing were just holes cut in the plywood discs, and greased - apart from the top one where we lashed out and got a proper thrust race.

We were deeply sceptical it would go, theory is one thing but practice is another! So were friends and family.

Came the day and we had 15 knots of wind coming from a very close-hauled (for a normal sailing craft) direction. We assembled the craft on the beach (to much speculation as to what it was) and then made the near-fatal error of test-driving the Villiers .. we forgot that it didn't like re-starting once warm! So after race start we drifted down wind for a while until the temperamental beast started again. By this stage we were halfway out into the main shipping channel and under the watchful eye of the Navy rescue craft. We assured them we were all OK, and when the Villiers finally sprang into life to our utter amazement the beast began to move ... quite well! We were making 4-5 kts, and pointing like we'd never experienced before as dinghy sailors. I think we made something like 30 deg off the direction of the wind and were on course to laying the finish line ... until a wave swamped the Villiers and it didn't recover. We tried padling (came prepared) but 5 or 6 people made very little headway with the ungainly hull. Then one of the team decided to spin the rotor by hand, and the raft started to move again - despite turning at a rate only about 20% of target. Speed about 1 kt but at least it was progress and we made the finish line.

And got the dozen beer prize for the most creative craft. About the only time the scientists beat the engineers at their own game 9and with the help of their own library).

This experience to me (the power of one person's hand rotating doing much better than the combined efforts of 6 paddlers) alone puts to rest the foregoing claims about efficiency, including that the power used driving the Flettner rotor would be better spent driving the vessel directly.

When one of my sons was 12 he had a free choice science project . I suggested the Flettner rotor, showed him my nold photos. He built a test righ comparing both conventional sail and flettner rotor, both of the order of a foot long. Results were impressive. And he won a prize too.

I've often toyed with the idea of fitting up a yacht and having a play. earlier this week, while visiting sons currently in Munich, we visited the Deutsches Museum and found their reference. This lead to my getting keen again, and this website. Maybe it might yet go even further.

 

[rwatson]

I've just discovered this thread and enjoyed reading it. .......

Great Paul.

What a great story. The transition from dream to reality is always the hardest journey.

Please do keep us informed of your experiments in the future.

 

[rwatson]

Lift from an aerofoil is always perpendicular to the direction of airflow though

If look at the diagram attached, see if you can spot some non perpendicular vectors. Thos is especially true of the high pressure areas.

and the same is true of the rotor (which is just another form of aerofoil).

Maybe not entirely. I remember reading a post that stated that this was an oversimplification, and that the rotor behaved more like a 'perpetual stall'. Blowed if I can find it now, but it stuck in my mind.

I've just dug out the emails I exchanged with Stephen back ...

I note with disappointment that Stephens web site and email address are no longer operational.

Come back Stephen - we need you to do a downwind run

I bet its those wicked oil companies who have kidnapped him, as a threat to world oil consumption



I am currently completing a scale model for another project. I think I will have to put a Flettner rig on it after it fulfills its primary function.

 

[Paul Walbran]

Because lift is the driving force for most conditions it's easy to get over-focused on it. But it is of course just a component of the total force which also includes the drag. If I recall correctly (being far from home and not having my notes hand to confirm) L/D varies with the rotor speed to wind speed ratio. If that's the case, then just as sail shape is trimmed for different wind angles so should the speed ratio. To point as high as possible, L/D should be maximised.

There may be a trade-off here in that this may occur at less than maximum lift (sorry, don't have notes again) but if that's the case then it's no different from the usual close-hauled situation in which sail shape is flattened (via outhaul, cunningham etc) to get the sailing angle which maximises VMG.

For downwind, a speed ratio which optimised drag would allow the total force on the rotor to be more in the forwards direction and allow sailing with the wind further aft. While this would allow sailing directly downwind, albeit with a bit of heel (hardly an unknown sailing phenomenon), the best course to sail would likely to be some form of down wind tacking which got the best VMG - again no different from conventional sailing.

When I get home in a coule of weeks I'll dig out a photo of our effort.

 

[Jeremy Harris]

If look at the diagram attached, see if you can spot some non perpendicular vectors. Thos is especially true of the high pressure areas.




Maybe not entirely. I remember reading a post that stated that this was an oversimplification, and that the rotor behaved more like a 'perpetual stall'. Blowed if I can find it now, but it stuck in my mind.




I note with disappointment that Stephens web site and email address are no longer operational.

Come back Stephen - we need you to do a downwind run

I bet its those wicked oil companies who have kidnapped him, as a threat to world oil consumption



I am currently completing a scale model for another project. I think I will have to put a Flettner rig on it after it fulfills its primary function.

That diagram has the lift vector marked correctly as being perpendicular to the airflow vector, though. The relative pressure arrows are just illustrative.

The force felt by the boat will be the sum of lift and drag and the relative angle and magnitude of that combined vector will vary as the magnitude of the two vectors changes with wind direction, wind speed and rotor rpm.

 

[Jeremy Harris]

Because lift is the driving force for most conditions it's easy to get over-focused on it. But it is of course just a component of the total force which also includes the drag. If I recall correctly (being far from home and not having my notes hand to confirm) L/D varies with the rotor speed to wind speed ratio. If that's the case, then just as sail shape is trimmed for different wind angles so should the speed ratio. To point as high as possible, L/D should be maximised.

There may be a trade-off here in that this may occur at less than maximum lift (sorry, don't have notes again) but if that's the case then it's no different from the usual close-hauled situation in which sail shape is flattened (via outhaul, cunningham etc) to get the sailing angle which maximises VMG.

For downwind, a speed ratio which optimised drag would allow the total force on the rotor to be more in the forwards direction and allow sailing with the wind further aft. While this would allow sailing directly downwind, albeit with a bit of heel (hardly an unknown sailing phenomenon), the best course to sail would likely to be some form of down wind tacking which got the best VMG - again no different from conventional sailing.

When I get home in a coule of weeks I'll dig out a photo of our effort.

If you dig back through this thread you'll find a spreadsheet I wrote a while ago to estimate L/D for various wind speed and rotor surface speed conditions. You can vary L/D a lot by varying the rotor rpm, so can use this to trim the "sail" for any particular conditions.

My plan was to use a relative wind vane on the top of the (non-rotating) rotor support mast to control the rotor rpm and direction. My thoughts (untested I have to say) were that such a system would be able to trim the rotor speed and direction for best performance, and also act as an automatic rotor brake/reversing system to counter the head to relative wind/stern to relative wind problem.

 

[Paul Walbran]

Thanks Jeremy, yes I did note the spreadsheet for further reference but have yet to have a good look at it, being basically on holiday with lots to do. (Though the reason I ended up with time to read the whole thread and look at other links was a pulled achilles tendon playing football with my sons. I should know better ... )

Given the electronic capabilities now, controlling things via a sensor at the top of the rotor sonds a very feasible plan.

 

[Paul Walbran]

Here is a photo of our effort in the Uni raft race referred to above.

 

[rwatson]

Wow !!! I am super impressed to think that it actually worked. It is very inspiring.

Thanks so much for the post Paul.

 

[BertKu]

Wow !!! I am super impressed to think that it actually worked. It is very inspiring.

Thanks so much for the post Paul.

Hi ,

Do you know a way to find out where the E-Ship1 is in the next view weeks. ? I will be in that area, Kiel and Emden in 4 weeks time. I would not mind seeing it in all its glory. Speed is approx 5 - 6 knots. Not that I am planning to build one, but it is certainly something I do not get to see in life everyday.
I tried Google, but it seems out of reach according to the below information.

Bert


Last Position Received
Area: North Sea
Latitude / Longitude: 53.3558° / 7.1978° (Map)
Speed/Course 0.5 knots / 195˚
Last Known Port: EMDEN
Info Received: 54d 2h 15min ago (AIS Source: 638)
Not Currently in Range
Itineraries History
Voyage Related Info (Last Received)
Draught: 6.9 m
Destination: EMDEN
ETA: 2013-01-18 09:00
Info Received: 2013-05-31 13:43 (54d, 2h 16min ago)
Recent Port Calls:
No Records Found
Ex Names History
No Records Found