Flettner Rotor for small craft - Design and Build Prototype

[rwatson]

After years of interest, I am finally able to think about building a working Flettner Rotor to try out.
I've started buying the gear and planning the structure, so its time to get suggestions and ideas from anyone interested.

I have a spare Canadian Canoe hull, which I will build an additional hull for as a stable platform, but FIRST, I am going to build the Mast Rotor structure, and test it on land.



The aim is to test results along these theoretical figures

If I can get 5 HP at 15 knots of wind, that will be terrific.

I plan to build a truss from Steel Square Tube, 25 x 25 mm, 1.6 mm wall. Four edges to the truss. It will weigh about 26 kilos. If I used 20 x 20 mm, it would save ~ 10 kilos, but my gut feel is that it wouldn't be strong enough.

I am going to have to disassemble to structure every launch, so maybe I will need to have the truss in two parts, to make it easier to erect.




I found that I can buy GoCart axle and accessories quite cheaply

It has the bearing blocks, sprocket, chain and brake system together, so this will form the basis of the mechanism.

The batteries and motor will be at the bottom, and drive a shaft to an aluminium plate at the top of the truss.


I plan to make the Cylinder in 2 sections, each section comprising 3 panels



I speculate that I will have a 4mm Aluminium circular plate at the top, and 2 more "rings", that the fibreglass panels will bolt to.
For the lower rings, I will need to attach some sort of bearing fitting to the Truss, to support and guide the Aluminium Rings inside the cylinder.

I've ordered the batteries based on an old Mobile scooter I have. It has two older 12V AGM batteries I can experiment with and I can use the charging system on it to charge the 2 new batteries I've ordered.




My logic is that if these two batteries give satisfactory longevity for Kart driving, then they should find running a small pedal assist motor pretty easy.
If the pedal assist electric motor isn't strong enough to spin, say 10 kilos of fibreglass and aluminium rings, I can either get a bigger motor, or give up

I've ordered a motor controller that I hope will work.


As well as a Tachometer, it is reversible, and steady speed controllable.
While I am waiting for it and the new batteries, I can start building the truss.

That will be enough to be getting on with. The excitement begins.

 

[rwatson]

The "Mast"

Doing some further analysis of the supporting Truss.

After adding up the webbing components, the whole thing would have been getting up in to the 40-50 kilo mark, which for a relocatable item, is just too awkward.
So, by reverting to a Tripod Truss, and using 20x20mm - 1.6 ,main lengths of square steel, and 15x15mm sq for the "webbing" , it should come in closer to 20 kg.


By looking at various commercial trusses, which are all built in Aluminium, I found load limits between 250 and 1000 kilos for this configuration.
If I can achieve minor deflection at a top load of say 130 kg, that would be adequate.

I will test the finished truss, and if it has too much deflection as a freestanding mast , I will have to look at using stays.

Applying stays seems quite straightforward. The plan was to have one end of a Kart axle spinning an aluminium "platter", so initially there was nothing to attach stays to.
If I make sure the topmost axle protrudes past the top spinning "platter", and have the end run in a bearing inserted into a "crane", there will be a stationary point to affix stays to.

I may have to make it a 3 or 4 armed crane, depending on what the whole system is attached to.
For initial ground erected testing, that will be very useful, to. It will save me having to build a robust foundation pad.

There is probably some software that could calculate the truss performance. I will have to do some research.

 

[rwatson]

Note for Build Diary -
Got quoted about a AU$1,000 to do an engineering case analysis for the steel tripod setup.
That would be extra to the cost of buying the steel and fabricating it.

Took a gamble, and followed my own advice from years a go, and bought a premade aluminium Stage Truss


It comes pre-rated

, rust proof and ready

It comes pre-rated, rust prof and ready built for the same price, for the same weight.
Trusst CT290-430S Box Truss 3m https://djcity.com.au/product/trusst-ct290-430s-box-truss-3m/

 

[mc_rash]

Hey @rwatson nice project, I would like to see it on the water.

1. Have you thought about using thin Aluminium panels instead of GRP for the hull of the rotor? It might be lighter and cheaper to wrap an Aluminium sheet than producing Fibreglas panels with molds etc. Or are there prefab GRP panels for this purpose?

2. How are you going to achieve enough stability for your canoe to prevent capsizing? Do you plan adding a keel?

 

[rwatson]

Hey @rwatson nice project, I would like to see it on the water.

1. Have you thought about using thin Aluminium panels instead of GRP for the hull of the rotor? It might be lighter and cheaper to wrap an Aluminium sheet than producing Fibreglas panels with molds etc. Or are there prefab GRP panels for this purpose?

2. How are you going to achieve enough stability for your canoe to prevent capsizing? Do you plan adding a keel?

Hi McRash
Thanks for the input.
Aluminium is a great contender, but has limitations if one can't Tig the light sheets.
The other nuisance consideration is that thin Aluminium tends to Tin can and bend in use.
My limitations mean I will be transporting the rig a lot, and having to reassemble at test locations.
To my mind this means small easily handled sections, that are self supporting.
Currently, I plan a small mould that will create 1/8 of the tower. That is, 4 sections make the lower half, 4 more sections make the top half.
That way, one easy to create, accurate mold will produce precise replicas for joining. Say 1.5 m tall, and from memory, under .7 m of curved surface.
Based on personal experience, I am way more confident attempting that with f glass.

Re, the canoe. Yes, that would be a project balancing that rig

In the fulness of time, I would create a 20ft outrigger for this canoe for testing, IF I can't come up with an easier solution.
There are always plenty of half finished cheap hulls for sale, and it may be easier to convert one of them.
My brother doesn't know if it, but his 26ft f glass yacht is in my sights

Initially, I am going to mount the entire rig on my 20 ft canoe Trailer.
This will allow me to do a huge amount of testing on land, and sort the bugs out there.

The trailer is 2.3 metres wide, so it has a good righting moment

The thing is, the testing will involve winds coming from the side, but that translates to fore and aft thrust vectors, so I can apply a lot of power to the land rig with zero chance of a capsize.

Put it this way, I will either look like a mad engineer with a skinny spinning merry go round, or, at 400 rpm, in 15 knots, I will be able to push the trailer and towing vehicle in a straight line at slow speed.

I may accept bets

But, on land I can solve the main risk
1. Excessive power use
2 vibration
3 mechanical loads
4 tower anchoring loads
Etc

So many mysteries to solve

 

[rwatson]

Answer to Inquiry
I had a question in another forum linked to this one, which I would like to include in this thread.

What are the expected Benefits of a Rotor Compared to Sails
1) Controllability - push button power and settings. No winches, tangled sheets and constant adjustment for wind shifts.
2) Reduced heel - the Rotor actually smooths out Roll and Pitch, and even tends to lean into the wind.
3) Safety - on a small one person craft, you can actually set a "Deadman's" switch, to stop the Rotor in a man overboard situation.
4) Safety - sudden squalls do not result in a knockdown situation.

To be proven aims
a) Cost of operation
b) Better performance. Downwind performance is problematic for Rotors, which can be offset by "tacking" downwind, like the AC50 catamarans do. I see no objection to flying as spinnaker in certain circumstances.

Edit - I include a lionk to a story of a case example to expand on these principles
Popular Science https://books.google.it/books?id=eAAAAAAAMBAJ&printsec=frontcover&hl=it&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false

 

[rwatson]

Adding another link to a valuable contribution - Working out the Operating power

"~ 400 Watts x 1.5 for mechanical losses (600 watts) = about 25 Amps on a 24 Volt system, is valuable information for my current project. Just by sheer luck, I bought 30 amp fuses.
BUT - I guessed a suitable electric motor of 250 Watts, which now looks way too small. It looks like my anticipation of having to buy a bigger motor is now reality.
The experimenter in me is inclined to initially fit the small motors, and see if they are overpowered in high winds. The power calcs estimate only 40 Watts required at 7mph (3.13 M/S) .
The 2 x 12V batteries are 34 Amp hours each (600 watts) so theoretically they could spin the rig for 2 hours ?"

Everything Old is new again - Flettner Rotor Ship is launched https://www.boatdesign.net/threads/everything-old-is-new-again-flettner-rotor-ship-is-launched.24081/page-40#post-941712

Big Thanks to Dolfiman

 

[Dolfiman]

For the rotor, have you think to look at the formwork tubes option : they are available in your diameter, very light, usually made of recycled fiber or equivalent, and proposed in length of 12' or more in standard. Numerous providers, like these ones :
Cardboard tubes for building and construction (abzac.com)
Sonotube, The Best Way to Form a Concrete Column | Form Tech (formtechinc.com)

 

[rwatson]

For the rotor, have you think to look at the formwork tubes option : they are available in your diameter, very light, usually made of recycled fiber or equivalent, and proposed in length of 12' or more in standard. Numerous providers, like these ones :
Cardboard tubes for building and construction (abzac.com)
Sonotube, The Best Way to Form a Concrete Column | Form Tech (formtechinc.com)

Hi Dolfiman.
Another great contribution.
No, indeed, I didn't think of that as a solution.
I have just been on the phone to Sonoco, to get specs. I'll put the details on record, for information for other viewers.

Their smallest practical size for my purposes is 305 mm Diameter, 3 metre length, 5mm thickness, which would be a snug fit for my 290 mm square truss.

At 12 kilos, with a 5mm wall thickness, it certainly fits within reasonable solution. I am looking at around $AU150 ex factory, but the distance from the factory is considerable, so it could end up being double that.
At that price, it would at least make a good prototype for initial trials, even if I had to build something else more robust for long term.



I ran the figures through Jeremy's calculator, and 300 mm does reduce performance drastically, almost half, by going from .5 metre to .3 metre.



It also adds another 25% of RPM to the equation for this lower output.

They also provide larger diameters, 356, 406, 457 and 508 mm
The cardboard tube at 508 mm dimensions, the weight is about 20 kg., and $AU300 approx, PLUS Freight ( say $300)
--------------------------------------------------------------------------------------------------------------
After much thought -
Conclusion - definitely a consideration IF I can get on locally, but at this stage, a custom fibreglass solution looks to be more optimal. I suspect I can build a f'glass cylinder to come in at around 12 kilos, that is disassemblable for ease of transport, more robust to fasten to bearings, main rotor etc, and more water/UV proof.


----------------------------------------------------------------------------------------------------------------
For anyone interested, this is how one other project solved the Tube construction.

 

[rwatson]

Stage 1 - Top Aluminium Disk built and Tested

The top disk, which has to support a fair bit of the structure, and act as a "flywheel" for the rotating cylinder, is made out of 6mm marine grade aluminium.
I had little trouble cutting it out with a metal power jigsaw, and I managed to get it within say, 7mm of a perfect circle.

I then mounted the top axle on a testbed of steel, attaching it with the bearings.

Initially, when I spun it up, I estimate it only got to about 100 rpm before severe vibration kicked in. This wasn't good, because I need a maximum of 600 rpm.
I used a disk grinder on a drill to try to get a better circle, but besides making a lot of aluminium dust, it wasn't cutting enough material

So, I made some rough mods to the crude workbench, bought a compressed air grinder and an aluminium carbide tool bit, and set up a primitive lathe.
I was able to get it to ~3 mm all round.



The testing had to wait while the Internet brought me my Tachometer.
Once the disk was within ~3mm round, I spun the disk up to 900 RPM, and there was very little vibration.


I am now able to manufacture the inner support struts for inside the "tower" now, and mount the top axle, with the disk on top.

 

[rwatson]

Progress report for a couple of watchers.

The major headache is designing and sourcing the correct bearing arrangements.
The two go-cart axles come with their own pillow blocks and bearings, but I have concerns about continual side loadings, as the axles will be supporting the entire weight of the "tower."
The provided pillowblocks are only light alloy, and I could see them shearing off from the sideways load.




So, I had to do a lot of research on Thrust Bearings, that would suit the two axles (25mm and 20mm ID) , and I came up with these.




The big problem is that Thrust Bearings don't normally come with any kind of Carrier. I could have had some made up by a machine shop, but that project has its own issues,

So, after a long internet search, I came up with these


I found a Bearing Flange with the same outer dimensions as the Thrust Bearings, and they finally arrived today.

The good news is that the Thrust bearings fitted really well.
One aggravation was that the Bearing Flanges come with a retaining circlip, (which fitted the Thrust Bearings, luckily) .
But, after spending two hours trying to locate a suitable tool for removing and re-installing the circlip (note, automotive Circlip tools that claim a 50mm spread, don't necessarily) , I foolishly let one "sproing" away into the hidden recesses of the shed, so I have had to order some more spare circlips.

Now I have the bearings sorted, I can go ahead and mount them to the aluminium tower on their shafts.
I will post some photos of the mounted bearings and shaft soon.

On a personal note, I have been messaging a friend who is out sailing in high winds and squalls, and has had to anchor for a few days to wait for the gales to die down.
I am hoping that a Flettner Rotor in place of traditional sails would make scenarios like this less of a problem.

 

[Paul Scott]

Is the Rotor essentially a free standing mast? Is the hull up to the Rotor and you twisting it?

 

[Paul Scott]

You may have seen this already, but it may be that reducing the span of the rotor and increasing the rotation speed works to reduce drag- at least that’s what this guy thinks. Might make the build simpler, lighter, and heeling more controllable? RC fun. (I still think that the top endplate will be a problem in higher wind, especially if you’re heeled.)

An interesting article on FRs- go down to the section on endplate size- the size of the endplate can be smaller at smaller Reynolds Numbers, or very high speed rotations- and it’s from an international journal of rotating machinery.

Flettner Rotor Concept for Marine Applications: A Systematic Study https://www.hindawi.com/journals/ijrm/2016/3458750/

 

[rwatson]

Is the Rotor essentially a free standing mast? Is the hull up to the Rotor and you twisting it?

Hi Paul, good to hear from you.

Yes, the mast is freestanding. I am not at all worried about the "twist" forces, but the standing forces will be a big concern. I have a few ideas on how to achieve this, which includes tensioned cables.

Re, the size of the rotor, size and rotation speed is a big part of the system. In the first post of this thread, I included a graphic which was based on a freely available rotor calculator, which works out the expected performance,
One of the objectives of this exercise is to verify the calculations, which can then be used for any desired configuration to suit a particular style of hull. Matching Rotor performance to a catamaran would be different to matching one on a houseboat.

As to end plate calculation's, there was a recent post where this came up
Everything Old is new again - Flettner Rotor Ship is launched https://www.boatdesign.net/threads/everything-old-is-new-again-flettner-rotor-ship-is-launched.24081/page-40#post-946046
There was even an analysis of end plate performance.

I also included an end plate test of a scale model that experimented with "slotted" end plates.

I will be pleased to check out the publication you included, to see if I can pick up further knowledge on the topic.

 

[rwatson]

Posting this Calculator for future reference.
Trying to revive the project, with a bit of fitting and turning.
More info soon.