Everything Old is new again - Flettner Rotor Ship is launched

Thank you Leo and Rwatson.
50% was pure luck and 50% was logic thinking. Next week I will finalize this story and then hopefully we should know what the story is.
Bert

 

Bert has been kind enough to send me a few snaps of the E-Ship1 on his trip .

meanwhile, Google has produced another study of rotor performance expectations done in 2011, with some more formulae to calculate expected rotor contributions, modeled on specific wind statistics over selected routes.

It may be of interest to some readers


addition - a very comprehensive review of all the different types of wind 'converters' and their uses, including the different application of the Flettner Rotor concept

 

Yes, it was an interesting trip.
Mr. Rolf Rohden was the driving force and project leader of this E-Ship1. In the meantime he has left the organization and has started his own business after some disagreements or differences in outlook on certain issues.
When they did the initial trips with this ship, they encountered Diesel/generator problems. As you know they had multiple diesel/generators and they used their own windmill generators. After they discovered that it was a problem, they moved the ship away from Kiel shipyard to Emden and as you can see, the ship is far out of the water, in comparison to previous pictures. Indicating that the diesel engines with generators are removed. Due to the slump in business, the company still has the financial commitment to pay for the initial items ordered and thus the ship will be one day again used, but nobody knows when. The principle was a success, just the problem when the ship is static and has to start moving. The coordinates of the E-ship1 >> 53 degree 21’ 19.41 North 7 degree 11’ 51.76” E .

Bert

 

Thanks for the papers.
In the first paper the estimated 16% contribution to power from the
Flettner rotors is certainly very attractive, but it is still not
proven on a real ship.

I'll be more favourably disposed to them when I see independent
measurements of the power generated (by, for example, the E-ship) and
other information on how the rotors perform in real conditions.
A little more on maintenance requirements and rotor vibrations would
be very welcome too.

Bert mentions that the project leader claims that "The principle was
a success" which is not quite the same as the system as a whole being
worthwhile to implement, run and maintain. Nor is it an independent
assessment.

Cynical, skeptical minds are still waiting.

 

Hi Leo, just to make a small correction. I did not speak to the project leader, but to some other people who have worked on the project. Maybe somebody in Germany can get hold of Rolf Rohden (he could have started a consulting engineering business in Hamburg or Bremen) and ask better technical questions then I could do. My knowledge on the subject is limited.

Could somebody explain to me, what possible could have gone wrong with those diesels/generators. If the ship is static, thus not moving, what energy would have been required to make the ship moving from 0. Were the generators to small and their exhaust fumes not enough to drive the turbines? We can only guess, as most people are reluctant to talk about it.
Bert

 

I think you have hit the nail on the head Leo. Mechanically, its enough of a proven solution that it 'works' as a concept.

The real life practicalities, acceptance of lower speed and pandering to the vagaries of wind, any mechanical ( vibration, power train, maintenance ) would eat into lower profits from slower performance for a ship like this.

Given that there is a big slump in seafreight demand, any economic limitations would be a major consideration.

Early on in the thread I posted an article about the "Tracker", a much more underpowered, smaller boat that actually showed a 'profit' on the rotor deal.

It would be interesting to see if the owner was still using the method, and if not, what the problems were

Once again, no-one advertises failure - even if it is of real interest.

 

The best economical example would be the Ferry we took from Kiel to Oslo. 20 hours, approx. 1000 people @ approx. 200 Euro = 200.000 Euro per night x 30 = 6 million Euro per month . The Ferry was doing 20 hours, while the E-ship1 would have done it in 60 hours. i.e. who would travel with an E-ship1 type of boat. Very few people and also only an income of 2 million per month with basically the same overheads. Who would really take an 60 hour ferry, while the competition does it in 20 hours.

The mistake the project team made, in my personal opionion, was that they had 6 or 7 small diesels/turbines/generators, instead of one massive diesel (or maybe 2) which could push the boat at 18 - 25 knots. Also they designed their own propellers.

I therefore guess that all diesels/generators/turbines have been removed and will at a later stage be replaced by 1 large Deutz diesel/turbine/generator.
Only then, will or can, Enercon announce their success story. Bert

 

I take the equations for L and D related to a Flettner rotor. It was, per unit height:
L = rho.RW.pi^2.d^2.s
D = 1/2.rho.RW^2.d.Cd
where rho is the density of air, RW the relative wind speed (m/s), d is the diameter of the cylinder (m), s the number of revolutions per second and Cd the drag coefficient, ranging around 0.8 for a cylinder. The result is in newtons.

A few notes:
- L is proportional to the relative wind and the rotation speed. Drag does not depend on the rotation of the cylinder and is proportional to the square of the relative wind .
- L is perpendicular to the relative wind, taking into account the movement of the cylinder, D is collinear to RW.
- The number of revolutions per second, s, must inevitably be limited by the speed at the cylinder periphery (m/s) and by the consumption of the engine responsible for rotating the cylinder .

The resulting L + D vector must be projected onto the axis of the boat, it can be positive or negative according to the angle of the relative wind.

I did not find interesting relationship on the speed of rotation of the cylinder, s (t/sec ): how to choose s according to the speed and the angle of the relative wind? rule of thumb?

Except miscalculation on my part, here's what I found for the projection on the axis of the boat of the resulting L + D. The relative wind is varying from 0 to pi, its speed is 11.11 m/s (40 km/h), rotor diameter is 2m, high is 5m and s equals one revolution per second. The x-axis is scaled in radians, the y-axis in newtons.

The force is negative headwind and finds its maximum crosswind, which is reassuring! and it is low but positive downwind.

Now my questions concern the choice of s.

Pierre-Yves

 

I have read on a paper about Flettner rotors:

<<When the wind strikes a vertical cylinder rotating about its axis, it tends to move laterally this cylinder, with a force which varies with the speed of rotation and which is maximum when the surface speed of the cylinder is equal to 2 pi (6,28) the speed of the wind.>>

From the equations, I have found that the force varies with the speed of rotation, but I have not found any maximum...

Where can I find a relation between the cylinder rotational speed and the relative wind speed?

Pierre-Yves

 

Did you look at the formulae in the spreadsheet linked to earlier, the one I suggested you look at and that was posted some way back in this thread?

That has details of the relationships between relative wind speed, rotor rpm, lift and drag and the results from it tally well with the two sets of wind tunnel data from tests on rotating cylinders.

 

Jeremy, I have looked at the formulae you have suggested and I can compute L and D. But I don't find this "maximum when the surface speed of the cylinder is equal to 2 pi (6,28) the speed of the wind". Is it thrue? If yes, how to find the equation?

Pierre-Yves

 

It shows the Lift and Drag falling off dramatically at 700 rpm for a rotor of .3m
As you increase the diameter, the lift peak falls back at 200 rpm.

It looks like operational considerations, like preferred motive power, convenience of thinner sections, bearing and vibration considerations will play a big part in any build

This graph is only lacking the vectors for wind angle to the boat instead of the optimum direction

I tried to create the table of vectors, but something went wrong.

Jeremy, or anyone with a younger more mathematical brain might be able to cure the bad figures.

 

There are bounds on the formula in the spreadsheet caused by the non-linear relationship between the various parameters (primarily the relationship between Cl, Cd and u/V, where Cl is the lift coefficient, Cd is the drag coefficient and u/V is the ratio of rotor peripheral velocity to relative wind velocity).

What I did was go through the wind tunnel test data and try to determine the mathematical relationship between Cl, Cd and u/V. When I plotted the data from the wind tunnel tests they seemed to conform fairly well to a 3rd order polynomial, as illustrated in the two plots in the spreadsheet that show the test data for Cl versus u/V and Cd versus u/V .

I then used those two polynomials to try and predict Cl and Cd over a range of values of u/V within the bounds of the experimental wind tunnel data (the spreadsheet does have hard limits that come into force when u/V goes outside a reasonable bound).

There is a strong possibility that Cl, in particular, may not follow the polynomial function I used for higher values of u/V, in fact I rather suspect that it may be more linear for values of u/V above about 3.5, but there isn't any test data that I have been able to find anywhere to indicate what Cl for a rotor may be when u/V gets above about 4.5.

 

Only by building a full scale unit and testing it. There is no wing-like formula applicable to these things. It isn't a wing or sail at all. It is a motor. It does not induce it's lift generating circulation as a result of geometry, but rather by creating an unbalanced separation bubble. That is difficult to predict when caused by skin rotation. And even if you can predict the separation and can determine that it is fairly stable, how do you convert that information into circulation? It does not scale with size or wind-speed the way a wing does. It is it's own thing and can't be engineered by analogy to a wing. Where is the aft stagnation point to be found for any given rotor at any given operational point and how stable is that point?

 

I'm afraid this is the right approach. However, German engineers of E-Ship One have certainly did the calculations!
P.Y.