Maritime Propeller Geometry

My name is Bernardo and I am a senior Naval Architect and Marine Engineer undergraduate student at the Federal University of Rio de Janeiro. I am currently developing a computer software that generates de 3D geometry of a maritime propellers from B series (wageningem) and Kaplan series. From this geometry I can export to a finite volume mesh comercial software called ANSYS ICEM-CFD and build a mesh fluid domain. Through that i can analyse the pressure and velocity field of the inflow passing the propeller and study thrust variations, increase in efficiency and cavitation areas of the blade using the comercial software ANSYS CFX 11 based on Computacional Fluid Dynamics (CFD).

I am attaching the software that i am developing so you guys can take a look in this new thread. It is still a work in progress and the visualization paintbox needs a lot of work. However the definition of the points representing the sections of the blade exported into the mesh software shows good accordance with the actual realistic geometry.

I would like some input on other conception of other propeller series that i can implement to the software as much as series as possible therefore creating a library of propellers into a simple software.

Does anybondy know any books containing that sort of information ? Or maybe web sites or articles ?

Thank for you time

Best Reggards,

Bernardo Xavier

Ps: The reference that i used to built the B series and Kaplan series is on the following book:

"The Wageningen Propeller Series" G.Kuipper, may 1992

 

Check threads by Rick Willoughby - he produces high efficiency propellors for his personal activities (hobbies?)

 

Bernardo
The visualisation is nice. The shade option does not work correctly for me but the others look good.

If you can set up a CFD model to analyse a simple prop you will have achieved a difficult task. You will have very complex 3D flow patterns to deal with.

If you can do this why not set up an optimising routine to generate the most efficient shape. Forget about standard designs other than to compare your model to reality and concentrate on producing a better prop and blade profile.

I am generating "optimum" props for low power applications using semi-analytical methods that gives reasonable results but it would be nice to see what a full CFD model could produce. I expect it will take lots of computing power.

You would probably find some useful insight into generating sections using NACA series foils. I like NACA 16-series for what I do and have some software to generate them but do not know the actual formula. I am sure you can find this information. 4-series are quite simple geometry.

Rick W

 

Reply to Rick

Rick,

The optimization ideia is very interesting, i have been getting very good results. Errors are in 1% to 10% depending on the rotation of the propeller when comparing the results from the numeric CFD model of the propeller to the experimental.

I am trying to invoke a DLL procedure and insert in to the mesh modeller software (ANSYS ICEM-CFD) so that i can built the whole mesh domain beguining with the global points of the blade imported from the software you´ve just looked. That way the domain mesh can be built with a single mouse click coz I already know the meshing parameters, the size of the domain and the correct turbulence model to be used to represent correclty the flow passing the propeller.

The mesh size grows about 2.5 million cells and it takes about 3~4 days to complete the simulation in an average PC (core 2 duo 3 MB RAM). To insert an optimization process into this analysis it would take quite a long time to complete, however the lab i work in has a cluster and the same simulation takes 2~3 hours to complete. As soon as i complete the DLL procedure construction, the process of building up the mesh will be very rapid and efficient. So thats one step closer to optimization.

Thanks for your help, im gonna try to look after those naca profiles you told me. I wrote and article about this simuulation but its in portuguese coz i wrote for a congress that took place here in Brazil. I am translating as we speak and i can send it to you when i finish it if you like.

Best Reggards,

Bernardo Xavier

 

Bernardo
I have attached an igs file of a little prop I had machined from aluminiun. The design conditions were boat speed of 3.3m/s, shaft at 360rpm and 35N thrust. The intention was to achieve high efficiency. If you have the opportunity I would be extremely interested in what your CFD analysis produces for this. It uses a modified 4-series section of around 8% thickness with slightly thicker section near the hub.

I already have methods for optimising but they are simplified for the low velocity ratio, low pressure coefficients and free stream conditions I work with.

I think the value with your CFD modelling would be to see how well you can simulate real world conditions for more heavily loaded props operating in wakes. Once you have the model proven then look to design optimum props for given conditions.

I also have 3d dxf if that is easier to work with.

Rick W

 

Rick,

Thanks for the iges file, as soon as i get to the lab this afternoon i´ll set up the mesh and run it this weekend. I believe i can get the results by monday coz there might be a queue in the cluster. As soon as i get the results I´ll contact you.

Best Reggards,

Bernardo

 

Bernado
You will find the prop is very unusual for a boat propeller. It has low loading so looks more like an aeroplane propeller. I have attached a photo so you are not surprised by the shape.

Rick W.

 

Rick,

What was the water specific density that you carried out your experiment ?

Bernardo

 

Bernando
The particular propeller pictured was used to set a water distance record under human power. This link shows photos of the boat.
http://www.adventuresofgreg.com/HPB/HPBmain.html

The attached photo shows the boat. Drag in calm water at 3.3m/s is 35N. The conditions are for fresh water at 10C.

Rick W

 

Rick,

Im sorry i took a while to respond, i prepared the simulation and left running during the weekend. To start with i generated a very poor mesh just to analyse domain independence, fortunately this first simulation i got a 31 N thrust. When i refine the mesh, all the pressure and velocity gradients will be captured with more precision and that will give more thrust, so im pretty excited to get something next to 35 N.

To refine the mesh im gonna need to make some adjustments in the geometry you sent me coz there were some holes when i opened in the mesh generator program. With big mesh sizes this holes are covered by the mesh, however they will be too big for small elements. I will take some time in adjusting the geometry. Im also quite occupied with my final thesis since im graduating in december so i may take some time to do this simulation remeshed. But as soon as I have some time i´ll jump right on.

I have some images of the simulation, but their in the lab, i´ll send them too you very soon.

Best Reggards,

Bernardo Xavier

 

Rick,

This boat looks very interesting, maybe we can investigate a hull propeller interaction with CFD.

best reggards,

Bernardo Xavier

 

Bernardo
I appreciate your efforts and understand your priority to get your thesis presented. It looks like you have done some good work.

The preliminary result for my little prop seems very good. I would be interested in any pictorial presentation from the CFD flow or pressure data if the software has that capability. Also interested to know efficiency it is indicating.

We try to avoid any prop/hull interactions but with no great science just a very clean hull design and prop mounted away from the hull.

Rick W.

 

Excuse me,
My name is Erick febriyanto, I am currently working my final project about Controllable Pitch Propeller.
Is there maritime propeller geometry to C series? and did anyone have reference how to draw propeller c series ?
Thank you very much.

Best regards