Fieldlines.com || Any CFD imagery of VAWTS in action?

Front Page - [Homebrewed Electricity-- (wind) (solar) (hydro) (steam) (controls) (storage) (mechanical)] - Classifieds - Site News
Everything - Newbies - [Remote Living-- (housing) (heat) (light) (water)] - Reviews - Diaries - Our Products

Any CFD imagery of VAWTS in action?
By Blue Tiger, Section Wind Posted on Mon Mar 20, 2006 at 10:23:35 AM MST
I need help in visualising wind flow around a spinning Savonius

Can anyone point me to a web site or text that describes the aerodynamic flow around the basic Savonius blade while it is motion? I'm thinking computation fluid dynamics imagery here - my maths is terrible but I'm good with the visual side of things. What is happening to the vortices on the trailing edge of the blade as it spins through 360 while the wind shoots through? And what would be the optimal 'face' shape for the front of the blade that leads into the wind? I've found bits and pieces discussing the familiar semicircular shape, but its use seems to be more out of constructional convenience than good design.
I'm looking to power an ancient Southern Cross reciprocal water pump (about 0.6 litres per cycle up to a head of 30 metres) for my farm in New South Wales (Australia) and found inspiration from all you do-it-yourselfers on this board. I know many of you hate the VAWTS, but they are comparatively easy and cheap to build. I could just put up the usual 44 gallon drum thing, but I feel the need to tinker. Wind load is very sporadic, but this doesn't matter - I just need the unit to tick over in the background topping up a holding tank.
Any suggestions? or do I need to download the open software CFD package at http://www.opencfd.co.uk/index.html
Blue Tiger

Any CFD imagery of VAWTS in action? | 19 comments (19 topical)

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#1)
by rotornuts on Mon Mar 20, 2006 at 07:30:03 PM MST

http://www.energy.ca.gov/2005publications/CEC-500-2005-084/CEC-500-2005-084.PDF

An alternate design but it helps you understand some of the issues and it has some CFD images.

Mike

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#2)
by Blue Tiger on Tue Mar 21, 2006 at 01:03:06 AM MST

Nice one Mike.

I'd come across the Benesh configuration mentioned in the paper while wandering through the VAWT patents. The cardboard model I made based on Benesh's instructions worked, and I was thinking of scaling up. The images in the paper you suggest help heaps - many thanks.

Anyone else with further suggestions on the CFD/image side of things?

Blue Tiger

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#8)
by Blue Tiger on Wed Mar 22, 2006 at 03:15:31 PM MST

Mike

I emailed Dr Rahai who is named as the Report's author. I asked if the exact geometry of the blade configuration was available in the public domain. He replied that it was not, and would only become so once the patent application had run its course. He hoped to have that done by your "summer".

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#3)
by tecker on Tue Mar 21, 2006 at 01:23:57 AM MST

What I did to help me understand the forces and flow is to place that rotor I was testing in water .At some point the rotor catches up with the positive directional pressure and at that time the turbulence is evident .Also in the area behind the working cycle. working on compressing down and in and using the upward and outward expand as the wind rises and falls back.

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#4)
by Blue Tiger on Tue Mar 21, 2006 at 03:15:16 PM MST

Nice idea Tecker, but is it legit? Yes, air behaves 'like' a fluid, but given that air is compressible and water isn't, it suggests to me that the behaviours seen in the water tank might not be mirrored by the high and low pressure zones forming and dispersing in the air.

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#5)
by dinges on Tue Mar 21, 2006 at 03:20:49 PM MST

From aerodynamics class, I remember that it is usually assumed that air behaves like liquid, i.e. is incompressible, like liquid. The reason for it was.... that it made the equations much simpler <evil grin>

Peter.

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#6)
by dinges on Tue Mar 21, 2006 at 03:22:20 PM MST

Seriously. The compressibility of air usually only becomes an issue at pressures higher than those normally seen in aerodynamics, so I'm told.

Peter.

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#7)
by tecker on Tue Mar 21, 2006 at 04:44:00 PM MST

I know the density is much diferent but you can see how the medium is displaced
around the surfaces when starting from dead stop fully submurged this would be simular to what's going on on the side away from the headwind. It plows through like a truck on the highway a cyclonic draft I think is possible.

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#9)
by Blue Tiger on Wed Mar 22, 2006 at 03:54:07 PM MST

OK tecker, I can see the merit in observing the 'start up' within a liquid. But in thinking things through I'm wondering how it is possible to observe much when the rotation is in full motion. Surely, once the rotor has turned full cycle once or twice, the turbulance will hide any 'markers' that might have been visible in the water.

I've emailed Dr Rahai - the author named in the California Energy paper that Mike mentioned above - and asked him whether their CFD analysis was on a static blade in a laminar flow, or on a rotating blade in turbulent flow. Seems to me that if the optimised design is based on a static blade then it might not be that optimised.

With two rotating leading edges chopping into the on-coming wind, then it would seem to me that the packets of air in and around the blades would be quite turbulent. Presumably any optimised design would be one that presented the greatest surface on the windward side for 'pushing' while at the same time provided the least resistance on the lee side of the blade. The optimised design would also need to shed the turbulance from its core.

Now if I could only just implant a CFD chip in my head ...

Blue Tiger

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#10)
by tecker on Wed Mar 22, 2006 at 08:55:58 PM MST

I hear ya

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#11)
by rotornuts on Wed Mar 22, 2006 at 10:04:20 PM MST

I'd like to hear the response if you get one. I haven't been able to place much weight on the flow visualizations done on vawt's because I believe they are static. Need to see a full visualization done at speed including global flow around the swept area. I believe the pictur would be much different.

Mike

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#13)
by finnsawyer on Tue Mar 28, 2006 at 10:00:15 AM MST

Supposedly air can be considered to act like a incompressible fluid up to speeds of 200 miles an hour. My own take on it is that since air is actually a compressible fluid and that since compression waves in air can be detected as sound, if the unit is running silently then you can assume the air is acting like a incompressible fluid. Of course, we don't hear all frequencies of sound. Don't you just hate these caveats?

Water has its own peculiarities. Propellers can create cavitation, which appear to be areas of partial vacuum. Again, though, that's not the same as a model wind mill operating in water. The bottom line is that the water test is probably valid as long it accurately reproduces a region of silent behavior of the VAWT or HAWT.
GeoM
[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#14)
by Blue Tiger on Thu Mar 30, 2006 at 03:27:00 PM MST

Finsawyer

I'm starting to think that using a water-based test bed is fine for basic interpretation about vortex behaviour. However, from my continued readings, it seems the only reliable thing to do is actually build and test our designs under operational conditions. While I like the idea of 'desk-topping' different designs, I'm yet to be convinced about the accuracy of the CFD models. Very few public papers discuss CFD VAWT behaviour, and those that do often fail to explain the operational assumptions made. By example, I've yet to find a paper that illustrates how the blade's efficiency varies depending on velocity through the VAWT core.

I happened across the TMA site the other day. They have a paper with some discussion about CFD analysis, but again I'm not sure if I believe the nice 'pressure' diagrams or not.
www.tmawind.com/papers/ASME%20AIAA%202004.pdf

Perhaps WindStuffEd shows the way. If you have an idea, build it, test it, upgrade it, and, hopefully finesse something that will help wean the world off hydrocarbons ...

Blue Tiger

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#16)
by finnsawyer on Fri Mar 31, 2006 at 08:31:50 AM MST

You mention the velocity through the core. How about the effect of the central (cylindrical) obstruction? This will modify the air flow through the VAWT, speeding it up along the sides of the central post. If the central post extends in radius out to the blades then the blades may see a faster air flow. Also, their presence may modify the flow behind the post reducing turbulence and improving effectiveness. When a fluid stream encounters a cylinder, it is forced to come to a stop at the point dead on. This results in an increase in pressure at that point, which in turn causes the fluid stream to split and speed up as it moves around the cylinder. Faster air has more power. The question is whether this increase in power can be captured by the VAWT. If it can this may mean the VAWT can violate Betz's Law as it applies to the incident air stream. (Betz's Law still applies to each blade all the time as it interacts with the air flowing over it). The effect should appear in water as well as air.
GeoM
[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#17)
by Blue Tiger on Sun Apr 02, 2006 at 03:57:30 PM MST

Finsawyer

I assume that the influence of a central cylindrical obstruction will be case-specific. It is clear from reading several papers that the various elements of 'performance' that people have measured from their respective VAWTs vary greatly depending on a dozen or so physical factors. While some of those factors are fixed, such as the gap between the two blades, or the area obstructed by the internal vertical axle, others such as wind speed are constantly changing. Therefore a machine's performance is also changing through time.

Part of our problem is that VAWT testing results haven't been compiled and presented in a consistent and accessible way. Ideally a centralised repository of test data, lodged under a standardised protocol would mean that a) collectively we don't keep on testing the same old thing, and b) we could induce some rigour into the actual tests which would then mean intercomparison of tests done by different folk would be more reliable.

Now, if only every guy on this board had access to a wind tunnel and testing gear, we could engage in 'distributed' research ... with each researcher doing a segment of the larger research picture ... it would be a bit like what the software guys do with the open systems software development (see SourceForge) for public interest rather than private gain.

As for the influence of the axle, there is specific discussion in Menet's paper at http://www.2004ewec.info/files/23_1400_jeanlucmenet_01.pdf

While you're at it, check out wwwsoc.nii.ac.jp/jscfd/cfds14/pdf/c06-2.pdf which has interesting cfd images of savonius and benesh type vawts. My adobe can't pick up the text, but the images should be visible. (Rotornuts, if you're reading this post, check it out).

Also, www.2004ewec.info/files/23_1400_michaelpercival_01.pdf is an interesting read for those of you interested in home-made savonius rotors. They used it to generate electricity, which seems a little futile given the rotational speeds. Nevertheless the field testing provided some good data that appears to correlate with that from the Sandia labs.

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#12)
by IntegEner on Tue Mar 28, 2006 at 08:39:25 AM MST

CFD, of course, means "computational fluid dynamics". Another growing subset of fluid flow study is PIV, or "particle image velocimetry", for serious applications. Notre Dame (the same ND of the "Fighting Irish") has hosted International Symposia on PIV in the past and getting into such nano-scale investigations, though regrettably delaying a look at operational wind turbines, has been the focus of many papers and proposals.

Anthony C.
www.integener.com

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#15)
by Blue Tiger on Thu Mar 30, 2006 at 03:29:09 PM MST

Anthony

I've not heard of "PIV". I'll check it out on the net - thanks for the lead

Blue Tiger

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#18)
by Blue Tiger on Sun Apr 02, 2006 at 04:01:48 PM MST

Anthony

Thanks for highlighting PIV - I hadn't heard of it until you raised it on this post. The net shows many examples. Some of the imaging is fantastic. Such a shame that the equipment is a little beyond the average joe. There again, I would expect many of the guys building rotors could build half-way decent PIVs.

Blue Tiger

[ Parent ]

Re: Any CFD imagery of VAWTS in action? (3.00 / 0) (#19)
by john j on Mon Apr 28, 2008 at 10:02:39 AM MST

Sure would like to see/hear about your findings.....

Cheers, john j

Any CFD imagery of VAWTS in action? | 19 comments (19 topical)

Menu

� create account
� How to use the board
� FAQs
� search the board
� Google search the board

Login

Make a new account

Username:

Password:

Total Views

��133 Scoop users have viewed this posting.

Related Links

� http://www .opencfd.co.uk/index.html
� Also by Blue Tiger

You must be a registered user to post here. It's easy and free, and the link is on the upper right side of your page.
All trademarks and copyrights on this page are owned by their respective companies. Postings are owned by the poster, but may be deleted or moved at the ADMIN's sole discretion. The Rest � 2009 Forcefield.
You can Email the board ADMIN here. PLEASE include the username you signed up with!