Hello Otherpowers Fieldliner's !
I'm sharing this because I'd like feedback on my validation process and because like most of you. We feel the wind passing, and we know that more ways to convert some of that energy to useful Watts, would be a good thing. I'm not against the current designs at all. I see though that many of the windiest places in the world are not accessible to modern turbines due to turbulence and wind shift (most of the worlds mountainous regions). As well I think these systems could be ideal for kinetic hydro, though that's a very different market.
I think I'm on to something that hasn't been well exploited in the past. Designing for high Reynolds number airfoils is a very well understood and developed process, but at low Reynolds numbers, where viscosity effects dominate flow, much less work has been done and less is understood.
I'm not sure that this is the right place in the forums for this discussion, and if a moderator wishes to move it, I'm fine with that.
One of the Moderators contacted me to let me know that if I supply them with some links, they can attach them to the post. I'll put together a list and hopefuly they will see fit to attach it here.
Because of the strong background I see here in Aerodynamics, I'll share my thinking about how these types of turbines work. I'm clearly standing on the shoulders of much more educated people than I and I'm not sure that I've got this all right.
The design I've focused on originates with Captian Savonius, who for good context, as I understand was trying to create a self powering Flettner rotor, and realized that it had potential far beyond that, but wasn't able to do good empirical work in assessing his prototypes.
The Blackwell report from the 70's is a mess, as far as I know no one has come close to reproducing it's work in the real world, and Ian Ross, with his paper “Wind tunnel blockage correction factors in high solidity turbines†around 2010 helps us to see why the data around VAWT's has been all over the place in wind tunnel testing. VAWT's have regularly tested very well in wind tunnels and then failed to perform in the real world, especially small VAWT's. Not clearly understanding the extremes effects of wind tunnel blockage correction factors, means that researchers from Blackwell to today have had a hard time understanding the discrepancy between tunnel and real world results. Another issue that was screened by wind tunnel testing is the very strong effects of even minor turbulence on small modern HAWT's and VAWT's with narrow blades.
One upside of Blackwells work though is that his AB testing showed that a 3 bucket system would have higher torque potential, but spin at a lower speed and have less total Watts collected.
Interestingly, in looking for technology that functions at low Reynolds numbers, Tesla's turbine is a boundary layer turbine that operates at very low Reynolds numbers. As far as I've been able to determine, Tesla's turbine has been tested to a maximum efficiency of around 50%, and the physics of scaling them up is poor. They are not a miracle device, but a device that shows another path from kinetic energy to shaft power. I now think about my turbine as a sort of 3D expanded version of a single side of a Tesla disk, operating as a boundary layer turbine, on the outside of the bucket.
The idea is that when a leading edge is pointing toward the wind, and the TSR is appropriately managed to between .75 and 1 (theoretical) then the wind adhears to the leading edge, with the difference in velocity causing torque on the leading edge. As the air (fluid) passes along that surface, it gets close to the center of the turbine, so the velocity of the surface is lower, and again this mismatch results in a pull, of energy being transferred to the turbine. As it reaches the center, it's velocity is quite low, and by that point the turbine has revolved 180 deg, and so the energy reduced wind is swept away by energetic new flow.
I did not originate this thinking or these idea's, (that a Savonius is not simply a drag based device), but encountered it in papers by Benesh, Modi, and Fernando, and probably others. As I'm not an educated person and learning this from my own research, I'm not great at keeping track of who to give credit too. The Benesh, Modi and Fernando team (at UBC?) produced a Savonius type in which the buckets were flattened, and this produced a turbine with much more power compared to the conventional Savonius as measured in their wind tunnel tests, where they unfortunately also were not fully aware of the blockage effects on the turbines. This meant that while they measured a substantive improvement vs the Savonius gemometry developed in the Blackwell report, they still were not able to reproduce these results in the real world.
It's true that companies like Windside, and Gus, have claimed Cp's of 0.21, real world results seem to be between Cp 0.05 - 0.12 which aligns with the wind tunnel work done in 2010 by Ian Ross. If anyone has results of wind tunnel studies where the blockage was below 5%, I'd appreciate a link or copy, or even better, real world results.
So this thinking brought me to the 2007 patent. At the time of filing that patent, I was getting between 10-15% efficiency, a bit better than a classic Savonius, but my device was much more labour intensive to make. I'd spent a lot of time, money and energy, getting a data logging system and sensors, and learning how to use them, and do the math to understand what it all meant. I can relate to the many experimenters trying to use bathroom scales and timing on wrist watches. I'd been building different versions of the same helical structure, when a friend asked me if I'd ever tried twisting the foil the other direction. Up to this time I'd only ever build them with a twist that resulted in the concave portion of the foil on the underside, and convex portion on top. I'd noticed in video of me allowing smoke from upwind flares to pass through the turbine, that smoke that entered the turbine at about 1/2 to 1/3 height, exited well below that, being displaced downward. I'd also noticed that as wind speed picked up, the turbines Cp went down substantially. This made me conclude that the turbine was pumping air downward, and that at some point, being as close to the ground as it was, it might create a pressure bubble that could divert air around the turbine. I created a new version, with the concave side up, in the twisted foil, and while it didn't self start reliably, it reached between 18-24%. It did start better once I added a flat base, but still would in some orientations, not start up till the wind shifted. I created the next version with a bit more twist, and larger height to width, and it self started reliably, and tested to above 24%. While it did use the cross sectional geometry of the 2007 patent, the idea of which way to twist the foil was not addressed in the patent, though the patent clearly taught only one orientation, some patent lawyers have argued that the patent applies to the new design, but not all agree. This geometry is downloadable from my thingiverse account.
After that it became a business and I wasn't empowered to continue to improve the design. I wasn't a very sophisticated business person and I made quite a few misjudgments, in the end allowing the company to be taken over by a very wealthy 90 year old who doesn't really believe in global warming, or see a need for alternative energy, other than to make him more money. He brought in other investors, I was pushed out, and then he and the others couldn't agree on what to do when their engineering contractor did an inadequate job of drawing up plans.
After that I stayed on with the build contractor for a while, but eventually that ran out of money and they stalled.
Of course during all this time, I continued to think about the design, and how it could be improved. I'd followed the work of Rahai, (at CalTech?) where in experimenting with slots in Savonius variants in coming up with the 2007 patent. He'd come up with extraordinary results, in a wind tunnel with a significant blockage factor. He wasn't able to reproduce his wind tunnel results in the real world. What I took from his work was a few things, that a bump near the center of rotation didn't hurt it's performance, that the height, or camber of the foil was very low, and that joined at the trailing edges (no overlap) was better than the 10% or so overlap which worked best in the versions tested by Blackwell.
After looking at many low Reynolds number airfoils, I concluded that the “bump†may not have been as important as I had thought. I also realized that while my turbine worked well, maybe better than any preceding VAWT of that size, it was likely that I got only a small part of it's total swept area “rightâ€, and that part must be dragging the rest around. I moved ahead with this thinking, and produced a curvy new turbine shape, but also realized that it was going to be just as difficult as my previous design to build. In factory production this wouldn't be a big deal, but for R&D with no budget, it was a stumbling block.
At the same time I'd been pushing forward a joined wing aircraft design I've been tossing around for many years. In doing this and the research around it, I came upon the Faucetmobile, a design inspired by Barnaby Wainfan's research on lifting bodies, and flying wings with very low Reynolds numbers. The Faucetmobile is a very startling boxy design, with no curved surfaces, and massive frontal area, yet flys more effieciently than most of the Cessna variations. Barnaby has been willing to discuss it on the homebuiltaircraft forum, and he reports that flow remained attached as it transfers over the hard chines in the aircraft skin, because of the very low Reynolds numbers forcing flow to obey viscously, rather than intertially.
With this in mind, I looked for ways to fauceting my new turbine design, especially ways that would allow it to be easily supported by an internal structure, and potentially skinned in fabric.
To get us right up to date, I've now built my first model of it, as you've seen in the pics and on YouTube. I'm just hooking up the instrumentation and getting ready to test. I'm short on cash, and there's a few things I'm afraid may break down, including my very old truck, which I'm using as a test platform. I'm also concerned that I've made the diameter too small to keep the Reynolds numbers low enough to make it insensitive to fauceting.
More will be known soon.
Thanks for all the help I've gotten here, both as direct answers to questions, and even more so as a repository of good questions and answers. I'll also apologize in advance, for not nessesarily being able to respond to questions in a timely way. I've got far more on my plate than I can comfortably handle, and I'm short on money, so I may not be able to respond as quickly as I'd like, as I may be caught up in other things.
