Author Topic: savonius - optimised (Read 7115 times)

chris psmith · « **on:** November 12, 2007, 07:35:45 PM »

Looking to put together a savonius to patent

http://www.google.com/patents?id=GUYrAAAAEBAJ&dq=4,838,757

I was wondering if anyone had done this before me that I could shre experience with?

resectioned. A little digging might have kicked out a few similar projects here lately.

chris psmith · « **Reply #1 on:** November 12, 2007, 02:00:51 PM »

have done lots of digging before posting and the closest i could find was a flat bladed savonius, nothing on this optimised design.

vawtman · « **Reply #2 on:** November 12, 2007, 03:14:38 PM »

"Looking to put together a sav to patent"

But the link leads to a patent,Confused in wisconsin

TheCasualTraveler · « **Reply #3 on:** November 12, 2007, 04:15:58 PM »

http://www.homesopen.com/wind/savonius.pdf

feral air · « **Reply #4 on:** November 12, 2007, 07:27:13 PM »

I doubt that there's anything you can add to a Sav. (or change) that would be worth patenting. The only addition that almost makes sense (a shroud [/wind directing device]) has been patented several times over in various forms.

You should really build a regular Sav. first (no shaft!). Matching the alternator to the rotor is the hard part (or so it seems) but new materials and methods for constructing the rotor could be handy too.

You could check out my "quick vawt" Diary entries for my take on a simple rotor design. I can't build for squat so the bar is really low there. I never finished it off with an alternator or anything but the pair of vawts that I made like that spin like crazy.

Adding a shroud or anything like that defeats the purpose, imo. Savonius rotors are supposed to be simple and easy to make. I dunno, take it easy

disaray1 · « **Reply #5 on:** November 13, 2007, 05:58:56 AM »

I think what Chris is looking for is the blade profile for Benesh or similar. He's looking at the patent, and want's to build according to that design. Yes? Chris, google monte350 and see what you get...

David

warpsta · « **Reply #6 on:** November 13, 2007, 12:58:18 PM »

I am in the process of building a Benesh profile Savonius using Picoturbine plans, it slow going with my other committments though (check out my diary to see what's going on)

Looking at that design it is similar to a Benesh, however it looks a good bit harder to construct due to the curve of the blades. What material were you thinking of using for the blades? How are you going to make the end pieces and attach the blade material to them?

I am using wire mesh and fibreglass, great fun but messy (I float out of my 'shop sometimes so watch the fumes ) I found it quite difficult to maintain the shape of the blade across the whole area. But I am getting it close enough for my first attempt.

Someone suggested that I use fibreglass sheets, on reflection this is a very good idea, but I wonder how much strength you would be able to achieve. My mill needs to stand up to some serious winds (>90mph last week) so I like my method, it seems very robust, time will tell though.

Something to consider is the fact that VAWT like to have a high aspect ratio (6:1 or 8:1) in other words tall and skinny, not short and fat. It would be better if you constructed more than one blade set, two would be the simplest (offset by 90 degs.) although I have seen 3 set VAWT's. You will be encountering the law of diminishing returns eventually though. My advice would be to keep it as simple as possible.

You would expect the mill to turn quite slowly, what are your thoughts on the PMA design? Big diameter disks with lots of magnets (high frequency) and coils with lots of turns (high voltage) would help you get the best from your mill, but that would depend on your budget.

I hope this helps.

feral air · « **Reply #7 on:** November 13, 2007, 09:00:21 PM »

"Someone suggested that I use fibreglass sheets, on reflection this is a very good idea, but I wonder how much strength you would be able to achieve."

The weakest link is probably in how you attach the sheets to the end plates. I would probably use steel or aluminum end plates with a good 3/4 inch lip and put the sheets on the inside, secured with VHB tape to the lip....in an ideal world.

"Something to consider is the fact that VAWT like to have a high aspect ratio (6:1 or 8:1) in other words tall and skinny, not short and fat."

Whoa, wrong, sorry...

High aspect ratio vawts are good for high wind regions and low aspect ratio vawts are good for low wind regions.

Unfortunately, I don't know of any good way to determine what's most efficient for a given site. That said, a best-guess should get you close enough. If 20mph winds are rare then a low(er) aspect ratio is probably better and if 20mph wind is pretty typical then go with a high(er) aspect ratio...and then there's the middle. That's my take on it, anyway. take it easy

chris psmith · « **Reply #8 on:** November 15, 2007, 04:27:47 PM »

Sorry all, been away on a business trip, not intending to patent anything but build to the patent listed. Made a savonius with two semi-circles, slow as predicted but interesting anyway.

Made a model to the patent and was surprised at the speed it would turn at (some photos loaded on my account).

Had a play with a shroud on the trailing edge that seemed to give significant speed increase and nocked up some plans of how I think i will proceed.

Regards

Chris

chris psmith · « **Reply #9 on:** November 15, 2007, 11:41:09 PM »

Bit more on plans so far:

Benesh turbine to patent 4838757.

Commercial generator - 500W, charge start 230rpm, max power 380rpm.

End plates, and center plate (to support blades) made of perspex (thoughts?)

Blades made of flat sheet (some sort of strengthened plastic sheet with ribs running top to bottom - still looking at what is around.

Blades shaped by feeding through top / center / bottom disc with blade profile cut through the discs - glue in place / support discs with threaded bar.

Cowl over trailing edge with wind positioning.

Rotor diameter 630mm (600mm blades) / Rotor height 750mm

Well that is where the planning has got o so far, welcome your thoughts.

Warpsta - you said the patent looked more complicated than the benesh plans, but it is a benesh patent?

Rgds

Chris

warpsta · « **Reply #10 on:** November 16, 2007, 02:40:36 AM »

The Benesh patent that I am using has a curved, aerofoil leading edge but it then continues straight towards and beyond the centre, the link to the plans is in my diary.

Looking at your dimensions you will have a swept area of 0.225 square meters, somehow I don't think that you will be able to extract 500W with this amount. The rpm's that you have quoted seem to be quite high for a VAWT, at what wind speeds do you expect to achieve these rotational speeds?

You can pretty much make the end plates out of anything you want, the lighter and stronger the better, I like the idea of cutting slots through the plates to form the blades then glueing them. You could use the ends of the threaded bar to mount the balancing weights.

The reason that I chose my dimensions was because that was the length of the wire mesh, you may want to source your blade material before you decide on dimensions so that you can reduce waste and maximise swept area.

chris psmith · « **Reply #11 on:** November 16, 2007, 09:11:37 AM »

Point taken on the size, I have no idea yet about the size needed to get the power but need to try and compromise on total size for various reasons.

I beleive that using a 1/4 shroud it will improve performance and can also be used to act as a brake if needed, but need more time to experiment.

On the subject of balancing, i was planning to have a play with a self balancing device, basically a loop of pipe round the outside diameter of the top pipe containing some ball bearings that move to balance out the rotor, while since i have seen the design so will need to do some surfing.

Regards Chris

chris psmith · « **Reply #12 on:** November 16, 2007, 12:26:07 PM »

Found the info on the balancing system, patent 6267450 or demo on link:

http://www.centramatic.com/Demo/demo.php

feral air · « **Reply #13 on:** November 16, 2007, 12:52:31 PM »

"Made a savonius with two semi-circles, slow as predicted but interesting anyway."

Do you have more details and a pic or two you could share?

It's pretty easy to increase the rpm of a simple Sav; Divide the cylinder diameter by 2, the result is the amount you can remove from the cylinder before you cut it in half. For example, with a 4 inch diameter cylinder...

Since I figured this out I see almost no reason to go all the way to a Benesh profile. If you're after absolute efficiency for a given size then ok, but otherwise, no....not when there's a gimme-more-rpms mod that is that simple. take it easy

chris psmith · « **Reply #14 on:** November 16, 2007, 12:59:13 PM »

Loaded a photo of my chocky tin attempt which you should be able to see on my details. I spent a while looking for large enough tube to use for savonius but i think it is actually easier to use flat sheet inserted through the blade profile cut into plates, well for me anyway, best tube i could find in any lengthe was about 8" diameter.

feral air · « **Reply #15 on:** November 16, 2007, 07:41:53 PM »

Yeah, the larger diameter culverts are pretty expensive and aren't as easy to work with. Using an 8" dia. pipe wouldn't be too bad if you could make it fairly tall.

Your tin sav would probably work better without the top and bottom plates since weight at that size is a real killer. On your benesh, is there a reason you didn't rotate one of the blade-sets 90°?

You might want to check these out...one and two. That's my so-called "quick vawt" made from 6" pipe. It hits "blur" speed in about 6mph wind but it's just yard art, I'm not trying to get any power out of it since it's so small.

Anyway, it looks like you've got a good start but I think you'll end up needing to go way bigger. At that size there's tons of fun but not much power unfortunately. take it easy

chris psmith · « **Reply #16 on:** November 17, 2007, 02:35:43 AM »

Good point on the chocky tin, might have another play.

Planning to keep bldes not inverted as the middle section is in place to keep the form of the blades, they pass straight through it, also planning to put a cover over trailing edge which wouldn't work if i split them.

No idea how to work out the size needed, can't realy go over 1 meter tall for simple reason that threaded bar is in meter lengths, i am off for a look round the hardware store today so might get a better idea of what size it will be.

Also, if i get time, i will have a play with the model to try and see what % speed improvement the cover gives to make sure it is worth the hassle.

Yard art looks like fun, might nock up a more permenant play one for hanging.

Rgds Chris

warpsta · « **Reply #17 on:** November 17, 2007, 04:47:17 AM »

I think that you will lose a lot of wind if you don't rotate one set of blades by 90 degrees, this will make starting a lot more difficult. You could use one bit of threaded bar for the bottom section and two bits for the top, offset from the middle by the same distance.

Does the cowl rotate?

chris psmith · « **Reply #18 on:** November 17, 2007, 10:25:30 AM »

Made a larger benesh this afternoon, 430mm diameter and about 400mm heigh. Proved the ease of making the blades by feeding flat sheet through the discs, took less time to build than the savonius.

Not had time to do much playing but seems very free to spin in light wind.

The cowel will rotate os the 1/4 is always covering the trailing edge to reduce drag, should be able to use my larger proto to see if its worth doing.

Photos uploaded as big benesh 1 to 3 on my files.

feral air · « **Reply #19 on:** November 17, 2007, 10:27:21 AM »

I don't think you lose much (usable) wind but it's a good idea to give 'em a 90° offset anyway. I like to equate it to piston engines...

You've got one big piston so there's an obvious (at least in low winds) chug-chug action. If it was two smaller pistons the chug-chug is a little smoother. The problem is that your one big piston could get stuck at TDC (top dead center=not catching wind)...with two pistons, when one blade-set is TDC the other is facing the wind to push it through the cycle.

That's a bit of a stretch analogy-wise but hopefully it makes sense.

From looking at the pics it looks like the cowl rotates to me. take it easy

feral air · « **Reply #20 on:** November 17, 2007, 10:42:10 AM »

And then of course I just realized that I probably screwed the analogy because tdc is the wrong position. Too much blood in my coffee stream...it's too early yet. Ignore that part and it makes sense...

The problem is that the rotor can rotate just enough that it's not catching wind anymore if the wind wasn't strong enough to get it around 180°. With a blade-set at 90° this is less of a problem. Or something...sorry, need more coffee. take it easy

TheCasualTraveler · « **Reply #21 on:** November 18, 2007, 08:13:49 AM »

Working on a Savonius myself. I read somewhere that funneling more air to them would make them work better so I am using the North facing wall on the second story of my house which commonly sees NNE to NE winds and funnels them around this corner of my house- hope this video shows up...

http://s241.photobucket.com/albums/ff74/compoast/?action=view&current=Savonius3.flv

chris psmith · « **Reply #22 on:** November 18, 2007, 09:44:10 AM »

looks good, what size is it and what sort of rpm?

TheCasualTraveler · « **Reply #23 on:** November 18, 2007, 11:50:53 AM »

Thanks Chris,

The tubes are 12 inches wide, the top and bottom are 24 inch discs and the hight is 60 inches (5 foot). The overlap in the center is about 3 inches. I read quite a bit on these and thought I would do one with 4 cups at 90 degrees from each other but I read several times that 2 cups was the optimum design and that the cups act as an airfoil when they are into the wind causing uplift that keeps it from stalling till the wind catches the next cup. It was a little stiff this morning since I just built it but once the breeze was steady at 5 mph it started to spin at about 50 to 70 rpm. That was using a watch and counting by eye. Since then it is loosening up and with the wind up to 7 or 8 mph it is doing (my estimate) about 150 and better. I had planned to use this one but I think it will just be my model for my final design. Using as a form the very light cardboard concrete forms from Home Depot or Lowes I want to lay one up in fiberglass and resin, same size. I also have some experimenting to do with placement. As I write this the wind is gusting and I'd est it is doing about 180 rpm in 10 mph wind.

chris psmith · « **Reply #24 on:** November 19, 2007, 01:55:53 PM »

Lost a few hours sleep after your comments on position and watching the video. Changed my mind on the design, now think I will wall mount in same way, will put two benesh rotors end on end with 90 degree offset as sugested. 3 pillow bearings (ends and middle), giving total length of 2 meters and about 450mm in diameter.

This gives about 0.5 sq meters in the wind.

I have open fields to one side of the house that get 90% of the wind, and that is the side of the house that is out of site so allows me to go bigger with no (or limited) complaints from my wife.

I beleive i will then need to gear it up to get the rpm i need (500w at about 300rpm).

Thoughts? Enough surface area to cope with about 1 in 3 gearing and turn a 500w generator in winds of about 10mph?

TheCasualTraveler · « **Reply #25 on:** November 19, 2007, 04:20:20 PM »

Whoa, before you mount it to the house...

Last night when the wind picked up to a steady 10 mph my turbine was screaming, I had to shut it down. The house was vibrating and thats without the load of an alternator on it. Now I knew from the start noise was going to be a problem and I intended to find a balance that would let me use the house with a minimum of noise and vibration. After last night I plan to scale down the size to only 3 or 4 feet tall and make it wider. That will lower the rpm's and increase torque. I need to really look at balancing and the bearings also. I intend to make it a solid fiberglass unit and instead of the half circles do that sort of ying yang, { 69 } design along with probably losing the big end pieces. I don't even know what I'll do for sure I keep getting different ideas but what I've learned so far is that I need to,

Keep the rpm's between 100 and 150. 200 tops. Do this by adjusting the size.

Make it a solid unit (fiberglass) so it has as little flexing as possible.

Balance it as close to perfect as possible.

Figure a really good top and bottom mount bearing.

Work on a hefty steel welded corner bracket with braces and maybe rubber bushings to dampen noise. (2x4's are fine for testing but won't cut it in the long run.)

All in all it's a work in progress, far from finished. I shot this video before the wind picked up. The wind was only about 5 or 6 MPH but you can see how the wall funnels the wind.

http://s241.photobucket.com/albums/ff74/compoast/?action=view&current=savonius4.flv

chris psmith · « **Reply #26 on:** November 19, 2007, 11:51:43 PM »

Andy,

Few thoughts, the 69 style is probably the benesh on the patent i listed, realy easy to do. The benesh is supposed to be smoother and quieter. reducing the free length of the blades with a disc in the middle will help keep everything rigid, looking at your video it looks like the blades have twisted.

Having two rotors with an offset of 90 degrees should smooth it out more as it is a more even mass.

Had a play with the tube of ball bearings idea mounted on the top of the rotor, it seems to work realy well and adapts for changes in the balance.

Using two rotors at 90 degrees end on gives the chance to stick another pillow bearing in the middle as well as the two ends so the whole thing should be quite rigid. Planning to screw it to a solid brick wall with steel mounts which should help as well.

Rubber mounting is a good idea that i will need to look into.

On the size, if you stick an alternator and some gearing on, will that not slow it down anyway? I am worried about reducing the size to a level where it just won't work.

I now have the basic frame worked out and priced up, looking at material, will have a look at fiberglass, with the benesh profile cut through 3 disks and a card blade pushed through it would probably be easy to coat it in fiberglass, i just fancy the thought of making it clear so was thinking perspex end plates and clear pocket typle plastic sheets, my worry is crack propogation so need to give this some real thought.

Thanks for sharing your progress.

Chris

TheCasualTraveler · « **Reply #27 on:** November 20, 2007, 09:46:39 AM »

Chris,

I don't know the name of the design I want to use. It is similar to the Benesh but with two teardrop shapes doing a 69. I see what you mean about the blades twisting but it is an optical illusion from the angle of the sun. I'd really like to see your photos of the bearings, that's an area where I'm clueless. As for the speed and gearing, I don't plan to do any gearing so as not to waste energy. The reason I have the audacity to think I can make power with a Savonius is twofold. First because I am using the house to funnel the air and second I plan to use a large rotor to compensate for fewer rpm's. If I figure right with an 8 inch rotor turning 100 rpm's the tip speed of the rotor is 3.5 ft per second. On a 24 inch rotor you get 10.5 ft per second (I'm rounding) so the large rotor is traveleing nearly 3 times the speed of the small rotor (at the outside edge) and unless I'm mistaken equivalent to a smaller rotor doing 3 times the rpm's. A 2 foot rotor on a HAWT would catch a lot of air and I suppose complicate the blade assembly but on the Savonius it lays flat to the wind under and out of the way of the blades. This no doubt is going to make for a lot of magnets and coils and probably a lot of wasted time and copper and hair pulling trying to figure out how to do the coils to prevent cogging, but hey, I'm having fun.

chris psmith · « **Reply #28 on:** November 20, 2007, 02:09:47 PM »

Andy,

Went to a bearing shop this evening, they wanted £16 for each 20mm pillow bearing, just got 3 from e-bay for a total cost of £18 delivered! Just do a search for pillow bearings on e-bay to see them, they are simple to use as they come with a mounting ready to fix to your wall mount and have a grub screw on the collar to make sure they grip the shaft. They are probably not the most ideal but in terms of simplicity and cost probably hard to beat.

I admire you for trying to do the electrics yourself, I decided early on that the fun for me would be in the rotor and not the electrics, I need to get about 300W from it to make it worthwhile so will be buying a generator.

Ordered the steel bar as well which i will be getting turned down to suit the bearings (25mm bar down to 20mm) which will make the end mounts for the rotors and screw directly to the threaded bar running through them.

If you have a look at the benesh patent I started this link with, he desribes an air guide to direct the air at the roto, seems to offer better efficiency and would be even simpler to do with the rotor on the side of the house.

Still no further on deciding what to make the rotors from yet.

Rgds Chris

TheCasualTraveler · « **Reply #29 on:** November 20, 2007, 04:05:11 PM »

<<Still no further on deciding what to make the rotors from yet.>>

Hmmm, looking at the patent drawing I'm inclined to think I would make the leading edge of the rotor cup out of a PVC pipe cut in thirds (using one third of the circle) It could be sanded to have a sharp leading edge. From the back edge of the pipe attach plexiglass cut and heated to form the right curve. Or, if your into building an oven and making a mold to shape the plexi, form both cups (or all 4 cups) out of .25 plexi. I thought of doing this, making an outdoor oven out of stacked concrete blocks and heating it with a plumbers propane torch. I think I read 350 degrees F is where plexiglass melts to where you can form it. I didn't trust my skills to get a nice form on the plexi so, thats why I think toward PVC for the hard curve...

Have you checked out,

www.windstuffnow.com

I believe he is the resident expert on VAWT's here.

feral air · « **Reply #30 on:** November 21, 2007, 02:12:52 PM »

It could be sanded to have a sharp leading edge

That's not really necessary since all the edges are (basically) trailing edges. Once it's de-burred you can stop sanding. ;-)

take it easy

chris psmith · « **Reply #31 on:** November 21, 2007, 11:43:01 PM »

Like the idea of the 1/3 tube, I will need to check the measurements to see if it is close to anything available, means i can use a stiffer matrrial for the rest of the blades.

Have thought again about going for steel, this would be quite a bit heavier but probably unbustable, the uncertainty i have is the pro's and con's of weight, if it is heavier it will need more force to get it going, but once moving it will keep a more constant speed if the wind is gusty, not sure if it would need significantly more wind to keep moving when it was at speed than a lighter version and i guess that would be the deciding factor, any thoughts?

feral air · « **Reply #32 on:** November 22, 2007, 12:37:48 PM »

not sure if it would need significantly more wind to keep moving when it was at speed than a lighter version and i guess that would be the deciding factor, any thoughts?

I'd try to keep it as light as possible since weight tends to just build up...sometimes it's like it comes from nowhere. If it turns out that a heavier rotor is better for your circumstances then you can always add weight later - it's a lot harder to pull weight out of a machine than it is to add it.

Anyway, if the rotor's skeleton is fairly rigid you could probably use those plastic signs that are (structurally) like cardboard. Those "Vote for WhoEver" or "Yes/No on Something" signs. They're pretty available and that would help keep the weight down if you don't go overboard on the skeleton. take it easy

News:

Author Topic: savonius - optimised (Read 7115 times)