Author Topic: Savonius Swept area vrs Overlap ?  (Read 3311 times)

0 Members and 1 Guest are viewing this topic.

gblass

  • Newbie
  • *
  • Posts: 15
  • Country: us
Savonius Swept area vrs Overlap ?
« on: June 23, 2017, 06:48:56 AM »
Hi All

I have 2 plastic 55 gal barrels I'm going to make a stacked savonius with.  It will be a test project so I plan on making it fully adjustable for swept area (diameter).  I was wondering why many designs have the barrels overlapped.  The savonius being a drag design the longer the arm the greater the torque and torque = power.  I'm sure there's some aerodynamic advantage to the overlap but I would think this would be far outweighed with the increase in torque.  Just wondering what others have found.

Thanks
George
5kw Grid Tied Solar.  1kw Solar with batteries and Midnite Solar charge controller. 600W Skymax Wind Turbine on 30FT tower.  80 gal Solar Water Heater.

mbouwer

  • Sr. Member
  • ****
  • Posts: 355
  • Country: nl
Re: Savonius Swept area vrs Overlap ?
« Reply #1 on: June 23, 2017, 07:37:39 AM »
Savonius is fun.

Fused

  • Jr. Member
  • **
  • Posts: 81
  • Country: us
Re: Savonius Swept area vrs Overlap ?
« Reply #2 on: June 23, 2017, 11:33:07 AM »
Id like to see that running.

Doug
System specs: 480 watts peak solar, 8- 6v duracell batteries at 24v controlled with tristar 45,, wind generator is axial flux
with 5 foot blades

if good guys don't carry guns, only the bad guys will.

SparWeb

  • SuperHero Member
  • ******
  • Posts: 3308
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Savonius Swept area vrs Overlap ?
« Reply #3 on: June 23, 2017, 10:51:08 PM »
Torque, yes.  Speed, not so much.
The longer the arm, the lower the RPM, so usually you don't want the buckets of the SAV to be too far apart.
Don't know what the overlap is for, but it may apply some pressure that cancels some of the drag on the forward-going bucket.  Speculating....
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

george65

  • Sr. Member
  • ****
  • Posts: 409
  • Country: au
Re: Savonius Swept area vrs Overlap ?
« Reply #4 on: June 24, 2017, 12:09:41 AM »

I wonder with the overlapped barrels if there is some advantage in re directing the airflow loike in a turbine engine  with the stators reflecting the gas flow onto the next set of blades?
I see the air coming into the forward facing barrel, moving through and " Exhausting" Out the back of the other blade.  This would be momentarily and the airflow would be constantly changing direction but there may be some small advantage to it.

I Usually see these sorts of turbines being 2 blade. Is there a benefit in 4 blades or is there some aerodynamic effect in drag or interruption to the airflow that makes them less efficient.

I'm thinking with the long arm design, having a couple of blades on a shorter arm so they sit inside of the arc of the other 2 90o apart may have some advantage. Would at least reduce the disruption of airflow to the following blade and maybe give an increase of speed.

oztules

  • Hero Member
  • *****
  • Posts: 1477
  • Country: aq
  • Village idiot
Re: Savonius Swept area vrs Overlap ?
« Reply #5 on: June 24, 2017, 04:16:06 AM »
http://www.sciencedirect.com/science/article/pii/S111001681200049X

Looks like 2 blades is  better than three, overlap increases efficiency.....

..........oztules
Flinders Island Australia

electrondady1

  • SuperHero Member
  • ******
  • Posts: 2856
  • Country: ca
Re: Savonius Swept area vrs Overlap ?
« Reply #6 on: June 24, 2017, 06:39:23 AM »
i've never built one but i thought the overlap was about 10%-15%.
the diameter of the center shaft is a factor as well.

« Last Edit: June 24, 2017, 06:54:14 AM by electrondady1 »

Gordy

  • Full Member
  • ***
  • Posts: 151
Re: Savonius Swept area vrs Overlap ?
« Reply #7 on: June 24, 2017, 07:59:52 AM »
gblass,

What are you hoping to do with this, yard art, electric, or water pumping? As this is a drag machine the tip speed will be limited to the wind speed. Since you asked about torque I can assume it's not for yard art. For electric generation a direct drive alternator needs to be built big, with lots of magnets and lots of copper to generate usable power at low speeds. Or you can gear it up for a high speed alternator, but that adds complexity cost and additional wear and maintenance items. I have not seen many claiming to generate electricity successfully built, but I have seen some long armed VAWT's pumping water.

Gordy

gblass

  • Newbie
  • *
  • Posts: 15
  • Country: us
Re: Savonius Swept area vrs Overlap ?
« Reply #8 on: June 24, 2017, 09:18:07 AM »
Thanks for all the replies everyone.  The purpose of this turbine will be to generate electricity.  I currently have a SkyMax 600 on a 30 ft tower.  I'm building the savonius to play with and see how much I can get out of it.  I like experimenting with mechanical things and own a full CNC machine shop.  My original question about overlap vrs swept area was to find out if there is more "power" in a larger swept area.  If so I have the ability to make gears to spin the generator at the correct speed to extract that power.
5kw Grid Tied Solar.  1kw Solar with batteries and Midnite Solar charge controller. 600W Skymax Wind Turbine on 30FT tower.  80 gal Solar Water Heater.

SparWeb

  • SuperHero Member
  • ******
  • Posts: 3308
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Savonius Swept area vrs Overlap ?
« Reply #9 on: June 24, 2017, 11:25:34 AM »
Quote
Gordy: ...a direct drive alternator needs to be built big...
Quote
gblass:...gears to spin the generator at the correct speed to extract that power...
No need.
Just make a generator that works at the same speed.
Direct-drive alternators do NOT need to be big, they just need to be sized to MATCH the driving force of the turbine.
That's what the Hugh Piggott DIY projects are all about.
If you hunt through the Scoraig Wind site you'll find plenty of step-by-step instructions to build what you need.
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

gblass

  • Newbie
  • *
  • Posts: 15
  • Country: us
Re: Savonius Swept area vrs Overlap ?
« Reply #10 on: June 24, 2017, 12:07:36 PM »
Sparweb - Thats a great idea.  I have Hugh's plans on how to make the axial flux.  Initially I'm going to use the servo motor I have to map the power output of the turbine in different configurations.  After I get that data I can make an educated guess on the axial flux. 
5kw Grid Tied Solar.  1kw Solar with batteries and Midnite Solar charge controller. 600W Skymax Wind Turbine on 30FT tower.  80 gal Solar Water Heater.

Gordy

  • Full Member
  • ***
  • Posts: 151
Re: Savonius Swept area vrs Overlap ?
« Reply #11 on: June 24, 2017, 02:03:29 PM »
Sparweb,

You are correct to a point. I failed to mention that I was referring to a discussion years ago about HAWT's vs VAWT's with the same swept area, and the desire to get the same power out of the alternator.

Gordy

 

george65

  • Sr. Member
  • ****
  • Posts: 409
  • Country: au
Re: Savonius Swept area vrs Overlap ?
« Reply #12 on: June 24, 2017, 07:58:22 PM »

Question remains, is overlap better and will more blades give more torque?

Adriaan Kragten

  • Full Member
  • ***
  • Posts: 158
  • Country: nl
Re: Savonius Swept area vrs Overlap ?
« Reply #13 on: June 25, 2017, 12:47:59 AM »
A normal Savonius rotor is not a pure drag machine but it works on a combination of lift and drag. If there is an overlap in between the blades, there is a flow through the rotor and this flow creates lift for certain positions of the blades with respect to the wind direction. Therefore a Savonius rotor can have a much higher maximum Cp (about 0.22) than for a pure drag machine (about 0.07, see KD 416). In my free public report KD 599 (see www.kdwindturbines.nl) I give an overview of reports with measurements performed on Savonius rotors which can be found for free on the Internet. In these reports the optimum overlap can be found. Most measurements are too optimistic because of tunnel blockage.

If different rotors are compared, we normally compare the maximum dimensionless power coefficient Cp. If you make a pure drag machine and if you take more blades at a higher radius, you will certain get a larger torque at a certain wind speed but the rotational speed decreases at increasing radius. The power coefficient can be very low even if the torque level is rather high (see formulas in KD 35).
« Last Edit: June 25, 2017, 12:59:58 AM by Adriaan Kragten »

george65

  • Sr. Member
  • ****
  • Posts: 409
  • Country: au
Re: Savonius Swept area vrs Overlap ?
« Reply #14 on: June 25, 2017, 03:11:23 AM »
In my free public report

How did I know I was going to see this before I opened the post?

Fused

  • Jr. Member
  • **
  • Posts: 81
  • Country: us
Re: Savonius Swept area vrs Overlap ?
« Reply #15 on: June 25, 2017, 08:40:07 AM »
hahahaahah
System specs: 480 watts peak solar, 8- 6v duracell batteries at 24v controlled with tristar 45,, wind generator is axial flux
with 5 foot blades

if good guys don't carry guns, only the bad guys will.

Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4574
  • Country: us
  • USA
Re: Savonius Swept area vrs Overlap ?
« Reply #16 on: June 26, 2017, 09:19:14 AM »
To help answer your two questions
Below is one of the best write-ups I've ever read about the "S" system.

http://www.motherearthnews.com/renewable-energy/savonius-super-rotor-zmaz74zhun

You'll notice that it shows them stacking the overlapped units, the overlap amount best suited and why.

Cheers
Bruce S

A kind word often goes unsaid BUT never goes unheard

Adriaan Kragten

  • Full Member
  • ***
  • Posts: 158
  • Country: nl
Re: Savonius Swept area vrs Overlap ?
« Reply #17 on: June 28, 2017, 05:34:22 AM »
In my free public report

How did I know I was going to see this before I opened the post?

Is this a serious question or are you joking? I thought that this report KD 599 might help you. If you want to read this report or any other of my KD-reports you have to go to my website and then to the menu KD-reports.

Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4574
  • Country: us
  • USA
Re: Savonius Swept area vrs Overlap ?
« Reply #18 on: June 28, 2017, 07:04:37 AM »
Adriaan;
I cannot speak for george65 or any others, however, what I believe this to mean is this.
 Rather than merely posting a short blurb about the question and then say "In my free report on my website ,,,,," . How about just answering the question?

Heck, from the years past where people were known for posting 1/2 a book's worth of information, you too could merely  cut-n-paste from your reports directly to their posted question, here.
With people's life's so busy these days jumping from one website to another, then back again for another question-n-answer session gets "complicated" and quite frankly; just too much.
Even when I get a return to duty office call and while assisting with a multi-district fire, I'm not wanting to jump websites either.
INMO YMMV

Bruce S 
A kind word often goes unsaid BUT never goes unheard

george65

  • Sr. Member
  • ****
  • Posts: 409
  • Country: au
Re: Savonius Swept area vrs Overlap ?
« Reply #19 on: June 29, 2017, 03:26:35 AM »

Speaking for myself....

I have looked at several of adrians  papers and frankly they are hard to read.  There is always way too much irrelevant information to wade through for the answer I am after and it is generally way too technical.

That said, I did look up the last report mentioned and found it to be nothing but a referal to some other ( old)  website's info and a bit of a informal summary thereof. Basically, not too helpful at all.

The long list of report numbers with no other information on the page is also a pain in the arse.  Why these papers have some made up number prefixed buy KD and no mention as to what they refer to is illogical.  There may be a lot to learn there but I can't see too many people wading through the list trying to find something of interest instead of being able to go just what they want to.  There must be a list of what these papers are because I can't imagine anyone remembering a bunch of numbers and their content like that.

Doing a cut and paste of the RELEVANT extracts would be far more helpful than this " In my free report" self promotion all the time.
Don't know about anyone else but for me it's got real old and fast becoming annoying being referred to another site when the question could be answered  where it is relevant. 

electrondady1

  • SuperHero Member
  • ******
  • Posts: 2856
  • Country: ca
Re: Savonius Swept area vrs Overlap ?
« Reply #20 on: June 29, 2017, 07:10:26 AM »
even Sandia labs fudged on the overlap question.
they did a lot of research on savonius mills back in the day  as a method of starting big darious machines  but when exploring the overlap they noticed when increasing the overlap, they were reducing the swept area.
i could never get a clear answer as to what was optimal so i went with 3 blades and no overlap.
the thing about a savonius machine is you get 2 torgue pulses per revolution.
with a 3 vane mill you get 3
keep the overall diameter small and stack the mill vertically
 

 
« Last Edit: June 29, 2017, 07:47:28 AM by electrondady1 »

SparWeb

  • SuperHero Member
  • ******
  • Posts: 3308
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Savonius Swept area vrs Overlap ?
« Reply #21 on: June 29, 2017, 01:02:26 PM »
Quote
gblass...Initially I'm going to use the servo motor I have to map the power output of the turbine in different configurations.  After I get that data I can make an educated guess on the axial flux....

Good plan.
Another thing to look into: "prony brake".
A way to measure torque somewhat directly.

Mechanical power = RPM*Torque / K

Efficiency = mech power in / elec power out

("K" is the imperial conversion factor that I can't remember right now)

This guy totally over-cooks the turkey: https://www.youtube.com/watch?v=u9vhGwwawQ8
But you should get the idea.
An old leather belt and two spring-scales also work fine.
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

george65

  • Sr. Member
  • ****
  • Posts: 409
  • Country: au
Re: Savonius Swept area vrs Overlap ?
« Reply #22 on: June 30, 2017, 09:30:26 PM »

the thing about a savonius machine is you get 2 torgue pulses per revolution.
with a 3 vane mill you get 3

Same thing as having a V8 engine Vs a 4cyl etc.

If you had 4 blades, ie, 2 Split drums, does that further increase torque or have consequences with interrupting airflow or increased drag etc?

MattM

  • Sr. Member
  • ****
  • Posts: 415
Re: Savonius Swept area vrs Overlap ?
« Reply #23 on: July 01, 2017, 08:12:25 AM »
I would think chord should play some issue with VAWT.  Thin chords are critical in HAWT, so flat plates should conceivably be fine in a VAWT.  I always see large swooping curves and hemispheres being used.  I would think the curve is counterproductive.