The farmer-built Savonius rotor: A low-tech approach to renewable power for farms

Project Overview

Project Type: Farmer
Funds awarded in 2010: $10,365.00
Projected End Date: 12/31/2011
Region: Northeast
State: Vermont
Project Leader:
Erik Andrus
Boundbrook Farm

Annual Reports

Information Products


  • Animals: bovine, poultry, swine


  • Energy: wind power
  • Sustainable Communities: sustainability measures

    Proposal summary:

    Small-farm scale alternative electricity generation is priced out-of-reach for most farmers. We propose to design and construct a prototype of a silo-mounted augmented Savonius wind turbine that could deliver substantial power output, yet is inexpensive and simple enough in its construction, installation, and electrical engineering to be easily built and maintained on a typical small farm. Nearly all farms use electrical power for daily operations. Though the general public is aware of the need for sound renewable alternatives to our current grid-based supply, the dialog is usually framed in terms of large, highly capitalized commercial wind, solar or methane projects. For small farm operations, generation of electricity is seldom made a priority because the residential and light commercial scale options on the market all demand high up-front cost, prolonged payback periods (generally well over 10 years), complicated technical issues, possible permitting problems, and the likelihood of expensive repairs requiring specialized parts and labor. Given these high costs, the less-than-ideal regional meteorological conditions for both solar and wind provide an additional disincentive to take individual action. An effort needs to be made to use the skills and resources of farmers to create a durable region-wide micro-generation infrastructure. The silo-mounted augmented Savonius Rotor described in this proposal is a part of such an effort. It is low-tech, inexpensive to build and maintain, and makes use of typical farm skill sets and a typical farm structure. 

    Doing all of our own flour milling in a given year (about 10 tons per annum at the bakery’s current rate of sales) would require between 3200 and 6400 kilowatt hours, costing between $500 and $1000 in electricity annually (at 15 cents per kilowatt hour). Based on his previous projects and estimated wind available at our site, Victor Gardy estimates the planned augmented Savonius rotor will generate about 17,000 kWh per annum, more than enough power to offset this economic cost to the operation as well as the environmental costs associated with the consumption of this quantity of grid-based electricity. This hands-on approach to reducing costs and environmental impact is a good fit for our farm and its mission.   If the prototype performs within anticipated parameters, it would serve as a model for farm-based wind power that could be replicated on a very large number of Northeastern farms.

    Project objectives from proposal:

    While the Savonisou Rotor does not compare to a state-of-the-art horizontal-axis wind turbine in terms of efficiency, it more than compensates with its low cost and simplicity. The resulting power can make a real difference in the economics of farm operations, whether it is used mechanically, in a stand-alone 12-volt system, or in a grid-tied net-metered system. Our search for others involved in low-tech wind power produced scant local activity. We consulted Gary Flomenhoft, a fellow with the Gund Institute for Ecological Economics since 2002. Sustainable energy production is, for him, both a personal and professional interest. He is familiar with the Savonius rotor and has found no local or regional initiatives using vertical axis rotors. He explains, “Everything is focused on achieving commercial generating scale because of high cost of horizontal rotors.” We have also made contact with Job Ebenezer, a mechanical engineer and former professor who now runs an Ohio-based project entitled “Technology for the Poor.” This project evaluates appropriate technologies to accomplish significant work in less-developed countries yet are easily built and maintained by local people. He built the Savonius rotor pictured in figures 1 and 2, during the 1970’s. which produced about ½ horsepower and cost $200 (at 1978 values) in materials. As energy costs dropped in the 80’s, interest in such projects has diminished and Ebenezer’s prototype has not led to widespread use of such wind collectors. Yet its simplicity and proven effectiveness suggests that such devices deserve a place in any consideration of renewable power generation. We also have been working closely with Victor Gardy, a local retired engineer whose career has been devoted to turbines, chiefly those of aircraft and power plants. As a personal hobby, he has also been researching and developing versions of the Savonius rotor for 29 years, including one installation in Charlotte, Vermont in the early 90’s. Gardy also shares our view that this useful device has been almost completely overlooked by the engineering community.

    Boundbrook Farm is a 110 acre farm in the town of Ferrisburgh, comprised of a small (1/2 acre) market garden, 15 acres of cropped fields, 60 acres of pasture, as well as woods and wetlands. The farm is worked primarily with draft horses, and produces grass-fed beef, pork, poultry, and grains. The grains are processed into feed or milled by Good Companion Bakery, our on-site brick oven bakery, and baked into bread or made into other goods. All our meats and baked goods are marketed nearby. Good Companion Bakery operates a unique “Bread CSA” with internet-based ordering of seven types of fresh baked goods, and delivery to five towns. The key people in the project are Erik Andrus, co-owner of the farm and bakery, Amos Baehr, resident farm employee, and engineer Victor Gardy. Prior to starting this farming operation, Erik Andrus was a freelance renovation contractor for 10 years. Amos Baehr also brings to the project many years of construction experience and a facility for data analysis. Victor lends his involvement from a personal commitment to bring about simple, effective, home-built generation devices to serve as a counter-example to the large and high-tech installations that dominate the public discourse on alternative energy. We began farming here in 2006 with the goal of creating an integrated diversified farm with an emphasis on sustainable, horse-powered grain production and European hearth breads. A key component of the operation is the milling of our grains. We have an Austrian-made 30 cm stone mill with a sifting apparatus, which can mill all the grain we as a farm produce, and can make sifted flour of various grades. Milling is, for us, the most energy-intensive part of our grain-to-bread process. Doing all of our own flour milling in a given year (about 10 tons per annum at the bakery’s current rate of sales) would require between 3200 and 6400 kilowatt hours, costing between $500 and $1000 in electricity annually (at 15 cents per kilowatt hour)

    We have already obtained analysis of the wind properties at our farmstead. Our average wind speed is 10.9 mph, just below the threshold at which conventional investments in wind power are considered in the least way viable. But given than the Savonius functions well at lower wind speeds, we have determined that our farm will make a good test site. The first stage involves design. The proposed rotor is an innovative application of a proven concept (first developed by the Finnish engineer Sigurd Savonius in 1922), and will make use of cheap, readily available building materials. Per Gardy’s suggestion, this prototype will feature features stators, or fixed panels, to reduce drag on the rotor cups as they return into the wind. Gardy’s previous research suggests a marked increase of nearly 100% in efficiency with the use of stators. A rough concept design is already in place, and we have good reason to believe that the proposed device will perform within the desired output parameters. However the builders (E. Andrus and A. Baehr) must work with Victor Gardy to complete a comprehensive design for both the rotor and the structural frame that houses it, as well to specify appropriate generators and metering devices. In addition a building permit must be obtained for the project.

    Since the proposed site actually involves reducing the height of an obsolete silo, the sort of permitting problems associated with erecting a tower are not expected. Both the designer and the builders are ready to proceed into this design phase. The second phase involves construction of the array. Using the working design, an array of augmented Savonius rotors comprised of two rotors, each 12 feet in height, will be assembled and placed into a bracing structure. All of this work will take place at ground level in or adjacent to the farm workshop, which is already well-equipped for projects such as this, over the course of about two weeks. The finished prototype will be a complete structural unit, which can be moved by crane to the nearby installation site. Installation is the third phase.

    In early July 2010, the silo will be prepared to receive the completed rotor array. We will use a hired crane to stabilize the upper portion of the silo as the bolts are loosened or cut at the 36’ level. We have determined that reducing the height of the silo from 60’ to 36’ is preferable to installing the array at 60 feet of height; at the lower height the rotor array will be both more secure and more accessible, and wind conditions will likely not be very different, since even an installation at 36’ would be clear of nearby barriers to wind flow. The upper portion of the silo will be removed and scrapped, and a pre-assembled flooring system including guardrail will be lifted into place by crane. Finally the rotor assembly will be lifted into place and secured with bolts and guywires. The final phase is evaluation. We will complete installation of the rotor by July15th, 2010. Once the rotor array is secured, it may be activated, and we will begin to monitor their performance electronically to evaluate its effectiveness.

    We will also proceed to use the generated power for grain handling, cleaning, and milling operations. Our long-term goal is to establish a gravity-flow grain cleaning and milling facility inside the body of the steel silo, but this lies outside the scope of our proposal, which concerns the rotor prototype and its performance alone. We will monitor the production of electricity from the rotor array 24 hours a day from July 15th-October 15th 2010, through the use of a data logger. The collected data can be exported to our farm computer and then analyzed and graphed. Since our goal is to develop a low-tech, cost effective generator, the production of power must be weighed against the cost of the power through both conventional grid-based sources and competing renewable/alternative power sources. The usefulness of the prototype is easy to assess once the output is known. The most useful measure is “payback time,” which is the cost of the turbine divided by its annual production of kilowatt hours at the going rate. We will also compare its payback time to various commercially available generation options on the market.

    At this point we estimate the array will be able to produce about 660 watts per hour under average winds, and should therefore yield an estimated 17000 kWh over the course of a year. At such a rate, given the $2450.00 projected cost of materials and components for the rotor itself, the project could pay for its own materials in savings in less than two years. The labor and other costs could be paid for in another three or four years, for a total projected payback period of five or six years. In addition we will also evaluate and graph variation in the prototype’s production of electricity, make note of the revolutions per minute of the rotors at various wind speeds, describe any noise or vibration associated with the unit, and evaluate its performance at very low wind speeds and its durability in high winds. If the test period proves to be more or less windy (based on airport reports) then we will adjust projections accordingly. A failed result might involve extremely low output, serious structural failure of the rotors, their bracing system, or structural damage to the silo base. Even such results could be instructive as they might point the way to improvements in design or installation. Damage to the silo base would not be catastrophic as our steel silos are otherwise obsolete. If the prototype performs within anticipated parameters, it would serve as a model for farm-based wind power that could be replicated on a very large number of Northeastern farms. While most commercial wind technology is only economically justifiable if one has an excellent wind site, the production cost and complexity of this prototype are so low that even farms such as ours with marginal wind availability may profit from installing one.

    Further, this project makes use of typical farm infrastructure. Silos are not suitable towers for conventional horizontal-axis wind turbines. However they seem to be quite suitable as platforms for the type of vertical axis rotor we are exploring, due to the lower speed and vibration associated with this class of windmill. This may mean that a typical northeastern farm eyesore, the obsolete silo, can (if still structurally sound) be reinvented for new use and profit. It should also be noted that this design does not need a prime wind site (or a silo) to perform. Low towers, platforms, or barn-roof installations and the placement of mobile rotors in open spaces such as pastures (generated power can be delivered to the farm by a reeled-out cable) are additional placement options that may work for other farms. The short payback time expected from this project means that farmers may see quick returns on their installation investment and dramatically reduce their utility costs. This can be true whether the resulting power is used mechanically, in a stand-alone 12-volt system, or in a 110/220 volt net-metered system where surplus electricity is sold back to the local electric utility. In our case we are pursuing both net-metering and the powering of some processing operations through a 12-volt system, notably flour milling. For farms that must have a backup to the grid, this design may offer a cost-effective alternative source of power for essential devices during blackouts. It also does not entail the noise, pollution, and fuel cost associated with gas generators.

    The prototype features a simple design; farmers will find it easy to construct using basic carpentry, mechanical and electrical skills and common building materials. This design therefore presents an opportunity for farmers to use their skills and talents to reduce their overhead, make a positive contribution to our regional energy economy, and become more self-reliant in a universally needed commodity.

    This project will deliver information on the augmented Savonius wind turbine through a variety of educational and outreach techniques to other farmers throughout the Northeast. Two mass media articles will be written over the next year. A “New Take on the Vertical-Axis Wind Turbine” article will be developed in July 2010 and published in a variety of local (i.e. Agriview) and regional venues (Country Folks). A second article will be published in these same venues at the conclusion of the three month trial period in October 2010 to share our results. The earlier articles will also announce a field day to be hosted at our farm in September 2010. This field day will feature a demonstration of the small scale wind-turbine. During the workshop, we will talk about designing and building the turbine as well as performance and practical applications for various farm settings. This workshop will be incorporated into the NOFA-VT summer workshop series. After the 3-month trial has been completed and data compiled, will produce a 10-page construction manual featuring complete plans, photos and diagrams, available for free download from the Northern Grain Growers Association (NGGA) website, or from the farm upon request, for a nominal $5.00 fee to cover printing and postage. We also will print 200 copies of this manual to distribute free of charge at the above-mentioned farm field day and at conferences at which we will present the project. In addition, an article on the project results will also be included in the NGGA quarterly newsletter (distribution to over 200 farmers). An Extension leaflet will be published and made available through the UVM Extension Publications Center and on the Extension website. Lastly, presentation of our results will be delivered at the 2011 NGGA annual conference, NOFA-VT winter conference, and the Vermont Renewable Energy for the Farm Conference. These annual conferences will distribute project information to more than 500 farmers.

    Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.