Simple Solar Pump for India
I will design a portable solar pump to be assembled and maintained in a village shop in Bihar.
Team Project My team partner Vivek offers a high volume solar pumping system to compete with flow-rates from diesel pumps. It is not part of this team project and needs to be judged separately from my proposal. Vivek's role in my project is to assist in marketing of my pumps. I will design a lower cost, lower output, solar pump to offer to the farmers who use less water due to their more precise farming methods. (see http://india.blogs.nytimes.com/2013/0...). Both of us recognize the advantage of submersible pumps which will still operate when the groundwater level drops in a drought. Suction pumps will begin to have trouble if the pumping water level drops to more than 6m below the pump inlet. BackgroundOver the past 30 years I have invented and marketed a line of small, portable solar pumps that operate directly from the solar panels with no electronics or high tech components. Our main market in North America is to water cattle on remote pastures. I also discovered a nice market for environmental cleanup projects in the local oil industry, as described at: http://www.sunmotor.ca/PDF_Files/Sunm.... Sunmotor pumps have also been supplied for remote farms and villages in Kenya, Zimbabwe, Mali, Nepal, El Salvador, Brazil, Bolivia and the Philippines. Pump Design (Keep it Simple) My guiding design principle is that solar pumps should be simple enough to be assembled and repaired anywhere in the world with ordinary shop tools. Pump Selection Sunmotor offers a wide range of solar pumps as shown in attached photos. Our pumps can be used with a float for surface water sources or without the float for shallow wells . The most useful model for this project will be an upgraded version of our SUB300 borehole pump, which can be modified to deliver higher flow at the low head required. This model will fit into a 100mm borehole and can deliver water from greater depths than the typical suction pumps presently in use in the region. We will designate the new pump as our model SUB500. My calculations show that with a carefully matched pump-end we can deliver ~30 m3/day at 10m total head, using a 572w solar array. Calculations Selecting a pump-end rated to 75 LPM at 10m head and assuming our easily achievable 'wire to water' efficiency of 30%, I calculate that we require ~ 440w of power to achieve full capacity output. Operating for minimum of 5 hours/day will deliver 22,500 l/day. With manual tracking of the solar array we can get up to 7 hours of pumping per day, delivering ~ 30 cubic meters per day. Solar Array Sizing We have learned from experience to oversize the solar array by ~30% to account for less than perfect sunlight conditions and for the decrease in power output at high temperatures, so I recommend a 572 watt array for this application. Backup Power Options The pump can be driven by any 12v dc power source including a vehicle battery. A 'plug-and-power' adapter cable with battery clips will be offered as an optional accessory. Security Our design for a portable solar array will make it feasible for a farmer to transport the system home every day and store it at his home. This will include the pump, solar array, and irrigation hose. Multiple Use Solar Array On days when no irrigation is required the solar array can be used to charge a battery in the home to power LED lights and other small electrical loads. If irrigation was only required every second day, the solar array would keep the battery charged so the home always has power. Technology Transfer for Local Manufacture I am in discussions with this company: http://www.rotomag.com/ which is interested to build our motors locally. We will work closely with our present manufacturer: http://www.flexcim.ca, to ensure efficient transfer of the technology required to build the motor and pump in India, including drawings and assembly instructions. Technology Transfer for System Assembly & Marketing Sunmotor will work closely with Vivek to provide training for assembly and marketing of our solar pump systems. We will offer a detailed, 'hands-on' training program in the operation, installation and maintenance of our pump. An inventory of spare parts will be included in the training seminar. Solar Module Selection We will work with Vivek to source the best and most economical solar modules. This may include unbreakable modules if they are available at a competitive price. Irrigation Design We will utilize a low cost, water- saving, irrigation system. Our design is based on supplying 3mm of water per day to the crop. The delivery will be a simple flexible hose that can be coiled up and transported to the farmer's home along with the complete pumping system. Total System Cost The target cost has to be no more than 100,000 INR (~$1,500 USD.) Here is our present estimate of costs when all components are sourced in India: Motor and pump ($500) 33,000 INR (Rotomag and Kirloskar) Solar array (572w) 30,000 INR (Investigating unbreakable modules) Trolley & frame 10,000 INR (Vivek) Cable and hose 10,000 INR (local suppliers) TOTAL 83,000 INR Operating Costs There will be no pump maintenance costs for at least 5 years. The solar modules require no maintenance except cleaning off dust and bird droppings. They will operate for at least 10 years without significant power loss. Updates Sept. 08 - I am delighted to announce a collaboration with vivek (http://greenpeacechallenge.jovoto.com...). We have agreed to collaborate on the assembly and marketing of solar pumps in India, starting in Bihar. I learned years ago that to work effectively in any country you have to partner with someone who "knows the territory". Vivek certainly meets that requirement, in addition to his impressive engineering skills. I look forward to working with him and his team. Also, our pumps complement each other. I specialize in small systems up to 1/2hp, while his system is twice as large. This lets us offer a range of pumps to the farmers to better meet their requirements. Sept. 10 - I have initiated a design of our SUB500 pump to deliver the specified volume of water. We will build two prototypes for testing and demonstration. Sept. 12 - Investigating the potential to use the pumped water to cool the solar panels. This would greatly improve the photovoltaic efficiency in a hot climate like Bihar. The dramatic drop in power output at high cell temperatures is well documented: http://research.iaun.ac.ir/pd/jjfesha.... One option may be to build a simple, narrow reservoir into a support mount for the solar array. Water would be pumped through the reservoir on the way to the irrigation system. The cool water would draw heat away from the panels and keep the p.v. cells near their peak efficiency. Sept. 13- Vivek's description of the bicycle cart solves the portability problem. The cart would simply be re-oriented as required for manual tracking. At the end of the day the solar modules are folded together for transport. Sept. 15- Reviewing the more detailed briefing notes provided by Anudeep, and communications between he and Windy, I realized that we need to increase the total head specification of our pump design to make it suitable for deeper water levels. Therefore I have re-worked the calculations based upon a total pumping height of 10m. The crop irrigation requirement information shows that the proposed SUB500 pump will be a good fit for wheat and vegetable production but will not be large enough for rice. This should still be useful since Vivek offers a larger capacity pump for crops requiring higher volumes of water. Therefore I will review with my design team the potential to upgrade the motor size in our borehole pump. Revised calculations show that we will require ~ 572w of rated solar array to deliver the 440w required by the motor. This 'oversizing' of the array is required primarily to compensate for decreased module output at high ambient temperatures. Sept. 18 The potential drop in groundwater levels in Bihar due to drought could threaten the widespread use of suction pumps. The present depth to groundwater of ~5m is near the maximum limit of surface suction pumps. This will become critical if the groundwater level continues to drop, as predicted (see http://timesofindia.indiatimes.com/in...). The physics of suction pumping says that beyond ~6m of lift you will start to see difficulty of keeping pumps primed and the potential for cavitation damage to the pumps. Sept.19-Most of the solar pump systems we sell for watering livestock in Canada include deep cycle batteries. The farmers like these because they make it easier to move the systems from one water source to another by reducing the size of water tanks required. I got wondering if deep cycle batteries might also offer some benefits to the Bihar project. Instead of transporting the solar panels out to the field every day and back at night, why not put the panels on the roof of the farmer's home and charge a couple of batteries which will be much easier to move. A second battery set would be charged for the next day's irrigation so there is always one battery unit under charge by the solar array. There are some significant potential benefits to this concept: 1. The solar array is more secure and less likely to be damaged. 2. More constant voltage to our pump means we can fine-tune it for higher efficiency. 3. The system could be sized to include LED lighting for the home which will eliminate fuel burning lights and improve the health conditions in the home. 4. In seasons when no irrigation is required the farmer would have a very robust home power system that could run small electrical loads and even small dc motors for crop processing, etc. This could offer an opportunity for one of the large battery manufacturers to lease their batteries and controllers as a sealed, mobile unit that would be towed behind a bicycle. At the end of the lease period it would be returned to them for recycling when required. With fine-tuning of our proposed SUB500 pump and motor combination I expect we can reduce the required power to ~400w while delivering 75LPM at 10m of head. If we pump for 7 hours we will use 400w x 7h = 2,800 w-hrs. of energy. This could consist of two 200 amp-hr. 12volt batteries connected in series to deliver 24 volts as required to operate our pump. Fully charged, these two batteries have a rated capacity of 200 x 24 = 4,800 watt-hrs, so they will still have ~ 40% reserve after 7 hours of pumping. This is a good fit for our project because it avoids complete discharge of the batteries which greatly shortens their life. Solar Array Sizing Assuming an input of 4.5 hours per day of peak sunlight, the required solar array is 2,800w-h/4.5h = 622w. I recommend increasing this by 20% to account for losses of conversion in the system, so we would use a solar array of ~ 750w. Controls The batteries are completely sealed and maintenance-free so they could be put into a locked box and rented to the farmers. The box would also contain an efficient charge controller for the solar input and a simple output cable with quick-connect plug for the pump. A low-voltage disconnect would also be built in to prevent excessive discharge of the batteries. Operation During the irrigation season the farmer would rent two identical battery pods. I visualize that they would each be mounted on a simple set of wheels that could be towed by a bicycle. The farmer takes a charged pod to the field to run his pump for the day while the other pod is being recharged by the solar array. In the evening he swaps the two pods so he has another fully charged pod to take to the field the next day. Home Power Applications During the irrigation season the farm family has to avoid connecting any large loads in the home, however a few LED lights could easily be accommodated without draining much power from the system. In seasons when no irrigation is required he only needs one of the battery pods to provides a significant power system for the home. It would operate many small electrical loads and even small dc motors for tasks like processing crops, etc. (Note: I doubt if this idea is achievable within the prescribed budget, however I wanted to show how deep cycle batteries can be a useful part of a solar pumping system design. This may become an attractive option once the new lithium ion batteries become widely available, offering much longer life and potential for recycling: http://en.wikipedia.org/wiki/Lithium_....) Sept.23- I now have a cost estimate for our proposed SUB500 motor. In small batches our cost in Canada is $657. I expect the cost can be reduced to ~$400 by manufacturing most components in India. The required pump-end will also be sourced in India and may add $100. I am awaiting cost estimates from various Indian manufacturers. Sept.27- I am pleased to announce that Kirloskar Brothers Limited (http://www.kirloskarpumps.com/) have agreed to work with us to provide an efficient pump-end to match my SUB500 motor. They are a well respected pump company in India. Oct.05- We have completed tests on our new motor proposed for use in the SUB500 pump (see uploaded efficiency curves). Next step is to match it with a suitable pump-end. Oct.08- Have completed preliminary design of the SUB500 motor. To reduce cost and weight we will utilize standard 3.5 inch aluminum pipe which is readily available in most countries. Oct.10- I have updated our cost estimate using recent numbers from Vivek for the solar modules and trolley. Shows we can achieve our target well within the specified budget. Oct.15- We have now sourced a 'first trial' pump-end for the SUB500 motor and will run some preliminary tests in the next few days. With the excellent efficiency achieved in our motor design we hope to get good results with the combined motor and pump-end. This may boost the overall system efficiency and let us reduce the size of solar array required. It is important to conduct careful testing find the best match. Engineering design really only sets a target that has to be verified by careful testing of the pump and motor combination. Oct.18- I have added a photo of our testing station that we use to simulate field applications for our pumps. It allows us to set the pressure gage to equal the total dynamic head (vertical lift + friction). Then we run the pump at the voltage of the solar array. By measuring flow and current we can calculate the efficiency of pump and determine its suitability for the application. As part of the technology transfer we will work with Rotomag to set up a similar test bed for the pumps and motors they will build for us in their India facility. This will ensure that all pumps meet the performance targets before being sent into the field. Oct.22 - Our tests have confirmed that the SUB500 pump will easily meet the target capacity of 75 LPM at 10m using 440w of input power to the motor. I believe that we can improve on this efficiency once we conduct tests on the wide selection of pump-ends available in India. I designed the mounting and shaft configuration to match the universal NEMA standard for shaft and mounting configuration used by most manufacturers of borehole pumps. This allows us to find a local source of matching pump-ends anywhere in the world. Our plan is to test several samples of suitable pump-ends from the leading India manufacturers and compare their efficiency and cost. Oct.23 - I am in on-going communication with Rotomag which has offered to manufacture our SUB500 Solar motor. Rotomag already has extensive experience in motor manufacturing for clients around the world. We will work with them to transfer the SUB500 technology to their shop and establish a quality assurance program for the manufacture, assembly and testing of each SUB500 motor produced. Oct.24 - I am continuing investigation of unbreakable solar modules. Possible sources in India and China at competitive prices. Oct.27 - The SUB500 solar motor will run any 4 inch (100mm) borehole pump up to 1/2 hp rating. This lets us source a pump-end from a local manufacturer in any country. It also provides an easy way to introduce solar pumping to thousands of drillers, plumbers and water contractors who are already familiar with borehole pump-ends from previous experience of installing ac pumps. They simply have to replace the conventional ac motor with our solar motor. The unit is then ready to operate by connection to a dc power source, which may be solar, wind or battery. No controller is required to connect to any of these power sources. Oct. 28 - The SUB500 motor + efficient pump-end will deliver the following daily volumes on a sunny day when powered by a 572w solar array: 6m TDH - 50 m3/day 8m TDH - 40 m3/day 10m TDH - 30 m3/day This assumes manual tracking of the array which will involve re-positioning the trolley a few times per day to keep the modules pointed at the sun. Oct. 29 -My design as presented is quite conservative in that I assumed a pump and motor efficiency of only 30%. With our excellent test results for the motor (83% efficiency) it is possible to increase our target motor and pump efficiency to ~40%. This will require a pump-end efficiency of ~ 50% which is certainly achievable. The overall effect will be to reduce the size of solar array to ~ 430w, reducing cost further and making it easier to transport. Additional improvement may also be possible by cooling of the solar modules with a small sprinkler, as proposed by Vivek. Oct. 30 - For existing wells with large enough openings we can offer our new model M40 pump which uses the same powerful d.c. motor we designed for the SUB500. This may be a lower cost option for large wells since fewer parts are required to build this type of pump. It is a higher power version of our Model M30 which we have manufactured for years (see http://www.sunmotor.ca/PDF_Files/M30.pdf). My calculations show that with the new larger motor it will deliver the following volumes: 6m TDH - 45 m3/day 8m TDH - 38 m3/day 10m TDH - 32 m3/day Oct. 31 - I am pleased to confirm that Kirloskar Bros. can supply a suitable pump-end for my SUB500 motor (see following email): "Dear Mr. Eric Jensen We have a suitable 4” bore well pump for 75 LPM @ 10 MTRS. We have a dedicated plant for the manufacturing of submersible pump sets located at Ahmedabad, Gujarat, India. This plant, with state of the art manufacturing facility, producing energy efficient submersible pumps. The facility also has a CED (Cathodic Electro Deposition) Plant, where components are coated for better life and makes our products energy efficient. MOHAN INAMDAR, DIVISIONAL MANAGER-IRRIGATION SECTOR,KIRLOSKAR BROTHERS LIMITED Nov. 01 - To market new technology to farmers is never simple. They may see many potential benefits in your product but most of them won't buy it for several years. They want to watch it in operation and see how useful (and durable) it really is. Equally important, they want to see if spare parts and repairs are readily available when required. The secret to introducing your product is to make the first sale to a local innovator who is willing to try new ideas and who is respected by people in the community. Once the innovator is satisfied with performance of your product the neighbors are more likely to buy one too. This article describes such an innovator in Bihar state: http://india.blogs.nytimes.com/2013/0.... The same article provides a lead on how to find similar innovators in other districts: "Ms. Devi is one of the 103,028 small farmers across 9 districts of Bihar who have adopted a new system of seed treatment and planting that uses no chemical fertilizers or herbicide. The system has been so successful that it will be rolled out across all of Bihar’s 28 districts in the next year." I suggest that this group offers a great opportunity to rapidly deploy solar pumps in Bihar. They are already proven innovators and they use water efficiently, so a smaller solar pump will be adequate for their farms. Perhaps we could partner somehow with the parent organization (S.C.I) to which they all belong and offer a micro-credit program to purchase our solar pumps. Nov. 02 - We can learn a lot about the challenges of marketing new technology to small farmers by studying the organization IDE (http://en.wikipedia.org/wiki/Internat...). Their project in Bangladesh set up a manufacturing and marketing network that has sold over 1.5 million treadle irrigation pumps. IDE founder Paul Polak has recently suggested that the same marketing approach could be useful to deploy solar irrigation pumps in India. Nov.03- I have uploaded a photo of our new M40 pump which can be used in wells with large enough diameter. The M40 utilizes the same powerful motor we designed for the SUB500 borehole pump. This will reduce overall cost by permitting greater volume production when Rotomag builds a batch of motors. I am also using many other common parts for the two pumps to make it easy for a local shop to assemble and market both pumps with a minimum of inventory. Nov. 04- The key to rapid deployment of our solar pumps is to invite local shop owners in each community to become assemblers and marketers of the pumps and solar arrays. This must be structured as a profitable venture for the shop owners. They will likely be the same shops that build the trolleys to carry the solar equipment to the fields. Most shops will already have the basic tools needed to assemble a Sunmotor pump. All they require is training. Our assembly procedure includes two simple tests that are conducted on each pump to ensure that it is correctly assembled before being sold to the farmer. These tests also form a critical quality assurance policy because they provide a check on the procedures being followed. Other assembly tasks include a 'quick-connect' plug to make it easy to connect the pump to the solar modules. Nov. 06 - One possible advantage of our submersible pumps which we have not mentioned to date is that they could be left in the borehole overnight. A simple cap with a lock would protect the pump from theft. This will save a lot of time in not having to pull the pump and hose out of the borehole each day and transport it to and from the farmer's home. (A similar strategy may work for a suction pump system to allow the pipe and foot-valve to remain in the borehole. This could greatly reduce the significant challenge of priming those systems each day). Nov. 08 - A potential supply chain for marketing my solar pumps in India includes the following: Technology Owner (Sunmotor - royalty option) SUB500 motor manufacturer (Rotomag - to build 100 units per run) Pump manufacturer (Kirloskar Bro. - 100 pumps per order) Solar module manufacturer (bulk orders to reduce cost) Distributor ( Vivek and company) System Integrators (local shops) Farmers For a sustainable business model not dependent upon government subsidies or NGO donations, everyone in the supply chain has to make a reasonable profit. The challenge will be to rapidly achieve a volume of production sufficient to allow each member of the supply chain to reduce their markup in prices as lean as possible so the farmer can still afford to buy the system and make a good profit from increased sale of crop (or reduced irrigation costs.) Nov. 09 - Here is a great website for finding renewable energy businesses in India: http://energy.sourceguides.com/busine.... By contacting companies listed on this site I received good information on local availability of deep cycle batteries. This person was particularly helpful: Mandar Rajadhyaksha <firstname.lastname@example.org> Nov. 10 - I just received the pump specifications from Kirloskar Bros. and have uploaded it as an attachment. It shows that their pump-end is 55% efficient, which is very good for a small multi-stage centrifugal pump. This lets us re-calculate the overall efficiency by multiplying that no. x the motor efficiency (83%) which predicts a 'wire to water' efficiency of 0.55 x 0.83 = 0.45 or 45%. Field testing will be required to confirm if we can achieve that target, but it offers a potential to reduce the size of solar array required. Combining this with Vivek's results from cooling the solar modules with a small water spray will make a very efficient system well within the specified budget.