The Basics- System Design Richard Perez System design harnesses a site's specific renewable power sources. System design precisely determines the right type and size of energy hardware required to meet the system's power needs. Here we decide which power sources to use- solar, wind, or hydro. We decide the type and size of the power sources- how many PV panels or how big a wind or hydro turbine. We decide the type and size of the system's battery. We decide the types and sizes of the inverter, controls, and instrumentation. A system's cost and utility are determined by these critical decisions. A Big Deal? Yes, system design is a big deal. Proper design requires the information generated from the system's power use survey and the system's site survey. The system's designer must match the available natural power sources with the specific electrical power requirements of the system's users. If there is no thorough power use survey or site survey, then it is impossible to design the system. So, before you can specify any hardware, you must do your homework. If you are vague on the details of the system's power use, or surveying a site for renewable energy potential, then see The Basics in Home Power #21 about use and site survey. The decisions made in designing a renewable energy system will determine if the system is effective or not. Each system is unique. Each system is located in a different location with differing power requirements. A system should fit its user like a hand-made pair of boots. Your system should be as unique as you are. Your system should be as unique as your site. Don't accept a generic, off-the-shelf system. Your RE system will provide a lifetime of effective power if it is properly designed. If not properly designed, then it will leave you in the dark and empty your bank account, all at the same time. System design is the place to discover any mistakes- in theory and on paper, BEFORE you spend money on the wrong hardware. System Costs The decisions made in a system's design directly effect the system's cost. The design phase of a RE system is a very good time to consult someone with current knowledge and extensive experience. The details of which PV module to use, which hydro to buy, how high to put up the wind machine, what kind of battery, how big of a battery to use, and myriad other details greatly effect a system's cost. If you want a cost-effective system, then work with someone who knows how to design a system, or learn the process yourself. The shortest path to a well-designed system points directly to your local, installing RE dealer. These guys are familiar not only with the latest hardware and how to apply it, but also your local renewable energy resources. Help! This article is designed to give you the Basics of system design. It provides enough information for you to discuss your situation intelligently with someone familiar with renewable power hardware and your local environment. If you are going to specify your system without the aid of a techie, then you will need more information than is provided here. Why don't I tell you all the information you need? Well, I've been involved in well over a hundred of these systems in the last twenty years, and I'm still learning new things daily. The hardware used in RE systems is changing rapidly. Consider that an entire book could be written on each hardware component in the system. That's a fact; I know because I've written one on batteries alone. Consider that the combinations of different types and kinds of hardware number in the thousands. Consider that a system's designer must select from these thousand of combinations the exact set that matches your use and site, and at the minimum cost to you. According to our correspondence, many Home Power readers have designed their own systems. They also tell us of many errors and many misspent dollars. It is possible to design a system using the trial and error method. It is also very expensive, very slow, and very frustrating. I don't mean to discourage anyone from learning or blazing new trails. I just want you to be aware of what lies ahead. If you have the commitment to do the job right, then the user-designed system is every bit as effective as one designed by a professional. In some cases, the user-designed system is better than can be done by any professional. After all, who knows your needs and your site better than you do? The Flow Being a digital nerd, I have concocted a flow chart for the decision making process that is system design. This flow chart shows what you need to know at each stage of the process. It also shows how to modify the system and the decisions required when changes are made. INSERTFLOWCHART Using the Flow Chart The major feature of the flow chart is the feedback loop connecting the decision "Can I afford the system?" with the Power Use Survey. If after all the hardware is specified the user cannot afford the system, then go back to the Power Use Survey and reduce the power consumption. In most cases, reducing power consumption is merely a matter of substituting efficient appliances for power pigs, or using of electric power only where it is cost-effective. In short, trim off the fat, and model the system again. Less consumption requires less hardware and this reduces the system's cost. Continue this feedback process until you have specified a system that you can afford. The Particulars of System Design Each renewable power resource is differently applied. A system sourced by hydro power for example will contain different kinds and quantities of hardware than a wind powered system. All systems, however, share a common design approach employing four basic classes of hardware. 1. Power Sources- these may be photovoltaic, wind, or hydro powered. The power source produces the electric power 2. Power Processors- these are inverters, controls, and instruments. The power processing equipment changes on form of power into another, controls the amount of power delivered to the batteries, and measures the electrical parameters of the power. 3. Power Storage- here power is stored in batteries for use when the source(s) aren't producing. 4. Power Users- these are the system's electric appliances. The appliances are why we're doing all this. The jobs they perform are the reasons why we need electric power. Almost all systems will employ these four classes of hardware. Almost all systems use this hardware in the same fashion displayed in the system block diagram that follows. As time passes, more of us are adding a second power source to backup our primary source. Wind and hydro are excellent compliments to a system primarily powered by PVs. When the sun isn't shining, it is often raining and the wind is blowing. In most cases, the secondary renewable power source feeds the system's batteries, just like the primary power source. INSERTSYSBLOCKDIA System DC Voltage In the past, almost all systems used 12 VDC as their base voltage. This was because the systems were small and extensively employed 12 VDC appliances powered directly from the battery. Now, with the arrival of efficient and reliable inverters, 12 Volt use has declined and 24 VDC is becoming the favored battery voltage. At this moment, the system's DC voltage should be determined by how much power the system cycles daily. Systems producing and consuming less than 2,000 Watt-hours daily are best served by 12 Volts. Systems cycling over 2,000 and less than 6,000 Watt-hours daily should use 24 VDC as a base voltage. Systems cycling over 6,000 Watt-hours daily should use 48 Volts. System voltage is a very important factor effecting the choice of inverter, controls, battery chargers, and system wiring. Once these components are bought, they cannot be changed. While some hardware, like PV modules, can be reconnected from 12 to higher voltages, other hardware like inverters, controls, and wiring is specified for a particular voltage and must operate there. There are specific exceptions to this rule that will covered in the material that follows. Batteries All systems, except large hydros and wind machines making 120/240 vac directly, use a battery. The function of the battery is to store power for use when the renewable power source is not producing. For example, a PV system is useless at night without a battery. Wind systems must use a battery to store power for windless periods. The choice of what battery technology to use is critical and greatly affects overall system performance and cost. There are now three battery technologies generally employed in home power systems: deep-cycle lead-acid, nickel-cadmium, and nickel-iron. Each battery type is differently sized and applied. For example, all lead-acid batteries will last longer if they are not fully discharged regularly. For this reason, lead-acid batteries are sized so that at least 20% of their rated capacity is never used. Other types of cells, like the alkaline nickel-cadmium and nickel-iron types tolerate deep cycling without premature aging and failure. The least expensive battery to purchase is lead-acid. Carefully used lead-acid cells will last about ten years. While alkaline batteries initially cost about two to three times as much as lead-acid types, they last between three to six times longer. This question of initial cost versus lifetime cost occurs repeatedly in system's design. Each case is specific, but a rule of thumb is, "Buy the best hardware you can possibly afford." Higher quality components always cost more, but they give better service and last longer. In most all cases, they are more than worth their additional cost. A word of advice, don't use automotive type lead- acid cells in renewable power systems. These cells are optimized to do only one job, starting a car. They are very ineffective and fail rapidly under deep-cycle service. Temperature plays a big role in choosing a battery technology. If your battery spends several months in temperatures below 45øF., then use alkaline types rather than lead- acid. Lead-acid cells have poor low temperature performance and must be kept warm in cold climates. Expandability is also a factor to consider when choosing a battery type. We have two years to add more cells to lead-acid battery. After this two year period, the original lead- acid cells have aged to the point where they will not function efficiently with newly added cells. In any case, all additions to an existing lead-acid battery MUST be exactly the same make, model and size of cells as the originals. In the case of alkaline batteries, this is not true. Alkaline cells can be paralleled to an existing alkaline battery at any time and need not even be of the same size and make. So lead-acid systems are only expandable within the first two years of operation, while an alkaline battery can be expanded anytime. A high quality lead-acid battery will cost about $0.12 per watt-hour of stored electric power. A reconditioned nickel-cadmium or nickel-iron cell will cost about $0.35 per watt- hour stored. New nickel-cadmium cells will cost about $0.80 per watt-hour stored. Inverters Five years ago, a good inverter lasted about a year. Now inverters have become ultra-reliable and are electrically bullet-proof. The criteria to use when choosing an inverter are, in order of importance, as follow: reliability, efficiency, utility, and price. Note that price is last on the list. A reliable inverter must automatically protect itself from the following conditions: output overloading, overtemperature, high battery voltage, and low battery voltage. These protection functions are included in every high quality inverter. Any inverter worth owning will have efficiencies in the 88% to 98% area over its entire output power range. Usually an inverter is over 95% efficient between 20% and 85% of its rated power output. Don't accept anything less because the wasted power must be produced and stored by the system. This costs more than an efficient inverter. Utility means how useful is the inverter. This is difficult to assess without living with the inverter in your system, and by measuring its output. For example, most inverters employ a high voltage shutoff circuit that turns the inverter off if the battery voltage is too high. While, this may not annoy a lead-acid battery user, but it can be very inconvenient for nickel-cadmium users whose battery undergoes larger voltage excursions while under charge. Other utility factors are technical and require a some knowledge of electronics and instruments to evaluate. So if you're not a techie, then ask a techie who has measured these parameters on a working inverter of the type and size you are considering. Some of these technical factors are as follows. Average inverter voltage output should be as close to 117 Vrms as possible (anything below 105 Vrms and over 130 Vrms flunks). The inverter's peak voltage output (Vpp) should be a close to 164 Vpp as possible (anything below 140 Vpp and above 200 Vpp flunks). Can the inverter deal with the inductive feedback produced by electric motors, transformers, and fluorescent lighting? The best way to assess an inverter's ability to handle inductive loads is to use an oscilloscope while powering these loads. If the inverter's waveform is radically different when it powers say a lightbulb (a resistive load) and a motor (an inductive load), then the inverter flunks. These utility factors are technical ways of defining the constancy and purity of the inverter's output. The instrumentation and tech knowledge needed to evaluate these inverter parameters are extensive. This is why we at Home Power test so many inverters. At this point, I will say that any inverter that has passed "Things that Work!" testing meets the above mentioned techie requirements. Nuff said... In terms of cost, an inverter worth having will cost between $.60 and $1.00 per watt of continuous output power. Don't even consider full square-wave inverters. They are less reliable, much less efficient, and less useful than the modified sine-wave types. I mention this here because the full square wave types are cheap and you will get what you pay for. An inverter that produces 1,000 and 2,000 Watts will power an entire household of careful users. This means that the users agree not to run everything in the house at the same time, thereby overloading the inverter. If the household must be able to simultaneously power several large appliances, then consider a 3,000 to 5,000 watt inverter. Or split up the loads into different circuits and power each circuit with a different 2,000 watt inverter. Surge capability used to be a big problem with inverters. Now, just about all inverters will deliver at least three times their rated output for surge demands. Surge output power is no longer a problem in well designed inverters. The Power Use Survey will detail all appliances powered by the inverter. List all appliances that will be running at the same time and add their wattages. This figure is the minimum size inverter to purchase. Some inverters can be expanded (like the PowerStar UPG models) or paralleled (like stacking two Trace inverters) for more output. This is a recent development and is greatly appreciated by growing systems. Different manufacturers rate their inverters in different ways. Ignore the ad verbiage and look for the inverter's CONTINUOUS output rating. This is the truest measure of the inverter's power handling capabilities. In the past, many of us used the battery chargers offered as options on inverters. This is still a very inexpensive way of getting a high output, electronically regulated, battery charger. For example, the Trace 2012 inverter has a battery charger option that will produce up to 110 Amperes up to about 15.5 VDC. This, however, requires the inverter to change into a battery charger when it is supplied with 120 vac from a generator or from the gird. This complicates the inverter's function and adds about $220 to its cost. A different approach is to let the inverter do only the job of making 120 vac from battery stored DC power. Use a separate component to recharge the battery. Modern switching supplies are being configured as effective (especially on generator power) and efficient battery chargers. They cost about the same as building the charger into the inverter and offer more utility plus greater reliability. Designing a Photovoltaic System PV systems are the easiest type of all to design. We already have an accurate idea of solar insolation from the Site Survey. We've done the Power Use Survey so we know how much electric power we require on an average day. All that remains is to specify the type and number of PV modules that will produce the power, and type and size of the battery. Choosing the right PVs For a 12 Volt system use PV modules that contain 36 series connected cells. Simply wire two panels in series for 24 Volt operation, and four in series for 48 Volt operation. Stay away from the "self-regulating" PV modules containing 30 or 32 series connected cells. Self-regulating models still require a regulator in most systems and make less cost- effective use of their cells. Look for PV modules that carry a full warranty from their manufacturer. Any panel worth having is now warranted not to lose more than 10% of its output power within a ten to twelve year period. Your PVs are a lifetime investment and not the place to save a few bucks on less than the best. The most cost-effective PV modules now use polycrystalline or "multicrystalline" silicon material. While the single crystal cells have higher efficiencies, they are also more expensive. Currently, high-quality polycrystalline modules are selling for about $7.00 per watt. Sizing the PV Array Size the PV array so that it will produce, on an average day at your location, AT LEAST as much power as you consume during an average day. For example on an average day here on Agate Flat, a 48 watt PV module produces 225 Watt-hours of power. We consume an average of 2,480 Watt-hours of power daily. Divide the power consumption by the average output of a single panel and the result is the number of panels required. In this case, 2,480 watt-hours per day / 225 watt-hours per day = 11.02 panels. We actually have twelve modules working now and hardly ever have to resort to a backup power source except during the dead of winter. Sizing the array so that it produces our average day's consumption is the smallest sized array to consider. There is no penalty for using more panels, if fact there are substantial benefits. The first is the amount of time it takes for the system to refill the battery after an extended cloudy period. The second is the system's ability to withstand unusual periods of high power consumption, like when city folks come to visit and leave all the lights on all the time. The third is that all systems seem to grow and use more power, so with more modules, power production is ahead of the system's growth rather than behind it. One of the very fine features about a photovoltaic array is that it is easily expanded. If we require more power, we can add more modules to the array at any time. This expandability is unique to photovoltaics and is not possible with wind or hydro turbines. Expandability makes it easy to start small with a PV system and have it grow as the family needs more power and/or can afford it. To track or not to track? A tracker is a device follows the sun's daily motion through the sky. In order for the tracker to be effective the site must have good access to both early morning and late afternoon sunshine. A site where the sun strikes only between 10 AM and 2 PM is not a good site for a tracker. Does the site have the unrestricted solar access necessary to make a tracker effective? This is determined during the site survey. If a site has enough sun to make a tracker effective, then the tracker will add about 25% yearly to the PVs power output. This means that the all modules mounted on the tracker will produce about one-quarter more power on an annual basis because they are following the sun's motion. Actually, the tracker's increase is about 45% in the summer and around 10% in the winter. Trackers cost money just like PV modules. At this moment, it is not cost effective to track less than eight modules. Under eight modules, we will get more power output for our dollar if we spend the money on more panels rather than a tracker. At eight panels in the system, the tracker starts to pay off. There are exceptions to this rule, for example array direct water pumps. If PVs are directly driving a water pump, without a battery in the system, then it is cost effective to track two or more PV modules. This has to do with technical details like the peak voltage required to drive the pumps electric motor. Choosing a PV controller and instrumentation A control is necessary to keep the PVs from overcharging the battery when it is full. There are many types of controls available, so consider wisely before purchasing. Here are some criteria to use when selecting a PV controller. Don't use controls that employ relays. Relays are electromechanical devices that are hundreds of times less reliable than semiconductors (transistors or FETs). Controls with user adjustability are more useful because we can adjust them to suit our system and needs. Controls that maintain a constant voltage while regulating are more effective than controls that totally disconnect the array. Sizing the PV system's battery Size the PV systems battery with a minimum of four (4) days of storage. Consider our system on Agate Flat that consumes 2,480 watt-hours daily. If we divide this figure by our system voltage of 12 VDC, we arrive at a daily consumption of 206 Ampere-hours from the battery. So four days of storage would be 4 days X 206 Ampere-hours per day or 826 Ampere-hours. If the battery is a lead-acid type, then we should add 20% to this amount to ensure that the battery is never fully discharged. This brings our ideal lead-acid battery up to a capacity of 991 Ampere-hours. If the battery is nickel-cadmium or nickel-iron, then this extra 20% capacity is not required because alkaline batteries don't mind being fully discharged on a regular basis. Our system at Agate Flat employs 750 Ampere-hours of lead-acid storage. This is really less than we need, and we often refuse power because our undersized battery cannot store it. Our battery is over ten years old and as soon as it dies, I will replace with a much larger alkaline type. Four days is the minimum storage capacity to install in a new system. Every day of additional storage will pay off in the system's ability to withstand long cloudy periods. In order for a solar system to be totally stand-alone and not require any backup power source, it should have at least seven days of storage in the battery. Those living in places where it is cloudy for extended period might well use an even bigger battery. If you want stand-alone performance from your PV system, then size the battery to withstand the longest cloudy period you ever experience. This can be expensive and it is usually much more cost effective to add a secondary power source, like wind, microhydro, or even an engine/generator. Designing Wind Systems Wind systems are at the mercy of their site survey. Without an extended site survey or real wind data for a specific location, it is really impossible to specify a wind turbine for the system. While PV and microhydro systems are often effectively designed by their users, wind systems should seek help from someone who really knows wind power. While a PV system may limp along with a bad design, wind systems can experience catastrophic and total failure due to poor design and/or installation. Choosing the Right Windmachine Again this is a job for someone with experience with all types of wind turbines. Not only must the wind turbine be well made, but it also must fit the wind conditions at your particular site and must produce the power that the system requires. Modern turbines usually produce some specie of low voltage and only the very large units make 60 cycle, 120/240 vac directly. Here is a list of the turbines that work well. This information is provided by Home Power reader and user reports. Any wind machine made by Bergey Windpower will not only work, but it will last. We've had very good reports on the inexpensive Whisper 1000 watt machines made by World Power Technology. In the smaller machines (under 500 Watts), the Windseeker models from South-West Windpower are receiving high praise from their owners. Well-constructed wind turbines will cost between $1.00 and $2.50 per watt of power. How high is high enough? As high as you can get it is high enough. The higher you place the wind machine, the more power you will get. Consider 50 feet to be a realistic minimum and after that, go as high as you can. Be sure that the tower is strong and well installed. Sloppy tower installation can bring the whole system crashing down. A tower's success or failure is a digital event- either the tower stands or it falls. Make sure it stands. Don't scrimp on the tower, this can cost you your entire wind machine, tower, and possibly any buildings underneath the tower. Guyed towers are more secure and less expensive than unguyed towers. Regardless of the type and brand tower you use, put it up right and make sure that it stays up. Anything less than the best is not even in the running. Choosing a wind controller In almost every case, the manufacturer of the wind machine also makes a regulator for that specific model. So, the user doesn't have to select a regulator because it is bundled in with the wind machine. These controls are shunt types that divert the turbine's output to maintain control of the system's voltage. Diversion regulator schemes are really the only type used, because unloading the wind machine will cause overspeeding and damage to the turbine. Sizing the Wind system's battery The size of a wind system battery storage is determined by the longest period of windless weather. This can be very difficult to determine in advance. For this reason wind systems usually have more days of battery storage than do PV systems. Shoot for a minimum of seven days of storage and extend this to fourteen days if you can afford it. Wind power comes in gusts and spurts, having a large battery makes more effective use of nature's least consistent power source. Designing Microhydro Systems Microhydro systems are defined as hydroelectric systems that produce less than 1,000 Watts (that's 24,000 Watt-hours daily). At the high end, microhydro systems produce enough to run three electrically efficient households. No other form of renewable energy is so reliable or powerful for what it costs. If you have microhydro capabilities at your site, then start smiling now. You have the most constant and inexpensive form of small scale, renewable power. But first, you must have the water. Again (and over and over again) a survey of a site's potential must be done before you can design the system. Fortunately a hydro site survey is easy to do and produces solid dependable data. In a microhydro system the length and diameter of the pipe must also be specified to suit the situation and the turbine. Consider that the pipe delivering water to the turbine is a conduit for power. Using long runs of small diameter pipe will make even the finest turbine ineffective. The situation with pipe in hydro systems is exactly the same as the tower in wind systems. Both provide access to the natural power source and neither are places to save money by specifying marginal components. Choosing the Right Microhydro Turbine Microhydro turbines come in two basic forms. One uses an alternator, just like an automobile. The other uses a permanent magnet (permag) generator/motor. The alternator based machines are for larger systems producing from 100 to 1,000 watts, while the permag units are best suited to systems producing under 80 Watts. Microhydro System Controls Larger systems will use shunt type diversion schemes for regulation. This prevents overspeeding of the turbine and premature wear of turbine parts. Smaller system can also use regulation schemes that unload the alternator when its power is not needed. In all cases, these controls need to be user adjustable. In all cases, the controls should not use failure-prone relays. Sizing a Microhydro System's Battery Microhydro systems are easy to fit with batteries. This is because the microhydro turbine is producing a constant amount of power all the time (24 hours a day). The battery acts as a "flywheel" to smooth out the inevitable peaks of consumption that occur during the day and when using large appliances. If a microhydro's battery can store one day's power for the system, then it usually works as well as many days of storage. Microhydros refill their batteries almost immediately after even a little power is consumed from the battery. As such, microhydro systems are "shallow-cycling" their battery and any ole' thing will last a long time. If you have microhydro potential, then spending your money on good pipe and an efficient turbine is more effective than spending it on batteries. Need more info about designing a system? If you doing this job by yourself, then you certainly do. I refer you to every issue of Home Power published to date. We've been talking about these specific technical details for more than three years and over one thousand pages to date. And we've just scratched the surface. Again, I urge those considering renewable energy systems to consult with someone in the business. I had to figure all this out for myself because when we started out there wasn't anyone in the business. Now, some twenty years later, many of us have learned a great deal about what works and what doesn't. So save yourself much heartache and many dollars, go out and make a new friend who's in the biz. While you still may wish to design and/or install your own system, the help of an adept is invaluable and will save many times more money that it costs. Access Richard Perez, C/O Home Power, POB 130, Hornbrook, CA 96044 ù 916-475-3179.