Storage Batteries
Most of those who build their own alternative energy system use one or more lead-acid storage batteries, which are widely available and seem to be most cost-effective at the present time. New battery technologies are out there, but the price tends to be very high. The most common 12 Volt battery around is the typical automobile battery, available in many stores and even at salvage yards in used condition. While this type of storage battery will certainly work (for a time, anyway) it is not an ideal solution for the home-built system. The reason has to do with the way the battery is constructed, using a large number of rather thin plates to present a large surface area to the dilute sulfuric acid electrolyte in which they are immersed. This vast surface area allows the battery to supply the large amount of instantaneous current demanded by the automobile's starter motor. In short, the car battery is designed to produce a large burst of energy for a few moments, then to be recharged almost immediately by the vehicle's alternator. The problem is that an alternative energy system discharges a battery at a relatively low rate but typically for hours at a time. Thus, the battery may be discharged to a fairly low level before it is once again recharged. This type of behavior, called "deep cycling", is hard on the thin plates of an automotive battery, which will last longest if the discharge is not more than ten percent of the battery's rated capacity. In short, they don't hold up well in deep-cycle applications. Fortunately, suitable batteries are available that don't cost much more than a regular car battery. These are known as marine deep cycle batteries and are specifically designed with thicker plates that can withstand repeated long discharges. They are often used by boaters for running trolling motors, fish finders, radios, lights and the like when a motor with internal alternator isn't being used for a continuous charge. There are also marine starting batteries, but unless they are an especially large deep-cycle style, they are not as suitable for our purpose. Marine batteries are available in several different size groups, priced accordingly. I originally purchased a small one in 2011 with a 75 amphour (AH) capacity for around $60 (plus deposit) and it worked well in my system. I should add that I connected this in parallel with a good, used car battery I had on hand for the additional capacity. Of course, I was careful not to draw the charge on the pair down excessively for the reason cited above. As the days became shorter and the need for additional storage became apparent, I invested in a second, larger marine battery of 115 AH. The three batteries provided me with plenty of power for lighting my home in the evenings and early mornings. I understand that the 6-volt batteries for electric golf carts, when wired in series, make an even better power source.

After having operated the system for about 11 months, the original marine deep-cycle battery went bad, apparently developing a bad connection internally. It was fortunately still under full-replacement warranty and I was able to obtain a replacement rated at 101AH for $10.00, the original part no longer being available. That original, well-used car battery continues to function fine and has been supplemented with a younger car battery, making a total of 4 lead-acid batteries on the system currently. It occurs to me that it is possible that if one generally only uses 10 or 20 percent of a battery's capacity by connecting enough of them in parallel, then purchasing automotive-style batteries having a long warranty might prove to be a better plan.

I ofter a piece of advice about something that is often not known or is overlooked. Lead-acid batteries will lose their capacity if they are left in a less-than-fully-charged condition for very long. During discharge, a coating builds up on the plates that over time serves to insulate them and will eventually render the battery useless. This process is called "sulphation" and it should be avoided. When purchasing a new lead-acid battery, be sure it hasn't been sitting on the shelf for very long. A battery will slowly self-discharge in storage, so check to see if there's a manufacturer's date code on the one you're thinking of purchasing. Some batteries may use a two-digit code in which a letter indicates the month and a number the last digit of the year. In this case, the letter "A" will stand for January, with each successive month being one letter furthur along in the alphabet. At home, recharge your battery as soon as possible after it has been used. Don't wait until it is getting low to connect it to the charging circuit. Unlike certain small consumer batteries we are all familiar with, the lead-acid type prefers to stay fully charged for a long life expectancy. Also, don't forget to check the battery's electrolyte level periodically and to add distilled water (available by the gallon in grocery stores) when necessary. Do not use tap water in a battery as the minerals and other impurities will damage it.

Monitoring Voltage and Current
You may wonder if there's a electronic device available to tell you how much charge is left in your storage battery, sort of a "battery gas gauge". To the best of my knowledge, there aren't any such devices that are inexpensive and actually work well because such an indicator would have to be microprocessor-controlled and probably calibrated by the user. The reason for this is that the voltage measured at the terminals of a lead-acid battery is the product of a number of variables. These include the following things:

• The battery's temperature. The measured voltage will be less when a battery is cold than when it is warm.
• The internal resistance of the battery, as determined by its capacity and internal condition, including age.
• Whether or not the battery is receiving a charge and the charge rate, if it's being charged.
• How recently the battery has been charging, if it's not presently.
• Whether or not the battery is under load (supplying power) and the discharge rate, if it is under load.
• How long it has been since the battery has been discharging, if it's not presently.

This is not to say that you cannot use an ordinary voltmeter to get a rough idea of the amount of charge left in a battery, but whatever voltage you read must be placed in the context of everything mentioned above. It is aid that a fully-charged 12-Volt lead-acid storage battery should measure nearly 12.7 Volts after it has been sitting idle, that is, without having been charged or discharged within the past several hours. While that is a good rule of thumb, it is not set in stone. My initial pair of batteries measured about 12.9V under those very conditions, and would still exceed 12.7V after a 3A discharge for an hour's time. When I added the third battery, this resting voltage rose to more than 13VDC, so apparently it is a factor of overall capacity and/or the number of batteries connected in parallel. You may also read that a battery measuring 12.06 Volts is at 1/2 charge and that when it reads 10.5V it is fully discharged. Again, this assumes there has been no activity for several hours' time. The problem is, we generally want to know the state of charge while the battery is doing something - either providing electricity, or being charged back up.

One time-honored way of determining the charge left in a storage battery is to use a battery hydrometer to measure the specific gravity of the dilute sulfuric acid electrolyte. This device, similar in appearance to a turkey baster with a short hose on the end, is used to withdraw some fluid from one of the cells. A weighted float inside the tube is marked with specific gravity readings that correspond to the surface of the liquid and the higher in the electrolyte it is able to rise, the stronger the battery's state of charge. Since the reading will vary somewhat according to the temperature of the fluid, better hydrometers will have an integral thermometer and a temperature-compensating chart for reference. Because the reading doesn't vary due to the load on or charging of the battery, it can be taken at any time. However, this method can be messy, inconvenient, and there is some danger from the acid. It also cannot be used on a battery that does not allow access to the cells.

What you can do to get a pretty good idea of a battery's state of charge is to keep a record of the voltage measurements you make on your own particular system, thus giving you something to go by. For example, when you know for a fact your battery is at full charge, write down the voltage you measure. Then, momentarily hook up the various loads you will typically be using and record the terminal voltage as each one is connected. That will give you a good idea of what your fresh battery puts out under different load conditions. Next, you can operate the battery with a known load, preferably of around one Amp, for a number of hours. Calculate the amp-hours of energy consumed and figure out what percentage this is of the total, rated capacity of your battery (or batteries). Connect up the different loads again and repeat the voltage measurement for each at this reduced state of charge. You can duplicate these steps several more times, always recording the results. What you want to end up with is a chart that shows what your battery puts out for voltage at each of several points in its discharge, for each of several load conditions. In practice, suppose that one evening you are running two CFL lamps that together are consuming 2.4 Amps and you see that your battery's voltage has dropped to 12.25 Volts. Having been through the procedure outlined above, you should now have a reasonably good idea of the state of the battery's charge by comparing it to the set of figures that most closely matches the present situation of voltage versus load. This is possible even though you have not made the measurement while the battery was at rest.

You must measure the battery voltage right at its terminals, not part way down a wire through which current is flowing to charge the battery or power some lamp or device. Since it may not always be handy to go right to the battery, it is quite okay to make the measurement from a remote location a reasonable distance away, provided the meter is the only thing presently connected to the interconnecting wires. In my system, I have a dedicated cable from the battery to a project box in my living room into which I can plug the probes of my digital multimeter. The 2-conductor cord you use for a remote monitoring cable doesn't have to be especially thick as long as it's a reasonable length, since the load the meter presents is very, very small. You do want to fuse it appropriately, however, since it connects directly to the battery.

Another thing you're probably going to want is a way to monitor how much current is coming to the battery while it is charging and leaving it when it is discharging. A dedicated Ammeter connected in series with the wiring to one terminal or the other is ideal, but again, may not be convenient. In any case, such a meter must be of the type that can read current flow in either direction and be adequate for the maximum amount of discharge current expected.

There is an easy way to make your own remote current monitor using the digital multimeter you already own (or probably soon will). This technique relies on the established voltage drop across a piece of wire of a particular material, cross-section and length. In this case, it will be a piece of 12-gauge copper wire, which can easily be taken from some Type NM or UF cable of the kind used in wiring houses. A 7.5-inch piece of this wire will yield a voltage drop of close to 1mV (one millivolt, or 1/1000th of a Volt) when a current of one Ampere flows through it. Since most digital multimeters have a 200mV scale, this is an ideal setup. Simply permanently connect one end of the piece of wire directly to the positive or negative terminal of the battery (it doesn't matter which), and the other end to all the wires that would normally connect to that terminal. In other words, this special piece of wire goes between the battery and the wiring connected to it. You will actually want to cut the wire a little long, because you want to be able to solder one conductor of a 2-wire cord close to either end of it in such a way that there are 7.5 inches of the 12ga. wire between the two solder joints. The cord for remote readings can now be run to a convenient spot where the probes of a meter can be plugged into the free end. Again, the cable you use doesn't have to be especially thick as long as it's a reasonable length. With this arrangement, you will be able to read the amount of current flowing in the battery circuit at any time, without investing in a dedicated ammeter. The multimeter's display will show a positive or a negative value, depending on how the meter is connected and whether the battery is charging or discharging. Just note which polarity indicates which condition. Using the meter's 200mV scale, you should have a resolution of 0.1A (or 100mA), and you read the number directly as Amps with no conversion necessary. In other words, if the display reads "1.6", you have 1.6 Amps of current flow.

One piece of test equipment you will probably want to invest in is a so-called "carbon pile" battery load tester. This device consists of a ventilated metal box containing a low-resistance wire element strung between insulators, a voltmeter and a momentary toggle or rocker switch. To operate it, one connects the very large alligator clips at the end of two short cables to the terminals on the battery, then presses the switch for a few seconds. This presents a heavy load to the battery, typically of 100 Amps, causing the reading on the voltmeter to drop to a lower value. The voltage under load will be directly proportional to the capacity and internal condition of the battery under test, thus giving a good indication of how well the battery is doing. The meter scale contains marked zones of different colors that show where the pointer should rest for a good battery of various ratings. Obviously, it is important that the battery be in a fully charged state prior to the test in order for the results to be valid. I purchased one of these for my system and use it every few months to make sure my batteries are up to snuff.

Charge Controllers
Unless your sources of alternative energy put out very little current, you're going to need a way to regulate the voltage supplied to the system and how much of a charge the storage battery or battery bank receives. There are many commercially-made charge controllers available in the marketplace, but I have no experience with them. Some styles have three sets of terminals to which the solar panel, battery and load (your wiring for lights, etc.) are connected. They are configured so that the solar panel is disconnected once the voltage at the battery terminals reaches a certain point and the load is disconnected when the voltage falls below a preset level. If selecting this type of controller, be certain it is rated for the amount of current your system is providing from the generating source as well as the load. Depending on what is connected to it, a small DC-AC inverter rated for 400 Watts may draw more than 40 Amps! It would also be nice if the cut-in and cut-out voltages are adjustable rather than fixed.

I am using a more basic control unit that I built myself from circuit information found on the Internet at the URL http://www.otherpower.com/images/scimages/69/CHARGE.GIF. It uses three ICs - a 7808, an LM339 and a 4001. It also has an IFR540 (or IFR510) transistor which drives (provides a ground connection for) a high-current 12V relay at the output that does the actual switching. The circuit includes trimpots for setting the voltage at which the relay is de-energized to connect the charge circuit to the battery and the voltage at which it energizes to open the charge circuit. The published circuit actually includes two FETs and two relays, but my system uses only one in a SPDT configuration. I set the former point at 12.6V and the latter one at 14.4V, which has worked out well. There are LEDs to indicate the present status as well as two pushbuttons to force the device to its opposite state. This feature is handy when, for example, I have reached full charge but am now connecting a load to the battery and wish to resume charging. A change to the circuit I found necessary was to power the LM339 from unregulated 12V rather than the 8 Volts, as otherwise I was not able to set the charge disconnect voltage any higher than 14. I suspect this problem could also be overcome by using a 7809 regulator in place of the 7808. I also added a power switch so that the charge controller can be shut off at night or when there is no danger of overcharging the battery. One reason I went with this design initially is that I sometimes had a micro wind turbine connected, and it's best with these not to disconnect them from a load for reasons of safety. Therefore, when the relay energizes at full charge the wiring from the alternative energy input was connected across a high-wattage bank of resistors to dissipate some of the energy. Were I not using wind power in my system, this wouldn't have been necessary. I suppose the circuit could easily be doubled to add the feature of removing any load from the battery if its voltage got too low, since only half the sections in the 14-pin chips are presently being used.

After using the controller for a few months, it became apparent that the batteries were not completely charged when their terminal voltage neared 15V and the relay activated. Therefore, I added a low-resistance, high-wattage resistor in series with a beefy diode as a bypass around the relay contacts when open. In this way, the batteries continued to receive a "trickle" charge. This modification improved things and I eventually paralleled a second resistor and diode to increase the current when the days became shorter. Still, I was not fully satisfied with this arrangement, so perhaps 22 months down the road a furthur improvement was made. This was done by eliminating both load and bypass resistors and instead sending the relay-controlled electricity through a voltage regulator and series diode configured for an output of 13.8 Volts. In this way, once the voltage at the batteries nears 15V and the relay activates, the solar panels are connected to the input of the regulator circuit to provide, through the diode, a constant 13.8V to the batteries. This voltage is sufficient to gradually top them off without overcharging. The high-power series diode is included on the output to keep current from flowing back into the regulator once the input voltage has dropped too low or is disconnected. Since the internal resistance of this diode causes a loss of roughly half a Volt under load, the regulator's output is actually set higher than 13.8V to compensate. I am using a regulator module available from Asian vendors on the Internet for as little as $10.00, shipping included. Good for 100 Watts (or more with fan cooling), the circuit board measures only about 60mm x 52mm and contains a small terminal strip on each end for input and output connections. A multi-turn variable resistor is provided for setting the desired output voltage. This product is referred to as a DC-DC "buck" or step-down converter and is of the switched-mode (or switching) variety. Whereas a linear voltage regulator essentially acts as a variable resistor and therefore converts the excess energy to heat, the switched-mode style uses a pulse of variable size to produce the desired output. This method is much more efficient and therefore produces less heat and is considerably less wasteful of energy. The same technology is being used in the compact and lighter-weight power adapters we see today. The only drawback is the potential production of radio-frequency interference (RFI). This new charging arrangement is superior, for given an adequate input, the voltage presented to the batteries (and therefore the charging rate) doesn't substantially diminish when loads of up to 7 or so Amps are connected to the system during daylight hours. This was certainly not the case under the previous arrangement, when even a small load would divert most (or all) of the charging current. I suppose one could simplify things by eliminating the charge controller entirely and just keep the solar panels or other source of power connected to the voltage regulator at all times. The downside to doing this would be a reduction in the initial charge rate due to internal losses in the circuitry, plus a maximum system current draw limited by the capacity of the regulator.

According to what I've read, it is best not to charge a lead-acid battery at a rate higher than 10% of its rated amphour capacity. In the case of a 75AH marine deep cycle battery, that would be 7.5 Amps. The typical micro-system with one or two 60W photovoltaic panels should pose no problem for a single battery, but do keep this in mind if your system is larger. Charging current will be divided among the batteries in a bank according to the amphour capacity and internal condition of each. I am inclined to believe that some of the issues regarding short battery life are related to an excessive rate of charge or overcharging the battery, so I do advise caution here. I had to add very little water to my batteries in a year's time, which is a good indication that they have not been overcharging.

Supply and Demand
It should be obvious that one cannot consume more electrical energy than the system provides, but not so common knowledge that all of the power going into a storage battery isn't going to come back out. That is due to less than 100% efficiency in the electrical and chemical conversions going on inside the battery. In fact, I've read that the charging efficiency of a lead-acid storage battery tends to decline the closer it gets to a state of full charge. This is a very important concept to remember, since it must be taken into account when you calculate charge vs. draw in your system. My personal feeling is to plan on a 15% loss of energy in the charge/discharge cycle.

Another thing to bear in mind is that the rate of charge from an alternative energy source is by no means a constant value. For example, a 60 Watt solar panel may only put out 0.6W in the early morning before the sun hits it, 42W at midday, and 28W by 3pm as the battery draws less current. The charging current will naturally decline as a lead-acid storage battery's internal voltage slowly rises towards the value of the voltage being supplied. Now, if it happens to be a rainy day, the same 60W panel may only produce 7.5W at it's peak output. Therefore, you can see that there is no way to precisely calculate the generation and storage of electricity, only some educated guesswork. One thing I'm hoping to see become commonplace is an affordable, microprocessor-controlled electronic device that can keep a tally of incoming and perhaps outgoing electricity, basically a DC watt-hour meter. Depending on your particular situation, you may find it necessary to add more solar panels and/or storage batteries to meet demand. Each of these items may be connected in parallel (positive to positive, negative to negative) to increase the capacity of electrical generation or storage. If you find that your battery (or battery bank) often doesn't get fully charged during the day, a second (or even third) source of charging current is probably in order.

Once the battery (or bank) has been brought to full charge, we at least have some idea of what the capacity is, assuming it is in new or excellent condition. Here again, battery capacity is likely to diminish some over time. According to what I've read, deep cycle batteries should not be discharged below 50%* of their rated capacity in amphours, and automotive batteries not more than 10% below theirs, if you want them to last for long. That means the total available output from a 75AH marine deep cycle battery like one of those I am using is really about 37 amp-hours. Assuming that you operate four 13W CFL lamps from an inverter connected to the battery, it would appear that you could safely keep them lit for a total of about 8 hours and 20 minutes (at a 4.5A** draw). This is somewhat of a simplification, however, as a battery's AH rating is often based on either a 1 Amp or a 20-hour discharge rate (check the label). The rated AH capacity decreases as the current draw is increased, so realistically we might be looking at only eight hours of illumination before reaching the halfway point in the battery's charge. A low ambient temperature will also reduce the battery's capacity below its rating. You may wish to do furthur research on the characteristics of lead-acid batteries so you have a better understanding of the principles I've mentioned here.
* Another source claimed only 20% for a marine-type deep cycle battery. While this figure seems rather low, bear in mind that the less you draw down a battery the more charge/discharge cycles you are likely to get from it.
** In this real-life example, 4.5 Amps is the actual, measured current drawn on my particular system. It takes into account the extra power consumed by the inverter's circuitry above and beyond that used by the lamps.