08-17-2014, 06:04 AM
I get by with 2,100 watts of panels (nine 240 watt nominal rated panels) and eight Costco golf cart batteries in series making a 48 volt 210 amp-hr bank. This amp-hr rating is for a nominal 20 amp discharge rate. It is desirable to match panels and battery bank such that the bank is charged at a rate between 5% and 13% of the rated amp-hrs, ignoring the mismatch in units. For example 5% of 210 is 10.5 amps and 13% is 27.3 amps. When I had only 3 panels in series in a single string putting out around 8 amps I was technically below the recommended minimum. Now with 9 panels in 3 strings I am up near the recommended maximum. Apparently you can go higher but it gets to be less efficient as you can only cram so much power into the batteries. Below the minimum rate of charge the batteries don't get stirred well enough which has longevity implications even if you make up for it by charging longer.
There is a school of thought that says going with a smaller battery bank and working it harder gives a better return on investment. Within reason if you cut the battery bank in half and discharge deeper you reduce the life span of the bank but not by half so there is a sweet spot in there if you can find it. Obviously discharging your bank by 90% daily will kill it pretty quick, but surprisingly many battery banks are killed with kindness, being feebly charged because they are not really drawn from.
When sizing solar systems one quickly progresses through the lower voltage battery banks. Almost no one should be using 12 volts. 24 volts is almost always a better choice for anyone using more than a few hundred watt inverter and frankly 48 volts seems like a no-brainer to me now that I went that way myself. When putting together the battery bank it is very preferable to buy larger batteries and keep them all in series rather than make parallel strings to gain amp-hours. As such there are a fixed number of permutations, like 2, 4, or 8 costco GC batteries or 2, 4, or 8 L-16 batteries. The L-16 batteries have the same footprint as the GC batteries but are twice as tall, have not quite twice the amp-hours and are somewhat more than twice the price. The benefit comes in that they are supposedly more durable and will last longer and that you can get nearly double the amp-yours without going to parallel strings. They have more room at the top for water, which seems like a trivial thing until you ruin a set of batteries by letting the level get too low. Things like that.
The bottom line is that the batteries are going to die sometime anyway and larger batteries cost more so you must sharpen your pencil to determine what works out for the best. Another alternative is adding more panels pointed east and west to start charging earlier and keep charging later reducing the demands on the battery bank. A tracker would do the same thing but panels are relatively less expensive these days so virtual tracking becomes an option. The point is that if the panels and controller charge hard in the morning and mid-day and go into float for the last few hours you could re-aim some of the panels without actually losing amp-hours into the batteries, allowing the panels to directly carry the misc. parasitic loads of the household and inverter earlier and later into the day, reducing both the actual amount that the batteries are discharged and the time they spend while discharged.
There is a school of thought that says going with a smaller battery bank and working it harder gives a better return on investment. Within reason if you cut the battery bank in half and discharge deeper you reduce the life span of the bank but not by half so there is a sweet spot in there if you can find it. Obviously discharging your bank by 90% daily will kill it pretty quick, but surprisingly many battery banks are killed with kindness, being feebly charged because they are not really drawn from.
When sizing solar systems one quickly progresses through the lower voltage battery banks. Almost no one should be using 12 volts. 24 volts is almost always a better choice for anyone using more than a few hundred watt inverter and frankly 48 volts seems like a no-brainer to me now that I went that way myself. When putting together the battery bank it is very preferable to buy larger batteries and keep them all in series rather than make parallel strings to gain amp-hours. As such there are a fixed number of permutations, like 2, 4, or 8 costco GC batteries or 2, 4, or 8 L-16 batteries. The L-16 batteries have the same footprint as the GC batteries but are twice as tall, have not quite twice the amp-hours and are somewhat more than twice the price. The benefit comes in that they are supposedly more durable and will last longer and that you can get nearly double the amp-yours without going to parallel strings. They have more room at the top for water, which seems like a trivial thing until you ruin a set of batteries by letting the level get too low. Things like that.
The bottom line is that the batteries are going to die sometime anyway and larger batteries cost more so you must sharpen your pencil to determine what works out for the best. Another alternative is adding more panels pointed east and west to start charging earlier and keep charging later reducing the demands on the battery bank. A tracker would do the same thing but panels are relatively less expensive these days so virtual tracking becomes an option. The point is that if the panels and controller charge hard in the morning and mid-day and go into float for the last few hours you could re-aim some of the panels without actually losing amp-hours into the batteries, allowing the panels to directly carry the misc. parasitic loads of the household and inverter earlier and later into the day, reducing both the actual amount that the batteries are discharged and the time they spend while discharged.