The Battery Bank
A group of batteries is used to store power so that the generator can have a rest.
It is also not very efficient to run a 10 Kilowatt generator to power one reading light.
Designing the battery bank depends on these factors:
1. Bank Voltage. This is dictated by the inverter you select. The D.C. input voltage of the
inverter must match the voltage of the battery bank. To get this voltage, we will
generally use multiple batteries in series (see below).
2. Total Bank Capacity. The total amount of electricity you want to store, measured in
Kilowatt-hours, will determine the total number of batteries you must use. The
capacity is additive, meaning that twice the number of batteries provides twice the
capacity, so total capacity is simply the capacity of each battery multiplied by the
number of batteries.
3. Battery make and model. Initial price, availability, longevity, and a number of minor
technical factors will influence your choice of batteries. These issues are discussed
in some detail below.
Bank Voltage
Commercially available inverters from Trace and Heart are designed for battery banks
of 12, 24, and 48 volts. Most of the batteries we will use in our off-grid power systems will
by 6 volt, so we must arrange the batteries "in series" to obtain these
higher voltages. Here&s a picture.
 |
The diagram shows two six volt batteries wired in series. This creates a battery bank
of 12 volts with a total capacity that is twice the capacity of one of the batteries.
|
Batteries can also be wired "in parrellel" to create banks that are twice the
capacity without doubling the voltage. We will generally combine parellel and series
wiring to create both the voltage and capacity that we want.
 |
Here we have four six volt batteries connected to create a bank that has four times the
capacity of a single battery at a voltage of 12 volts.
|
Bank Capacity
The number of batteries you will need is governed by how much power you want to be
able to use before having to recharge the battery bank. You must also accept that the
batteries can not be fully discharged without permanently damaging them, so while
a battery my have a total capacity of 1200 watts, only a portion of that, say about half
or less, is usable without reducing the life expectancy of the battery. The issue of
life is taken up in the next section. For now, let us assume that only about half of the
bank total capacity is usable capacity. This implies a recharge point of 50% discharge.
What usable capacity do we want to have in our system?
If we take our usable capacity and divide by our average hourly power consumption we
get the number of hours our power lasts without running the generator. In general, we will
want enough usable battery capacity to last at least one entire day, preferably more.
For example, let's say that we expect to use 6 KWH each day and want to run the
generator no more than once a day. This means we must have one day of usable
juice in the battery bank. The total capacity of the bank must
be twice this, 12 KWH, as we have assumed a 50% recharge point.
If we use Trojan T-105 batteries, which have a total capacity of about 1.35 KWH, then
we will need 9 batteries. This is one too many. If we use 8 batteries we will be a bit
short, but may be alright so long as our estimated usage is a bit high. We could also
go to either 10 or 12 batteries depending on the voltage of our inverter.
Battery Specifications
Batteries are generally specified in terms of ampere-hours instead of kilowatt-hours. To
convert one to another requires the battery voltage. Watts, the electrical measure of
power, is equal to amperes times voltage. A 200 amp-hour battery rating for a six volt
battery is a power of 1200 watt-hours, or 1.2 kilowatt-hours.
The table lists some key specification data for a variety of Trojan batteries. I am in no
way associated with Trojan. They happen to be leading manufacturer of deep cycle
batteries and they are the batteries I use in the systems I have built. For the advanced
student, the capacity shown in the table is the 20 Hour rate--if you don't know what that
means, don't worry about it, it is not important for our purposes.
| Model | Amp Hours | Life Cycles | Weight | Cost |
| T-105 | 225 | 754 | 61 | $58.75 |
| T-125 | 235 | 650 | 66 | $64.65 |
| T-145 | 244 | 625 | 71 | $94.05 |
| J250 | 250 | 650 | 72 | $97.60 |
| J305 | 305 | 625 | 91 | $127.00 |
| L16 | 360 | 650 | 113 | $161.10 |
| L16HC | 395 | 650 | 121 | $183.00 |
There are a number of interesting relationships that can be observed from this data.
First, capacity increases with increase in weight. This is because the design of these
batteries are largely the same, it is just the physical size that changes with model.
The table shows amp-hours per pound for each of the batteries from the table above.
There is a small variation in this number between models, but as a general statement a pound of Trojan battery is a pound of Trojan battery, no matter which model
you buy, you are buying the same "stuff." Intuitively, the lower this number, the more durable the battery design should be, since the primary weight of a battery is
the lead plates.
| Model | AH/lb |
| T-105 | 3.7 |
| T-125 | 3.6 |
| T-145 | 3.4 |
| J250 | 3.5 |
| J305 | 3.4 |
| L16 | 3.2 |
| L16HC | 3.3 |
Another observation we can make is the cost of an ampere-hour of capacity.
The table shows the computation of amp-hour per dollar for each of the Trojan
batteries listed previously. The figure shown is the number of amp-hours of
capacity purchased by each dollar. The T-105, by far the most common deep
cycle battery in the world, provides 3.83 amp-hours for each dollar compared
to the 2.16 amp-hours provided by an L16HC. The advantage of using higher
capacity batteries is the reduction in number of wires and connections, but
if cost is an issue, the T-105 provides 77% more capacity per dollar compared
to the L16HC.
| Model | AH/$ |
| T-105 | 3.83 |
| T-125 | 3.63 |
| T-145 | 2.59 |
| J250 | 2.56 |
| J305 | 2.40 |
| L16 | 2.23 |
| L16HC | 2.16 |
A final number worth computing is the life cycle cost. The measure of this is the
product of the capacity and the number of cycles divided by cost. This is a measure
expressed as AH/$ as in the table just completed, but this table relates the cost
of amp-hours stored and returned over the entire life of the battery, rather than the
capacity of the battery bank as in the previous table.
Again, the mule T-105 wins hands down against the thoroughbred L16. Cheaper really
is better!
| Model | Life AH/$ |
| T-105 | 2887 |
| T-125 | 2363 |
| T-145 | 1621 |
| J250 | 1665 |
| J305 | 1501 |
| L16 | 1453 |
| L16HC | 1403 |
One caveat. As the number of batteries in a bank is increased a number of undesirable
features begin to crop up. Because L16's can build a higher capacity bank with fewer
batteries, the more expensive batteries may be the right choice in some situations.
When? Generally, when you can avoid parellel strings it might be a good idea and
where the maintanence personal will be poorly trained.
Battery Life Expectancy
Batteries will last
a given number of cycles of charge and discharge for a certain depth of discharge.
As the depth of discharge is increased the life of the battery (measured in number
of cycles) is reduced.
The table illustrates the life expectancy of the Trojan T-105 with respect to average
depth of discharge in each cycle.
|