[Power] Charge controller ?

[Richard A. Smith]

On 03/06/2010 06:23 AM, Mikus Grinbergs wrote:

>> The answers to such questions always depend on what you want to do with
>> the system. Supplying power to 2 XO's can mean you want to fully
>> recharge the XO's from empty or it can mean that you just want to run
>> the XOs without using much of the XO's battery.
>
> I had hoped for a guideline to a system "good enough" for practical use.

I haven't put much thought into a system for powering one's and two's 
beyond the individual 10W panel use.  That's part of what this list is 
for and the procedure is mostly the same so here goes:

> Putting the above advice into a calculation (the XOs being connected to
> a large external battery):  8 watts * 2 XOs * 6 hours_used_per_day = 96
> watt-hours required each day to keep those XOs running (discounting
> losses).  Suppose the solar panels are rated at 24 watts (96 / 24 = 4
> hours of charging needed per day) -- let's say the controller is rated
> for 10 amps.  Is that a reasonable capacity for the controller ?

Wh/day is the critical first number.  Ballpark rules of thumb will get 
one into trouble but as a general guideline for back of the napkin type 
calcs after you account for the rest of the system your necessary PV 
array wattage will need to be around 1/3 of your Wh/day.

The controller size will be based on your max PV wattage under max 
insolation.  24W array under 1 sun @ 12V is 2A.  @9V (assuming your 
controller will operate that low) its 2.6A.  You have to think about the 
range of voltages because if you try to draw more power than you can 
source the voltage will drop but you still may be able to source 24W. In 
which case your current will increase.

Making a wild guess about your solar insolation levels lets say you 
range from 0-2 Suns (ie max of 2000W/m^2) so in the middle of summer 
high noon your 24W array is a 48W array.  If during that period you try 
draw more than 48 and your PV array drops to 9V then its 5.3A.  So yeah 
a 10A controller would do fine.  Controllers are usually specified in 
wattage since many of them be set up for operation on a range of output 
voltages most common are 12, 24 & 48.

The exact answers to the cals above need the detailed specifications of 
the PV panel and the charge controller.

All that said 24W is too low.  Pin needs to be > Pout.  Without that if 
you have half a day without sun and use up an extra 48 Wh then you will 
need to decrease your usage the next day by 1/2 to allow your battery 
bank to full recharge.

>> Next question is how
>> long do you want to be able to go without recharging from the sun?
>
> I thought how long the XOs could keep going without recharging from the
> sun depended upon the size of the external battery - not upon the rated
> capacity of the charge controller.

Yes exactly.  Its the size of the external battery that we need to 
determine and that depends on how many days you can live with without a 
full recharge.  In your above system you need max output for 4 hours and 
you need it _every_ day without fail.  During the peak of the  summer 
that's probably possible but what about winter when your insolation 
drops by 1/2 or 2/3rd's or when it rains or snows. (depending on your 
location).

So you batteries have to be sized for how long you want to be able to go 
under the worst of conditions.  Lets say its 1 day.  So when it rains 
you still want all day operation.  Thats 96 Wh used [1].  The next day 
you need 96 + the 96 you lost on the day of no Sun.  So for a full 
recovery of the bank in 1 day you really need an array of 48W.  If you 
can accept recharge in 2 days then you can decrease the necessary 
wattage to about 36.  Hmmmm... 96Wh * 1/3 = 32 so you can see where the 
1/3 ballpark guess comes from.

If you want to go all week and you have crappy sun then you can see that 
your both your battery bank and PV array would have to be huge.

Once you have sized your PV array such that you can keep your batteries 
charged where you want them then you plug those numbers up in the 1st 
section and size your charge controller.

Once you have that set you can figure up your Ah needed for the battery. 
  Its not an exact conversion but Load Wh/system V = Battery Ah.  96Wh = 
8Ah. But you lose 10-20% of that energy in the process

Here's the kicker usually makes larger solar setups very expensive.  You 
need 8Ah of battery but you really have to provide a bank that's a 
minimum of 2x that size or 16Ah.

The reason is that battery cycle life is based on % Depth of Discharge 
or %DoD.  The most common battery is lead acid and it really hates it 
when you go into a large %DoD.  If you take a LA battery below 50% DoD 
on a regular basis you will destroy it.  So for 8Ah of juice you need a 
battery bank of 16Ah and you only go down to 50% DoD.  50% is the bare 
minimum.  Going to full 50% DoD a lot will greatly shorten your bank 
life so thats expected to happen a lot then you need even more than 16Ah 
something closer to 30% DoD which would be ~27Ah.

If however you were to get a non-LA chemistry then %DoD can be less of a 
concern. Say if you used LiFePO4 like what we use in the XO you can go 
to 100% DoD without much concern.  Your batteries + controller will be a 
lot more expensive on the front end but you won't have to replace the 
batteries as soon.

If you google a bit for solar calculator or solar array sizing or visit 
many of the renewable energy sites they have online calculators or 
downloadable spread sheets to help you play the what if games.

[1] Its actually more than that.  96W assumes perfect energy conversion. 
  There will be some loss in the charge controller, distribution, and 
due to the charge efficiency of the battery.  Guessing for this is hard 
but 10-15% is probably a reasonable number. That makes your load during 
battery charging in the 106-113 Wh range.

-- 
Richard A. Smith  <richard at laptop.org>
One Laptop per Child