[Power] [support-gang] solar power design for 5-to-10 XO-1 laptops in Haiti (or elsewhere!)

[Richard A. Smith]

Re-quoting the original question for the power list:

 >     Will a 33V input break a 10-Amp Charge Controller like this? I'm
 >     working with 12V batteries and I would like to use a 136W 24V
 >     (actually outputs about 33V). I have anther charge controller with
 >     similar specs but I'm afraid to rely on it because of the same
 >     uncertainty. Can you recommend something which will definitely and
 >     reliably work to step the voltage down?

 > 
http://amazon.com/Sunforce-60031-Digital-Charge-Controller/dp/B001DZONCW

On 07/23/2013 12:29 AM, George Hunt wrote:
> I think you must be talking about a solar panel which puts out 33V open
> circuit. Am I correct?

33 Voc seems about right for a 24V panel.  However it doesn't have to be 
totally unloaded for it to reach that voltage.  All you have to have is 
the current draw from the device be very low.  Say like in the float 
phase.  You need to look at the VI curve for the panel.

> The output power from a solar panel may be considerably less, if there
> is a lot of mismatch between output voltage and load voltage.  That is
> why MPPT (maximum power point tracking) controllers are used.
>
> The controller that I have played with is basically a switch which
> connects the panel to the load, until the battery voltage (12v) rises to
> 14.6. After that it disconnects, and reconnects again when the battery
> voltage falls to 13.5.

Here's the way most cheaper controllers (ie non-MPPT) work:  The output 
is a Pulse-Width-Modified (PWM) signal. Think of PWM as a very fast 
switch.  So just like George says its directly connecting your battery 
to the PV panel and watching the resulting voltage on the battery. Its 
doing this many, many times per second.  You can think of the battery as 
big capacitor that will smooth the voltage spikes.  So even though you 
are hitting the battery with a bunch of ~24V spikes the over all time 
average of the voltage is kept to the 13-15V range.

So as long as you don't exceed the voltage rating for the parts on the 
charge controller and don't have a panel with more wattage than the 
controller can handle it should work.

The controller is rated as a 12V device.  I'm guessing that the 
capacitors and such in the device are only going to be rated for 25V. 
Unless the controller data sheet specifically says it can handle higher 
voltages I highly doubt it can take a 33V input.  Its even worse than 
that.  Because you are cutting the signal off and on you will actually 
have over 2x the input voltage due to AC effects.  So for a 33V input to 
the PWM you need a system that can deal with 66V.

> I think the best thing for you to do is to  put the solar panel out in
> strong sun, and measure the current into a somewhat discharged 12 volt
> battery.  You don't want the current to be more than the current rating
> of the controller.  I expect that it will work fine.  I've never seen
> more than 65% of rated output from my system.

I'm not as optimistic as George.  My guess for that setup is that it 
would work for a while and then fail prematurely due to capacitors on 
the charge controller that fail due to voltage spikes beyond their rated 
voltage.

You should check with the manufacturer and see what the maximum 
operational input voltage is.

> Because of the relatively high voltage, I'd expect that you may be able
> to get less of the available power into your battery than if you used
> something like
> http://www.amazon.com/Tracer1210RN-Solar-Charge-Controller-Regulator/dp/B008KWPGS6/ref=sr_1_5?ie=UTF8&qid=1374553109&sr=8-5&keywords=solar+controller+mppt.

You will absolutely get less than the rated power.  Perhaps a _lot_ 
less.  It depends on the VI curve of the PV panel.

PV devices have an output thats based on the voltage across the panel. 
In order to get a specific amount of current you have to have a voltage 
specific voltages.  If you look at the VI curve for the panel you will 
see a peak where the voltage and current are such that the panel 
provides maximum power output ( power = voltage * current ).  This is 
called the Maximum Power Point or MPP.

In the PWM controller setup you are forcing the system to operate at the 
voltage that you are charging the battery with because you are 
connecting the battery directly to the panel (for brief moments).  So 
when your controller charges your battery at 13.7V then that means thats 
where your PV panel is operating at.   So to know how much power you can 
get you have to look at the VI curve find where the curve intersects 14V 
and then look at what current the panel will output at that voltage then 
current * 14 will give you an approximation.

I say approximation because in reality rather than a constant 13.7V its 
a complex waveform of the unloaded PV voltage ramping down to 13.7 
during the PWM on time.  14V however should be close enough.

So to maximize your power from a PV panel using a PWM based charge 
controller you really need PV panels that match the battery charging 
voltage.  The panels you are thinking about using would be much better 
matched for a 24V system rather than a 12V system.

> I don't know what the specs are for the controller that you have may be.
> The one at Amazon at the url above says it can handle PV inputs as high
> as 100V.

The URL George gives is an Maximum Power Point Tracker (MPPT) 
controller.  MPPT controllers have an extra front end DC-DC converter 
that converts the input voltage into the battery charging voltage.  They 
also have smarts that adjust the allowed current draw such that PV panel 
is allowed to operate at its MPP.

So long story short is that if you want to use a PV panel with a voltage 
that is over about 2x your battery voltage you need a MPPT controller. 
While the Sunforce controller might work its going to be sub-optimal and 
I suspect will fail prematurely.

Hope this helps.

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