Maker Pro
Maker Pro

Solar Panel Configurator

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
I have a solar storage setup composed of six 80 watt panels, a cheap ebay regulator, and a 12V leisure battery hooked up to a 12VDC--->240VAC inverter.

I've connected all six panels together in parralel. The voltage output from the array varies from near zero (16 millivolts) in the dead of night, to around 21 volts in full direct midday sun.

At dawn and dusk, there is a fairly extended period where the voltage output from the panels transitions through the 12V point.

The regulator will only start charging the battery when the input voltage rises above 12V.

What I'd like to do, is try connecting two sets of three panels each in parralel, to give two seperate strings which each vary 0-21V. And then, depending on the voltage, connect the strings together either in series (to give a higher voltage when the sun is low) or parralel (keep the voltage down but take advantage of extra current when the sun is high).

I've figured out how I could do this pretty easily using a relay with two sets of ganged contacts and some creative relay wiring - but how would I accomplish automatic control?

I was vaguely thinking about placing two high value resistors across the output of one string of panels to form a potential divider, then feeding the mid point of the divider onto the base of a transistor, which would then be used to switch the relay coil. Or, maybe, some form of operational amplifier based comparator.

My design skills aren't sufficient to come up with a workable circuit though. Does anyone know of an existing design, or could suggest a basic workable schematic?
 

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
attached: rough schematic of my idea so far (no doubt not a workable circuit)
 

Attachments

  • solarconfig.jpg
    solarconfig.jpg
    62.5 KB · Views: 134

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Firstly, the relay is wires appropriately to switch the panels between series and parallel

However, the simple circuit you use to determine when to switch the relay has some issues. It won't switch quickly, and has no hysteresis (other than any in the relay itself). When the output is near the switching point, it may switch back and forth repeatedly. This is especially true if the load changes.

I would probably either leave them in a series configuration and use a buck converter to reduce the voltage, or (and I've seen this done) leave them in parallel and have a boost converter which is enabled when the voltage falls below a preset value.

In either case you add an additional diode drop, but in the latter case the regulator is only used when the voltage falls below some lower limit.

edit: the amount of power you'll actually get in this situation is very limited.
 

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
Thanks for your reply Steve. Yes I had wondered about what might happen near the switching point.

Buck converters and boost converters - these are animals I have not yet met, I will go and do some reading. If anyone can point me further in the right direction for these, that would be handy.

What would be the nature of the power limitation?

Very happy to consider a wildly different solution to the one I'd originally envisaged. Maybe should I ditch the cheap ebay regulator altogether?

It's one of these:

http://www.ebay.co.uk/itm/Solar-Cha...lator-Safe-10A-12V-24V-UK-STOCK-/360684563165

The main reason for doing this is the UK's dull winter days. With panels connected in series, doubling the voltage means that I would reliably get sufficient voltage on dull days. I originally said >12V but looking at the regulator specs it looks like it might actually need to see > 14V from the panel array to charge the battery. But if I leave the panels connected in series permanently, and there's a bright spell, I worry I might fry the regulator.

I tested the panel output yesterday, a pretty dull rainy day, and I was getting around 9V from 3 panels in parralel. So doubling it would get it up to 18V and hopefully let the regulator start to fill the battery albeit at only half the current. I figure half a dull day's current, is better than no charging current at all..
 
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Are you using an MPPT charger? (It doesn't look like it).

Getting an MPPT charger will improve efficiency greatly.

If you get one that is rated for the voltage of all your panels in series, even better.

Once the open circuit voltage falls substantially the total power available is low -- you won't get anything near the rated current from the panel.
 

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
No I don't think it does MPPT, at least it doesn't say so. I looked briefly into MPPT & just about understood what was going on, seems to be something about automatically adjusting the load on the panels to try and keep them operating at the 'knee' of their IV curve?

I'll go read up some more on the MPPT, if you think that's the way to go :)
 

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
Would this be suitable?

www.ebay.co.uk/itm/MPPT-10A-Solar-Controller-Regulator-12V-24V-100W-240W-10-30-more-Power-Boost-/171104980485

My only worry would be that sometimes, I'm likely to be trying to draw pretty heavy currents. I have a 2000 Watt / 4000 Watt peak, pure sine wave inverter which I'm planning to use for powering small mains appliances in the kitchen - 400 Watt steamer, 1000 Watt mini oven, 750 Watt mini kettle. Would there be anything to stop me running heavy cables direct from the battery to the inverter, rather than using the MPPT's built-in load terminals?

i.e. would the MPPT still operate correctly if it couldn't "see" the load directly?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
The unit probably has a mosfet to switch the load on and off and they have picked one capable of 10A.

In this case, you would be better off using the "load" terminals to switch a relay capable of turning your load on and off.

This will allow the regulator to turn off your load when the battery is discharged.

Yes, run heavy cables direct to the battery (via the relay).

One possible caution is that the heavy load you draw may drop the battery voltage sufficiently to drop the battery voltage to the point where the charger thinks the battery is flat. When it switches off the load, the battery voltage will recover, the load will be switched on, the load drops the battery voltage... Bad news. If this happens you may have to wire the load direct to the battery and monitor the charge level yourself.
 

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
Thanks. I've been doing some reading & understood sufficient to know I probably need a ready-made MPPT controller which features a built-in boost converter as part of the design.

Unfortunately most of the cheap far-eastern stuff on ebay falls a bit short on reliable in-depth technical info.

The best I've found so far is a German design which features a buck converter, but it's relatively expensive.

I do know and understand properly what it is I'm looking for now though, so 2 thumbs up for pointing me right.
 
Top