Newbie PV System using MPPT

[estel3an]

Good day,

My name is Esteban Gonzalez and I'm from Guatemala, I have a little bit of experience designing PV Systems but I never have used MPPT Charge Controllers. I Was looking at the TS-MPPT-60 of Morningstar and I belive its the charge controller I'm looking for. The PV System I'm designing is for supply a need of 13.5kWh/day with 3kW of Panels with 2 days of autonomy. But I have a few doubts, how is the best way of sizing a PV System using a MPPT charge controller?
The panels I want to use are evergreensolar 215W and 24V (nominal) (Imp = 7.43, Vmp 29, Voc = 35.6 and Isc 8.12).

According to my calculations with 2 panels in series (48V) and 7 rows of this configuration I've got 57Amps peak. But I dont really know how to size the batterys for this system. if some one can guide me with that. What could be the best configuration I could make for my battery bank?



Best regards.

[Cariboocoot]

Welcome to the forum.

Your need of 13.5 kW hours per day is quite large for an off-grid system. I do not believe you will be able to achieve this with a 3kW array. Normally we would expect:
3000 Watts * hours of good sun (say 6 where you are) * 0.52 over-all system efficiency = 9.36 kW hours per day AC.
You would probably need 5kW to 6kW of array to achieve that much daily 'harvest'. Keep in mind that hot panels will not produce as much power as cool panels, and Guatemala is hot!

The battery bank would also need to be substantial for that much power, even on 48 Volts. 13.5 kW hours AC is going to be 15 kW hours DC due to conversion losses, and then you need to add in the inverter consumption. You are very close to needing 16 kW hours DC capacity. On 48 Volts that is about 334 Amp hours. At 25% depth of discharge that would be a very large 1336 Amp hour 48 Volt battery bank. It would need about 134 Amps of peak charge current potential. This would mean around 8kW of array and two large charge controllers such as the Outback FM80 or MidNite Classic 150.

Now about those panels. You have fourteen of these 215 Watt Evergreens. Their Vmp is too low to work on a 48 Volt system with just two in series: Vmp 29 * 2 = 58 Volts; a 48 Volt system needs a Vmp around 70 to work properly, especially in a hot climate because it is the Voltage that will fall as the temperatures rise. If you put three in series this problem is solved, but you will not be able to use all fourteen panels; you either have to drop two or buy another one (I'd opt for the latter). Likewise more than three in series and you will have trouble with the array Voc being too high for the controller unless you opt for one of the MidNite Classic 200 or 250 models. Again, though, the number of panels you have does not divide up evenly into this configuration.

So you have two options: reduce the load demand considerably or increase the array size significantly. If I were you I'd try for the first option before looking at the second.

[estel3an]

Thanks cariboocoot,

I will try to recommend do changes in the loads, replace them with more efficient aquitances. The place where the instalation will be made is in a cold place ( lower than 20°C ) its in the mountains. I can get more panels too but do you think is possible to use 3 of the panels evergreen in series and 7 rows? because the Nominal Maximum Solar Input is 48V and the Max. Solar Open Circuit Voltage is 150V for the TS-MPPT-60.

I have more questions of the MPPT. How can you size the battery bank using this kind of charge controllers?

Best regards.

[Cariboocoot]

Let's do some basic math and see what happens.
Three of the panels in series would give you: 645 Watts, 87 Vmp, 106.8 Voc, 7.43 Imp, 8.12 Isc per string.
If you had seven of these strings in parallel: 4515 Watts, 87 Vmp, 106.8 Voc, 52.01 Imp, 56.84 Isc for the whole array.

That array on a 48 Volt system using an MPPT controller has an output current potential of about: 4515 Watts * 0.77 (panel & controller efficiency) / 48 (minimum system Voltage) = 72 Amps. One 60 Amp charge controller would be stressed with that much current; it would 'clip' the 12 Amps above its output maximum and that power would be lost.

The best way to set up a system is to size the battery bank first, based on the daily load demands. Then you calculate the amount of panel & charge controller capacity necessary to recharge that bank. The good "middle-of-the-road" rule-of-thumb number is to try for a peak charge current of 10% of the battery bank's Amp hour capacity (based on the "20 hour" rate). This is how I arrived at the 1336 Amp hour 48 Volt battery bank in the previous post, its need for 134 Amps peak charging current, and the resulting 8kW array (134 Amps * 48 Volts / 0.77 efficiency = 8353 Watts).

Frankly I do not think the Morningstar MPPT 60 is going to work in this application. The power demands are quite large and would have to be reduced by more than 50% before a single 60 Amp controller becomes viable. You should look at the MidNite Classic line: http://www.solar-electric.com/misocl.html One of these can handle 80 Amps @ 48 Volts. For the previously mentioned set-up you would still need two such controllers as the combined current is so high.

By the way, it is also sometimes possible to split systems into multiple smaller set-ups which can handle the total power requirements without having to rely on any heavy current situations as above. This would depend on what kind of loads there are and whether some can be grouped together separate from others.

[vtmaps]

I have more questions of the MPPT. How can you size the battery bank using this kind of charge controllers?

You are doing this backwards.
First figure out your loads.
Next pick an appropriate battery voltage and capacity.
Then, and only then, select the appropriate number of panels and the electronics needed to keep the battery happy.

Cariboocoot explained that a 13.5 kwh daily load requires a 48 volt battery with about 1400 ah capacity and about 8000 watts of panels. You will need to put the panels in strings of 3. Then you can choose a couple of large charge controllers. Cariboocoot gave you valuable design advice.

--vtMaps

edit: I see that Cariboocoot beat me to it again

[stephendv]

The good "middle-of-the-road" rule-of-thumb number is to try for a peak charge current of 10% of the battery bank's Amp hour capacity (based on the "20 hour" rate). This is how I arrived at the 1336 Amp hour 48 Volt battery bank in the previous post, its need for 134 Amps peak charging current, and the resulting 8kW array (134 Amps * 48 Volts / 0.77 efficiency = 8353 Watts).

Esteban, before you get a fright at the 8kW of panel number, bare in mind that this "rule of thumb" is a very VERY rough shortcut to determine the panel size and does not take your location into account. As you can imagine, 8kW of solar in Norway does not provide the same amount of power as 8kW in Nigeria or Guatemala.

Do you require those 13kWh/day in winter as well as summer? Using PVWATTS http://rredc.nrel.gov/solar/calculat...ATTS/version1/ you can estimate the amount of kWh you can expect for each day or month from a given solar array. The worst solar month seems to be June where 1kW of panel will produce roughly 2.5kWh per day. So to get to 13kWh in June, you'd need roughly a 5.2kW array.

[Cariboocoot]

Esteban, before you get a fright at the 8kW of panel number, bare in mind that this "rule of thumb" is a very VERY rough shortcut to determine the panel size and does not take your location into account. As you can imagine, 8kW of solar in Norway does not provide the same amount of power as 8kW in Nigeria or Guatemala.

True, to a certain extent.

The extra-long days near the Equator add to the total number of hours of sunlight, but they are not all usable hours for panels on a fixed mount. Further to that the higher temps found in those areas midday reduces panel Voltage and thus total output power with any MPPT type system (including grid-tie). Finally, total Watt hours harvested per day is not the same as creating enough current to bring a battery up quickly. If you do not have sufficient current available the battery bank won't recharge properly and its life will be shortened. It's a rarity, I'll grant you.

Winter and Summer aren't too different near the Equator, btw. Guatemala has some interesting climate due to elevation differences (just like B.C.) and being between two oceans. Some areas are desert, in fact. Normally the temperatures don't go below 20C and can go above 40C. If you're high enough up a mountain you can get below freezing. So he will have the fact that elevation helps panel output (areas 3,000 to 6,000 feet if that's the location) whereas the heat will hurt. Likewise if it's in a dry area that will help, but in some spot that gets doused by hurricanes - major problem.

Ain't it fun trying to figure this stuff out by remote control?