Comparing Mono & Polycrystalline Panels output based on the lowest months irradiance

Ray,

Here is the estimated sun for your area using PV Watts and 43 degree tilt:

http://pvwatts.nrel.gov/pvwatts.php

Month	Solar Radiation ( kWh / m2 / day )
January	6.32
February	6.75
March	6.43
April	5.96
May	4.51
June	4.40
July	3.95
August	4.96
September	5.63
October	6.03
November	6.32
December	6.38
Annual	5.64

Your power needs:

• 227 Watts * 12 hours per day 2,724 Watt*Hours per day

Note--Your loads may be voltage dependent. I.e., during charging, your batteries will be at ~58-60 volts during the day--And can even be as high as ~62 volts during equalization/very cold weather. If your load is an AC inverter--They are a "constant power" load... I.e., follows Power=Voltage*Current. As the voltage rises, the current draw actually falls.

2.7 kWH per day is not a big load. However, you do have to decide if you will turn off the loads during bad weather (very sunny area on average), will use a genset, or you want to oversize the array to minimize genset as much as possible. If this is your "base load"--You should plan on only using ~65% to 75% of your system's predicted output--Or 1/0.75 to 1/0.65 "oversizing" of your array:

• 2,724 Watt*Hours per day * 1/0.65 oversizing to support base loads = 4,191 WH per day base load oversize

Battery bank supplying base load for 1-3 days of no sun--2 days seems to be a good cost/performance pick. 50% maximum discharge:

• 2,724 WH per day * 1/48 volt battery bank * 2 days storage * 1/0.50 max discharge = 227 AH @ 48 volt battery bank

You have not talked about the loads here--Surging loads (like starting a well pump) can affect the sizing of the battery bank (also, AC inverter losses, inverter running 12 hour vs 24 hour per day, etc.).

Running loads during daytime vs night time... If you run the loads during the day, the batteries are not cycled very deeply (just during early/late during the day). This can help them last longer. However--The daytime loads take energy directly from the solar array and reduce charging current (i.e., a day or two of bad weather recharging can take longer to bring the battery bank backup). If you run the loads during the day, you still want to see ~5% to 13% minimum rate of charge available during the day. However, if you are planning on oversizing the array to reduce genset usage (and solar panels are at historically cheap prices)--This may not be a big issue. Sizing the array to your battery bank (larger battery bank "needs" a larger solar array):

• 227 AH * 59 volts charging * 1/0.77 panel+controller deratings * 0.05 rate of charge = 870 Watt array minimum
  
• 227 AH * 59 volts charging * 1/0.77 panel+controller deratings * 0.10 rate of charge = 1,739 Watt array nominal
  
• 227 AH * 59 volts charging * 1/0.77 panel+controller deratings * 0.13 rate of charge = 2,261 Watt array "cost effective" maximum
  

If you plan on using power during day--Then "adding" to the array above:

• 227 Watt load * 1/0.77 panel+controller dratings = 295 Watt "additional" solar array to support daytime load+charging

And then sizing the array based on your amount of sun (and loads). If you have more than 3-4 "hours per day" of sun during the winter--It is very easy to size the system for the amount of loads during winter without needing a genset (other than possible stormy weather). Assuming this is a "required" daily base load (cannot drop usage during bad weather), and a "pretty large" base load derating of 65% (just another "fudge factor"):

• 4,191 WH per day (base load upsizing) * 1/0.61 typical DC system eff * 1/3.95 hours of sun (July) average = 1,739 Watt array "oversized" for reliable base load supply
  

Plus ~295 Watt "extra" array/solar panels if you run loads during the daytime (loads+battery charging).

This is a pretty "massively" over sized system (solar array) because of the 65% base load / minimize any genset runtime. During summer this thing will produce much more power than you will need on average:

• 1,739 Watt array (+295 Watt daytime charging???) * 0.61 panel+controller derating * 6.0 nominal non-winter sun = 6,365 Watt*Hours per non-winter nominal day

The next issue is the voltage rating of the solar array... You say you have 36 volt panels... We need to make sure that these are Vmp~36 volts. This voltage rated panel is around--But many panels are Vmp~30 volt rated. Your array should be a minimum of Vmp~72 volts (2x 36 volt or 3x 30 volt panels in series).

And the maximum Vmp-array (cold weather Voc panel voltage) depends on the brand/model of MPPT charge controller. The typical Outback/Schneider/Midnite/etc. 140-150 VDC max input controller would have Vmp~array of ~100 VDC maximum (where you may have sub freezing weather).

Because you are in a very sunny region (even during winter)--A 1,739 Watt array will certainly work well. If you are OK with using a genset during bad weather/deep winter (or turn off loads during these times), you could possible go with as low as a 870 Watt array (or 870+295 Watt array for daytime loads). And see how it works out (add panels later?).

These days, with panels being relatively cheap (Australia, perhaps not so cheap)--Over paneling will save you on battery cycling and genset+fuel usage. Also, you will be spending much less time (during bad weather) "managing" your system (loads, genset, etc.).

-Bill

Regarding panel tilt... The solarelectrichandbook does the angle measurement from "vertical".

The PV Watts program (and most people) do the measurement from horizontal.

So, 34 degrees from vertical would map to (90-34=) 56 degrees from horizontal with PV Watts.

DC to DC converters are similar to AC inverter in day to day operation (constant power devices--Although, there are different designs for DC converter control circuits--And other operational modes are possible).

56 volts charging voltage (aka 14.0 volts for a "12 volt" battery bank). That is a pretty low charging voltage for a deep cycle flooded cell battery (typically ~14.4 volts for AGM and ~14.75 volts or so for flooded cell). And that is at ~25 C -- Room temperature. As batteries get colder (for cold climates/outdoor banks), the charging voltage will increase at ~ 5mV per degree C per Cell under ~25C. (24 cells for a 48 volt bank or 24*5mV=0.120 volts per degree C for your bank).

For a 48 volt system--Your "loads" should be able to take at least 60 VDC (ideally >62 VDC) for charging (and equalization).

PV Watts uses (I think) a combination of latitude (physical/seasonal solar position). And the local weather (PV Watts picked the location as "closest" (or similar) to your location's weather) to estimate your location's "hours of sun".

Many MPPT charge controller mfgs. have online string calculator tools to help ensure that the solar panels (and their series/parallel wiring) will be appropriate for their controller (and "worst case" weather conditions).

For various reasons, I would highly suggest that you use 2 days of storage and 50% maximum discharge. If you discharge a battery to 50% state of charge daily--At 10% rate of charge it will take ~8-10 hours of charging to fully recharge. And with solar, it is difficult to get that many hours of "usable sun" from solar. Although, you can use a physical tracker to get more charging time during the day, or a split array, one half pointing North East, and the other pointing North West.

Automated Gensets--If you can avoid the requirement and operate it manually--I would highly suggest that. The more you automate your system, the more things that can go wrong.

-Bill

Just was replying to:
Automation, while sometimes needed for remote sites/etc., are adding more points of complexity/failure.

I am suggesting even a smaller genset to carry through more than ~2 days of bad weather (dark clouds can reduce output to as low as 5% of typical daily output--almost nothing).

It is always a trade-off between loads, panels, batteries, and possible battery bank damage from over-discharging/lack of charging vs the costs/hassles of a genset as backup power.

And Ray, did I miss replying to something? Of course, you have to give me a bit of time to make the longer replies--We are all volunteers here.

Take care,
-Bill

That is why I try to show my work and the fudge factors I use for the design.

I obviously do not know your needs and desires. All I can do is make guesses from my side.

One of the problems with "engineering" a design--Everything tends to towards worst case design. Not knowing earlier that you can reduce your loads to 33% during poor weather--I was using 1/0.75 and 1/0.65 as "fudge factors" to derate for minimum base loading.

Of course, because you can reduce your base loads dramatically--Then you really do not need the 1/0.65 (my worst case guess) at all... You can drop that factor from the daily solar power estimate.

However--If you keep the 2 days + 50% max discharge as a suggested battery bank--Then you look at the 5% to 13% typical rate of charge--Which does "push" for a larger solar array if you keep the 10% minimum recommended rate of charge for most deep cycle flooded cell batteries. You can get away with 5% minimum rate of charge (you need around 2.5 to 5% rate of charge for equalization) in a very sunny climate (which is why two array calculations--One based on battery bank size and the second based on loads+hours of sun).

Take care,
-Bill