Building new Off Grid Home in Panama

Re: Bulding new Off Grid Home in Panama

We are only talking about a couple of percent when looking at 57.6 or 58 or 59 volts charging. Realistically, if you are withing 10% of the "real numbers", you are usually doing well.

The difference between 57 volts / 48 volts = 1.1875 or ~ 20% larger array--That is significant.

When doing these calculations from an "engineering" point of view, I usually want to make sure your system will not under-perform vs your expectations. But, frequently you end up with a bunch of derating factors that may tend to push you towards oversizing your system... For example, as string of "typical" deratings:

0.81 panel derating * 0.95 controller derating * 0.80 battery derating * 0.85 inverter efficiency = 0.52 end to end derating.

0.81 panel derating and 0.95 controller eff are pretty real. 0.80 is used for flooded cell and 0.90 can be used for AGM. In real life, a flooded cell can be >0.90 efficient and an AGM can be upwards of ~0.98 efficient. But, a lot of this depends on your particular usage... For example while flooded cell batteries are >>0.80 efficient when cycled from 0.50 to 0.80 state of charge, they probably drop below 0.40 efficient at >90% state of charge (equalization is close to 0.0 useful efficiency).

For the Battery Bank, the neat thing is when you have more loads, the average battery SOC drops, and therefore becomes more efficient--sort of supporting your heavier loads (poor weather, guests, miss-planning, etc.).

Solar panels on a MPPT charge controller are less efficient in hot sunny weather--but fortunately, that is when we get the most sun. They are somewhat more efficient in cold water (near freezing and below by 10-20%)--which helps out in winter. Etc.

Anyway, we could add 3% wiring drop, another few percent for the "hot" copper wiring losses on the roof (copper resistance rises with temperature), another couple percent for dirty panels, etc...

But, that would drop our 52% typical derating even lower (or force even a more expensive system on you). That is why I almost always bring up the discussion of alternative power (utility power, generator, etc.) into the discussions... Making the array 50% larger to account for a few days of worst case loads and/or a few more days of pure solar during bad weather--The additional arrays (even if you double them) are not going to generate appriciable power during a week of dark/stormy weather. On my 3.5 kW array which will generate around 15-19+ kWH on nice days, will drop as low as 0.5 to 1 kWH per day during stormy weather.

And, you should not plan on "needing to use" 100% of your predicted power every day... Some days you will have more sun/energy available--other days less. You will just have to plan your usage and be flexible (when to wash, when to irrigate, etc.).

For me, I have grid power (pure GT system--So my "solar numbers" are always maximum power the array can output--no full battery bank in the afternoon reducing solar array production). But for an "off grid" person, there are going to be times when the weather is bad. And your choice is to either cut back power usage dramatically (so you don't damage the battery bank). Or spend a few dollars (or ten's of dollars) per day on fuel to get you through the dark times. (and/or install alternative power sources like wind or hydro to help).

What you do during the dark times is your choice... My suggestion is to accept that you will get close to 100% of your electricity from solar for ~9 months of the year and size the solar array (and battery bank) to support that. And accept that you will be using the generator/alternative power/conservation at times during the 3 months of "winter".

When I use PV Watts, I look for the poorest 3 months of production, toss those, and then find the fourth poorest month and use that as the "break even" number. For example, use Managua Nicoragua as being close and I can use PV Watts to give us 12 months of seasonal sun averages. Fixed array facing south with 12 degree from horizontal tilt:

Month      Solar Radiation (kWh/m2/day)
1      5.61     
2      6.10     
3      6.29     
4      6.22     
5      5.12     
6      4.40     
7      4.70     
8      4.84     
9      4.81     
10      4.71     
11      4.96     
12      5.06     
Year      5.23

Their lowest fourth month is October at 4.71 hours of sun per day (still pretty good, usually we use ~4 hours of sun for planning in North America). Lets say you want 8kWH per day during most of they year (if you have heavy summer A/C or other seasonal power needs, need to check into those numbers too):

8,000 WH * 1/0.52 typical system efficiency * 1/4.71 hours of sun = 3,266 Watt array minimum suggested

But, you also need to look at the battery bank charging requirements too (large battery bank needs large charging current). Assuming an 800 AH @ 48 volt battery bank:

800 AH * 58 volts charging * 1/0.77 panel+charger derating * 0.05 rate of charge = 3,012 Watt array minimum
800 AH * 58 volts charging * 1/0.77 panel+charger derating * 0.10 rate of charge = 6,026 Watt array nominal
800 AH * 58 volts charging * 1/0.77 panel+charger derating * 0.13 rate of charge = 7,834 Watt array "cost effective maximum"

Because you have pretty high amount of "average sun", your array does not need to be too large. However, if you have a lot of power use and need a larger battery bank, the battery bank charging requirements push you towards a larger array.

Again, a lot of this depends on your power usage... If mostly during the day, and you can cut back on usage during bad weather, then a smaller battery bank may be a better deal. If you charge during the day and use power at night, and cannot cut back on usage too much (fridge, lighting, business), then a larger battery bank is going to be the curse.

In reality, you may only have a week or two of generator run time during winter (may cost you a few hundred dollars in fuel)--Because for many locations, even during the winter you will still have more sunny/clear days than heavy cloud days. You will need a backup genset anyway--So just have enough fuel to support those days when it is needed.

After a year or so--You can always revisit the Solar Array/Wind turbine vs genset question and decide if you want to invest in more panels/RE power or continue with the present fuel usage.

Sizing the battery bank to the loads, for me, is the most critical step. Batteries are there for 5-8 years and you don't really want to add a mix of old and new batteries over time--You end up with batteries failing randomly over time and the mess to diagnose/prevent a bad battery from killing others/and replace every few months (kind of like a mix of old and new tires on a car--you are having to always futz with them).

Don't get too deep in the few percent side of the calculations. You can easily affect your power usage by a factor of 2x (larger or smaller) based on your energy habits and electrical appliance selection. I carry out the calculations to 3 or 4 places, not because my calculations are that accurate... I do it so you can repeat my calculations (and catch my errors) and follow where each set up numbers comes from and is used next.

For the most part, your power measurements (from array, load usage, etc.) will be, at best, 5-10% accuracy. You will never be able to measure the improvement by using heavier copper wire with 1% loss vs 3% voltage drop loss. But--Everything adds up as seen above into the "terrible" 52% end to end AC system efficiency. So every point in the project from panels to wiring to inverters to loads--You are looking to get the most efficient devices/configurations you can--Because the losses all add up very quickly.

-Bill