Mitsubishi Air conditioner for 10hours

You ought to look into solar panel powered mini split air conditioners, some use a big 48V
battery for limited operation after sunset. Others have a line backup. Bruce Roe

Air conditioning is going to be a really big load for solar. What hours (10am - 8pm ?) do you need to run the AC ? How much overlap with the sun, daylight hours only, or run it from 10pm - 6am for sleeping? These factors will greatly influence what can be done. What about a cloudy day, will you need to run the AC without sun at all ?

I understand your point but right now I have only this option

Hello Sir,

Load timing is 5hours day time 11am-4pm, night timing is 9pm - 2am

We get 6-7hours bright sunshine everyday throughout year.

We hardly get cloudy days, maybe 1 or 2 days per year so this is not problem.

Find big 2 volt batteries would be big help...then size balance of system around those batteries...not the other way around...

One sensible way to start is to calculate the actual design cooling load the A/C needs to meet and then, if the A/C unit is already in place, determine how much power and energy the A/C will draw to meet that load. Then, with what sounds like fairly consistent solar energy availability, design and size the PV system and ancillary equipment including batteries to meet the design load and design conditions as best suits your lifestyle and budget.

You need more information than a forum can give you.

1.) Determine a design load first.
2.) Get a book on off grid residential solar design.

Use a forum to fill in knowledge gaps created by your self education.

Well, if you take 2600W for 10 hours per day, with 2 days of reserve battery capacity and ~40% DOD. Your're looking at between 48 and 72 very expensive batteries. Not to mention the large solar array you'll need to charge them, and the ~5kW or so inverter.

See attached PDF.

You can model it yourself at


If you have gasoline, diesel, or natural gas available, this might be a less expensive option for the first several years.

Once you work out the exact load (wattage and duration) then you can begin to estimate the size of the solar array you would need to keep the batteries charged. But charging batteries is very inefficient. So if you're consuming ~30kWh per day (10hrs @ 2600W). You may need at least 2x that in charging capacity.

Based on your latitude, array orientation, and inclination, you can estimate the productivity of your PV array using PVWatts (https://pvwatts.nrel.gov). Since it sounds like you are somewhere hot and sunny, I'd expect you'd need an array in the 7 to 9kW size, roughly speaking.

You're probably looking at $40,000 to $60,000, perhaps more, for such a system.

Wholesalesolar dot com, Renvu dot com, Outbackpower dot com should all have off grid system estimators to help you estimate the inverter and PV pieces.

JPM is right (as usual) -- I just did Watts x time. With no consideration for how often the unit is cycling on and off. I just assumed the compressor was running 100% of the time. If you have data as to how often the compressor is actually running, then you may be able to cut your battery bank size significantly -- if the compressor is only running say 50% of the time. BUT, given your daily temps -- 38-50C. I would expect that little AC unit is very busy.

It is going to be to your great advantage to:

1) Insulate as well as you can first (including low-E glass if applicable)
2) Get a more efficient, load matched A/C first. (Load matched means that at the highest load/hottest day it is at 100% duty cycle)

All that will save you a lot of money on any solar system (and on your power bills right now.) Then once you do all that:

3) Get a grid tie solar power system. There are ways to do net zero with an A/C or a whole house load.

That will give you the most benefit. However, if you have extra money you don't need, then you might consider:

4) Go hybrid (battery+inverter.) This will be very expensive and require a lot of care, but if that's your thing, go for it.

Will this be a ground mount solar array? You'll need commercial or at least residential grade panels. So 72-cell or 60-cell panels. Strings will be several hundred DC volts each (in series). Lots of potential panel options. You'll want 300W to 400W panels, and you'll probably need 20 to 30 of them. Lowest cost per watt is probably the best choice -- but something with a better thermal coefficient might be helpful. In no particular order -- Jinko, LG, Panasonic, Hanwha, Trina, etc.

Why PWM? I'm not sure they even make those in this scale. I would stick with MPPT.

On #2: For residential A/C equipment the 100% duty cycle is usually reduced to around 75 -80 % run time. That increases equipment size. Also, and partly as a reason, but not the entire reason for the 80% run time rule of thumb, it's necessary to include the latent (condensing) load in the calculations. That latent load, depending on location and design dew point, can add as much as 30-40 % to the cooling loads.

If the OP is in a location with a low design dew point, the design dew point will lower the latent HVAC loading, but will still be a consideration. Anyway, a 100 % design point runtime is probably better reduced to 90 % or less.

That does not mean however, that oversizing is a good idea for A/C. As a matter of design fact, oversizing often produces a "clammy" feel to a conditioned space because equipment of conventional design but large(r) capacity will not run long as long at less than design conditions - most all of the time - In those cases conventional equipment may not run long enough to condense as much water vapor from the air being cooled during what will be shorter runtime cycles.

With most A/C, there is a goldilocks size, but an 80 F outside temp. with a, say, 75 F dewpoint will present a different cooling load and scenario than a 100 F outside temp. with a 30 F. dewpoint as might be encountered in the desert SW. Fortunately, the sizing usually has a rather wide window (no pun intended), but more or greater oversizing is never a good idea.

Best idea for the OP, just like every other HVAC design situation is to start by calculating a design heat load, seeing what/which lifestyle changes are possible/manageable, and then what sealing/insulating measures make sense, and then recalcing a heat load after those changes, and then properly sizing the A/C equipment.

Then, after all that, begin a PV design, hopefully grid tied if possible because in terms of long term or just about any scenario, if available, grid tie will beat off grid hands down for cost effectiveness and ease of operation.

I would also add that a variable speed fan can help you manage relative humidity vs temperature on less-than-worst-case days.

"Find big 2 volt batteries would be big help...then size balance of system around those batteries...not the other way around..."

Bingo

AC on solar lol. Buy I nice generator an your set.