diy solar

diy solar

Run 5k BTU Window AC on solar during daylight only

jwh92020

New Member
Joined
Jan 22, 2020
Messages
97
I have a friend that has property in the desert outside Las Vegas, NV. He has grid power to the house and is currently using a combination of a swamp cooler and window ac to cool the house. The swamp cooler works ok (at best) to keep the house "tolerable" 90+ inside on a 115 degree day, and he uses the window unit at night to cool his bedroom. His electric bills ran him $400 in August and September, then dropped to $325 in October. I have no idea how efficient the swamp cooler motor is (it was on the house when he bought it). I'm trying to help him configure a system to run a 5000 btu ac the draws 450 watts. He would only run this unit during daylight hours in the main living area, then run the bedroom unit ad swamp cooler at night. Like a lot of folks, he doesn't have much budget to put toward this. I figured he could pick up 4 - 6 San Tan Solar 250 watt panels (37v) and a small grid tie inverter. He works all day away from the house, so the only thing that would run on the panels would be the air conditioner. Any advice on where and what to pint him towards would be appreciated.
 
Probably can be done for ~$1200 using an AiO.

Assuming $0.12/kWh, it uses $0.36/day. 1200/.36/365 = 9 years to break even assuming he uses it 365 days per year. He doesn't. Let's say he uses it during the hottest 6 months. That's a 4.5 year break even.
 
You have to have a battery for it to work. Even if it is a small battery (24v 50ah?) it has to be there to keep the power flow when the solar dips below the power needed to run the AC. Unless you can guarantee 500w of constant output from the panels (You can't, trust me) it will not work.
 
Probably can be done for ~$1200 using an AiO.

Assuming $0.12/kWh, it uses $0.36/day. 1200/.36/365 = 9 years to break even assuming he uses it 365 days per year. He doesn't. Let's say he uses it during the hottest 6 months. That's a 4.5 year break even.
I'll show him your numbers. What is an AIO?
 
You have to have a battery for it to work. Even if it is a small battery (24v 50ah?) it has to be there to keep the power flow when the solar dips below the power needed to run the AC. Unless you can guarantee 500w of constant output from the panels (You can't, trust me) it will not work.
You have to have a battery for it to work. Even if it is a small battery (24v 50ah?) it has to be there to keep the power flow when the solar dips below the power needed to run the AC. Unless you can guarantee 500w of constant output from the panels (You can't, trust me) it will not work.
With 1500 watts of panels, you don't think he would get a constant output of 500 watts (as long as the sun is on them)? I was looking at grid tie inverters. I don't understand much about them - still researching.
 
Ok, that I understand. A friend suggested a grid tie inverter. He said they don't need batteries(?). I know nothing about them, but a quick search showed 1000 watt units in the $150 - -$250 range.

The problem with those units (I have two) is that if you ever actually back feed power to the grid, the power company will stomp your a$$. I have one that I use for a specific purpose that can feed back about 350W; however, based on my whole-house energy consumption I'm always using more than that.

With your proposed array, it's likely likely that he'll back feed power to the grid at peak solar.
 
Feeding back wouldn't be a good thing. When people have whole house solar and produce more than they use, how is danerous feed back prevented? I don't understand how a single 24v 50 amp hr battery (as suggested by another member) can service the ac.
 
With 1500 watts of panels, you don't think he would get a constant output of 500 watts (as long as the sun is on them)?
I can't speak for Vegas, but here in Central Texas we get occasional clouds and all it will take is one cloud to drop your output below 500w and that will cause the AC to cut out. You don't want to be doing that often or the compressor will suffer.

To give you an idea, I have a 2250w array and some days a single cloud on an otherwise completely sunny day, will knock my output down below 500w, sometimes as low as 200w. With a battery that will get you at least an hour of run time (500wh of usable), it will keep this from happening.

The easiest (not least expensive) setup is similar to what I have. You take a LV2424 or similar Growatt ($700) and hook it to a simple 24v battery ($500), and the panels ($500ish). You can run up to 9 250w panels (3S3P). You hook the AC Grid into the unit to charge the batteries if there is no solar and then you can run up.

As mentioned above though, it will take a while to get your money back. The nice thing though is as long as the sun is up, you will have AC regardless of the condition of the grid. And eventually you can add larger batteries and power more stuff, for longer.
 
Feeding back wouldn't be a good thing. When people have whole house solar and produce more than they use, how is danerous feed back prevented?
It is not a good thing, unless you have it setup with the power company and wired properly. I believe it is actually illegal to do this as you can fry the lineman if the power goes out. Most good AIO units will not feed the grid as they are isolated.

I don't understand how a single 24v 50 amp hr battery (as suggested by another member) can service the ac.
The battery is not running the AC unit, it just gives you a bit of cushion in case the panels are not putting out enough power. Basically if the panels are putting out 300w, the other 200w will come from the battery. Once the sun pops back out, it will provide more than the 500w needed for the AC, so the rest will charge the battery back up. It is basically a reserve when you need it. 24v 50ah = 1000wh, so it will run the AC for maybe 2 hours before it dies. So you have 2 hours of wiggle room if the sun is not cooperating...
 
Thank you for the explanation. Makes sense now. His long term plan is to add more solar for less dependence on the grid. Another friend suggested using a 60 amp mppt controller, 2000 watt pure sine inverter and a 200 ahr Lifepo battery and 1000 watts of panels as an inexpensive way to accomplish his immediate goal.
 
It is not a good thing, unless you have it setup with the power company and wired properly. I believe it is actually illegal to do this as you can fry the lineman if the power goes out. Most good AIO units will not feed the grid as they are isolated.


The battery is not running the AC unit, it just gives you a bit of cushion in case the panels are not putting out enough power. Basically if the panels are putting out 300w, the other 200w will come from the battery. Once the sun pops back out, it will provide more than the 500w needed for the AC, so the rest will charge the battery back up. It is basically a reserve when you need it. 24v 50ah = 1000wh, so it will run the AC for maybe 2 hours before it dies. So you have 2 hours of wiggle room if the sun is not cooperating...
I'm learning a lot right now, thank you. The battery use makes sense now. Another friend suggested a basic 12v system - 1000 watts of panel, 60 amp mppt, 2000 watt inverter and a 200 ahr Lifepo battery as a stand alone system.
 
Another friend suggested a basic 12v system - 1000 watts of panel, 60 amp mppt, 2000 watt inverter and a 200 ahr Lifepo battery as a stand alone system.
1000W of solar at 12V is up to 83A of charging current (assuming you could actually get 1000W from the panels). A 60A charge controller on a 12V system means any power over 768W will be wasted. A 70A charge controller will use up to 896W.
 
I'm learning a lot right now, thank you. The battery use makes sense now. Another friend suggested a basic 12v system - 1000 watts of panel, 60 amp mppt, 2000 watt inverter and a 200 ahr Lifepo battery as a stand alone system.
24v is better in most cases, especially if there is no vehicle involved. 48v is even better, but that requires a bigger $$$ battery setup. The higher the voltage, the smaller the wire size can be and the less amperage is involved.
 
1000W of solar at 12V is up to 83A of charging current (assuming you could actually get 1000W from the panels). A 60A charge controller on a 12V system means any power over 768W will be wasted. A 70A charge controller will use up to 896W.
So for a bit of future proofing, a 100 amp SCC?
 
The rule of thumb (not sure how scientific it is):

12v = 1200w
24v = 2400w
48v = 4800w

You can go higher, but the numbers above are 100a which is a lot of juice running through the wiring. The more amps, the bigger the wire has to be, so there is a real advantage to using higher voltage. 1000w of panels requires 83a of scc at 12v, but at 24v it is only 42a. The same controller can run more panels and more wattage at higher voltage.
 
The rule of thumb (not sure how scientific it is):

12v = 1200w
24v = 2400w
48v = 4800w

You can go higher, but the numbers above are 100a which is a lot of juice running through the wiring. The more amps, the bigger the wire has to be, so there is a real advantage to using higher voltage. 1000w of panels requires 83a of scc at 12v, but at 24v it is only 42a. The same controller can run more panels and more wattage at higher voltage.
Makes sense. Thank you
 
So for a bit of future proofing, a 100 amp SCC?
If you are looking at a 100A SCC then you really should be looking to increase system voltage instead.

1000W of solar on a 12V system requires a 70A or 80A SCC.
1000W of solar on a 24V system requires a 40A SCC.

If you go with the original idea of a 60A SCC but with a 24V system then you can have up to 1500W or so of solar.

And as recently shown, the higher voltage allows for a bigger inverter or smaller wires/fuses.
 
If you are looking at a 100A SCC then you really should be looking to increase system voltage instead.

1000W of solar on a 12V system requires a 70A or 80A SCC.
1000W of solar on a 24V system requires a 40A SCC.

If you go with the original idea of a 60A SCC but with a 24V system then you can have up to 1500W or so of solar.

And as recently shown, the higher voltage allows for a bigger inverter or smaller wires/fuses.
That makes sense. He said he's got 2 grand that he can invest in it for now, so we'll have to figure what he needs to get this done in budget.
 
Back
Top