easy, DIY solar powered AC for small room??

I have a small 10x10 foot room I am considering cooling with a small 12000 btu AC, but I am wondering if it's possible to set this up to be powered by solar? The AC would not be run continuously, just long enough to cool the room off for sleeping, maybe 2 hours per day. I'm thinking a car battery with an inverter and the battery charged by solar cells, but I don't know if this would work or even how to begin. I also want to keep this as low cost as possible and stick with a simple design that works.

Any suggestions would be appreciated.

Thanks, AA

Reply to
AA
Loading thread data ...

Some considerations, The 12000 BTU AC unit listed here:

formatting link

..shows that it has a rating of 115VAC/60Hz with a draw of 10.2 Amps. They also quote a rating of 1110 Watts.

If you were to use an inverter such as the following:

formatting link

..the "optimum" efficiency is listed at ~90%. It would be more likely that you may see 85% efficiency. That would take an input power of

1110 Watts / 0.85 = 1306 Watts.

Converting the wattage requirement to Amps at a nominal 12VDC would be

1306W / 12V ~= 110 Amps. This amperage delivered for the two hours that you specified is 220 AmpHour.

A possible battery may be the Trojan model T-145 shown in this brochure on page 11 of the PDF. The quoted 5-Hr rate is 215 AH.

formatting link

To re-charge the battery from a solar array will require an array capable of restoring the 220 AmpHour of energy during one summy day period of time.

The solar panel showen here:

formatting link

280_watt_panel.htm

..is good for 280 Watts on a good sunny day. One could estimate that you have say 8 hours of sunlight on days when you would like the air conditioner to function. Also let us guess at solar battery charger efficiency of 85%. Some simple math shows that:

(1306W / 0.85) / 8 hr = 192

It looks like one of the above panels could keep your battery charged. Two panels would give you a nice extra capacity in case you wanted to swipe some system power for something else besides the AC unit.

--
Michael Karas
Carousel Design Solutions
 Click to see the full signature
Reply to
Michael Karas

I would look at direct solar absorption cooling, like the kerosene or propane refrigerators use. The solar heat collector would supply the energy. There used to be a clever little 12V 'fridge available, "Norcold" I think. It used a "vibrator compressor" and could be used on a small boat. Perhaps something could be adapted from that.

Reply to
Wond

WOW, this looks doable.

What would be used to connect the solar panel too the battery ?

I would think there would be some sort of a protection circuit between the solar panel and battery.

Thanks

Reply to
hamilton

Thanks. Although this design looks interesting, it far exceeds the budgetary amount I had in mind ($500 max, sorry I didn't specify initially). I thought maybe I could put something together quick, simply, and cheaply but doesn't really look that way. Unless someone else has some other suggestions.

AA

Reply to
AA

Yes, but you won't like the price. Let's grind some numbers and see.

At 12,000 BTU A/C will draw about 11amps at 117VAC or about 1300 watts. Assuming you need A/C for about 6 hrs per day, and it runs about 50% of the time, you're looking at: 1300 watts * 3 hrs = 3900 watt-hrs

I don't know your location, but if you're in the SF Bay area, solar insolation is about 6 hrs during the summer for a non-tracking collector

Assuming no obstructions, that should give you about 4 hrs charging time. A typical solar panel puts out about 10 watts per square foot. To obtain enough power, you'll need: 3900 watt-hrs / 10 watts/sq-ft / 4 hrs = 98 sq-ft of panels That's about 10ft x 10ft of panels. Note that this assumes no wire, connector, charge controller, or DC to AC inverter losses. My guess(tm) is about 80% yielding about 120 sq-ft of panels.

If you shop around, you can find panels for less than $2/watt. The typical 200 watt panel is about 17 sq-ft, so you'll need about 10 panels at $400/ea for a total of $4,000. Something like this:

formatting link

You'll also need to store the power to deal with startup surges. The battery system will need to be big enough to not discharge more than about 20%. Very crudely, you'll need 80% more stored power than you're draining. 1.8 * 3900 watt-hrs = 7020 watt-hrs capacity. for a 48v battery system, you'll need 147 amp-hrs of batteries. That will be at least 4 of these batteries for about $1,000 total:

A 1300 watt 48V inverter and charge controller combination will also be needed. Here's a 3KW model for $1800 that might work.

I made quite a few guesses and did no price shopping, so the numbers will vary somewhat. However, I think they're in the ballpark.

Are you sure you want to do a solar powered A/C? You would probably be better off with a solar concentrator, running a boiler, which drives the A/C compressor motor directly. Just one problem... it's really noisy.

Recommended reading:

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
 Click to see the full signature
Reply to
Jeff Liebermann

Adapting the solar panel to the battery requires the use of a high efficiency switch mode converter/charger unit. Those are available from various solar equipment suppliers.

It may very well be more cost effective to find an integrated solution that combines the solar cell / battery management with the sine wave inverter to the 120 or 240 VAC.

The OP's idea to try to support his 12000 btu AC unit for $500 is just not in the game plan. There is a portable 1800 Watt solar power gererator available at Northern Tool for ~$1800 that may be of interest for those that want a more turnkey solution.

formatting link

Do check carefully as to your application with the above unit as it is not clear if it even produces a simulated sine wave output or a full sine wave output. Some loads will requre full sine output.

--
Michael Karas
Carousel Design Solutions
 Click to see the full signature
Reply to
Michael Karas

Really?

The first (discounted) price I find for a T-145 battery is $209 and it is

6v so you need 2 of them for $418.

Cycled to 80% discharge they claim a life of around 700 cycles.

Replacing around 2.2kWh of mains electricity with a cycle of these batteries gives a kWh life of around 1540kWh.

Where I live 1540kWh of mains electricity costs around $300 (and that is after political interference with the market to heavily subsidise eco green bollocks power generation)..

Is constantly loosing money on just the batteries never mind the capital expense and depreciation of the panels, control electronics and installation doable?

Only if mains electricity is really expensive where you live or if you can steal the equipment.

Reply to
nospam

A *big* assumption here is that the inverter can supply the compressor starting current.

Assuming PF=1.

Double that for 2 hours per day.

It's not going to work.

Reply to
krw

Heat pumps (like an air conditioner uses) are a poor choice for power-stingy applications. If your room has one or two occupants, you'll need to pump at least a few hundred watts of heat (the metabolism of living persons guarantees that) as well as dehumidifying...

So, consider a heat-storage unit (a big water tank?) and some heat exchangers (like car radiators, with a fan) so you can cool the water in the night, and use that reservoir of cool material to relieve the hottest hours of the next day.

So-called swamp coolers (basically a big wick kept damp with water, that evaporates outdoors) can also deal with a bit of heat, without burning big motor-power.

Reply to
whit3rd

Nice, but the two panels are only 45 watts. Assuming an 1800 watt draw for 1 hour, it would take: 1800 / 90 = 20 hrs of continuous sunlight to recharge the battery (assuming perfect weather and probably tracking the sun).

A few years ago, I built a spreadsheet to size the solar array and battery pile needed to run a local ham radio repeater.

Just replace the repeater drain characteristics with the A/C numbers and see what happens. I'll try it when I get back from service calls. Incidentally, the basic assumption is that the solar array must recharge the batteries within a 24 hr period or the load will eventually run the battery to zero.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
 Click to see the full signature
Reply to
Jeff Liebermann

Are you sure you need 12,000 btu's for a 10x10 room? I'm pretty sure my bedroom is bigger and uses only about 5,900.

Reply to
Tom Kupp

Well, it isn't so simple. You would need a solar tracking system to keep the panel pointed at the sun all day. If fixed in position, you will be lucky to get half the stated power averaged over the day. Next, the 280 W rating is in Arizona, not necessarily where you are located. Also, there are a bunch of tradeoffs.

For charging, you need a maximum power point controller that is a DC-DC converter that optimizes the energy taken from the solar panel. Otherwise, if you just connect a 12 V panel to a 12 V battery, you will get less that the maximum power out of the panel. Finally, you would get greatly better efficiency out of the whole system if you went to a higher voltage, such as 24 or 48 V. 12 V to 120 V inverters are terribly inefficient.

Jon

Reply to
Jon Elson

Hmmm, you got a well out there? Pump well water up and through a heat exchanger, and a fan blows the air through the heat exchanger. This can probably be built mostly from scrap yard parts. You can dump the warmed water back down the well.

This kind of nearly passive ground water cooling used to be used a lot. It certainly requires less energy than mechanical refrigeration. If your groundwater is not cold enough it won't work, but a lot of places have really COLD groundwater.

Jon

Reply to
Jon Elson

y).=20

=20

$500 at $0.12/KWHr is 4200 KWHr or 4200 Hr at 1KW. So at 2 Hr daily that ge= ts you 2100 days of operation. Then assuming it is only used 4 months out o= f the year or 120 days, the $500 would power the unit for 2100/120 =3D over= 17 years. IOW your /idea/, even at the your fantasy cost, wouldn't start t= o pay off until the 17 year mark. But you will have probably run through al= l the components at least once, maybe two times, during that period. IOW, r= unning it off the line is unbeatable. The other suggestions only begin to m= ake sense /if/ you don't have line power.

Reply to
bloggs.fredbloggs.fred

Not one net watthour of PV solar has ever been generated. To date, pv anything remains a gasoline destroying net energy sink.

Net pv energy can be expected eight to ten years after the fully burdened, subsidy free costs drop below twenty five cents per peak panel watt.

more at <

formatting link
>

--
Many thanks,

Don Lancaster                          voice phone: (928)428-4073
 Click to see the full signature
Reply to
Don Lancaster

Give it up. If you have access to the power grid, there is no, zero, none, not any cost-effective PV generation scheme. Unless you can steal the components, that is. There were schemes where the government stole money from me to subsidize your PV scheme, but those programs are pretty much exposed and defunct.

PV is great for school crossing signs where the high cost of the solar component is still WAY less than the cost to dig up the street to get power from

100 feet away.

Put your budget into better insulation. Needing a 12000 BTU unit to cool 100 square feet at night suggests that your insulation is deficient. Consider ground-source cooling. Depends a lot on where you live. Right now, it's 86F outside, 107F in the attic and 65F under the house. Stays about that temperature under the house year-round.

I once had a fantasy about reworking an electric blanket to pipe chilled water thru it. Never got past the concept stage. Maybe with a heat exchanger under the house... If you've ever tried to share a waterbed, you're aware how tiny differences in water temperature make a huge difference in comfort level. Wouldn't take much energy to keep the water cool enough.

RE the other suggestions... if you do attempt this, consider that it can take a LOT more power to start an AC compressor than to run it. Make sure your inverter can handle the surge.

Reply to
mike

I have some 95Ahr AGMs and they are ball busters. The biggest Odyssey battery is 125Ahr. You can always put batteries in series like they do in data centers, but that won't drive the repeater directly.

It isn't clear from your spreadsheet if you are taking into account both the loss of the charger and the charge efficiency of the battery itself.

Based on an APU battery evaluation project for a company I can't name, the Odyssey batteries were the best, with Dekka a close second. They are selling APUs for military use, basically to remove generators from sensitive locations. An additional product line is to replace APUs in trucks with battery storage.

For example:

Reply to
miso

See line 33 for Charger Efficiency. However, I forget to include battery charging efficiency. Thanks. Also, the NREL link seems broken. I'll fix those the next time I need to use the spreadsheet.

Incidentally, I modeled two solar powered radio system with the spreadsheet. My numbers were rather optimistic, mostly due to battery ageing and filthy solar panels. With new batteries and clean panels, I think it's slightly optimistic, but close enough.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558
 Click to see the full signature
Reply to
Jeff Liebermann

Agreed. There's no cost effective way to do it. Just like wind farms, pv arrays only give rated output under ideal conditions, so you might need 2 or 3 times the expected number of arrays to get reliable operation. They are also very expensive and you could run an ac for a very long time for the cost of arrays, inverter and associated control and monitoring. The arrays don't heve an indefinate life and neither do the batteries, which will quickly become degraded under daily high rate charge / discharge cycles. For best efficiency, the inverter needs to run at a much higher voltage as well. It's a really neat idea, but the technology just isn't ready yet in terms of cost.

A better idea might be to invest in small gas (not gasoline) powered combined heat and power plant, which will provide electricity, heating and cooling...

Regards,

Chris

Reply to
ChrisQ

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.