diy solar

diy solar

My system I bought waiting on mailing.

GGameBoy

New Member
Joined
Feb 22, 2021
Messages
63
I am jumping right into solar and making my own solar generator. I’m a complete newbie with batteries electronics and electricity. however I have been getting a crash course on Wills videos and feel maybe moderately informed where I should be. Can an expert look at the parts I have already purchased and make sure I didn’t make a 4 figure mistake? Lol

8 catl 310 amp hr lifep04 cells for 24 volts


EPEVER MPPT Solar Charge Controller 40A​


Overkill solar 100 amp 24 volt bms.

debating to go with the 2000 or 3000 watt Giandel 24 volt pure sine inverter. Bms can only handle 2560 watts. My thinking is do I want to stress the bms or the inverter? What’s ur thoughts? A 2nd bms?

at the moment I have 1.6 kWh of solar panels.

finally I bought this to run an window ac unit the keystone 5000 btu ac seer 12.1 I think max 415 watts continuous? I want to run the ac unit during the summer 24/7. I know the battery bank is the absolute minimum but I’m hoping the unit thermostat will kick off the ac long enough to save battery juice to save it to charge up the next day.

I’m also considering getting a 2nd Everett mppt to charge up at a max 2kwh.

I have breakers and wire and shunt and what not that i am in the process of finalizing the details with.

Side note, I live in an extremely hot environment like Will I live in imperial ca. I plan on putting this into a plastic tool box maybe 50 gallons. I know I need to keep the system cool.
 
As I understand SEER ratings (it's possible that I don't understand them), that 415W is an average that includes cycle off time and higher peaks.

415W/.85 * 24h = 11,718Wh A/C requirement (includes 15% inefficiency in converting DC to AC)
310Ah * 3.2V * 8 * .8 = 6349Wh Battery capacity (includes a 20% reserve to improve cycle life)

Estimate inverter burns 20W, so 20W * 24h = 480Wh

Total demand: 12,197Wh

40A * 27.2V = 1088W typical peak charge power.

Since you're clipping your array, it's very difficult to estimate the energy your array can produce; HOWEVER, it's an absolute certainty that your array can't produce anywhere near the total daily demand.

Your solar availability is a critical piece of missing information. Here's PVWatts for a 1.6kW array at 180° and 10° tilt (for peak summer production) in Imperial, CA (no panel shading all day):

1618591702143.png

June is often the hottest month in Phx, so I'll assume the same: best case of June, 276kWh/31 = 9.2kWh/day

9.2kWh/day is notably short of the 12.2kWh/day requirement, and this assumes you can actually use your entire 1.6kWh array, which you can't.

Adding a second 40A SCC and having an array size of at least 2.2kW should make this work.

The battery capacity is very questionable, but without knowing a consumption profile on a 24 hour basis, it's just a guess.
 
Damn that was a fast response and very detailed.

Ok. So add a second 40amp SCC, double the battery size. If I arrange two 8s batteries in parallel with a 2nd bms I could get that 12 kw. And then add solar frankly solar panels r dirt cheap since I live near Phoenix from suntan solar. So I can add panels for nothing.

is this a logical solution? Or is there better one? Possibly cheaper?
 
Last edited:
The battery MIGHT be sufficient because your array is so big, and cooling demand is lower at night. Your solar will power it for a very large portion of the day with a lot of surplus for charging and likely a lot of wasted solar through the day. This was a conservative "traditional" analysis based on some pretty coarse assumptions.

Your proposed solution should take care of it in a brute force approach. Another 8 cells is going to cost you another $1K. If you're okay with that, then it will probably work. My bigger concern is that I've been too conservative.

Personally, before I commit the big $ on a second battery, I would buy the A/C unit and put it on a kill-a-watt with it running full blast 24/7 with the thermostat as low as it will go to keep it running as much as possible - do it for a week and see what the kill-a-watt says.
 
Last edited:
Back
Top