diy solar

diy solar

Sanity Check & Comments Requested for 24V System Build

WoodyInAz

New Member
Joined
Feb 1, 2020
Messages
21
Here is the setup so far (click image to expand):
24V-Solar-Arizona.jpg
Questions: How many panels can I have for this setup? How many batteries? Is the 100amp battery fuse sufficient?
Goal: Run 3 window AC units (5 amp 450 Watts) plus laptop and monitors. (I can start with 2 AC units).

If I can run these ACs from 9am to 8pm weekdays I can save about $1,500.00 a year.

This is an off-grid setup, but I have APS connected to the house.

Solar panels ($110 ea) and batteries ($140 ea) were purchased at Santan Solar in Gilbert AZ and have 4 year warranties.

Let me know what you think! Thank you
 

Attachments

  • S12V550NGF.pdf
    2.8 MB · Views: 12
  • SunPower ds-e20-series-327-residential-solar-panels.pdf
    461.5 KB · Views: 6
How many panels can I have for this setup?
The number of panels is principally determined by the capabilities of your SCC. From what I can make out from your image, you have an Epever Tracer model #8415AN, which has a max PV input rating of 2,000W delivered into a 24V battery system, so the max number E20-327 panels you can connect is 6 (6 x 327W = 1,962W is OK, but 7 x 327W = 2,289W BUST!).

Because your max PV voltage is 138V (at 25℃), you can only connect up to 2 panels in series (2 x 64.9V = 129.8V is OK, but 3 x 64.9V = 194.7V BUST!) so the only configuration option you have is 2S3P.

How many batteries?
As many as you like, or rather, as many as you need for your off-grid autonomy requirements.

Is the 100amp battery fuse sufficient?
Two aspects of solar system design determine the fuse protection requirements: (1) The anticipated load current +25% and (2) The rating of the cable.

Your anticipated load current will be 104A at 24V (2000W / 24V +25% = 104A) so fusing below this runs the risk of nuisance blowing. 4AWG cable is rated for 120A so fusing this cable would demand a fuse nothing higher than 120A, 2AWG is rated for 170A, so nothing higher than a 170A fuse, 0AWG 200A etc.

I would recommend using 0AWG cable to reduce resistive losses now and future-proof the design for a higher rated inverter (200A x 24V = 4,800W) and fusing the line now at 125A.

Goal: Run 3 window AC units (5 amp 450 Watts) plus laptop and monitors.
Whilst operating, your inverter is rated properly but AC units have a very large surge demand, maybe 2 or even 3 times the run-rate. You need to be sure that your inverter can cope with this.
Note: surge current will not blow an automotive-style fuse.

If I can run these ACs from 9am to 8pm weekdays
That is a huge demand! o_O

450W x 3 units x 11 hours per day = 14,850Wh energy consumption per day, plus your other loads. 4 x 139AH at 12V is only 6,672Wh of battery and, because you shouldn't regularly discharge a lead-acid below 50% DoD, this is in effect only 3,336Wh. You'd only be able power this load for a couple, maybe three hours.

The Tesla packs come in 5kWh modules and around $1,000 each (used) you would need at least three to support that level of consumption.

Furthermore, your 1,308W (4 x 327W) array would only generate 6,540Wh on a good summer's day (1,308W x 5 hours good sun = 6,540Wh) , quite insufficient to replenish your expected consumption.

You would have to expand your current design to just run one of these AC units.
 
The top description says 6000W at 48V, but in the details I see the rated input is 4000W at 48V (Tracer 8420AN) - this seems a bit misleading?

So if I configure for 48 volts I can get 4000W of solar input. Then get another 4 batteries and have 2 sets of 48V battery charge.
APS "peak" rate is between 3pm and 8pm - these 5 hours are what I am targeting for AC usage. I can get 4 more batteries and 4 more panels for $1000. Max panels would be 12 for 4000W. Thanks Again tictag!
 
The top description says 6000W at 48V, but in the details I see the rated input is 4000W at 48V (Tracer 8420AN) - this seems a bit misleading?
I do agree, in fact, none of the Tracer range are rated for 6,000W PV input, but that being said, I would always consult the manufacturer's website to confirm specifications - marketing is just another a word for lying. Your datasheet is here, or linked directly here.

So if I configure for 48 volts I can get 4000W of solar input.
I personally would never recommend a 48V DC system due to:
  • The shock hazard a 48V (nominal) battery system presents*
  • The difficulty in buying components rated for 48V DC
But there is nothing stopping you buying two (or more) SCCs to increase your max PV input. This is a personal choice.

The Tracer 8420AN supports up to 180V (at 25℃) and 4,000W delivered into a 48V battery system, this would allow you to connect a maximum of 12 x EP20-327 panels up to 2 in series in a 2S6P configuration (see above for calculation) and generate 19,620Wh per day in ideal conditions.

Note: the described 200V max PV voltage is only rated thus at the minimum operating temperature (i.e. -25℃)

Then get another 4 batteries and have 2 sets of 48V battery charge.
8 x 139AH (at 12V) would still only give you 6,672Wh of storage maintaining a 50% max DoD.

But with that said, your main consumption i.e. these A/C units, would only be active during the day. If you could match PV input to their consumption you would strictly speaking only require storage for your other loads at night ... but this would give you absolutely no off-grid autonomy for 'poor weather' days. But then, would you need A/C running on 'poor weather' days? (rhetorical).

If you could provide, say circa 2,000W PV input this would cover your A/C demand during daylight hours (3 x 450W = 1,350W + battery charging = circa 2,000W).

I've never considered this usage scenario before, normally you just work out your daily consumption then plan a PV array to replenish it every day, but this may well work for you. :)


* anything more than 30Vdc is generally considered an unsafe voltage for installation within living areas. This is a rule of thumb, not a code requirement.
 
Update: With my 4 panels and 4 batteries I am able to run 2 5000BTU window a/c's for 6 hours (peak APS power is 3pm-8pm). Mission accomplished. I will have a $100 (or less) monthly bill from APS. BUT - I am installing 4 more panels and 4 more batteries! :)
 
I also have APS. If all you wish to accomplish is to run the AC's during those 6 hours, You might also wish to install your panels differently to get maximum exposure during those hours, or break them up. Have some for maximum exposure during those 6 hours, and some for maximum energy generation. You also may find that having more solar panels might be a better solution then more batteries. Since You have available grid power it may not be worth the substantial cost of building a system that always will guarantee power to the A/C's. Efficient 5000 BTU window A/C draws about 450 watts each. There's at least one unit, I heard, that will start under a 1,000 watt load. So, if You turn A/C's on one at a time. You shouldn't have to worry about needing extra capacity for the startup (and hopefully the compressors wouldn't all stop and start at the same time while running). If it was me, I would wire things up, so I could plug the A/C's in Outlets from the Solar system, or the existing outlets getting the energy from the grid, or some sort of switching system. Make sure, You understand the capabilities and limitations of Lead Acid Batteries. They have particular requirements when it comes to discharge, charging, efficiencies, etc. Play with the combination of watt numbers/batteries/panels/controllers, etc to see what works out best. You may find out that it is more beneficial to buy solar panels then batteries and having two controllers, one for each string of panels on the roof in different locations.

I looked at those batteries. Seems like a good price, if they are just slightly used and have not been abused. I passed on them primarily because of the weight and the natural limitations of lead acid based batteries The solar system I'm building is semi-portable, and also used to experiment for future larger systems, so weight is important. Good Luck.

P.S. I really like the people from Santan Solar. I'm hoping they will be able to find some Lithium Batteries to sell.
 
Check out cybo energy. 4 panel mocroinverters will self-consume and pass the grid through to make up the difference seamlessly. With or without batteries.

cyboenergy.com
 
Incidentally if you run heat pumps you may pick up some efficiency. The inverters were made for ac units basicslly as far as i know amd transmogrified into a slew of options.
 
Hey Dash! Thanks for the comment! I have basically doubled the size of the system. 4000W Giandel and a 100amp EPEVER. So I will have 8 panels (327w) and 8 batteries. And yes, I am placing the 4 additional panels to maximize afternoon sun around 3-4pm. These Frigidaires only pull 500 watts when I turn them on, no surge - and that is how they are advertised. :) These AGM batteries never go below a 15 percent discharge - and I will be switching to Lithium as they age. And yes - I do plug the units into APS power after 8pm! .052 cents Per kWh off peak with the Saver Choice Max. I will be keeping an eye on Santan Solar for those Lithiums! Thanks again.
 
Hey WoodyInAz. That's great. I may do something similar, but will probably just use one A/C unit in one room (maybe larger then 500 watts). 4-5 330 watt panels, and one 12 v 100 ah lithium battery. The idea is for the solar to generate the power to run the AC nearly all the time (3-8pm), and the lithium battery to use as a buffer, and also when the sun starts going down. The really big unknown is how much real usable electricity can be generated by the solar panels that are mounted. We know what they are rated at, but how much will we really get out of them. I also don't know if I can adjust the controller to assure that the maximum amount of energy going into the battery is 1C, maybe less, so as not to damage the lithium battery or trip the BMS.
 
My 4 327w panels can generate a max 1,100 watts. They are fixed, 2 lay flat and 2 are at a 20 degree pitch facing south. I imagine the other 4 will not do as well because I plan to set those facing west to catch the afternoon sun angle, maximizing the sun at 4pm in the afternoon. I may even put these 4 on a tracker if I get ambitious. lol
 
Kernal came up with this idea, but still would need a few batteries to supply power when it gets dark. Any extra power, after the batteries have been charged could be used to heat the hot water heater, (the next greatest use of electricity, in a electric home.) , or send to the grid.

1588826053781.png
 
I'm just going to throw this out there. For what it's worth. I thought it was common knowledge, but apparently it's not. APS allows you to run a 1kw grid-tie system with no notification, no contract, no separate meter and disconnect, etc. They don't even want to know about it. As long as you don't exceed 1 KW. I told several friends, and we have been doing this for over a year, no problems with any of us. We are all getting around 5 -6 kwh per day. We have those cheap and readily available used 250 watt panels and used Enhase inverters (5 M215's). Less than five hundred bucks for the whole system. Permit is up to you. And with time-of-use metering, it pays for itself in less than a year. And no power goes to waste. No batteries, no charge controller, no nothin. Of course, if the power goes out, you're out of luck. But you are already knew that.
 
I'm just going to throw this out there. For what it's worth. I thought it was common knowledge, but apparently it's not. APS allows you to run a 1kw grid-tie system with no notification, no contract, no separate meter and disconnect, etc. They don't even want to know about it. As long as you don't exceed 1 KW. I told several friends, and we have been doing this for over a year, no problems with any of us. We are all getting around 5 -6 kwh per day. We have those cheap and readily available used 250 watt panels and used Enhase inverters (5 M215's). Less than five hundred bucks for the whole system. Permit is up to you. And with time-of-use metering, it pays for itself in less than a year. And no power goes to waste. No batteries, no charge controller, no nothin. Of course, if the power goes out, you're out of luck. But you are already knew that.

That's exciting.
 
Bill Taylor - so you send your power to the grid for free and allow APS to make $$$? Of course they don't care! lol
 
Bill Taylor - so you send your power to the grid for free and allow APS to make $$$? Of course they don't care! lol

Oh crap, you caught me. I am an undercover agent for APS. And I'm trying to get everyone to give us free power. Suckers!
But seriously. They might get a few stray watts here and there. But the panels are favored Southwest because time-of-use kicks in at 3 pm. The house always pulls a few hundred watts, and anytime I can offset their 25 cents per kwh peak rate, I'm in. I have CT's on the mains and every branch circuit. They don't get too much from me. One of my friends declined their offer to install a smart meter. His meter actually does turn backwards. Ahh, the good old days....
 
Oh crap, you caught me. I am an undercover agent for APS. And I'm trying to get everyone to give us free power. Suckers!
But seriously. They might get a few stray watts here and there. But the panels are favored Southwest because time-of-use kicks in at 3 pm. The house always pulls a few hundred watts, and anytime I can offset their 25 cents per kwh peak rate, I'm in. I have CT's on the mains and every branch circuit. They don't get too much from me. One of my friends declined their offer to install a smart meter. His meter actually does turn backwards. Ahh, the good old days....
Well in theory, if APS can reduce their costs by getting some electricity back that doesn't cost the solar user any money, then APS's operating costs is reduced and the corporation commission will reduce the price the consumer pays for the electricity. (I did say in theory, ha)
 
Question: My evening rate for APS is 5.23 cents a kWh. Does it make sense to charge up my batteries immediately after discharging 10 percent, or is waiting for the sun to come up to recharge ok? If this makes sense, what kind of charger should I get?
 
Well now, I'm just doing some math in my head. Eight batteries at 139 ah each at 10% DOD. That's 111 amp hours. Figure in about 10% inefficiency, at least, equals about 120ah x 12 v = 1.5kwh. And at roughly $0.05 per kilowatt hour equals about $.08/day, or $2.50/month worth of juice. Just ballparking those numbers of course, for the sake of discussion. So, assuming 6 months per year of air conditioner use, it's about $15 a year to prevent any possible life reduction in the batteries because of being partially discharged overnight.
So I started doing some more math in my head. My AC used 8.1kwh between 3-8pm yesterday. (High was 109°, low was 80). If I charged those batteries at night @.0523, and used them during the day to avoid paying the .26, even with 20% inefficiency, it would save me almost $50/month. So, again assuming 6 months per year of air conditioner use. You can buy a really nice battery charger for $300, and it would pay for itself in 6 months. No solar panels, no charge controller, no wire, fuses, mounts etc.
Oh boy, is that going to open up a can of worms.

This may be in the running for the longest answer given to a simple question. Then again, I have read these boards for a while. Not even close.
 
Last edited:
That is what I was thinking Bill. What battery charger were you thinking about for $300?

Bill - what plan do you have? I just switched to Save Choice Max and will pay .0523 Off Peak and .087 On Peak (as long as the 1 hour of most usage is not "On Peak" - then I get a $17 penalty for that day).
 
Last edited:
APS is making it easier and easier for all of us with APS to get off the grid. I so much miss SRP.
 

diy solar

diy solar
Back
Top