diy solar

diy solar

From Harbor freight special 12 Volt Off Grid learning system, to new ground up build off grid solar build plan.

Justkeepitsimple

New Member
Joined
Mar 12, 2024
Messages
9
Location
Southwest USA
Hi my name is Eric. My wife and I life far off and away in very off grid AZ. About 6000’ elevation so milder summers (about a week or two it pushes over the 100 Fahrenheit mark, and we see an average of about 8 to 12 inches of cumulative snow throughout the winter. about 3 to 4 months of the year the nighttime temperature dips into the 30’s-40’s °F at night). All things considered pretty mild for Arizona.

I’ve been researching and learning about solar for about a year now. Some one gave me a turnkey system to tinker with experiment/ learn on. (3) 100 watt harbor fright (freight) panels, a cheap Amazon PWM charge controller (that little blue 20$ infinitely rebranded one with simple digital readout), (2) medium- large Walmart flooded lead acid deep cycles (nominal 100 Ah each) and a modified sine wave 2000 watt harbor fr[e]ight inverter.

I went into this knowing almost nothing about basic electronics or solar.

What I’ve learned:
  • panel efficiency degrades greatly when it gets hot out.
  • Cheap PWM charge controllers kinda suck
  • Don’t run flooded lead acids lower than 12.5 V (after many cycles down to 11.5 V… well 2 years and they’re about toast now).
  • Check/Top off flooded lead acids with distilled water at least every month.
  • Modified sine wave kind of sucks.
  • Amps/ watts/ volts formula.

With that said we’re some learned up and getting ready to build a new system from the ground up. I like things to be reliable, durable, simple, modular, and good cost-benefit ratio.

Our loads are pretty minimal we generally power down the inverter overnight.

I’d like to build and design a system that’s easy to troubleshoot, somewhat redundant, and preferably compatible with the little Honda 2000 inverter generator to top of batteries on cloudy days.

I’ve had the circumstance come my way to acquire some used 40 W panels. I know I know…old stuff…be a lot of work to string together… a lot of new better options ($$$) but hey! they still work and they free!

Brought a bunch home. 11 out of 14 tested still good. 440w nominal. There is no manufacturer or voltage rating listed anywhere on them so I’m guessing but by their size they appear to be 40 W nominal panels. I cleaned them all off and put them in perpendicular to the sun at high noon and tested the open circuit voltages.
Maximum measured 23.4 V
Minimum reading was 18.6 V
Average over the 11 good ones was 21.95V

I am planning to also add the (3) 100W harbor freight specials to the parallel string feeding the charge controller for a total of 740W nominal solar going in to charge control/ battery bank.

Then I did a thing and bought (4) 12V nominal (4D group size) 290 Ah nominal flooded lead acid deep cycle batteries (DEKA). I know I know… I’m kind of doing this ass backwards but bear with me.

I pretty well have my heart set on building A 12 V system. I know… I know the wiring is gonna be expensive! But for the following reasons mostly to do with redundancy and ease of troubleshooting/repair:

1) I will string panels together to use solar panel input in parallel to provide voltage/ current to the charge controller (i’ll get to my selection criteria for the controller/ inverter toward the end of this post).
Because I:
A) Don’t know if these panels can handle being run together in series

B) Don’t want any funny undrevoltage going to my charge controller if one of them happens to bite the dust (ie in a 24 or 48 V configuration).

C) If one does go out it’s a pretty small percentage of my system and I can eliminate and/or replace with another 40W panel (I have access to about a dozen more free ones where these came from).

-If a 12 V battery kicks the bucket I can just take it out of the parallel bank and replace when I get to it instead of limping along on two batteries (if 24V system) or being so out of luck (ie situation in a 48V system with 3 out of 4 Twelve Volt batteries).
Our sunlight hour loads are pretty minimal and our overnight loads are even smaller for the proposed new 12 V system.

200w for deep freezer (off at night)
75w for Starlink router (18hr on 6hr idle)
200w for laptop/ phone chargers (all powered down 2-4 hours after sunset)
100w for box fan (run 24/7 in the summer)
150w for nighttime lighting (2-4 hours post sunset, could be even less if I switched over bulbs to LED).
Water all gravity fed. No water pumps. No well to run. Don’t do TV. Our luxuries are our 3 ft.³ chest freezer and star link for the smart phone email, wifi calling/ YouTube. Burn wood for heat.

Apx 3800 w/ h on the high side consumed during average 12hr period of darkness.
Out of our 13920 w/h battery bank equals//
About 27% DOD on battery bank.

My design target depth of depletion on these batteries bank is 15 to 30% for longevity.

Which leads now to the back asswards figuring of the inverter and the charge controller.

I am a big fan of overbuilt reliability and durability. As far as I’m concerned cost is no object when durability and reliability is concerned.

I’m thinking of going with:
Morningstar TS-MPPT-60 > TriStar 60 Amp 12/24/48 Volt MPPT Charge Controller

My thinking for this is that at some point when we want to expand at that point yeah I’ll probably go to $$$LithiumFe 48 V system and fancy new 300 W panels… Bam I’ve already got a charge controller that can handle that.

Also thinking of going with the largest 12 V rated morning star pure sine wave inverter they have available (looks like the biggest one they have for 12 V is 700W continuous with some surge capability up to 1300 W under 100°F).
Morningstar SureSine 700 Watt 12VDC 120VAC Pure Sine Wave Inverter with North America Type B Outlet .
if I decide I want to install a mini split I’ll either get one that can run straight DC no inverter needed day time only running straight off a string of dedicated panels, or get another 700W inverter dedicated for a small sized mini split and up the panel count and battery bank some. I know some people say 700 W inverter isn’t all that much but we pretty simple people and I know it’s sorta fiddly and convoluted but if I have to we can always throw the 2000watt modified sine wave harbor freight special on the battery bank and use that to run our lights and chest freezer anyway. I don’t know. I’m just impressed by MorningStar reliability record and the biggest 12Volt inverter looks like they make is their 700 W model.

I know 12 V isn’t considered very optimum by most people these days but for the reasons that I listed mainly to do with reliability and ease of troubleshooting that’s what I’m inclined to go with for my current circumstance.

My question for the community TLDR is have I picked an appropriately sized charge controller for this load? I know at 12 V DC theres a lot of current (amperage) potentially going into the charge controller and I’m not sure if that Morningstar 60 amp charge controller I’m looking at can handle it or not?

Quick stats (11) 40 watt panels 21.95 V avg.
(3) 100 watt panels not sure about the open circuit voltage on these panels. I need to put the multimeter on them I guess. They the harbor freight hundred watt specials. But assuming on the average side they’re about 20 V output each.

So that’s Approximately 740 watts of panels nominal. @20 V avg.

That’s about 37 Amps. Well that answers that question lol sometimes you just gotta write things out, think about it, and look at it twice and you can answer your own questions! Thanks forums for listening to me think out loud.

Going to keep batteries, charge controller and inverter in their own earth sheltered concrete ventilated storage box, earth sheltered in ground installation and shaded by static solar panel ground array. Big diameter cables shortest runs possible from panel parallel sequence to inverter input. May not keep batteries at perfect operating temperature but will keep them warmer in the winter and cooler in the summer. Some battery sacrifice performance maybe during the more mild months but really it seems the extreme high or low temperature that murder them. At least the inverter and the charge controller will stay happy and cool.

Any thoughts criticisms or suggestions? I’m open to hearing all but pretty well made up my mind I am going caveman simple 12Volt system at least for this stage in our life.

Got the panels already… Already ass-backwards bought the (4) 12 V flooded deep cycle lead acids, they seem like a good value for the money as long as you’re diligent to check the water levels at least every month more often in hot weather, and keep depth of depletion to a maximum of 30%. I don’t know. Change my own oil. Cut my own firewood. What’s checking the battery water level once a month?

Oh another thing I side forgot to mention I would really like a charge controller that can handle being hooked up to a generator to charge batteries on cloudy days. I’m assuming the 60A morning Starr MPPT should be able to handle that I’ve read the documentation up and down on it but don’t see it specifically stated there how one goes about wiring that up? I’m not looking for anything fancy like an auto Gen set kick in. Just want to get the alert on the charge controller Bluetooth app: “battery voltage below 50%” stroll out to the shed pull start the honda 2000 genny, drag the ole 10gauge extension cord out and plug it in to the charge controller. It’s probably right in front of me. But my brain is so fried from crash course in solar design and amps ohms watts I probably missed it somewhere.

No possibility of grid tie now or in the future. Nearest powerlines are 20 miles away and it just ain’t gonna happen for electricity to get brought out here. so that simplifies things at least from a design choice standpoint.

Thanks for those 2 bearing with this long ramble. Good community I’ve learned a lot from searching through the forums here.
 
While your heart is presently set on 12v, 48v may be easier to maintain over time as the trends of home batteries are all 48v that I've seen.

That said, You should figure out peak demand (watts) and daily usage (watt hours) and build around that.
Example: If you use 30KWh/day, you need to be able to support that on a combination of battery and solar (and grid if you're grid tied) If you want to go strictly offgrid, you have to plan for multiple days with no sun or an alternate means of power (generator) to top off batteries.
 
Last edited:
Morningstar makes good stuff BUT they make both types of charge controllers , pwm and mppt. Not knowing your panel specs and mixing types, better go mppt. That being said, they probably have the most expensive 60A charge controller on the market... not sure its worth it, but you shouldn't have any trouble with it.

For an inverter, morningstar gets even more expensive for the wattage and they dont have a ac charger built in for your honda 2000.

Perhaps consider a LF 2000w inverter/charger. I like Aims but theres lots out there. Samlex, magnum and outback would be a solid choice too. Aims has a manual adjustment for generator load right on the unit. The others require programming or an external control sold separately to not overload the suitcase generator.

A stand alone inverter can work but youd need a seperate charger to go with it. IOTA and Samlex are decent options for independent charging.

Im a lead acid fan, you should be OK with those 4D 12v. They cant handle the cycles like some of the purpose built 6V deep cycles. So keep them charged best you can. 4-5 years is realistic if you are gentle on them.
 
  • Don’t run flooded lead acids lower than 12.5 V (after many cycles down to 11.5 V… well 2 years and they’re about toast now).
  • Check/Top off flooded lead acids with distilled water at least every month
FYI: the electrolyte expands as it charges, so you should only top batteries off when they are fully charged.
 
FYI: the electrolyte expands as it charges, so you should only top batteries off when they are fully charged.
New thing I learned! Thank you for conveying that important knowledge. I love some old school forums. So much better than short attention span non-archivable/ non-searchable Facebook groups.
 
Yup, that controller has no provisions to connect a generator.

Keep in mind this point in your calculations for storage needs...

The inverter has a standby draw that is ALWAYS on... so, figure 25W if you have not measured it.
Thats 25W every hour, or x 24 or 600Wh to get storage Wh burned from your battery.
 
You can buy a single 550 watt solar panel for about $200. You might spend more than that in connectors, fusing and mounting hardware for those crappy, old 40 watt panels..
You can pick up a 3000 watt, 24 volt AIO inverter/charger for under $400.
This would be quite cost effective and simple to install.
 
You can buy a single 550 watt solar panel for about $200. You might spend more than that in connectors, fusing and mounting hardware for those crappy, old 40 watt panels..
You can pick up a 3000 watt, 24 volt AIO inverter/charger for under $400.
This would be quite cost effective and simple to install.
Yes, panels are cheap (per watt) nowadays. Some good deals can be found locally as well if only looking for a few.

I suspect once hes up and running, he'll catch the bug and end up buying a pallet within a year 😅
 
Hi my name is Eric.


My question for the community TLDR is have I picked an appropriately sized charge controller for this load? I know at 12 V DC theres a lot of current (amperage) potentially going into the charge controller and I’m not sure if that Morningstar 60 amp charge controller I’m looking at can handle it or not?

...

...

Oh another thing I side forgot to mention I would really like a charge controller that can handle being hooked up to a generator to charge batteries on cloudy days. I’m assuming the 60A morning Starr MPPT should be able to handle that I’ve read the documentation up and down on it but don’t see it specifically stated there how one goes about wiring that up? I’m not looking for anything fancy like an auto Gen set kick in. Just want to get the alert on the charge controller Bluetooth app: “battery voltage below 50%” stroll out to the shed pull start the honda 2000 genny, drag the ole 10gauge extension cord out and plug it in to the charge controller. It’s probably right in front of me. But my brain is so fried from crash course in solar design and amps ohms watts I probably missed it somewhere.

...
Hello Eric.

I think you have some misunderstanding of a SCC (Solar Charge Controller). A 60 amp charge controller charges your batteries at 60a if sufficient PV power is available to convert to battery voltage at 60a. It charges at less if you have less PV. So if you have a 12vDC nominal battery this translates to ~720w. If you have more PV than the ~720w it can not make use of it but it could max out for longer periods during the day. Teh largest concern with SCC is not to exceed Voc level.

SCC are not able to take AC voltage from a generator and charge your batteries. For that you need either a independent battery charger, a inverter/charger or get an AIO (All in One) that incorporates SCC/inverter/charger into one box.

My advise is to research a bit more and possibly consider ditching the 40w panels. I know you say you want 12vDC system but it is another area you might consider upgrading to a higher voltage. You can still create 12vDC with a converter if needed. But the future expansion capability of 12vDC setup is minimal.
 
Yup, that controller has no provisions to connect a generator.

Keep in mind this point in your calculations for storage needs...

The inverter has a standby draw that is ALWAYS on... so, figure 25W if you have not measured it.
Thats 25W every hour, or x 24 or 600Wh to get storage Wh burned from your battery.
OK. That makes sense I will have to look into a separate battery charger module. I am assuming that if I want to charge the batteries from the generator I need to be able to disconnect on the solar charge controller while the generator is charging batteries?
 
I echo the suggestion to look at new panels. They're so cheap today and they last much longer than batteries, inverters, etc. Over a 20 year period they're <10% of your total cost. It doesn't pay to sacrifice overall system performance by rigging up mismatched panels.

If you ever go to a nearby city you might also find some used ones. Like this DFW ad for a couple of $65 Jinko 395Ws. A guy in Phoenix gave away a dozen or so 300s last year. And so on.

Not sure the 40s are worth the hassle. Your planned ~37A array exceeds commonly used wire/connector ratings. So you'll spend extra on wires and a combiner box or whatever. And the lowest voltage panel in your array will pull the others down, hurting overall output.

I'd probably save the 40s for random 12V projects like a powered gate or security system or ???. Or give them away as door prizes :)
Consider using the old 3x100W system to power some daytime loads, maybe the freezer. Dead battery should be OK for that.
I also vote for an AIO if you want generator backup.
 
OK. That makes sense I will have to look into a separate battery charger module. I am assuming that if I want to charge the batteries from the generator I need to be able to disconnect on the solar charge controller while the generator is charging batteries?
No the solar charge controller can stay hooked up, it doesn't care.
 
Hello Eric.

I think you have some misunderstanding of a SCC (Solar Charge Controller). A 60 amp charge controller charges your batteries at 60a if sufficient PV power is available to convert to battery voltage at 60a. It charges at less if you have less PV. So if you have a 12vDC nominal battery this translates to ~720w. If you have more PV than the ~720w it can not make use of it but it could max out for longer periods during the day. Teh largest concern with SCC is not to exceed Voc level.
OK. I am planning to run all the panels in parallel to the charge controller. I know it will be more expensive to do that way as far as wiring cost is concerned but I like the redundancy/ reliability aspect should a panel in the array go bad. Wired in parallel my voltage output from solar array should be in the neighborhood of +\-21 V. Totaled up I have 740 watts of panel at 21 volts is about 35 amps current. If I wired the panels up in series that would increase the voltage going to the charge controller. If one of the panels in the array fails though then the output from the array is 0. The charge controller im looking at is rated for up to 120VoC -60 amps total max input. Maybe it’s retarded doing it in parallel as I’m gonna have to do a lot of ever-increasing wire gauge junctions to combine the solar array wiring string to the charge controller , but for me the redundancy and reliability of that method is more important to me than the cost savings of series wiring the panels and saving money on copper. Or at least that’s my line of thinking. I may be missing something?
 
At some point I’m very open to considering new/better/bigger panels. Like I said I’m not averse to spending extra money for reliability durability and simplicity. I know how it goes sometimes free doesn’t add up once everything is said and done.
 
Uh.. it might be rated at 120Voc but the 60 amp is the output to the battery rating. W=VA
OK. Thank you for your response. Sorting through all this stuff is kind of nebulous sometimes. How do I make sure I don’t exceed the maximum input to the charge controller? Is The maximum input to the charge controller strictly limited by the top end VoC number Number rated for the charge controller or does it matter how many amps the panels are delivering to the solar charge controller as well? Perhaps I have a fundamental misunderstanding of what voltage open current actually means.
 
Actually I don’t thing I’m not misunderstanding at all. The Nominal wattage from the panels will be 740 W. Minus the resistance of the cabling from the panels to the controller that will deliver approximately 35 A of current at +/-21 V to the charge controller. Charge controller will convert that (~35A @ 21V+\-) into the appropriate trickle charge the batteries depending on their state of charge and temperature. As long as I don’t exceed 120 total panel voltage supply going into the charge controller The charge controller can handle the rest.
 
OK. I am planning to run all the panels in parallel to the charge controller. I know it will be more expensive to do that way as far as wiring cost is concerned but I like the redundancy/ reliability aspect should a panel in the array go bad. Wired in parallel my voltage output from solar array should be in the neighborhood of +\-21 V. Totaled up I have 740 watts of panel at 21 volts is about 35 amps current. If I wired the panels up in series that would increase the voltage going to the charge controller. If one of the panels in the array fails though then the output from the array is 0. The charge controller im looking at is rated for up to 120VoC -60 amps total max input. Maybe it’s retarded doing it in parallel as I’m gonna have to do a lot of ever-increasing wire gauge junctions to combine the solar array wiring string to the charge controller , but for me the redundancy and reliability of that method is more important to me than the cost savings of series wiring the panels and saving money on copper. Or at least that’s my line of thinking. I may be missing something?
Keep in mind, anytime you put over 2 strings in parallel, you have to fuse that string.
Fusing a 40W panel wastes 5 watts from the connections to the fuse resistance...
So...
Series 3, parallel 3, only 3 fuses...
But you need a real mppt controller.
 
Lots of thoughts, mostly since I'm in a somewhat similar environment in NE Missouri and am probably all of about 2 months ahead of you on the learning curve. That said:

When you kill your Pb based batteries, which you will in a year or two, invest in LiFeP04 ones. Cost about 2X your Pb ones (if you buy cheapies with Bluetooth BMS, say $200 with shipping), to 4X for name brand - with $900 ones out there that just seem like price gouging. In any case, they have 100% drain capability, although many only do 80-90%, but even that is double what your doing on Pb, and they are likely to last you at least 10 years, maybe 20+. Its just the financially smart investment. 100amp is the standard size, although larger are available. I'm running (2) Amazon no-names in parallel, expecting no chance of warranty claims, but bought them for my own learning exercise at the time.

Go MPPT vs. PVM. Somewhat pricy but rock solid for small systems are those sold by DIYSolarForU.com (note, shy the final "m" that this forum uses). Want to move up in the world, check out Victron gear. Lots of data on their victronenergy.com site, suggest currentconnected.com as a retailer - great rep, great service, friendly to newbies (just make sure they know you are one). Your going to get something like 35% more power out of the MPPT units vs. PVM (more-or-less).

MPPT chargers can (and should) be configured differently than PVM. The key to understanding MPPT is to focus on their maximum open voltage capability, and to come in slight below that, after temperature adjustment. As you noted, the colder the panel, the better they do. Typically MPPT charges have 150 or 250 volts max inputs, until you get into ~$1000 stuff. For me, I'm using a 1.25 factor (good to about -25C) (yeah, learn to think in C, not F, or at least find one of the many converter websites). Hit the web and find the lowest historical temp for your area, and buffer that a bit with climate change in mind. So, if you have a MPPT controller rated at 150 VoC, you can feed it with panels in series up to 150 VoC... exceed that and you will likely smoke your unit. Because of panels reactions to temps, that is most likely going to happen at dawn on the coldest day of the year... Say you are using panels with VoCs of 25V. You would think 150/25 = 6 in series and would fry your MPPT controller on that cold day. 25V temp corrected is 25*1.25=31.25. Divide that into your 150 VoC controller maximum and you get 4.8 panels, so 4 as the max, not 6. Of course, by mixing panels, your might squeeze in another one, just keep the CORRECTED total VoC safely below 150.

Personally, after my experimental phase, I go with "buy once" in mind. A Victron inverter from CurrentConnected (they double the 5 year warranty) is warrantied for 10 years, and will likely last 20+. Cheap Chinese knockoffs may last 5 if your lucky.

To help with the multitude of Victron units, their inverter charger naming convention is V/VA/Charge-Max current in/Voltage out. So a Multiplus II "12/3000/120-50 120V" means 12 volts in, 3000 VA (not watts, figure 80-90% of VA in watts with a mixed load), 120 amp charging, 50 amp maximum AC input, 120V output. That unit is what I'm playing with now. You can restrict AC input down to a bit below 10 amps, allowing me to safely AC feed it from a standard 15amp circuit. It allows me to current limit my battery charge rate to 40 amps (0.2C being considered "normal" for my specific LiFeP04 batteries), and will "assist" with battery power if my demand exceeds the 10 amp input limit. All nice things. I find the Victron Community web pages a bit brash, but the folks here are great, as is Current Connected support.

Depending on your space, and with panels as cheap as they are now (recently saw some at $0.33/W), perhaps consider bifacial panels oriented vertically east/west. Controversial with a lot of pros and cons and likely not as efficient as south facing if your just interested in recharging your batteries daily, but worth educating yourself on.

Cheers!
 
Last edited:
Back
Top