diy solar

diy solar

General Battery advice

SSC00

New Member
Joined
Feb 9, 2024
Messages
192
Location
USA
I am in the process of building a smaller learning system and I am trying to figure out what battery system I should use without spending far too much. So far I have about 1.4 kw of panels and will get a 3kw 24V PowMR AIO inverter. for now I dont think I will need beyond 2-3 kwh of battery for my learning set up. What type of battery should I use for a learners system?
 
2-3 kwh of battery
3000Wh / 25.6V nominal = 117Ah

I'm wondering if you should start with 12V for ease of first system as well as usefulness down the road. Or going straight to 48V and your starter AIO becomes your backup if it survives your learning process.

Down the road, 24V equipment will likely all be unusable.
 
a pre-built 25.6v/100ah battery with 150a+ BMS or a couple of big AGMs with a 24v balancer. The Lifepo4 will prove to be better value (cost/kwh) in the long run. AGMs are simple/easy and BMS-free.
 
Last edited:
I'd start with two 12v100ah batteries ... allows for your intended 24v AIO configuration now, or fallback to 12v inverter later (if your chosen AIO craps out, and threads abound about reliability and troubleshooting of various AIO's).

You can explore battery-bank design (we use 2s2p of 12v300ah, busbars & such), investigate the various bms designs offered for each brand/model and which has higher amp draw than others, and ah capacity (of each battery and of your bank design). Warranties, support & service model, etc.

We use LiTime (formerly AmpereTime) 12v batteries for the house battery bank(s), great support/service, distributed & serviced in the US, etc. We also have SOK 12v batteries for other projects ...

Nothing but 12v & 24v designs for various projects ... haven't yet gone down a 48v path.

Hope this helps ...
 
I built a starter/learner system somewhat similar to yours, with six 210 W panels for a total of 1.26 KW, and a 24 volt / 3000 KVA (basically 3000 W for discussion purposes) Victron inverter. I started with one Amperetime 100 AH battery, then added another about four weeks later. After another six months or so I added a third. I'm very happy with them for a stater/learner system and have no regrets. To power the "shed" buildings for which I built the system, I doubt I'll need to change anything for the next decade.

Some here say 24V will be useless but that's a gross exaggeration and overstatement that I wouldn't worry about if you've already purchased the equipment. That said, I would go 48V if I were starting over, simply because it will more easily scale and the prices now aren't that much more than 24V. There's absolutely no way I'd go with a 12V system or buy 12V batteries and connect them in series. But if you've already purchased some stuff, you should be fine. Take a look at @Will Prowse battery tear-down videos and go with a brand he likes, and you'll likely be as happy as I've been. BTW, I do advise adding in a good shunt (Victron is awesome) as it's the only accurate way to measure state of charge with lithium batteries.
 
3000Wh / 25.6V nominal = 117Ah

I'm wondering if you should start with 12V for ease of first system as well as usefulness down the road. Or going straight to 48V and your starter AIO becomes your backup if it survives your learning process.

Down the road, 24V equipment will likely all be unusable.
I selected the AIO that I have for now because it seemed to be the cheapest that wouldn't crap out on me in like a few months. Not saying that its the best one for my situation because I dont know whats all out there. I'm sure there is a list of all the AIOs somewhere here on this forum but I haven't found it yet. I haven't bought it yet though. I am always open to hearing something new though. My plan is to just go with it though.

I do agree with 48V seeming to be the standard for a larger system. It seems like every battery - AIO package is basically 48v. same with all the server rack batteries and everything. At least my mindset is to just have this whatever system now to learn, then move on to one of those systems where virtually everything would be nicer and I can have a fully fleshed out plan from what I learn essentially tinkering around.


a pre-built 25.6v/100ah battery with 150a+ BMS or a couple of big AGMs with a 24v balancer. The Lifepo4 will prove to be better value (cost/kwh) in the long run. AGMs are simple/easy and BMS-free.
That is an option. So I could just buy a couple AGM or even LiFeP from somewhere like home depot and then use a BMS with them?


I'd start with two 12v100ah batteries ... allows for your intended 24v AIO configuration now, or fallback to 12v inverter later (if your chosen AIO craps out, and threads abound about reliability and troubleshooting of various AIO's).

You can explore battery-bank design (we use 2s2p of 12v300ah, busbars & such), investigate the various bms designs offered for each brand/model and which has higher amp draw than others, and ah capacity (of each battery and of your bank design). Warranties, support & service model, etc.

We use LiTime (formerly AmpereTime) 12v batteries for the house battery bank(s), great support/service, distributed & serviced in the US, etc. We also have SOK 12v batteries for other projects ...

Nothing but 12v & 24v designs for various projects ... haven't yet gone down a 48v path.

Hope this helps ...
This does help. I think now i am more or less debating chemistry and brand. I have seen some of Will's videos on battery break downs and how he recommends against lead acid. I would avoid any company he actively dislikes, but be ok with ones he was at best neutral on.

I built a starter/learner system somewhat similar to yours, with six 210 W panels for a total of 1.26 KW, and a 24 volt / 3000 KVA (basically 3000 W for discussion purposes) Victron inverter. I started with one Amperetime 100 AH battery, then added another about four weeks later. After another six months or so I added a third. I'm very happy with them for a stater/learner system and have no regrets. To power the "shed" buildings for which I built the system, I doubt I'll need to change anything for the next decade.

Some here say 24V will be useless but that's a gross exaggeration and overstatement that I wouldn't worry about if you've already purchased the equipment. That said, I would go 48V if I were starting over, simply because it will more easily scale and the prices now aren't that much more than 24V. There's absolutely no way I'd go with a 12V system or buy 12V batteries and connect them in series. But if you've already purchased some stuff, you should be fine. Take a look at @Will Prowse battery tear-down videos and go with a brand he likes, and you'll likely be as happy as I've been. BTW, I do advise adding in a good shunt (Victron is awesome) as it's the only accurate way to measure state of charge with lithium batteries.
Cool that you started out with a system similar to mine. My goal is to move non critical loads onto it. and see what happens. lol.

I haven't purchased the AIO yet, but I already have the panels, so I know what the AIO needs to be capable of. I think for the system I would be building in the long run I would go with 48v, but I have somewhat settled on using this AIO that does 24v. I have watched several of Will's battery break downs and have been looking at some of the brands he didn't actively recommend against. It may be good for me to get batteries that make me do bot series and parallel just so I can have the experience or maybe not. I could technically do the same with far smaller batteries too.

I am looking for a way to monitor the system information too. I would like to be able to combine it with home assistant, but its not clear how I would do that with the AIO that i currently have selected. I have a lot of respect for what Victron is doing, but their stuff is just so expensive. lol I am not sure if there is a more budget friendly energy monitor out there though.

wow that battery is fairly cheap. I found this on amazon and this is a company that Will was somewhat neutral on in a video I watched. doesn't mean he would necessarily recommend using them but he didn't say to avoid them. What you posted would be cheaper than these https://www.amazon.com/LOSSIGY-12V100Ah-Lifepo4-Trolling-Off-Grid/dp/B0BHVCC4NM/ref=sr_1_3?dib=eyJ2IjoiMSJ9.DGJQMVJ97909BWWKeCuYYFm6TsgJ_AMy5SHoJuhwlaRWPRx3nC8ovLDABq6BOdIHVHWeQEv1jBv9UuqIkZrVGJGTcTZIu073lE4JRamFxaqlIT-Bm7yOImLblLdm5SjFhNxMGl02G9FnYXG4b85iignx-LZUOgqlr1VXxGdxavoQBJHXXGjfQAzcU2fjZDY2x_JHi2boh7aHQ5u1QvDRiUfadaZD7Hsx4QI0B4qN1oM.3cOUMUNeQHoQrCsuolJPdpfj-uWeNesznAjZrLK-JRI&dib_tag=se&keywords=lossigy+12v+100ah+lifepo4+lithium+battery&th=1


Also, thank all of yall for the suggestions. this is great.
 
If 48V a possibility at all, I’d look at the EG4 6000XP. I have yet to see a better value AIO based on specs and what others are generally reporting (I don’t have one - yet).
 
If 48V a possibility at all, I’d look at the EG4 6000XP. I have yet to see a better value AIO based on specs and what others are generally reporting (I don’t have one - yet).
EG4 equipment is certainly on the board for my full system build. I don't want to buy one like that and screw it up, hence why i'm generally going for cheap stuff for the learners build.
 
Im probably the only one on here that recommends golf cart 6v lead acids.

I also love my 24v system and would do it all over again.

Golf cart 6v are widely available, generally around 220ah, can handle cycles just fine, local options to warranty or replace, can be configured to any voltage, dont care about temperature and the best part is no bms.

My recommendation would be Trojan T-105, probably easiest to find at your local Club Car dealer. Or check Trojans website for dealers.

Heck even the Duracell line at Sams Club hold up good. If you really screw up within a year, go trade in for new under warranty. You cant do that on Amazon.
 
Im probably the only one on here that recommends golf cart 6v lead acids.

I also love my 24v system and would do it all over again.

Golf cart 6v are widely available, generally around 220ah, can handle cycles just fine, local options to warranty or replace, can be configured to any voltage, dont care about temperature and the best part is no bms.

My recommendation would be Trojan T-105, probably easiest to find at your local Club Car dealer. Or check Trojans website for dealers.

Heck even the Duracell line at Sams Club hold up good. If you really screw up within a year, go trade in for new under warranty. You cant do that on Amazon.
that seems like it could get pretty expensive. from what i have understood looking around, you can't get the full capacity of a lead acid battery without causing big time damage and this could reduce lifespan a lot. I may be wrong about that
 
What happens if you add a battery with the same voltage but a different Ah rating? For example, a 12v 100 ah battery with a 12v 200 ah battery?

What happens if the solar panels are still generating and the batteries are full and there is no load to take that energy? The charge controller would stop that right?
 
What happens if you add a battery with the same voltage but a different Ah rating? For example, a 12v 100 ah battery with a 12v 200 ah battery?
They become 300ah @ 12v
What happens if the solar panels are still generating and the batteries are full and there is no load to take that energy? The charge controller would stop that right?
Solar panels only generate when they are loaded by the charge controller. No load = no generation (or little to keep things @ float voltage).
 
It's not very obvious to people who think inside the box but your learner system shouldn't exist. There's far too much info out there proving that simplicity is built right into new products. Just don't let marketing catch you up in making your system more complex than it has to be. You should have two systems. One is a 12-volt with 4 group-29 marine batteries from Walmart. Don't waste your money on the MAXX. It's a slightly higher Ah rated battery that is advertised as a slightly lower Ah with a deeper rate of discharge. This is deceit intended to get you to spend 10% more money for 2% more usable potential. Parallel the batteries and run your fridge on a 1000-watt inverter while using the batteries directly for lighting anything you can illuminate with RV LEDs. Remember to connect the positive to the battery on one end of the bank and the negative to the battery on the other end so that no battery is drawn on too much more than any other.
Your other system should be a 48-volt AIO. This is the standard of the industry and is the highest voltage that provides low enough risk of breaching the resistance if your skin and ending your life. It will also carry the lower amperage for equal wattage on a smaller wire that will save you money. You may feel a tingle when completing the circuit. Be sure to tap the final connection on a battery terminal a few times until the spark is very minimal. Then make the connection. Like some people elude to, you will outgrow your system. I have about 6000 watts of panels charging 70,272 watt hours of Lead acid storage giving me 5000 watts per hour for up to 35,136 watt hours before the next full topping of the batteries. I use Lead-acid for my reasons. If you like something else, use it but your 12-volt system powering your fridge should be lead-acid for reliability's sake. Just keep in mind that I described a system having 5,856 watt hours but only 2,928 usable. Some fridges can go through that in 2 days. You may want to make sure your incoming power exceeds the potential daytime use of the loads plus the battery bank capacity. Being off-grid means occasionally not having use of all your conveniences due to long periods of overcast skies. You can compensate for this but the system needed won't have saved you money over being on grid or running a 48-volt generator to charge your bank. Solar is as much a personal choice as a logical one.
 
that seems like it could get pretty expensive. from what i have understood looking around, you can't get the full capacity of a lead acid battery without causing big time damage and this could reduce lifespan a lot. I may be wrong about that
You're not wrong. Prolonged storage under 11 volts causes irreversible calcification and no amount of 35-Amp welder boosting is gonna fix that.
 
What happens if you add a battery with the same voltage but a different Ah rating? For example, a 12v 100 ah battery with a 12v 200 ah battery?

What happens if the solar panels are still generating and the batteries are full and there is no load to take that energy? The charge controller would stop that right?
Understand the difference of being wired in series and in parallel. Also do not apply the same standards for lead acid batteries to LiFePO4 batteries. Each requires their own approach when using. Since you eventually intend to create a whole house solar setup I really recommend you just start with LiFePO4 and avoid lead acid. If you are interested in understanding batteries the website https://batteryuniversity.com/ is a good place to start.

I will mention once again it is not a good practice to connect in series 2 or more LiFePO4 batteries that have their own BMS. It can be done with types that communicate but the simpler BMS's found in inexpensive pre-built batteries don't have that feature. Connecting in parallel is no problem.

Solar panels only generate power when a load demands it. A SCC job is to properly charge the battery and it stops charging once it determines the battery has reached full charge based on voltage and current flow. There are two stages of charge. Bulk or Boost, known as CC or constant current, is where the SCC outputs up to the max amperage that it can based on its design and settings, if the panels/sun can generate enough to reach it. Once it brings the battery voltage up to that level it switches to CV, or constant voltage, where it holds that voltage for a period of time by supplying current enough to do so. That current drops as the battery gets towards full charge as the constant voltage supplied by the SCC and batteries internal voltage equals out. Eventually the SCC says the battery is full and no more charging happens until battery voltage drops enough to start the process over.
 
It's not very obvious to people who think inside the box but your learner system shouldn't exist. There's far too much info out there proving that simplicity is built right into new products. Just don't let marketing catch you up in making your system more complex than it has to be.
Fair point. I am not a total idiot, but this is my first time doing something truly electrical, so i am bound to make some kind of mistake. I would rather ruin cheap equipment than expensive equipment. A warranty is only good if its the a defect in the product causing failure, not if I screw something up. I don't know if i'm making it more complicated than I need to or not. I have no idea how easy it is to fry an AIO or any other bit of equipment. I have my sights set on a particular cheaper AIO that runs on 24v, but i haven't bought it yet (because I am asking them questions now). It seems like that would be cheaper anyways than buying a separate charge controller and inverter.

As far as my experience goes, I have wired up wall plugs and switches before and I wired up my current electric water heater, but that was so easy that it didn't feel like doing much anyways. Not sure that building a solar system will truly be like that. I didn't have to adjust any settings and I didn't have to pick out compatible materials, etc. Hardest part was selecting the proper wire and breaker size for the water heater and that wasn't hard at all.


Your other system should be a 48-volt AIO. This is the standard of the industry and is the highest voltage that provides low enough risk of breaching the resistance if your skin and ending your life.
Like some people elude to, you will outgrow your system.
My plan is to have a 48-v system afterwards anyways. I am working on sizing what that system should be. Because of what utility territory I live in, it is absolutely imperative that I do not let a single watt-hour go to the grid. lol I live in a suburb, so going fully off grid does not make sense. Just replacing a part of my load does. It would probably be super expensive to handle the peak load that could be created by my range anyways, since it is the piece of equipment that has the potential to draw the most load in the house.

I will mention once again it is not a good practice to connect in series 2 or more LiFePO4 batteries that have their own BMS. It can be done with types that communicate but the simpler BMS's found in inexpensive pre-built batteries don't have that feature. Connecting in parallel is no problem.
I shouldn't add the cheaper LiFePO4 batteries in series, even if they claim they can be added in series?
 
Back
Top