12V DIY solar power station

[Bacchus]

I recommend watching some videos by this guy.

DIY Solar Power with Will Prowse

Want to build an awesome off-grid solar power system? My videos will teach you everything you need, no experience necessary :)

youtube.com

Has some good stuff and tear downs to find the better quality inexpensive lifepo4 batteries...


Also, going to pickup a 50a minimum lifepo4 charger to run off AC as when the power goes out, my existing 500w of solar may as well be 50w in a storm..

As such, my 2550w inverter generator can recharge my batteries at in a few hours and given the draw off the 50a charger, say 4 hour of generator to recharge the batteries at 200ah, 40hrs run time on a 5gal gas can, 10 full recharge cycles of the battery.

5 gallons gas with no solar input will give me 10days of run time of a 200ah lifepo4 battery bank running refrigerator.

I've seen a lot of Will's videos and I've learned a lot from some of them. At some point he gets annoying due to his disdain for smaller systems.

My new favorite is Everyday Solar. He covers things I'm interested in, he performs convincing testing, he explains fundamentals that apply to whatever scale you are working with, and his terminology is always correct.

 

[Bacchus]

My favorite Will Prowse site. Lots of system information and also details on where to buy and which components are recommended.

I also found two solar power forums, although there are probably more:

DIY Solar Power Forum. (Will Prowse)

NAZ Solar Electric Forum.

A while back I was confused by some of the test results I was getting regarding my Inverter reactions towards extreme voltage inputs. Today I dragged my variable power supply out to the garage and hooked it up to the inverter inputs. Starting at 14vdc I gradually decreased the voltage and recorded the inverter responses. At 10.0 vdc the inverter activated an alarm. At 9.5 vdc it shut off the inverter output.

I reset to 14vdc and gradually increased the voltage. At 15.6 vdc the inverter activated an alarm and shut off the inverter output.

I expected the inverter to act in my battery's behalf, no surprise there, but it is helpful to know exactly how much margin I have for voltage deviation. I guess there are some inverters that allow you to set the trip voltages, but mine does not.

This really emphasizes the need to reduce as much as possible the voltage drop between the battery terminals and the inverter inputs. I had a 1.5 vdc drop for a while, and that meant that at a battery output of 11.0 vdc the inverter would only see 9.5 vdc at its inputs and would shut down early.

 

[Bacchus]

I recommend watching some videos by this guy.

DIY Solar Power with Will Prowse

Want to build an awesome off-grid solar power system? My videos will teach you everything you need, no experience necessary :)

youtube.com

Has some good stuff and tear downs to find the better quality inexpensive lifepo4 batteries...


Also, going to pickup a 50a minimum lifepo4 charger to run off AC as when the power goes out, my existing 500w of solar may as well be 50w in a storm..

As such, my 2550w inverter generator can recharge my batteries at in a few hours and given the draw off the 50a charger, say 4 hour of generator to recharge the batteries at 200ah, 40hrs run time on a 5gal gas can, 10 full recharge cycles of the battery.

5 gallons gas with no solar input will give me 10days of run time of a 200ah lifepo4 battery bank running refrigerator.

I finally got around to looking at the video you linked. That's an excellent resource, I saved the link.

When I was using my power supply to survey the inverter response, one of the things that surprised me was that the inverter was drawing over an amp from the batteries even with no load on the inverter. 12W ! Not really a problem for me, I always switch the inverter off when I don't have anything plugged in.

 

[Bacchus]

Not finding the 100Ah used car batteries I want. The attraction for me was that I could get them for about $60 versus $300 for a LFP battery, which put me ahead with the used battery even with its 50% of nominal energy capacity.

Or so I thought. I discovered that LFP batteries can now be had for $150, which makes them the better deal since for about $140 I can get roughly 1200 Wh with either one LFP or two lead acid.

But even that is not right. I decided to do an apples-to-apples comparison of the batteries I had on hand, and I discovered that my used batteries had a much more reduced capacity than I would have guessed.

I tested them with a box fan set to its highest output which I measured at 72W, almost exactly what my refrigerator averages. I charged them up and then ran them on my solar station until the inverter quit. I also had a Delta 2 unit available, 1024 Wh, that I ran for comparison. All tests were run in my garage at temperatures ~ 55F.

Test 1. My "new" refurbished AGM, 95Ah: 800 Wh, 11:31 h:m.
Test 2. My used truck battery, 90Ah: 310 Wh, 4:35 h:m.
Test 3. A used battery from my car: 170 Wh, 2:34 h:m.
Test 4. EcoFlow Delta 2, 100 Ah: 740 Wh, 10:46 h:m.

The battery in the Delta 2 is an LFP, so I am guessing that its inverter was more protective of the battery than is my solar station inverter.

From these pitiful results I decided it was time to go with LFP, and I bought two of them.

Test 5. LFP #1, 1280 Wh: 1160 Wh, 16:46 h:m.
Test 6. LFP #2, 1280 Wh: 1220 Wh, 17:40 h:m

 

[Bacchus]

Next I tested using my refrigerator with the two LFPs from Tests 5 & 6, with the system inside the house and no solar panels. The total energy used was 2430 Wh, and the total time was 30:31 hours:minutes. That comes to about 80W average, usually I see about 73W. However, I was testing long enough to see two defrost cycles 600W @ 3 minutes, if that's what they were.

That achieves my goal of running my refrigerator for 24 hours with a system cheaper than a comparable commercial unit. My system is now $649.24 (w/o solar panels). To get a 2400 Wh EcoFlow Delta 2 today on Amazon, it costs $700 for the basic 1200Wh system plus another $600 for a 1200 Wh external battery, or $1300 for both. I don't know why their extra battery costs so much. My 2400 Wh system is almost precisely 50% of the EcoFlow.

My next step is to get a couple more solar panels and verify they will function as expected and to insure I have all of the connecting cables and adapters.

And then I'm done with this particular solar station, and I can start thinking about the next one with enough energy to power my gas furnace, withs its 1/3 HP blower fans, or maybe even my window air conditioners, but those are 400, 600, and 1000 W, so I might decide they are out of reach. My next solar station will likely be wired in a 48V configuration, use a MPPT instead of PWM controller, and more batteries.

I can almost build a smaller, 1200 Wh system, from the leftover parts from this project, using that AGM battery. The leftover parts were parts that I replaced in my 2400 Wh system when I determined those parts were insufficient.

 

[Bacchus]

By the way, I'm certainly not saying my $650 system is equal to the EcoFlow $1300 system, just that mine meets my limited requirements for half the price.

The EcoFlow is certainly the better system, with its multiple outlets, far better system status displays, and its sensational 2-hour recharge time using house 120VAC. I'm using a 10A battery charger, and it takes about 14 hours (each!) to recharge my batteries with my cheap charger. Much faster with four solar panels and a sunny day, of course.

If somebody offered me either system for free, I would certainly choose the EcoFlow, but since I'm buying I like my $650 system.

At some point I had the thought, what if I were to get the EcoFlow system and then use my own LFP battery ($150) instead of buying the EcoFlow "Extra Battery" for $600. Can't I just plug my battery into the EcoFlow external battery connector? Not very easily, I read where somebody claimed the EcoFlow extra battery port is for 48V, and it has a connector that I've never seen before. The EcoFlow specs that I looked at online for the Delta 2 system and the Extra Battery don't mention the voltage at the extra battery port. I think they are trying to force the customer to buy their extra battery.

Some Redditors had a discussion about this. There seemed to be two schools of thought: "Yes" and "No." Several claimed you could connect your own battery into the EcoFlow PV port (not the extra battery port), using splitter cables to also include the solar panel input. I'm not sure about that, it seems to me the battery would be delivering 12V and the solar panels might be delivering much more than that; after all, we're on the input side of the EcoFlow controller. (How come I never see diodes in any of these configurations?)

Somebody else suggested you could just connect one item (solar panels or external battery) at a time to the PV port, and I don't know why that wouldn't work, but you would have to switch them twice a day. Connect the panels during sunshine and then disconnect the panels and connect the battery between sunny periods. Even that isn't quite the same thing because there is no opportunity for the external battery to get recharged. I guess you would just get extended time until your extra battery depleted, kind of like having an extra can of gas for a generator.

 

[Bacchus]

I just saw another YouTube video where the guy extended his system capacity by plugging his system into another solar station. The idea was the second station would function the same way as charging from the wall. So that's a third possible input port.

 

[osprey]

I finally got around to looking at the video you linked. That's an excellent resource, I saved the link.

When I was using my power supply to survey the inverter response, one of the things that surprised me was that the inverter was drawing over an amp from the batteries even with no load on the inverter. 12W ! Not really a problem for me, I always switch the inverter off when I don't have anything plugged in.

Hence my comment about getting a dc powered fridge/freezer to not have the parasitic draw of an inverter. Just running my 45qt dc fridge I am sure I could go two weeks or more on my 140ah battery before needing a charge. If I added 200W of solar I probably could go a month or more at least in the summer.

 

[Bacchus]

Price war on commercial power stations on Amazon.

For example,

EcoFlow Delta 2 $414 today.

EcoFlow Delta 2 Expansion Battery $449 today.

A commercial 2048 Wh system for $863.

 

[Bacchus]

I decided at the current low Amazon price to get my own Delta 2 and Extra Battery. I can use the Delta 2 for the refrigerator and continue to expand my DIY system, aiming for furnace fans and possibly air conditioners.

So I tested the Delta 2 + Battery (2048 Wh) on my refrigerator, everything inside the house and no solar panels attached. Testing without panels gives me a lower limit of what the system can do, testing with solar panels on a sunny day will give me the upper limit.

Yikes! The Delta 2 + Battery only powered the refrigerator for 20 hours. I got 30 hours two weeks ago with my 2048 Wh DIY system. The Delta 2 only scored 1480 Wh/2480 Wh = 60%. I paid about 30% more for the Delta 2 (on a really good sale, about 35% off) and only received about 2/3 of the performance as my DIY system.

I also put quite a few hours into developing my DIY system, but that was fun and not a chore.

 

[coop44]

A chest freezer with an independent temperature controller may work. The biggest problem with a standard refrigerator is the loss when the door is opened, the cold falls out. A controller will allow you to keep the lower portion of the chest frozen while the upper part is above freezing, depending upon where you place the temperature sensor. Lots of off grid folks use this setup.

A chest freezer will draw less power, in general. They use smaller compressors, and run for less time and are generally better insulated than a refrigerator.

Use a Johnson temperature controller.They make a few different ones. Try this,

View: https://m.youtube.com/watch?v=IFV_5B32aMs

If you use lead acid batteries, go to deeper cycle ones, 6 volt golf cart or 8D diesel starting batteries. Either would be superior to automotive.

 

[Bacchus]

I picked up a new pair of solar panels. My one solar panel wasn't sufficient to recharge my 1800Wh refrigerator each day. Three panels won't be enough either, at least not until there are six good sunny hours in the day.

The two new panels were fine. I also wanted to ensure that I had enough cables and adapters and splitters for the new panels. And that part was not fine.

When I first bought parts for this project I bought very inexpensive items, and the panels and cables and controllers and ac chargers all had SAE connectors, and so I bought long SAE cables to reach outside from the station to the panels. As I upgraded most of these components I kept the SAE components in my station for legacy reasons. My two new solar panels have MC4 connectors, and so I used MC4 splitters to put the panels in parallel, and then I used MC4-to-SAE adapters to connect to the long SAE cable, connecting to the controller with SAE.

My first problem was that the SAE adapter connected the MC4 (+) to the black wire in the adapter and the MC4 (-) to the red wire. I didn't notice this because all of the MC4 cables were black and black instead of black and red, and I assumed the MC4 gender that connected to the SAE cable red wire would be the (+), but it was the (-). There is a tiny little red indicator on the MC4 (+) cable, but it wasn't obvious and I didn't think I needed to look for it. So I was wiring the voltage backwards from the panels to the controller!

I was afraid that maybe I damaged the controller, but luckily the controller was smarter than me and ignored the faulty polarity at its inputs. So I grabbed one of my SAE polarity inverters and put it inline so that the SAE cables' red wires would always be (+); makes it easier to keep track of and to recognize at a glance.

Still didn't work; I was getting no voltage at the controller input. I went back to the panels and disconnected its cables from the splitters and checked: +21V, good. But for some reason tracing back to the controller for the missing voltage was harder than it should be; it should be easy! And, of course, the answer was that one of the cable segments had an intermittent connection. Sometimes it was there and sometimes it wasn't. I've debugged a lot of intermittent electrical and electronic connections in my storied career , and this intermittent was in the long SAE cable. I replaced that cable with a different SAE cable and everything worked fine.

I only used that cable maybe six times previously in all of my testing, so I'm surprised that it went bad so fast. Not worth my time to open, inspect, and replace an SAE connector, so I'll cut off both connectors and save the 16AWG cable.

 

[teflon6string]

Fridges are great but they suck up lots of power. I am listening to ours running right now. Honestly it feels like it runs more than it doesn't.

Refrigerators and freezers actually make heat (and magically expel it out back somewhere). Anything that makes heat sucks power. A laptop (or CPAP machine) on an inverter will sip at a battery for days while a fridge or stove will drain it pronto.

 

[OrPackrat]

Refrigerators and freezers actually make heat (and magically expel it out back somewhere). Anything that makes heat sucks power. A laptop (or CPAP machine) on an inverter will sip at a battery for days while a fridge or stove will drain it pronto.

They transfer heat from inside to outside.

 

[icemanb105]

A short while back I started my current project, building a 12V DIY solar station that would 1) power my refrigerator every day 24 hours a day as long as the sun was shining, on occasions when we have a power outage; and 2) be cheaper than commercial units with similar power (W) and energy (Wh) outputs; and 3) provide an education to prepare me for a followup power station that can keep my air conditioner running.

My current status is that at $338.91 (excluding solar panel) I have a system that is 2000W power and a nominal 2400Wh energy capacity. And since I am using lead acid batteries the energy capacity is closer to 1200Wh, compared to the commercial systems that are using LFP batteries. My system has two car batteries, an inverter, a controller, a battery state monitor, a power input monitor, and an AC charger. Most comparable reputable brands seem to cost about $600 for that power and capacity.

But, the name brands include more, such as LFP batteries that will last at least triple the number of power cycles; much nicer and more informative power status displays; maybe an app to control thru Wi-Fi; more connectors, including USB support, and more. Since I don't expect to use this more than about once a year (not many power outages in my neighborhood), the cheap lead acid batteries are fine.

And, neither my system in its present state nor the $600 commercial systems meet my refrigerator criteria: not enough capacity to run my refrigerator for 24 hours.

During my last test (with two batteries and one solar panel) my station ran for 16 hours and 44 minutes and delivered 1370 Wh of energy (my refrigerator needs about 1750 Wh in 24 hours). My one solar panel contributed 353 Wh from 9am to 1:30pm (sunny hours) before the sky clouded up.

Nice. Good investment.

 

[Bacchus]

Today's news is that certain power inverters made in China have been discovered to contain cellular communications circuitry. The threat is that the Chinese could remotely turn off the inverters, disabling power, and thus impair at least parts of the grid.

I looked at half a dozen stories on the Internet but none of them identified the brand of inverter. Chances are that this applies to inverters with a much larger power capability than my dinky little DIY system inverter. But if I knew my brand of inverter was under commie control I would replace it asap. I'm petty that way.

 

[OrPackrat]

Today's news is that certain power inverters made in China have been discovered to contain cellular communications circuitry. The threat is that the Chinese could remotely turn off the inverters, disabling power, and thus impair at least parts of the grid.

I looked at half a dozen stories on the Internet but none of them identified the brand of inverter. Chances are that this applies to inverters with a much larger power capability than my dinky little DIY system inverter. But if I knew my brand of inverter was under commie control I would replace it asap. I'm petty that way.

Put your inverter in a Faraday cage

 

[Bacchus]

Put your inverter in a Faraday cage

Inverters are cheap enough I would rather replace it if it came to that.

I thought of a Faraday cage, although I was thinking of one of the "locking" foil bags and I didn't have confidence it would have enough ventilation. Anything bigger would be a pain for my portable system.

 

[2nd Amendment]

A short while back I started my current project, building a 12V DIY solar station that would 1) power my refrigerator every day 24 hours a day as long as the sun was shining, on occasions when we have a power outage; and 2) be cheaper than commercial units with similar power (W) and energy (Wh) outputs; and 3) provide an education to prepare me for a followup power station that can keep my air conditioner running.

My current status is that at $338.91 (excluding solar panel) I have a system that is 2000W power and a nominal 2400Wh energy capacity. And since I am using lead acid batteries the energy capacity is closer to 1200Wh, compared to the commercial systems that are using LFP batteries. My system has two car batteries, an inverter, a controller, a battery state monitor, a power input monitor, and an AC charger. Most comparable reputable brands seem to cost about $600 for that power and capacity.

But, the name brands include more, such as LFP batteries that will last at least triple the number of power cycles; much nicer and more informative power status displays; maybe an app to control thru Wi-Fi; more connectors, including USB support, and more. Since I don't expect to use this more than about once a year (not many power outages in my neighborhood), the cheap lead acid batteries are fine.

And, neither my system in its present state nor the $600 commercial systems meet my refrigerator criteria: not enough capacity to run my refrigerator for 24 hours.

During my last test (with two batteries and one solar panel) my station ran for 16 hours and 44 minutes and delivered 1370 Wh of energy (my refrigerator needs about 1750 Wh in 24 hours). My one solar panel contributed 353 Wh from 9am to 1:30pm (sunny hours) before the sky clouded up.

Might check out the power draw on a refrigerator designed for off-grid use and see how they compare to yours:

Category: Appliances

backwoodssolar.com

 

[Bacchus]

Thanks for that link.

Most of those refrigerators are only 2 to 4 cubic feet. I didn't look at every single one of them, but the biggest I found was 16 cubic feet. My existing refrigerator is 39 cubic feet, and sometimes I have to do some creative rearranging to get everything to fit. Mine was also much cheaper, but that was more than ten years ago when I bought it.

I can see some value in having a smaller unit that I can store in the garage to keep more expensive items in case of a power outage, particularly a small chest freezer. This one, for example. 8 cubic feet is sufficient for about half of my freezer items, and it only draws 495 Wh per day used as a freezer with 90F ambient temperatures., compared to the 1800 Wh (70F inside temperatures) of my refrigerator-freezer; and only 195 Wh if I used the chest freezer only as a refrigerator. I'm gratified the seller provides that information.