diy solar

diy solar

Off grid without battery storage - Draft Design Reviews welcomed

carleriksen

New Member
Joined
Aug 28, 2021
Messages
15
I am interested in building a system that is only intended for me to tinker with and to have a source of back up power. I know the better answer is a generator; and I do have a generator. But, you know, you could run out of gas...or it could be smart to extend the gas.

I will not tie it to the grid. I will probably not even attempt to have a transfer switch. So a completely separate system.
I'd like it to function like the generator, to be able to plug a few things in and run them when the sun is up. Generator if there is no sun.
I am not interested in trying to store power. Batteries quickly become expensive and I don't want to maintain them. Cheap, simple system is desired.

My initial question is: Do I need batteries to make it work? What I have come to understand is that I probably do - for startup/smoothing out power. Is that correct?
Can you use a tiny battery if you don't care about storage? Car battery? Starter batteries have great short duration amp draw.

I have 2kw worth of panels; 10 ea 224w nominal 30v. I also got 10 microinverters but they are utility type so no good. Nothing else purchased yet. Panels were a deal that literally showed up at my door.


EDIT:
I live in SoCal - Inland Empire area - sun is plentiful and cloudy days rare.
Loads will be same as I have tested to run with my Pulsar 2300/1800W inverter generator - refrigerator, some lights, one window AC, my sons PS4 (of course!) and if I can a microwave. I don't expect to run all at the same time - in particular microwave, that is right at the edge of the Pulsar's capability. All this is emergency load planning.
Solar would be to get whatever power I can and spare the gas.

Why do all this? I've decided I like the idea of being a resilient citizen - someone who can handle up to several months without government assistance given a major crisis. Insurance in a way. We pay insurance for so many things, I decided I wanted some element of power insurance. Not power at a daily normal level, just power to handle an anomaly.

EDIT 2:
Installation will be on a patio cover in the back of the house, regular residential neighborhood.
 
Last edited:
Welcome! Interesting set up. We will need the specific details on the micro inverters. They may be useful.

There are several ways to proceed. Best to not buy anything for a bit while you find your way.

Others will want more detail and have more answers.
 
I believe MPP Solar has several models that will operate without batteries, according to their website that only applies if its 1 inverter. If 2 or more are stacked for split phase then batteries are required. Enphase IQ8 microinverters are designed to provide back up power without having to install the batteries. Not sure if the Enphase system can be operated without their Smart ATS which you wouldn't need if not connected to the grid. Either of the above may be what you are looking for, however, some additional research is needed.
 
Wow, thank you for the welcome and fast replies! That was quick.

Well, I am happy to provide any detail and to listen to others experiences and opinions.
I have attached, I hope - if it works, a picture of the panel name plate and the microinverter name plate
 

Attachments

  • Sharp panel.jpg
    Sharp panel.jpg
    97.6 KB · Views: 15
  • Enphase microconverter.jpg
    Enphase microconverter.jpg
    170.9 KB · Views: 14
Last edited:
Just my 2 cents but including even a small lead Acid battery in a system like 100 Amps will make things so much more pleasant to use. You really need that buffer for when the clouds pass over to keep thing s going even if it's for 15-30 minutes.
 
Good point, I had to some degree gathered that a battery might be needed as a 'starter' for loads, but you are very correct that buffering for variation in output is a consideration too. I will say that my initial No Battery stance really is mostly that I don't want to have a real bank. I am certainly willing to include some minimum to make things work; truthfully that is what I hope to learn, how to do this minimum.
 
My initial question is: Do I need batteries to make it work? What I have come to understand is that I probably do - for startup/smoothing out power. Is that correct?
Can you use a tiny battery if you don't care about storage? Car battery? Starter batteries have great short duration amp draw.

As mentioned you'll want at least SOMETHING in there to smooth out the power flow, even if it's just a little battery. What you'll want to make sure to do is get a Deep Cycle FLA, AGM, GEL, or LiFePo4 to do it with. You Could do it with a bog standard car battery, but you'll probably destroy the battery pretty quick as it's not designed to discharge and recharge multiple times in a row, it's more a 1 big OOMPH! and some time to take a break and recharge. Fortunately, a nice size DC29 122Ah FLA Deep Cycle at WallyWorld is about $100 out the door with core charge and taxes and plays well with solar. That'll be cheaper than anything you can get at the auto parts store and give you a nice buffer space.

The best bang for the buck would be a Chins/AmperTime/whatever 100Ah LiFe's if you can afford it. They're about the same as an AGM but give you twice the usable amp hours and you still get all the smoothing and buffering your solar system wants.

The big banks come into play when you want to be able to run the PS4 all night long and have krappy weather all week. :)
:)
 
Last edited:
Another vote for the Chins/AmperTime/whatever 100Ah LiFe's. A good start. Will also need a charge controller.
 
Welcome! Interesting set up. We will need the specific details on the micro inverters. They may be useful.

There are several ways to proceed. Best to not buy anything for a bit while you find your way.

Others will want more detail and have more answers.
I agree wholeheartedly.
I currently have 16 panels grid tied using Enphase micro inverters that are looking for 240vac nominal with <5% total harmonic distortion and most of the inverters I'm seeing provide power @ <3% thd so I should be able to trick them into thinking they're on a legit grid if I have battery storage capacity to send excess power they produce.

What I'm working toward is very similar to what you @carleriksen seem to be also working toward. I started my thread on the subject yesterday too. You might find the discussion there to be helpful?

 
Thank you for the replies and insights so far!

Reading more I now understand a bit about the micro inverters; in particular that they simply dump the power indiscriminately. Sufficient batteries to absorb available power would be needed so my take to not use them seems to be correct. If there is something I am not understanding right with regards to these I will gladly listen and learn.

Regarding choice of minimal battery. Given that I would not use this system on a daily basis, it is a solar backup generator so to speak, would that change the outlook on battery of choice?
The battery would live on a trickle charger till needed. I've heard that deep cycle batteries does not like to be stored like that. I understand that starter batteries do not do well with discharges but they do well living on a maintainer.

The Walmart size 29 marine is impossible to beat for price, that is for sure. 59lbs of 12V battery for about a hundred.

@Ozark Tinkering Yes, it looks like we have similar thoughts. You are ultimately building much bigger than I am but same basic thought process
 
Reading more....

I am having to 're-calibrate' as 2 years ago I spec'ed out a bigger system with batteries and all and came to the conclusion that it was not making sense. Too much money for the insurance I was going for. Back then I never considered anything but FLA but now I am learning that AGM might be the right thing. Doesn't have the cycle life, doesn't have the maintenance, does have high discharge current, lives well on float....

It all looks so different now that I am thinking of the gas generator as the short/medium term 'battery bank'
 
I am interested in building a system that is only intended for me to tinker with and to have a source of back up power. I know the better answer is a generator; and I do have a generator. But, you know, you could run out of gas...or it could be smart to extend the gas.

I will not tie it to the grid. I will probably not even attempt to have a transfer switch. So a completely separate system.
I'd like it to function like the generator, to be able to plug a few things in and run them when the sun is up. Generator if there is no sun.
I am not interested in trying to store power. Batteries quickly become expensive and I don't want to maintain them. Cheap, simple system is desired.

My initial question is: Do I need batteries to make it work? What I have come to understand is that I probably do - for startup/smoothing out power. Is that correct?
Can you use a tiny battery if you don't care about storage? Car battery? Starter batteries have great short duration amp draw.

I have 2kw worth of panels; 10 ea 224w nominal 30v. I also got 10 microinverters but they are utility type so no good. Nothing else purchased yet. Panels were a deal that literally showed up at my door.


EDIT:
I live in SoCal - Inland Empire area - sun is plentiful and cloudy days rare.
Loads will be same as I have tested to run with my Pulsar 2300/1800W inverter generator - refrigerator, some lights, one window AC, my sons PS4 (of course!) and if I can a microwave. I don't expect to run all at the same time - in particular microwave, that is right at the edge of the Pulsar's capability. All this is emergency load planning.
Solar would be to get whatever power I can and spare the gas.

Why do all this? I've decided I like the idea of being a resilient citizen - someone who can handle up to several months without government assistance given a major crisis. Insurance in a way. We pay insurance for so many things, I decided I wanted some element of power insurance. Not power at a daily normal level, just power to handle an anomaly.
As some said here, try to find a small battery or old one with a high C-rate; like from an old car. Something you can use as a "capacitor" to feed your loads for 10 minutes while that single one annoying cloud in the clear sky day passes in front of the sun.
 
Thank you for all the help on the battery question, I now feel confident that I can try with something simple/cheap to at least be able to pull power while power is generated.

To keep it very cheap, at least for testing to prove out the concept, I am now thinking to go 12V instead of 24V. My panels would ideally be used in a 24V system but I got them for only $250 so treating them as if they were 12V appropriate while technically a waste isn't a big financial waste.

So for really cheap I could use a 12V PWM charge controller, a 1000W/12V pure sine inverter and a 12V car/or marine battery.
I am looking at Rich Solar's PWM 30 amp

My next question is to see if I have understood PWM controllers. Their main drawback is upstream efficiency, is that correct?
As I understand a 12V PWM will reduce the voltage to about 13/14 volts. That would drop the generated wattage from my panels to half or even a little less. Am I getting that right?

Am I correct in thinking that if I obey the max charge current and max voltage spec on the pwm I am good to go?
In case of Rich Solar pwm it states 30 amp and max 50V

Given the panels I have I think I can connect 3 in parallel for 22 amp and 30V; I can only get up to ~13.8V 22 amp charge current on the other side of the controller. Are both of these thoughts right?

If all this works out I could make 3 of those strings, each parallel to the battery, and I'd have up to 3*13.8V*22A = 910W on the battery side for charge or load. Is that right?


Now please, shoot some holes in this where I am wrong! I am here to learn

EDIT
I am looking at 12V rather than 24v, even though the PWM can do 24v and my panels are better suited to 24v, due to cost of 24v pure sine inverters and having less battery.
 
Last edited:
First of all, I am very enthusiastic to see what you are doing.

Second, I think you understand well what MPPT is and does. Yet I explain here too:
MPPT easy explained is a DC-DC converter that tracks the highest power point of solar panels. This is interesting even when you have both panel and battery at 12V because the battery limits the voltage of the panel and that translates into lower power output. MPPT escapes to that obligation because of the DC-DC step; the input side can work at the ideal panel voltage.

Eg. 12V solar panel. Nominal output: 18V 5,55A = 100W, so If battery voltage is 13V; 13V 5,55A = 72W max. output from solar panel.
The higher the SOC, the higher the energy output from the solar panel.

When you use a 12V battery with a 24V solar panel, MPPT I would say is essential because with 12V you are reducing the possible output 50% directly. So effective you may only get 36% of the nominal panels power when using PWM.

Third, costs of a pure sine inverter (off-grid inverter, right?) should not differ much for 12V or 24V; what makes the price differ is the wattage. So I believe buying 12V inverter or 24V inverter should not make a big change. But maybe you found something else? (Not an expert here)

Placing a 24V inverter does require you using 2 batteries for the 24V; so indeed, here you will have extra costs.


What you aim to do is nice, specially considering the lower price you paid for the panels; but it will limit significantly your max. output. Perhaps you can explore further if somebody is selling used car batteries at junk price to get the 24V and invest near to the same money for a 24V inverter. On the other hand, If your loads will NEVER be more than your possible 12V output, then indeed, no need to bother.
 
So for really cheap I could use a 12V PWM charge controller, a 1000W/12V pure sine inverter and a 12V car/or marine battery.
I am looking at Rich Solar's PWM 30 amp
Instead of a battery you can use a 12v battery charger or even convert a old PC PSU to start with to keep things going when the sun goes down.
PWMs are the ones for cheap panels. Just add another one for every couple of panels you come across.....I'm up to 22 panels, all converted to 12v.
The only things not on the system are the microwave and inverter cooktop.
 
Back then I never considered anything but FLA but now I am learning that AGM might be the right thing.

The occasionally adding distilled water is about the only difference between FLA and AGM, that and being able to lay the AGM's sideways if you wanted. If you're on a serious budget, don't discount the FLA's. Yes, you have to do occasional maintenance, but they're 1/3 the price on average and that buys a LOT of distilled water.

If the system were someplace remote and unattended then the maintenance aspect becomes much more important, but it sounds like that's not the case.

If you can swing $100, I've got 2 of the Older PowMr 60a MPPT units going on 4 years in service. Maybe not the most efficient or feature rich, but it'll give you plenty of room to grow to 24v or 48v down the line and handle 60a of charging. Most importantly, with your panels, CHEAP and EFFECTIVE which is a rare combination in solar stuff. :)
 
Great comments!
Thank you very much for the interest and time to help think this through. It is very much appreciated and helps me clarify what I am doing.

Taking into account the difference it makes going to 24V.... it is very true, I would get much more usable power not under driving the panels... after listening to what you guys are saying I realize I really nearly can have my cake and eat it here. The PWM's I am looking at auto switch between 12v/24v simply by the battery voltage reference they get. The MPPT mentioned is capable of same. So I don't have to decide until the battery what the system voltage really is. My panels lends themselves well to PWM at 24V nominal. Actual charge voltage would get real close to panel nominal

Taking wiring into account I think I now have a workable plan. I did not mention my installation site in my opening post, I should have. I plan to install on a (very solid, in-ground concrete anchored) patio cover. It gives me some great options. I will have access to the underside and I can use the cover provided by the panels to mount things very close to the panels, be protected from weather and still have easy access.

SYSTEM DESIGN:
What I can do is to create a distributed system with minimal wire runs. Three PWM's right by the panels, 3x3 panels parallel wired to PWM. Then have a junction box right by the panels and join the three PWM outputs. Then one heavy gauge run to battery below. Should not matter if I join the controllers at the battery or just before.

If I connect a 12V battery, I have a (~1KW) 12V system. If I connect 24V worth of battery I have a (~2KW) 24V system. Only real commitment is buying the inverter.

Given that the system is intended as backup/power insurance it makes further sense to have a distributed system. One controller failure would not end the day; just reduced power. The inverter does still create a single point of failure though...if only you could parallel inverters as you can with inverter generators; two smaller ones joined would remove single point of failure.
 
Last edited:
Back
Top