diy solar

diy solar

Undersized solar charge controller question

AussieKev

New Member
Joined
Jun 3, 2021
Messages
18
I have a 310W household solar panel. I don’t have the exact specs or data sheet for the panel. Assume it’s around 36v and max output of 9A. I have this panel connected to a Renogy 20A MPPT charge controller charging a 12v 280AH AGM battery.
In full sunlight during the middle of the day the max output of the charge controller is 20A.

The battery is often drawn down to 12.1v so I seem to use most of the available capacity. Now I assume that if I had a larger capacity solar charge controller I assume that the panel would be able to output more than the 20A that I currently get.

Just wanted to confirm here how much power I’m probably missing out on here. I assume that I’m probably missing out on the 5-7A for 1-2 hours around midday. Is that about right?
 
I think with the current setup, you are already 3-6A over the charge controller output limit. To prevent AGM battery to fail early, it is suggested to use only 50% of its capacity. You need to calculate the cost of charge controller and battery replacement since you are over worked your system.
 
Last edited:
So your panel pointed at the sun and angled correctly should give you at least 80% . At least that's what I get when point my panels ideally, and sometimes I get up to 100%.

I doubt you'll get more than 20 amps off your panel. 310 watts divided by a charging voltage of 14.5 gives you 21.3 amps, so you're pretty much at your panels capacity already. At best you'll see 2 more amps. So may not be worth it.

If the real problem is battery going to low, if charging stops before the sunsets and closer to noon, then I think you need to add another battery in parallel which assumes there's enough hours in the day your panel is not working.
 
Before we start: I think 310w on a 20A mppt is a great fit, but I am biased toward such arrangements. Read the following pre-coffee math with critical eye and healthy mistrust. :)


I assume that if I had a larger capacity solar charge controller I assume that the panel would be able to output more than the 20A that I currently get.

Probably a bit, but how much would each extra watt cost?

Let's assume that after conversion losses, panel temp derating, wiring losses, etc, that the MPPT could be expected to extract 83% of the panel's rated wattage at local solar noon under good conditions. So ~251w (310w x .83).

At 12.1v (morning) this would be ~20.7A, a tiny bit of clipping (8.5w) . But it would be a strange morning when you could make 251w.
At 13.6v (guessing at Vfloat) this would be ~18.5A - no clipping.
At 14.6v (guessing at Vabs) this would be ~17.2A - no clipping.

If the 30A controller cost $100 (chinese mppt) each highly unlikely extra watt would cost $11.76. If the existing 20A controller were priced that way it would cost something like $2,800, which none of us would pay.

Caveats:
  • If the controller is getting hot or otherwise doesn't like running near rating then I'd move up
  • if you are getting exceptional yield (high altitude, cold temps, tilted panels) then the incoming wattage and resulting clipping would be higher
  • If you resell the 20A to recoup costs the $/W of the extra harvest will be less
Here's a blog post of me working through the same decision on my own system so you know I have skin in the game rather than being strictly armcharir. :)
 
Thanks everyone !!
Sounds like it's not really worth upgrading the solar charge controller. I think I'll stick with my current MPPT and put any money towards a new lifepo4 battery ! Cheers.
 
If you only draw your battery down to 50% DoD at night, it will take an entire day's PV production to recharge the watts consume. Maybe more given the inefficiency of LA battery. You may not have enough absorption time.
This leaves no additional power from PV for daytime loads.

You could add a second PV panel in parallel with the first. If oriented the same, middle of the day the SCC would be clipping production.
If you oriented one panel ESE and one WSW, with a 60 degree acute angle between them, area presented to sun would never be more than what a single panel presents. Peak current generally no higher than single panel except under unusual cases of light reflected from clouds.
This wouldn't deliver a full 2x the Wh/day of a single panel due to reduced light intensity morning/afternoon and panel orientation vs. horizon, but it would produce considerably more than 1x what a single panel produces.
 
I doubt you'll get more than 20 amps off your panel. 310 watts divided by a charging voltage of 14.5 gives you 21.3 amps, so you're pretty much at your panels capacity already. At best you'll see 2 more amps. So may not be worth it.
I'm not sure that 14.5V is the best number for that calculation. What I envision happening is that if his battery is deeply depleted as is suggested above, then what you will likely see is bulk charging starting at about 12.5V, with the voltage peaking at around 14.5V as the battery reaches full charge. So, the most amps would be injected into the battery at the lower voltage, with amperage tapering off by the time the voltage goes up to 14.5. I would use (310W/12.5V) X .85fudgefactor = 21A. Right around max for your controller.

When this would really be an issue is when you have depleted the battery overnight, then you have a cool cloudy morning, and the clouds clear right around noon. Suddenly the panel is putting out max amperage and exceeds the controller's limit.
 
If you only draw your battery down to 50% DoD at night, it will take an entire day's PV production to recharge the watts consume.

Good catch.

In another post OP mentioned the used 280Ah AGM bank was capacity tested at 130Ah, so maybe he's using ~65Ah overnight. If that's the case the bank is charging even more inefficiently than normal for lead, turning a nontrivial amount of Abs/Float wattage into heat. It would be interesting to know what current acceptance looks like in Float and late Absorption.

In that same post he mentioned he is shopping for LFP, so PSOC issues go away and charging efficiencies will be greatly improved over injured AGM. If he's really only using 65Ah overnight I suspect the 310w panel will be enough on the LFP bank. I'd like to hear your thoughts on this.
 
I think to really "know" what is going on, you need to put some measurement devices on your system. They could be fancy, but even just a means to measure net current in/out of the batt and PV would tell quite a bit, and load measurements couldn't hurt too. Even a cheap setup would do, although you'd likely have to write down measurements on a regular basis to get an idea of what is going where, when. Once you know that, it should be fairly simple to identify the limiting factor(s).

Though from what has been said by you and the other posters, unless your batt is very inefficient in its charging (is it warm to the touch often or all the time), it sounds like you are limited by the size of your PV vs the loads you are running. If that's the case, a different charge controller or batt isn't going to help much. But I would reserve final judgement until you really know how much power you are producing, using and how well/poorly your batt is running.
 
Back
Top