diy solar

diy solar

How to calculate oversizing array for clouds and to minimize battery needs?

drink5

New Member
Joined
Jul 17, 2021
Messages
1
The process is:
  1. determine the number of what-hours the equipment consumes in a day (say 800 wh/d).
  2. Divide that number by .8 for losses (800 / .8 = 1000 wh/d)
  3. Divide that value by .5 to get the base battery size if lead acid or .9 if LiFePO4 batteries (100 / .5 = 2000 wh/d). Divide by system voltage to get battery amp-hours (e.g., a 12V system would be 2000 / 12 = 166 Ah battery).
  4. Multiply the watts in step 3 by the number of days of backup (e.g., 3 days is 3x2000=6000 wh, or 500 Ah).
  5. Get the value from an insolation map for your location that represents hours-of-sun-at-100%-output (say 4).
  6. Divide step 3 by the value from step 5 and again by .8 for losses (2000 / 4 / .8 = 625 watts) to get the minimum number of watts the panels must supply to meet your needs (minimum as clouds/shade will reduce output). If you expect 50% clouds, double the panel wattage. Keep in mind that in winter the cold weather will negatively impact the battery and solar output, see the battery FAQ for more information. If you get a larger inverter than you'll think you need (or use microinverters) you'll be able to add more panels later if needed.
So, I'm a newbie, trying to learn fast because I'm about to buy an off-grid cabin in Maine. I've worked through this and used the number I think fit my situation, but I have a question...

I understand that the idea is to get the minimum number of watts the panels must supply to meet someone's needs, but that would be charging batteries up during the day, and draining the batteries at night (and day depending on usage). So it's essentially a yo-yo effect, which uses batteries and therefore cuts down on the life of the battery. I know, I know, that's what they are there for. But my question is this...

Would it make more sense to buy more panels so that it charges the batteries faster, and all my daily usage is taken care of from the over abundance of power coming in from the array, while the batteries sit there topped off? Also, when it is a little cloudy and you're only getting 25%-50% from the panels, you have so many that your batteries are still topped off and extra power for daily usage is there? I get that at night, depending on usage, you will be using the batteries...just trying to think of a way to use them less on a daily basis, and have them for longer periods of down-time when it's really needed. Trying to think of extending battery life, and making sure that I have a good sized array so that even on cloudy days, I'm pulling in enough power.

Even if there were 3 days of super cloudy days, if I had a big array of panels pulling in that low percentage rate, added up it would still be a fairly good amount that might even cover all my daily needs, therefore not really needing such a large battery bank...which is the most expensive part of all of this. Does that make sense?
 
Welcome to the forums!

As you've already read through https://diysolarforum.com/threads/h...and-inverter-is-needed-to-charge-a-battery.73 and your question is something else that's very interesting, I'll move your post into a new thread so more people will see it and hopefully chip in their ideas.
...off-grid...Would it make more sense to buy more panels so that it charges the batteries faster ... Trying to think of extending battery life
There are a lot of things off-griders do to reduce the battery size by utilizing daytime power. For example, only run appliances (e.g., not-water, washing machine) during bright sunny times of day. Planning your day around power usage takes discipline and a good weather report, but it can save you a lot of $$.

There are also a couple of threads talking about automation... that is make decisions for how to consume power based on current SoC and incoming solar energy.

..Even if there were 3 days of super cloudy days, if I had a big array of panels pulling in that low percentage rate, added up it would still be a fairly good amount that might even cover all my daily needs, therefore not really needing such a large battery bank...which is the most expensive part of all of this. Does that make sense?
It does make sense.

It's true that on overcast days you'll get some power, but on super cloudy days it's pretty small. It's not easy to calculate and I haven't seen the information on datasheets. I posted my cloudy-light data here just so folks would have something to play with (post above that might be useful too). Basically, for me, efficiency started falling at 200 W/m² and tanked at about 60 but I've got some great quality monocrystalline panels.

What I'd say is use a nearby weather-underground weather station that reports solar energy. Look outside to get a feel for the incoming power based on conditions - that way you'll have an idea about your "super cloudy day". From that, even if you just used my data, you could calculate a max-size array. Then I'd say plan for max size plus some percent, think big. But, first pass, only install what you think you need. Then measure the system performance based on actual incoming performance and your life-style/needs. Once you have real data you could expand the system towards your "max" plan as needed. Because your plan incorporates the maximum, you can grow to it as you need but yet not spend any extra $$ unless you need to.

Hope that helps!
 
Last edited:
I think you're on the right track - For sure you need a battery bank large enough to cover your overnight needs and if you don't have enough solar to meet your daytime needs on cloudy days *plus* recharge what you used overnight, then you either need a bigger battery bank to last two or three consecutive nights, or you need more solar. You either buy more batteries, which are expensive, or you add more solar than you think you need so you can get by with fewer batteries

You could start out with twice the solar you've calculated you think you really need and then see how well your batteries do with extended periods of poor sun. You can always add more battery power later, or maybe even more solar if you have room on the roof, but starting out with less battery power and more solar would seem to be the best bang for your buck. A few months of actual usage will tell you which way to go from there

Don
 
Even if there were 3 days of super cloudy days, if I had a big array of panels pulling in that low percentage rate, added up it would still be a fairly good amount that might even cover all my daily needs, therefore not really needing such a large battery bank...which is the most expensive part of all of this. Does that make sense?
Its all a numbers game and at what point you want to design the system around.
You can use various websites to list the amount of sun your going to get in Jan vs July. For me its about half the sun in Jan as July. Which would mean twice the size of the array to power things if I sized it "right" for peak solar.

(to throw some arbitrary numbers in for the sake of the discussion) Sure you could spend $1000 more to double the size of the array, and spend a $1000 less on half the battery. but at the end of the day it still cost you the same amount of money,
Would it make more sense to buy more panels so that it charges the batteries faster, and all my daily usage is taken care of from the over abundance of power coming in from the array, while the batteries sit there topped off?

As to the battery life, its universally accepted that the slower the charge/discharge rates the longer the batteries last. (with old fasion lead being a thing that wants to be charged at a given rate and rates below that cause issues)
So no I see no benefit to having them charged fully in the first 2 hrs of my 5 peak sun hrs and having them sit "topped off" for the other 3 hrs of the day.
 
So, for overpaneling for me, I need to watch out that my max charge rate of 13% of C20 is not exceeded for my current 12 V flooded lead acid batteries. So, I can overpanel as much as I want, but I can not push more than 60 amps into my batteries. As was mentioned, I installed as much as I thought I would need, and found out I maxed out at about 45 amps. I then installed more panels, but found out with these additional panels I max out at 73 amps, which exceeds the charge rate of my batteries. When I've used my RV, I have a portable panels that I would disconnect if the charging amps ever got above 60 amps for me. As it actually worked out, I was done charging from my overnight usage before there was enough sun to exceed the 60 amps.

Besides physically disconnecting my portable panel bank, each of the three Victron charge controllers I have can have a max amps out set. Unfortunately, I can't set a total of 60 amps for all three, Victron does not have an option for that. I could set any amp cutoff limit to the Victrons I chose, but not a total for all three combined.

I am upgrading my system to two 24 volt batteries at 280 ah with lithium cells, and the battery charging amps will climb to 140 max per battery, or 280 total, but the BMS will limit charging to 100 amps per battery, or 200 amps total. So with these lithiums, even with me adding more solar panels, I no longer have to worry about pushing too much in. I may add 70% more panels, but I expect my max amps charging at 24 votls to be around 50 amps once I add those panels, and about 38 amps with the panels I have.

With my current 12 volt set up, on a sunny day I push max of about 6 charging amps to the battery, but on the overcast days I've seen, this drops to about 1 amp. So for me, about 1/6th the power on a cloudy day.
 
Would it make more sense to buy more panels so that it charges the batteries faster, and all my daily usage is taken care of from the over abundance of power coming in from the array, while the batteries sit there topped off? Also, when it is a little cloudy and you're only getting 25%-50% from the panels, you have so many that your batteries are still topped off and extra power for daily usage is there?
I believe so. I over paneled my van for the same reason. Does anyone know, are partially charged lead acid batteries for an extended time damaged? Are idling solar panels because the batteries are full, harmful?
 
My home's array on a good day will produce over 70kWh. A few weeks ago we had a pretty poor solar day. 3.5kWh, 1/20th of the output of the good day.

Oversizing an array makes sense but there comes a point where over sizing is over spending just to cover a few occasions when there's a run of poor output, when just having a backup source of energy would be better, e.g. a generator.
 
. Does anyone know, are partially charged lead acid batteries for an extended time damaged? Are idling solar panels because the batteries are full, harmful?
My Trojan Battery guide says to store them fully charger. They should always be charged when they drop below 75%If it's hot, check specific gravity every two to four weeks. I have FLA stored in hot weather, and check that the battery monitor is still 100%. For extremely cold, battery charge also needs to be monitored to make sure they don't get discharged and freeze.

Idling solar panels are not harmful, but if I am storing for more than a week, I disconnect everything both loads and SCCs. I do leave the battery monitor connected. I can't remember why I decided to store these disconnected and not with a float voltage.
 
Does anyone know, are partially charged lead acid batteries for an extended time damaged?

Yes.
FLA especially needs to be recharged full soon.
AGM may be more tolerant. At any rate, it can left unattended longer after a full charge (like 6+ months) with its lower discharge rate.

Are idling solar panels because the batteries are full, harmful?

No.

It's true that on overcast days you'll get some power, but on super cloudy days it's pretty small. It's not easy to calculate and I haven't seen the information on datasheets. I posted my cloudy-light data here just so folks would have something to play with (post above that might be useful too). Basically, for me, efficiency started falling at 200 W/m² and tanked at about 60 but I've got some great quality monocrystalline panels.

I have about 12kW of panels. On a lightly overcast day it might put out 2000W peak. Heavily overcast, 300W.

So, for overpaneling for me, I need to watch out that my max charge rate of 13% of C20 is not exceeded for my current 12 V flooded lead acid batteries. So, I can overpanel as much as I want, but I can not push more than 60 amps into my batteries. As was mentioned, I installed as much as I thought I would need, and found out I maxed out at about 45 amps. I then installed more panels, but found out with these additional panels I max out at 73 amps, which exceeds the charge rate of my batteries. When I've used my RV, I have a portable panels that I would disconnect if the charging amps ever got above 60 amps for me. As it actually worked out, I was done charging from my overnight usage before there was enough sun to exceed the 60 amps.

Here's one situation where automatic control is especially convenient.
SCC, inverter, and battery which don't talk together means excess PV charges battery too fast. If sized smaller, inverter may take all production and discharge battery for more.

With Victron equipment, a battery shunt can be monitored with control messages to SCC so it adjusts output to maintain charge current.

With SMA or other AC coupled systems, battery inverter charges at programmed rate and uses frequency shift (or other method) to tell PV inverters to adjust output.

Hybrids, I think most of them regulate battery current because they have knowledge of where PV current is going.

With this control, way over-paneled is great. Battery stays full all day while appliances are run. Orient panels two directions for flatter production during the day.

PV panels and AC coupled inverters or SCC can cost about $0.025/kWh amortized over 10 years. With rack mounts and electrical, maybe $0.05/kWh.

Battery inverters have a wide price range, and utilization is a big factor in $$/kWh.

Batteries can dominate price of system unless undersized:

FLA $0.25/kWh
AGM $0.50/kWh
Name-brand lithium $0.50/kWh (price is 5x AGM and cycle life too, we hope)
DIY LiFePO4 $0.05/kWh (if it lasts.)
Off-brand lithium can be between those two.

What I have is AGM, usable capacity around 1.5 hours PV production or one night's usage. This is for grid-backup.
If you're off-grid, DIY LiFePO4 sized for one night's usage would be most cost effective. Whether or not convenient depends on how frequently you get storms. An ICE generator will help you get by with small battery.
 
Oversizing an array makes sense but there comes a point where over sizing is over spending just to cover a few occasions when there's a run of poor output, when just having a backup source of energy would be better, e.g. a generator.
And the same can be said for oversizing your battery bank. A $400 2Kw inverter/generator makes much more economic sense than buying way too many panels or batteries. Sam's Club has a nice one on sale right now


Don
 
I have 4500 watts of solar and get 3700 wats of charging , in July I charge up buy noon even if it’s cloudy .
In December on a sunny day I can see 1800 watts but the sunny days are few and far between .
On a dark cloudy day I get 300/400 watts if the sun pops up I get more power
If I doubled my panels I could get 600/800 watts on a dark day and this would get my power up to full .
There are just days I get no power.
No power is no power, it dosent matter how much solar you have .
It’s better to charge longer and slower this time of year , so I like to be done with absorb by 500
I use 3 900watt strings this time of year .
 
Would it make more sense to buy more panels so that it charges the batteries faster,
Yes:
If you’re using lead acid have enough battery that you can go three days above 12.04 volts, and without exceeding your charge controller- having extra watts so you’re recharged early in the day will help those batteries last a very long time
 
Back
Top