How do i know when battery is @ 50% discharge

[dishxpert]

Hello Guys,

I am a newbie to this forum and to PV in general. Just had my system setup and trying to fully understand how to manage it.

I am located in the Caribbean where i get about 5 sun peak hrs/day (930am - 2:30pm).
Additionally, i asked a company to design a system that would give me about 150Kwh/mth (40% of total consumption).

This is what they sold me:

5x 230W Canadian Solar 24V panels (7.78A/29V ea) in parallel
8 x 6V/225Ah US-225 Ah connected as 24V system -- total 450 Ah
1 x TS-60 Tri Star Morning Star Charge Controller
1 x Samlex 3000W 24V inverter
1 x Gentran 30A Automatic transfer switch with 6 circuits pre-wired.

Here are my concerns/questions:
1) My charge controller never reads above 29A though 5 panels in parallel should be giving me 7.78A/panel x 5 panels = 38A. Why is my total current so low?
Panels are mounted facing south with an inclination of about 22o.

2) In a given day i collect 140 -180Ah depending on weather conditions. Isnt this low based on the specs of the panel?

3) My battery voltage went as high as 29.5V one day. Problem is, i am not sure when i am at 50% discharge hence when to switch over to o my utility company. Do i look for a particular voltage? For now, i am using 24.2V but not sure if this is too high or low.

4) Does my battery bank of 450Ah match the output from the panels? If not, what should i change?

5) For now, i am running my refrigerator and Office from the system, a total of 380VA/260W (reading from my kill-a-watt). For how many hours should i run these before 50% discharge?

6) I am thinking of upgrading my system before Christmas to produce 220Kwh/mth, up from 150Kwh. How many more panels and batteries would i need?

I will stop there for now but would appreciate detailed answers/comments to my questions. I am continuously reading the articles online to find some answers in the interim.

Thanks much.

Dishxpert

[Cariboocoot]

Welcome to the forum.

Might as well get this part out of the way first: you don't have enough panel. Now let's answer the questions and deal with the 'why'.

Here are my concerns/questions:
1) My charge controller never reads above 29A though 5 panels in parallel should be giving me 7.78A/panel x 5 panels = 38A. Why is my total current so low?
Panels are mounted facing south with an inclination of about 22o.

Panels do not put out their rated power level except on rare occasions. What we would typically expect from five 230 Watt panels is:
1150 Watt total array size * 77% efficiency (typical - would be worse if the weather is hot) = 885 Watts effective / 24 Volts (minimum Voltage level you should allow) = 36 Amps potential peak charge current. If your panels are hot or the batteries do not need more current the current will not be there.

2) In a given day i collect 140 -180Ah depending on weather conditions. Isnt this low based on the specs of the panel?

What you can expect from 1150 Watts of panel in terms of daily AC Watt hours:
1150 * 5 hours equivalent good sun * 0.50 (over-all system efficiency) = 2.875 Watt hours. Multiply by 30 and you get 86.25 kW hours per month; not what you were looking for.
With a battery-based system you will only see the Amp hours used/replaced, not the total capacity of the panels. If there's no place for the power to go (loads or charging batteries), they just don't 'harvest' it.

3) My battery voltage went as high as 29.5V one day. Problem is, i am not sure when i am at 50% discharge hence when to switch over to o my utility company. Do i look for a particular voltage? For now, i am using 24.2V but not sure if this is too high or low.

Batteries ought to hit around 28.8 Volts every day for at least as long as it takes to reach that level. This is the Absorb stage. You can get a good idea of what the batteries should be doing from the deep cycle battery FAQ's: http://www.windsun.com/Batteries/Battery_FAQ.htm
The best way to know the SOC of your batteries is with a battery monitor.

4) Does my battery bank of 450Ah match the output from the panels? If not, what should i change?

In my opinion, no. Two things to look at: peak potential charge current and daily Amp hours.
As mentioned before the best you can expect for current is 36 Amps. You would be better off with 10% of the total capacity: 45 Amps. You're at least one panel short of that figure. Your 29 Amps observed is about 6% of the battery capacity, and that's before any loads take their toll. That is barely above the 5% recommended minimum.
In terms of Amp hours, if your DOD is 25% that's 112 Amp hours. Over five hours of sun you would need to average 23 Amps to make this up. If the DOD is 50% you couldn't do it with a 29 Amp peak. That 112 Amp hours is rough 2.2 kW hours AC; far below what you hoped to achieve.

5) For now, i am running my refrigerator and Office from the system, a total of 380VA/260W (reading from my kill-a-watt). For how many hours should i run these before 50% discharge?

You would be doing good to get ten hours at that rate. But that is a rough calculation and quite a number of things will affect it. I based it on 50% DOD and 85% inverter efficiency.

6) I am thinking of upgrading my system before Christmas to produce 220Kwh/mth, up from 150Kwh. How many more panels and batteries would i need?

If you really want the 220 kW hours per month potential:
220 / 30 = 7.3 kW hours per day. That's about a 3kW array and around 1350 Amp hours of battery.
This is large for an off-grid system and it would be expensive. You should first spend the money doing whatever you can to reduce the consumption figure; it will pay back better.

Frankly I think whoever design your system neglected all the loss factors that eat up power. This would be panel temperature, wiring length and size, battery and inverter efficiency, et cetera.
For instance, with that size system there should be six panels wired three in a string and two strings in parallel feeding an MPPT charge controller. That would be more efficient right there.

[Vic]

One additional thought,

The Canadian Solar panels that I find on line have a 29.6 Vmp. IMHO, this is a bit too low for use "in parallel" on a 24 volt FLA battery.

Perhaps, the battery is always HOT (and the panels COOL), and perhaps there is another charge source that can perform an EQ ... but ... seems a bit too close to the edge in my book.

Havind the Panel and Battery P/Ns would help.

Perhaps I am being too literal, reading just what was posted by the OP. Whatta I know ?!? Vic

[dishxpert]

Thank you for the reply so far.

Am here trying to make enough sense out of them. What i hear is that my designed system (1150W) wont give me what i want due to loss of energy from various sources.

The current (of max 29A) is lower than anticipated value of 36A plus too close to the battery voltage of 24V.

So i need two strings of 4 panels in series, then both strings connected in parallel? This will give me what? About 81V and how many amps? 15?

How many 6V/225Ah batteries will i now need?

This is draining me!!

SO, can i stick with a 24.2V to achieve 50% DOD?

Thanks guys and please send me additional pointers.

Dishxpert

[petertearai]

SOC is best checked with a hydrometer and a battery monitor. Buy a battery monitor, you will never regret that purchase. Also more PV is required .
You have found a great site for help with solar power.

[BB.]

There are two major classes of battery charge controllers... Simple PWM (Pulse Width Modulation)--Where there is basically a "switch" (older ones could be relay, newer ones are transistor) that simply turns on and off... Solar panel with Vmp=35 volts and battery that needs ~29 volts to fully charge (and around 30 to 31 volts for equalization).

Switch closes, current flows from solar panel to battery bank (basically at the charging voltage of the battery). Switch opens, charging stops. Turn the switch on and off several times to hundreds of times a second, you get a reduced average charging current. Maximum current from a PWM controller is simply the maximum current from the solar panels/array.

MPPT charge controllers are much more expensive and are based on (typically) some sort of Buck Mode switching regulator.... For the sake of argument, the charge controller can behave like a DC version of a variable step down transformer. If the Vmp-Array Vmp~60 volts, the firmware in the charge controller Down Convert (at ~95% efficiency) from high voltage/low current (solar array) to low voltage/high current (charging battery bank).

The PWM controller, for best use of your solar panel power, needs a Vmp~17.5-18.x volts (12 volt battery) to be able to charge the battery in most conditions (remember, Vmp falls as the panel gets hot in sunlight--The panel may be upwards of 180F in full sun on a very hot/windless day). Vmp can fall below Vbatt-charging if Vmp is not around 17.5 volts or higher (12 volt battery).

If you put your panels in series (or series parallel), the charge controller during maximum charging current (battery less than ~90-80% state of charge) is a constant power device--or in math:

• Vmp-array * Imp-array = Vbatt-charging * Ibatt-charging (ignoring losses)

So, if you have Vmp array of 60 volts and 2 amps, you will have at the battery 15 volts and 8 amps charging current:

• 60 Volts * 2 amps (array) = 120 watts = 15 volts * 8 amps (battery)

So, if you have non standard solar panels (over ~200 watts), then it usually makes sense to get a MPPT charge controller to match the solar array Vmp/Imp to the battery bank Vcharge.

Also, a MPPT controller is nice because you can run Vmp~100 volts and charge a 12/24/48 volt battery bank. This allows you to use much smaller wire from the solar array to the charge controller. And/or you to put the solar array farther away from the battery shed (better sunlight, etc.).

All About Charge Controllers
Read this page about power tracking controllers

Anyway, you are sort of in the middle and trying to grab all of the "Facts" of designing a off-grid solar system... And it is confusing because, depending on specific choices/needs, the design can be done with different components/sizing of the parts/wires/batteries/array/etc...

I would suggest that there are two ways to design your system... The "best" is to measure your loads and define your needs (on grid, off grid, emergency backup, etc.). Then size the battery bank. And lastly the charge controller. If you have the need for a backup generator/AC mains as backup--then we finish off with that.

The second is to pick your core "hardware" you already have on hand (or is the only components you can obtain locally). That may be your battery bank, or your solar panels, or other expensive part(s) of the system that you do not want to part with.

With the core parts picked, we can then suggest the balance of system and estimate how much power you will get from it for your area. At that point (before you have spent any more money), you can decide if the system will meet your requirements or not. And proceed or not...

We can help either way--But it will be less confusing for you if we walk through one paper design first.

Also, battery wise, here are a couple good FAQs:

Deep Cycle Battery FAQ
www.batteryfaq.org

The battery is really the "heart" of your off grid system... If treated well (operated within its capabilities), the batteries will last you many years. If under or over charged, left setting for months without charging, run dead by somebody forgetting and leaving the lights one, etc.--The battery bank will only be so much scrap lead.

That is why we spend so much time discussing the batteries and how best to charge and discharge them. And the maintenance required.

-Bill

[stephendv]

The Canadian Solar panels that I find on line have a 29.6 Vmp. IMHO, this is a bit too low for use "in parallel" on a 24 volt FLA battery.

Nice catch Vic. So I'll state it a little less humbly Those panels are not suitable for charging a 24V system. Since you have a PWM controller, the panels should have a Vmp rating of about 36V. Now if you had an MPPT controller, then you could rewire those panels by putting 4 strings of 2 panels in series. You'd have to check the Voc rating of those panels, but at a glance I don't think you can put 4 in series as this would take you over the Voc limit of the MPPT controller (140V).

But first things first, whoever sold you that system should correct their mistake and either change the panels for some that will actually charge a 24V battery - or they need to swap out your PWM controller for an MPPT version (which is more expensive) and then you can rewire your panels.

[dishxpert]

So, to charge my 24V, 5 x 230W panel system i need a MPPT charge controller? What voltage rating?

Say i get a 6th panel, how should i wire them? 3 in series + 3 in series, then make parallel the two strings? How many Amp and Volt will i end up with and why is that a better system?

Pardon the ignorance.

[stephendv]

With a 24V flooded lead acid battery, you'd typically have to charge it at 28.8V - and occasionally, you'd need to equalise it at about 32V. Assuming you have the "CS6P-230" model of panel, they have a Vmp rating (their voltage at normal standard test conditions under load) of 29.9V, that's at 25 degrees C panel temperature. This voltage will usually be lower by the time it reaches the battery because of:
1. voltage drop in the cabling
2. voltage drop in the charge controller
3. voltage drop due to temperature (the panels also have a Normal Operating Cell Temperature rating of 27V for those panels)

So with all in parallel, you'll never be able to charge at 28.8V, meaning the batteries will never fully charge and will sulfate in short order.

If you want to keep those panels, then yes you'll need an MPPT charge controller. Most of those available in the 60A range have a voltage limit of 140V. If you get 1 more panel then you can put them in 3 parallel strings of 2 in series or 2 parallel strings of 3 in series. The "CS6P-230" panels have a Voc of 37.1V so a 2 x 3 configuration gives you a Voc of 111V and a 3 x 2 would give you 74V - both of which are below the 140V limit of the controller so both will work.

[niel]

3 in series x2 gives the benefit of overcoming v drop losses better (or smaller wire) and no need for fuses/circuit breakers being you only have 2 strings here.

2 in series x3 gives a bit more efficiency on the controller and any future expansions would be in multiples of 2 rather than 3 making it cheaper in the cost of the pvs.