Help with 48V AGM battery bank
anbus
Registered Users Posts: 7 ✭
Hello everyone!I need your advise!It's the first time i will install such a 'big' off-grid pv system.As you can see at the attached image, i will use 18 pv modules 230 each, 2 Mornigstar Mppt 60A charge controllers,16 x 12V/200Ah AGM batteries and one Victron 48/5000 Multi.My questions are:
a)Is my sketch correct?
b)Will the 80Amps that the chargers will deliver to the battery bank, be enough for such battery bank?
c)Is there any better way to wire the batteries together for a 48V system?
d)and where exactly should the cables coming from the chargers and from the inverter to the battery bank, be connected?
I think that i have a mistake with my batteries setup.Could anyone saw me the way?
Thanks a lot!!!
a)Is my sketch correct?
b)Will the 80Amps that the chargers will deliver to the battery bank, be enough for such battery bank?
c)Is there any better way to wire the batteries together for a 48V system?
d)and where exactly should the cables coming from the chargers and from the inverter to the battery bank, be connected?
I think that i have a mistake with my batteries setup.Could anyone saw me the way?
Thanks a lot!!!
Comments
-
Re: Help with 48V AGM battery bank
The first thing I notice is a potential 19 kW hours of power. That makes me think someone skipped the #1 step in off-grid design: conservation. That's almost as much power as my old all-electric house used. Why is this system being planned so big?
The second thing is all those batteries. Four parallel strings of four in series. If you really do need 800 Amp hours @ 48 Volts you should consider different batteries. Concorde makes some very big AGM's, including 2 Volt units like these: http://www.solar-electric.com/pvx-9150.html Twenty-four of those in series would provide about the same power and simplify wiring tremendously (and thus reduce possible problems with wiring). If you must use the sixteen batteries, you'll need to make the parallel connections using bus bars: all battery strings having equal-length positive and negative cables to the bus bars and the charge controller and inverter connections made there. In addition, each battery string should have its own fuse.
As for the array, if that's eighteen 230 Watt panels in total that's 4140 Watts. That isn't going to produce 80 Amps of charge current. More like: 4140 * 0.77 = 3187 / 57 Volts charging = 56 Amps. That would still be approximately 7% charge current, but that before loads. In my opinion this array would be marginal for that much battery bank. In addition it falls significantly short for replacing the total potential Amp hour use. You'd have to keep the DOD down to about 25%. If that's two separate arrays of 4 kW each it would work.
Were it me, I'd start by re-examining the issue of "why so big?" -
Re: Help with 48V AGM battery bank
i'll comment on the general questions as they appear. i am not discounting coot's comments excepting that you already own the batteries and i doubt you'll just dump them to go with 2v cells now.
a> i don't know if the sketch is correct as to how your system actually is in reality.
b> if it is 80a then yes as that would be a 10% rate of charge. we recommend 5%-13%.
c> the batteries are fine for series/parallel if you keep the balance by making the string wires identical to one another to prevent inequality. all wires interconnecting the strings should also be identical to keep equality. i am confused why you have a fuse on the negative wire from the batteries as that isn't usually done or needed.
d> the connections for the the ccs and load should be connected at opposing strings. that is, the + can be connected for say the inverter on battery string 1's + and the - for the inverter can be connected at the - on battery string 4. they all need not be wired to those 2 particular posts as you may wish to have battery string 4's + for the ccs + output and battery string 1's - for the ccs - output.
this may help you some,
http://www.smartgauge.co.uk/batt_con.html
i should add that using the proper wire gauge is important too as you did not specify what it is you have for the wires or their distances. this is determined by the voltage drop % and we like to keep that under 2% or 3% total. you can read more on that in the link in my signature line. -
Re: Help with 48V AGM battery bank
On a battery system that large--have you thought about a battery monitor (Victron makes a couple--looks like you can use on your system--may need different shunt if > 500 amp @ 48 volts).
If you do install a battery monitor, you would have to rewire the solar/inverter negative connections...
Basically, you need to bring all the battery negative leads to one common point, attach the shunt (high power precision resistor) to that point, then attach the negative bus leads (to rest of chargers/inverters/loads) so that all charging/discharging current flows through the one current shunt.
-BillNear San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset -
Re: Help with 48V AGM battery bank
Hmm. Expanding on Bill's suggestion... Perhaps a separate battery monitor on each of the 48 Volt strings? Has anyone tried that before? I'm just thinking it might give at-a-glance early detection of any imbalances that might arise. -
Re: Help with 48V AGM battery bankCariboocoot wrote: »Hmm. Expanding on Bill's suggestion... Perhaps a separate battery monitor on each of the 48 Volt strings? Has anyone tried that before? I'm just thinking it might give at-a-glance early detection of any imbalances that might arise.
interesting thought coot. i think separate monitors would be too costly. you may get away with having the shunts for each string and just switch the shunts to 1 monitor. -
Re: Help with 48V AGM battery bankinteresting thought coot. i think separate monitors would be too costly. you may get away with having the shunts for each string and just switch the shunts to 1 monitor.
I don't think that would work as they have to be connected to the shunt all the time to keep track of the current flow in and out of the battery, and thus arrive at a SOC figure at any time.
Yes, the Trimetric is $140 each not including shunt. So a bit pricey for four.
But hey, when I'm spending other people's money the budget is unlimited!
Maybe four of the Midnight Solar battery meters for about $60 each? No - no way to isolate one's reading from another. Rats. -
Re: Help with 48V AGM battery bank
you're right marc as my idea would only be good for present conditions for 1 at a time and could not total each area. -
Re: Help with 48V AGM battery bank
First of all thankk you for your advices!I design again the system, this time with 24 2V OPzV batteries.Almost 29KW of battery bank, which will give us about 3 days autonomy at 50% dod especially at winter.Plz check the attached image to guide me.
The house that the system will be installed, is an all season house, that's why i think that it will be need 4KW of solar modules....for the winter!
I still haven't cleared in my mind one thing...The +/- from charge controllers should be connected on the battery bank, with the +/- of the cables going to the inverter?Thus the energy coming from the sun is converted directly to ac power,right? No need to be stored in chemical form in the batteries and then going to the inverter from the batteries.
Waiting for your guidance -
Re: Help with 48V AGM battery bankI still haven't cleared in my mind one thing...The +/- from charge controllers should be connected on the battery bank, with the +/- of the cables going to the inverter?Thus the energy coming from the sun is converted directly to ac power,right? No need to be stored in chemical form in the batteries and then going to the inverter from the batteries.
Waiting for your guidance
Okay, let's not think like that. It's a "form vs. function" thing. In other words what the wiring looks like against how it performs.
You want to think like this: the panels, charge controller, and batteries are the charging circuit. If you have two parallel banks you connect the (+) of the controller to the (+) of one of those banks and the (-) of the charge controller to the (-) of the other bank. This means the charge current flows "in one end and out the other" hopefully fairly equally through all batteries.
The batteries to inverter is another circuit. Again the (+) and (-) leads for the inverter would go diagonally opposite each other on the battery bank. Yes, the (+) connections go to the same post as the charge controller (+) and likewise with the (-).
Just don't even think about panel production "going directly to AC" or you're going to get a wrong sort of picture in your head. No matter how it's wired the current flow is the same and depends on the state of charge not where the wires run. Technically, any time panel production can exceed the batteries' charge needs the power used for loads is coming from the panels.
If you're going with one set of really big batteries it all becomes easier.
I suggest you check out PV Watts http://www.nrel.gov/rredc/pvwatts/ for your location and get a fairly accurate prediction of how the seasonal changes will affect your power production. You might have overkill with that set-up (in which case you're spending too much money) or you might not have enough (in which case you may want more). Remember; back up generators are much less expensive per kilowatt hour than redundant PV and battery capacity. Greece is usually quite nice weather, isn't it? And you sure don't have the daylight hour difference we have in the Cariboo! -
Re: Help with 48V AGM battery bank
In your drawing, you say 2v @ 60 AH battery cells... I assume you intended to type 600 AH?
You should also have a fuse/breaker for the + cable to the Inverter.
All fuses/breakers should be installed as close as possible to the current source... In this case, close to the battery bank.
And high current 48 volt cables should be short and heavy between Charge Controller to Battery bank (somewhere around 0.40 to 0.20 maximum voltage drop (for optimum battery charging/voltage sense by charge controller). If you use the MorningStar TS line of charge controllers, they include Remote Battery Voltage Sense Connections. Connecting those up will make the Battery to Charge controller less sensitive to excessive voltage drop.
And you should review the voltage drop for the Battery to Inverter connection. Too long of battery to inverter leads (and/or too small of diameter) can cause the inverter to shutdown (under-voltage) with heavy loads/starting surge current.
Also, I would suggest a Battery Monitor so you can watch the state of charge of the battery bank (AGM batteries--you cannot measure specific gravity of the electrolyte with a hydrometer). People drawing too much energy from a battery bank and never getting the bank back to full charge is a common form of "battery abuse" and will shorten the life of a battery bank to a year or even months (in severe cases).
lastly, get the Remote Battery Temperature Sensor for both charge controllers. This too will help insure the batteries are quickly, and safely charged (cold batteries require higher charging voltage; hot batteries require a lower charging voltage).
Off-Grid inverters need the battery bank to store energy for use.
However, during the day, your Inverter power for the AC loads will come directly from the solar panels--so you will avoid battery losses during sunny days for those loads (i.e., AGM batteries are around 90-98% energy efficient).
Many people have setup manual or automatic monitoring of the battery bank... For example, when the charge controllers go to "float charging" (drop voltage a bit and no longer are pumping current to the batteries--i.e., batteries are full), they then turn on optional loads (water pumping to cistern, irrigation, etc.).
-BillNear San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset -
Re: Help with 48V AGM battery bankFirst of all thankk you for your advices!I design again the system, this time with 24 2V OPzV batteries.Almost 29KW of battery bank, which will give us about 3 days autonomy at 50% dod especially at winter.Plz check the attached image to guide me.
Hi Anbus, the 2V gel cells are a big improvement over your previous setup with the 12V AGMS. But they will be very expensive. It's quite commonplace to use 2V OPzS cells for such large installations, and you could even consider PzS traction cells (for forklifts) - both of these options would be considerably cheaper than OPzV. They would require some maintenance though, topping up with water every year or so for OPzS and more often for PzS.
There's a well known manufacturer of batteries in your neck of the woods: Sunlight.gr. (where I bought my 48V bank)
The lowest charging voltage you'll have will be 54V with an array of 4.14kW that's a max charging current of 76A. So instead of two 60A controllers you could use a single 80A outback FM80 controller - or two 45A MPPT controllers from morningstar.
In the last diagram there is no battery fuse on the bat+ connection, which you had in the first. -
Re: Help with 48V AGM battery bank
Guys, thanks a lot!!!
Now, as you can see at my updated sketch i added somethings.The batteries will be OPzV 600Ah from sunlight.I don't care about the cost.
At the sketch i marked with 'P' the Positive battery contact, where the '+' cables from the charge controllers and the '+' going to the inverter cable, will be connected.The same for the '-' with 'N' mark.
Is this kind of wiring work properly?I also added the BMS from victron and i can add 2 battery temp sensors.
Sorry for my not so good english....
Waiting for your advices again!!!:D -
Re: Help with 48V AGM battery bank
Looking good! I would move the 250A breaker so that it is the master breaker for the whole battery. I.e. install it as close as possible to the battery + terminal. Then all the other connections to the battery such as the inverter and the charge controllers will be connected on the other side of this fuse.Battery positive -------- 250A breaker ---------- inverter | |____63 A breaker _____ charge controller | |____63 A breaker _____ charge controller
-
Re: Help with 48V AGM battery bank
Dear stephendv, after your suggestion i update my sketch again.I hope that this will work fine.I don't want my customer to get angry:D
A question about Tristar....which is the minimum MPPT?My array voltage will be at about 70V at 70 degrees Celsius.And even lower on really hot days.Anyone knows what temperature can a module reach during a 40 Celsius ambient temperature? -
Re: Help with 48V AGM battery bank
With the gel cells you will never need to charge at more than 2.35V per cell, so 56.4V max. You would have more than enough room with 70Vmp at 70 degrees cell temp.
EDIT: Ambient temp to cell temp calculator: http://pvcdrom.pveducation.org/MODULE/NOCT.htm -
Re: Help with 48V AGM battery bank
I would suggest that you connect the fusing/breakers like this (normally, you would not want one breaker opening to take out everything--solar array+inverter+other loads):[FONT=Fixedsys]Batt+ ----350A fuse------- 250A brkr ------ inverter | |__63 A breaker __ charge controller | |__63 A breaker __ charge controller [/FONT]
I do not know European breakers and wiring--but in North America, wires and circuit breakers are rated to carry 80% of their "rated load".
If you have a 100 amp load (or charging current, you need to make the circuit breaker 1/0.80 or 1.25x the continuous current:- 100 amps * 1.25 Safety Factor = 125 amp fuse/breaker/wire minimum rating.
Also, make sure you use fuses/breakers rated for 60 VDC minimum and 10,000 AIC (amp interrupt current)--or whatever the standard for your region would be.
DC current is much different than AC current when used for switches/circuit breakers. DC Current tends to sustain arcing (sparks) across the contacts much better than AC current. So DC switches/breaker are usually much larger than the same AC rated device.
60 VDC, high interrupt current circuit breakers are not easy (or inexpensive) to find. Make sure they are rated (voltage/current) for your DC application.
Also, the Battery Monitor Shunt needs to be connected:[FONT=Fixedsys] |--------To Battery Monitor | |---To Battery Monitor | | Battery Minus====BM Shunt======Inverter Minus | |= charge controller Minus | |= charge controller Minus [/FONT]
You want all battery current flowing through the shunt so you can keep track of all the current flow (charging / discharging) to monitor the battery bank state of charge.
You might want to look for wiring diagrams from your vendors... I know that Outback power has a nice library of wiring diagrams. You should probably look through them. They have many samples and will show much more detail (for simple and complex installations).
The maximum temperature rise (panel mounted near roof with little air flow) is typically assumed to be 35C rise.
So, 40C air temperature + 35C rise = 75C panel/cell temperature.
Ideally, your Vmp-array should be around 70 VDC minimum. However, with MPPT charge controllers that have ~150 VDC maximum input voltage (like the TriStar MPPT), you should be aiming for a Vmp-array of around 100 VDC maximum.
If you installation is in an area that does not get very cold (like an island or similar), you might be able to go above 100 VDC Vmp-array.
There are two conditions you are looking to meet. 1) is Vmp-hot (minimum panel voltage when array is hot to charge the battery bank; and 2) Voc-cold (maximum solar array voltage when array is very cold and no current flow).
What is the Brand/Model (Vmp/Voc) rating of your solar panels you are planning on using?
You might try sending a PM or Email to user "peterako"==He is from Greece and has a solar and wind power system (he has not been here since April of 2011, so he might not read this thread).
Your English is pretty good and very readable/understandable. Δεν θα θέλατε να διαβάσουν ελληνικά μου. (Google Translate).
Stephendv is out of Spain, and Russ is out of Turkey--And we have some other members from Europe that can probably give you some suggestions on sources for the DC parts if you have trouble locating supplies locally.
-BillNear San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset -
Re: Help with 48V AGM battery bank
The modules i am going to use have a Vmp 29,75V and Voc 37V.
Another question now...
Since the Victron inverter can supply continuous 4500watt at 48V, we are talking for 98Amps sucking from the battery bank.Assuming that sometimes we are going to need the peak of the inverter which is 10.000watts/48V=208Amps.....how many Amps should my fuse be? 250A is ok?Of course i should choose a breaker that can sustain these 208 Amps right?
You have helped me very much till now:D -
Re: Help with 48V AGM battery bankThe modules i am going to use have a Vmp 29,75V and Voc 37V.Since the Victron inverter can supply continuous 4500watt at 48V, we are talking for 98Amps sucking from the battery bank.Assuming that sometimes we are going to need the peak of the inverter which is 10.000watts/48V=208Amps.....how many Amps should my fuse be? 250A is ok?Of course i should choose a breaker that can sustain these 208 Amps right?
In my opinion a 250A breaker is fine for that application. Bear in mind that the victron can only sustain 10kW for a few seconds, so even if 208A * 1.25 = 260A, 10A over the theoretical limit for a few seconds really doesn't matter. -
Re: Help with 48V AGM battery bank
Victron should tell you the ratings for the fuse/breakers. But, the math I start with:- 4,500 watts * 1/42 volt battery cutoff voltage * 1/0.85 inverter eff = 126 Amps
- 126 Amps * 1.25 NEC safety factor = 158 amps minimum circuit
Note--Remember that Inverters are "constant power" devcies... As the battery voltage falls, they take more current to maintain their output load requirement--ore or less, the equation is:- Power = Volts * Current
Typically, most inverters will have a a cutoff voltage or ~42 VDC--so that sets the low voltage limit. However--at 42 volts, the battery bank is dead and possibly being damaged by over discharging (the weakest cell in the 48 volt string actually hits zero VDC and starts "reverse charging"--pretty much will kill that cell.
And there is inverter efficiency. I used 85%--Your inverter may be better (or worse) at heavy loads.
Surge current--that is always a big issue... Normally, in the US, we assume ~5x surge current for a very short period of time (2 seconds or much less).
If your surge currents exist for longer periods of time (over a few seconds), then you would need to up rate your fuses/breakers/wiring. Some inverters are rated for 2x Rated Power for 10 minutes--There you would need to plan on a 2x larger electrical circuit/fuses.
Again, get the information from Victron--they should have an installation manual that lists the minimum/maximum wiring requirements (as part of their warranty and possible regulatory filings/requirements). I am just guessing here.
A few people here have recommended not taking a battery bank below ~46 VDC to avoid "killing" the bank. However, surge current, voltage drop on wiring, etc. may make you "adjust" that 46 volt requirement for your system.
And that is where a properly setup Battery Monitor will help. It does not estimate charge by battery voltage--It measures current into and out of the battery bank... Normally, the recommendations are:- Avoid going below 50% state of charge very often--Batteries will last longer
- Avoid ever going below 20% state of charge--Cells will be dead/possibly "reversed charged" and immediately damaged
- Avoid spending much more than 1 day below ~75% state of charge--The longer a Lead Acid battery spends below ~75% state of charge, the more it will "sulfate" (fluffy lead sulfate crystallizes--and is no longer part of the battery chemistry).
- Estimate battery bank/cell state of charge with no loads/charging current after a several hour "rest". Estimating charge by voltage while under load/charge is not accurate.
- Battery Voltage (and electrolyte specific gravity) is temperature dependent. If the battery bank is not ~25C (77F)--Your voltage values will need to be "adjusted" because of temperature (near 0C banks will have higher voltage; near 35C banks will have lower voltage)
- Batteries do not like to be hot, and do not have as much capacity cold. Watch the battery bank temperature and try not to let it get too warm. Batteries will age if hot (rough engineering rule of thumb--for every 10C rise, battery life will be 1/2... 20C rise 1/2*1/2=1/4 battery life; also cold batteries will last longer by the same formula... -10C, battery will last ~2x as long).
Hope this helps.
-BillNear San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
Categories
- All Categories
- 222 Forum & Website
- 130 Solar Forum News and Announcements
- 1.3K Solar News, Reviews, & Product Announcements
- 189 Solar Information links & sources, event announcements
- 886 Solar Product Reviews & Opinions
- 254 Solar Skeptics, Hype, & Scams Corner
- 22.3K Solar Electric Power, Wind Power & Balance of System
- 3.5K General Solar Power Topics
- 6.7K Solar Beginners Corner
- 1K PV Installers Forum - NEC, Wiring, Installation
- 2K Advanced Solar Electric Technical Forum
- 5.5K Off Grid Solar & Battery Systems
- 424 Caravan, Recreational Vehicle, and Marine Power Systems
- 1.1K Grid Tie and Grid Interactive Systems
- 651 Solar Water Pumping
- 815 Wind Power Generation
- 621 Energy Use & Conservation
- 608 Discussion Forums/Café
- 302 In the Weeds--Member's Choice
- 74 Construction
- 124 New Battery Technologies
- 108 Old Battery Tech Discussions
- 3.8K Solar News - Automatic Feed
- 3.8K Solar Energy News RSS Feed