Battery capacity sizing

tkc100
tkc100 Solar Expert Posts: 67 ✭✭✭✭
It's been awhile since I have been on the fourm.
Everything has been working and I have had other things taking up most of my time but now it's time to give some consideration to one of the solar installations I am responsible for.
I need to purchase a new set of batteries and I am struggling a bit with just how much capacity I need to buy.
I know the proper way to size up a system is to first start with the load requirements but suffice to say in this particular installation there will never be enoungh given the present array.
Help me if you can with the mental - math - common sense and experiance to come up with some kind of an idea as to how much battery I should purchase. Currently the 24 volts system is operating on 16 - 6 volt batteries. They are Amp hour rating 186 AH C-20 golf cart batteries. Worked pretty good when they were younger but now its time to replace them.
I have never done the brain damage to figure out what would be the best ballance between battery AH capacity and the equipemnt I currently have.
I have attached the specifications
Attachment not found.
Even though the current configuration is 16 x 6 I really don't need to stick with that.
Your help and shared experiance will be greatly appreciated.
Thanks!

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Battery capacity sizing

    What the array wattage?

    Any idea of watt hours or amphours per day?

    Is this charge during the day and use battery at night or is this using a fair amount of power during daylight too?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    Re: Battery capacity sizing

    I believe you have 4 parallel sets of batteries, if so you may be getting a poor balancing on each of those 4 strings. Something to investigate and possibly replace with 2 strings of higher Ah rated batteries.

    hth
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • tkc100
    tkc100 Solar Expert Posts: 67 ✭✭✭✭
    Re: Battery capacity sizing

    BB it is good to see you are still around!
    I know this is not the correct way to go about this but I am not certain what the actual watt hour or amp hour draw is no this particular installlation.
    Safe to say we can't make more than they will use. There is a auxillary generator at this site. It didn't get used too much when the original batteries were new but now they are using it quite a bit. I figure the old battery had about 750 Ah when they were new.
    All the specifications are in the attachment (link Specifications.pdf)Attachment not found.
    The pannel are rated (advertised as 200 watts each) There are four panels in a series parellel configuation so total wattage is 400 watts.
    I guess what I am looking for is if I were able to save all that can be produced how much battery would that take?
    This is South West Texas so there is plenty of sun but also a lot of heat in the summer months.
    I currently have the pannel facing a bit to the SSE to take advantage of the morning sun when it still cool.
    I understand that without knowing what is being used it is difficult to calculate what needs to be put back but is it possible to work from this premise.
    The batteries are taken to 50% each and ever night. Given the equipment listed in the specification how much battery would I need.
    I have got to get these batteries purchased by the end of the month. It's and end of the fisial year budget thing. Otherwise I would do this in a more logical fashion.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Battery capacity sizing

    If you have 800 watts of panels (4*200 watts), assuming you are somewhere around Austin using PV Watts, fixed array tilted to latitude above horizontal:
    Month    Solar Radiation (kWh/m 2/day)
    1      4.32     
    2      4.96     
    3      5.47     
    4      5.52     
    5      5.54     
    6      5.93     
    7      6.21     
    8      6.22     
    9      5.77     
    10      5.65     
    11      4.60     
    12      3.96     
    Year      5.35
    

    Lets say that they get OK performance for 1/2 the year at ~5.5 hours of sun--Then from the solar panels, I would guess they are using:
    • 800 watts * 0.52 system derating * 5.5 hours of sun * 1/0.85 inverter eff = 2,692 Watt*Hours per day average from battery bank

    Based on using the 24 volt battery bank to 50% state of charge per day:
    • 744 AH * 25 volts nominal * 0.50 discharge = 9,330 WH per day from battery bank

    So--something does not match...

    If they are using AC power mostly--Trying to get a Kill-a-Watt meter on there would be interesting.

    My guess is they are closer to using:
    • 800 watts * 0.52 system derating * 5.5 hours of sun * 1/0.85 inverter eff = 2,288 Watt*Hours per day of AC power

    The battery bank needed to support such a load--2 days no sun and 50% maximum discharge:
    • 2,288 WH * 1/0.85 inverter eff * 1/25 volts * 2 days no sun * 1/0.50 max discharge = 431 AH @ 24 volt battery bank suggested

    The battery bank would should be charged at 5% to 13% rate of charge:
    • 431 AH * 29 volts charging * 1/0.77 panel+controller derating * 0.05 rate of charge = 812 watt array minimum
    • 431 AH * 29 volts charging * 1/0.77 panel+controller derating * 0.10 rate of charge = 1,623 watt array nominal
    • 431 AH * 29 volts charging * 1/0.77 panel+controller derating * 0.13 rate of charge = 2,110 watt array "cost effective" maximum

    So--I would suggest a battery bank about 1/2 the present AH capacity and double the array to ~1,623 watts. And see how things work.

    If they need more battery bank, they could double to 860 AH and still meet the 5% minimum rate of charge. But they would have to be more careful about energy usage (don't want to keep cycling the battery bank around 50% state of charge and never getting it back to >90% SOC a couple times a week).

    They probably would be better off (at least for the short term) with a smaller battery bank and larger array (10% rate of charge). Even if they use the battery "a lot" (cycling from 100% to 50% SOC at times), at least they have a large array that will get it recharged fairly quickly over the next couple days. The bank may not last quite as long--but they only have 1/2 the batteries to replace at a time (saving money).

    Once they get a better understanding of their loads and how to operate their system (I would suggest a battery monitor), then a couple years down the road they can replace the "cheap set" with a correctly sized bank (and solar array--if appropriate).

    You can give them some hints with the battery monitor--Something like:
    • >75% State charge, everything OK
    • 50% to 75% SOC--cut optional loads, start the genest next morning (especially if using more power and/or poor weather forecast)
    • <50% SOC--Start genset right now, turn off when >~80-90% SOC. If battery SOC does not start rising, call you next day for help.
    • <20% SOC--Turn off all loads--Call you for help. Battery damage imminent.

    2,288 Watt*Hours per day is not a lot of power--I would be suggesting a system capable of ~3.3 kWH to 4 kWH per day as something that probably can run their home if they are fairly conservation minded (but that is a lot of money to spend "on a guess")--What kind of loads do they have (refrigerator, fans, computer, well pump, etc.)?

    Not a lot of answers--Hope it gives you some useful ideas (basically, take money that would have been spent on large battery bank and shift funds saved to add solar panels).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • tkc100
    tkc100 Solar Expert Posts: 67 ✭✭✭✭
    Re: Battery capacity sizing

    Bill

    Thanks so much for taking the time to walk me through the mental process. I know this has to have been a subject discussed many times before in different variations on this forum.
    I hesitated to start a new thread but yesterday I did not have the time to do the research. I work for Texas Parks and Wildlife and maintaining several solar installations is only one of many job responsibilities.

    We are a bit further south than Austin. Think the Rio Grande and almost Mexico. Big Bend Ranch State Park. N 29 degrees 28 minutes.
    The four solar panels in use are 200 watts each. From the get go this installation has been a problem. I sort of inherited it. The SX3200B solar panels were not really made for a 24 volt nominal system. They are much better suited for a grid tired system. So rather than replace the almost new panels I wired two of them in series and then the 2 sets of 2 in parallel. I then replaced the PWM controller with the Xantrex XW-MPPT60-150 MPPT controller. It is able to make use of the higher voltage. I guess with the MPPT controller it's a little more difficult to calculate the total achievable wattage but normally in a series parallel hook up these 4 panels would have 400 watts of power with a nominal voltage of 32 volts. Vmp of 48.8, Voc of 61.6

    I have ordered and plan to install the Victron BMV 600 Battery Monitor so as time goes along I will be better able to answer all those elusive questions about the power consumption but for now here is what I am up against.

    The current set of 16 - 6 volt batteries are history. The physical installation and the cabling is in good condition and is configured for the 16 batteries.
    It is the end of our fiscal year and I have got until the end of this month to purchase the best compromise I can.
    Perhaps next year we can do some new panels but for now it's the batteries.

    What do you think?
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Battery capacity sizing
    tkc100 wrote: »
    Bill

    Thanks so much for taking the time to walk me through the mental process. I know this has to have been a subject discussed many times before in different variations on this forum.
    I hesitated to start a new thread but yesterday I did not have the time to do the research. I work for Texas Parks and Wildlife and maintaining several solar installations is only one of many job responsibilities.

    We are a bit further south than Austin. Think the Rio Grande and almost Mexico. Big Bend Ranch State Park. N 29 degrees 28 minutes.
    The four solar panels in use are 200 watts each. From the get go this installation has been a problem. I sort of inherited it. The SX3200B solar panels were not really made for a 24 volt nominal system. They are much better suited for a grid tired system. So rather than replace the almost new panels I wired two of them in series and then the 2 sets of 2 in parallel. I then replaced the PWM controller with the Xantrex XW-MPPT60-150 MPPT controller. It is able to make use of the higher voltage. I guess with the MPPT controller it's a little more difficult to calculate the total achievable wattage but normally in a series parallel hook up these 4 panels would have 400 watts of power with a nominal voltage of 32 volts. Vmp of 48.8, Voc of 61.6

    I have ordered and plan to install the Victron BMV 600 Battery Monitor so as time goes along I will be better able to answer all those elusive questions about the power consumption but for now here is what I am up against.

    The current set of 16 - 6 volt batteries are history. The physical installation and the cabling is in good condition and is configured for the 16 batteries.
    It is the end of our fiscal year and I have got until the end of this month to purchase the best compromise I can.
    Perhaps next year we can do some new panels but for now it's the batteries.

    What do you think?

    Let's try this again:

    1. Yes, to get the maximum benefit from the panels when they are too high in voltage for a PWM controller you can put them in series parallel (or just series if you have a CC with very high voltage input. But that is also more typical of a grid-tie inverter than a battery CC.) In your case, with what you have series-parallel should be best, although just all 4 in parallel would also work as long as the amperage below the maximum for the CC.

    2. Regardless of how you interconnect them (within reason....) if you use an MPPT controller you will get close to the nominal 800 watts out. In the case of a series-parallel set of four panels, the Vmp will be twice that of a single panel and the Imp will also be double that of a single panel. So the power would be 4 times that of a single panel. You only get into a problem if the output current corresponding to that power level is more than the CC can handle.
    Even in that case, as long as it is not outside the CC manufacturers recommendations, the CC will just not take any more power from the panels (using a current output less than the MPP current.) In the case of a 24 volt nominal system, 800 watts into the batteries would be about 33 amps from the CC.

    3. To make up a 24 volt system, the ideal would be to use all of your new batteries in a single series string. The old installation with 16 6-volt batteries and a 24 volt output configuration would probably have had 4 series strings of 4 batteries each, with those series strings in parallel. Four strings in parallel is getting to be high to be able to balance the charge and discharge currents. You would be better off, as you buy new batteries, to look for 2 or 4 volt batteries which can be put into a series string of 12 or 6 batteries respectively to give you the total watt-hour capacity you need. The maintenance would be easier and would probably make up in the long term for the cost of adapting the mounting and wiring scheme.
    SMA SB 3000, old BP panels.