Minimum Battery Bank for XW6048

Altnrgy
Altnrgy Registered Users Posts: 6
Hello All,
I’m not sure if this is the best place to post this question so please accept my apology if it’s not.

I’m designing a grid tied dual XW6048 system. The manual says a minimum of a 100 AH battery bank for one inverter which I know is ridiculously low. Since the system is grid tied I want to use the smallest feasible battery bank for several reasons. The obvious one is initial cost. The second reason is efficiency as in daily delivered AC KWH’s. As the battery gets larger the higher the charge/discharge and self-discharge losses get, especially as the bank ages. We only need a max of one day of autonomy since we are grid tied and have a generator.

I’ve never had good luck with sealed batteries. We can locally source Trojan L-16H-AC’s with a 435AH capacity @ C20 rating. We are planning to use just one string unless you folks can help me understand that is going to be problem. I don’t like multiple strings of batteries and a single string of the something like Surrette or HUP is totally overkill for a grid tied system plus cost 4x’s more.

The main reason we need the dual XW6048 is because of the 2HP well pump that is grossly oversized but suggestions to replace it with a soft start models are falling on deaf ears. I’ll look over the load calcs again but it might make more sense to go with a dual XW4548 instead. I see that CEC rates it with slightly higher efficiency as well.

I’ve zeroed in on an array of 7.5 to 8.5 kW. That’s all the roof we have so it’s unlikely the array will ever get larger. I like what I see on these MPPT600-80’s charge controllers but have some questions about them which I’ll ask in a separate post.

Any suggestions or requests for clarity are more than welcome.
Go easy on me. It’s my first post.
Thanks,T

Comments

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Minimum Battery Bank for XW6048

    Welcome to the forum.

    You're right: 100 Amp hours on a 6kW inverter is absurd. What they meant to say is 100 Amp hours per 1kW. I believe Solar Guppy came up with that as being necessary to get the grid tie to function properly. You can't be him for accurate info, and he' sure to correct me if I've gotten confused about it. Again. :roll:

    So you need at least 600 Amp hours. If you know you have a need for 'X' Watt hours for one day autonomous, the battery bank has to be able to meet that requirement as well.

    There's one thing you won't like about the XW 600-80 MPPT: it's really expensive.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Minimum Battery Bank for XW6048

    Note that is 100 AH @ 48 volt battery bank per 1kW of sustained load (or solar panel, which ever is larger).

    You might look into Variable Frequency Drive (VFD) for your 2 hp pump. Basically a variable frequency inverter that can soft start and adjust motor RPM's by varying the frequency (assuming pump is seriously over sized, you can crank way back on the power requirements). Will work with 3 phase pump motors and single phase 3 wire motors (those with an external capacitor connection--as I understand).

    There are a lot of VFD manufacturers out there (google), but I don't any experience with them--Perhaps some people here can recommend brands/vendors that they have had good luck with.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Minimum Battery Bank for XW6048

    i'm not a xantrex expert, but regarding the use of the l16s, if the 1 day of autonomy would be no more than 1/2 of the battery capacity it may be ok from a battery capacity use point of view (48v x 435a/2=5220wh), but may be too low for this inverter and you propose running 2 inverters so i doubt it will work so i'll let those that know better in utilizing these inverters comment on this.

    if the pumps are the main source of your problem then try to address this before you purchase anything for a solar system. alas, i am not an expert here on pumps either.

    that would be a very large array with a potential in the many 10s of kwhs per day and i understand this is to offset your loads, but for 1 day of autonomy based at this rate of power being fed daily will need a much larger battery bank to sustain that kind of indicated draw. to keep a backup of, for example, 1000w you need a minimum of 2000w in batteries to keep some lifespan to the batteries.

    is there a reason you a going to this very high voltage for the array that would necessitate this specialized high priced mppt controller? be advised with an array as large as you propose you would need multiple controllers due to the high current that array would translate into at 48v output (remember the inverter is 48v and so must the batteries be as well as the cc output to charge them). controllers do have current limits. 8500w/48v=178a and this is far above any single mppt controller. it may be more cost effective to go more strings of pvs at lower voltage to keep controller costs down.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Minimum Battery Bank for XW6048

    Or minimize the battery-based GT to just enough for emergency back-up purposes then feed the remainder of the PV array to a standard GT inverter to get the benefit from the maximum number of panels possible.
  • Altnrgy
    Altnrgy Registered Users Posts: 6
    Re: Minimum Battery Bank for XW6048
    BB. wrote: »
    Note that is 100 AH @ 48 volt battery bank per 1kW of sustained load (or solar panel, which ever is larger).

    You might look into Variable Frequency Drive (VFD) for your 2 hp pump. Basically a variable frequency inverter that can soft start and adjust motor RPM's by varying the frequency (assuming pump is seriously over sized, you can crank way back on the power requirements). Will work with 3 phase pump motors and single phase 3 wire motors (those with an external capacitor connection--as I understand).

    There are a lot of VFD manufacturers out there (google), but I don't any experience with them--Perhaps some people here can recommend brands/vendors that they have had good luck with.

    -Bill

    Thanks for quickness and variety of replies. We're stuck with the pump. It's new and their well/pump guy is adamantly opposed to anything other than what he's already put in there. I'm coming a long a little late in the process. I've politely tried to point out there is a variety of pumping options but I'm the only one who seems to care how much it's going to add the cost of the system.

    Thanks, T
  • Altnrgy
    Altnrgy Registered Users Posts: 6
    Re: Minimum Battery Bank for XW6048
    niel wrote: »

    is there a reason you a going to this very high voltage for the array that would necessitate this specialized high priced mppt controller? be advised with an array as large as you propose you would need multiple controllers due to the high current that array would translate into at 48v output (remember the inverter is 48v and so must the batteries be as well as the cc output to charge them). controllers do have current limits. 8500w/48v=178a and this is far above any single mppt controller. it may be more cost effective to go more strings of pvs at lower voltage to keep controller costs down.

    Good question. Wire size and cost is the quick answer I came up with. The array is about 150' away from the battery bank. Of course, I was planing on more than one charge controller. Using the Xantrex online Sizing Tool I determined I need 3 of the of MPPT-60's. But to stay under a 2% voltage drop I need a pair of 1/0 copper wires for each charge controller. PV prices have really dropped these last few years so maybe 5% is more tolerable but even if I could tolerate that I'd still need #4 wire.

    Unfortunately they do not have a sizing tool for their MPPT-80 model but if I did the calculations right I can easily get that many watts on two 80's. Paying attention to the maximum and minimum voltage as in allowing for heat to lower the MPPT voltage and the cold to raise the open circuit voltage I can do 9 modules in series That's put's my MPPT voltage somewhere over 225 on the hottest days. #8 wire results in less than 1% voltage drop. So the answer is even though I'm paying more for the controllers I'm making it up in wire cost.

    Thanks, T
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Minimum Battery Bank for XW6048

    Another controller you might consider is the MidNite Classic 250: http://www.solar-electric.com/misocl250mpc.html
    With a 250 Volt input limit, it's something between the typical 150 V in MPPT and the very expensive 600 Volt Xantrex unit.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Minimum Battery Bank for XW6048

    I'd like to know why you need a 2hp pump ? I'm irrigating about 1 acre, and running a house, with 1/2 hp, and 160' lift.

    How much water do you need, and what is the lift you require ?

    The PV array to run a 2hp pump (4,000w consumed) would have to be really big !

    Does it have to run at night, do you have elevated storage, or pressure tanks ?

    Making several starts a day (every time you turn on a tap) is really hard on the pump and the starting surge uses a lot of power.
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Minimum Battery Bank for XW6048
    mike90045 wrote: »
    I'd like to know why you need a 2hp pump ? I'm irrigating about 1 acre, and running a house, with 1/2 hp, and 160' lift.

    How much water do you need, and what is the lift you require ?

    The PV array to run a 2hp pump (4,000w consumed) would have to be really big !

    Does it have to run at night, do you have elevated storage, or pressure tanks ?

    Making several starts a day (every time you turn on a tap) is really hard on the pump and the starting surge uses a lot of power.

    I don't think he has a say in that; somebody else put in the pump and all he gets to do is try and find a way to power it. You know what plumbers are like: always stomping on goombas and spoiling Bowser's plans. :p
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Minimum Battery Bank for XW6048

    The main factor for battery AH size is the PV power of 8000 watts.

    What many folks don't recognize is single phase 60 Hz AC is not constant power profile. The power follows a sine-squared profile. It looks like two camel humps each cycle or 120 Hz of ripple.

    For 8000 watts from inverter injecting into grid with a battery float voltage of 53.2 vdc this 120 Hz ripple current will vary from zero amps to 212 amps peak. (actually a bit higher to account for inverter efficiency loss)

    The PV charge controller(s) at 8000 watts will be delivering a constant 150 amps D.C. to the battery connection.

    The difference between the inverter AC ripple current and the constant DC supplied by charge controllers is flowing through batteries. This means there is 150 amp peak flowing into the battery and 62 amps peak flowing out of the battery at a 120 Hz rate. The average of this ripple current equals zero but the battery is stressed by this ripple current.

    If you have 3 milliohms net series equivalent resistance within the battery system there will be 0.25 vpp ripple voltage on the battery. The actual Rs of the battery depends on the AH size of the battery and the age of the battery.

    The rule of thumb is 100 AH (for 48v inverter) for each 1kW of PV power to batteries.

    You can also figure your battery AH capacity need based on your off grid kWH needs. Whichever, PV power or off grid kWH's, is larger is the battery capacity needed.
  • sub3marathonman
    sub3marathonman Solar Expert Posts: 300 ✭✭✭
    Re: Minimum Battery Bank for XW6048
    RCinFLA wrote: »

    What many folks don't recognize is single phase 60 Hz AC is not constant power profile. The power follows a sine-squared profile. It looks like two camel humps each cycle or 120 Hz of ripple.

    For 8000 watts from inverter injecting into grid with a battery float voltage of 53.2 vdc this 120 Hz ripple current will vary from zero amps to 212 amps peak. (actually a bit higher to account for inverter efficiency loss)

    The PV charge controller(s) at 8000 watts will be delivering a constant 150 amps D.C. to the battery connection.

    The difference between the inverter AC ripple current and the constant DC supplied by charge controllers is flowing through batteries. This means there is 150 amp peak flowing into the battery and 62 amps peak flowing out of the battery at a 120 Hz rate. The average of this ripple current equals zero but the battery is stressed by this ripple current.

    If you have 3 milliohms net series equivalent resistance within the battery system .

    OK, I'm hoping we can go over this again real slow. I kept hearing Solar Guppy and others talk about the "ripple current" but I always figured I didn't know enough to even ask a most basic question.

    After reading RCinFLA, I think I'm actually beginning to see things. BUT, what I can't see is that on the XW6048, even at high noon, putting out 3 kw from the 4.7kw array, the panel will show some tiny number like -2.1 amps. So RCinFLA is saying the "ripple" is being "smoothed" by the battery? And how "stressful" is it for the batteries? I've never seen a battery specification saying it can withstand something like 100 amps "ripple."

    And as always, thanks for the help.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Minimum Battery Bank for XW6048

    I've always wondered about adding a nice tantalum "ripple cap" at the inverter input. gets rid of the cable inductance, and removes the reliance of the batteries to supply 120HZ surges. Even a large one, should help the situation.

    It'd have to be fused though!
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Minimum Battery Bank for XW6048
    mike90045 wrote: »
    I've always wondered about adding a nice tantalum "ripple cap" at the inverter input. gets rid of the cable inductance, and removes the reliance of the batteries to supply 120HZ surges. Even a large one, should help the situation.

    It'd have to be fused though!

    Just a little 4 or 5 Farad cap. That is Farad...
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Minimum Battery Bank for XW6048
    OK, I'm hoping we can go over this again real slow. I kept hearing Solar Guppy and others talk about the "ripple current" but I always figured I didn't know enough to even ask a most basic question.

    After reading RCinFLA, I think I'm actually beginning to see things. BUT, what I can't see is that on the XW6048, even at high noon, putting out 3 kw from the 4.7kw array, the panel will show some tiny number like -2.1 amps. So RCinFLA is saying the "ripple" is being "smoothed" by the battery? And how "stressful" is it for the batteries? I've never seen a battery specification saying it can withstand something like 100 amps "ripple."

    And as always, thanks for the help.

    Was based on OP's original statement of "I’ve zeroed in on an array of 7.5 to 8.5 kW." with two XW6048's.

    So I used 8kW of PV maximum power and assumed the two inverters and charge controllers are tied to a common battery bank. Just think of the ripple current as if it is battery peak loading and charging. If AH is large enough it can take 100 amps of ripple current.

    The battery is acting as a very large filter capacitor.
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Minimum Battery Bank for XW6048

    As to soft starters, I have used soft starters on single phase 240 vac motors.

    Syncronous motors look pretty much as locked rotor amperage load until they reach greater then about 90% of their final run rate rpm's.

    My air conditioner compressor is about a 4 HP motor. The Lenox A/C scroll compressor design seems to have a greater startup current then piston/reed valve compressors. My run current on compressor is less then 14 amps at 240 vac. The startup current is a wopping 120 Arms, lasting for 400 msec.

    With the soft starter I was able to bring down the startup current to 40 amps but the length of the startup period just extended out correspondingly. To inverter it makes little difference as they can take a higher amperage for a short period.

    I ended up putting in two SW+5548's in series stacked arrangement which is able to start and run the A/C compressor. The inverter acts as a soft starter. Only associated issue is the voltage sag that happens at startup may effect other equipment on the inverter. I use a separate inverter for other loads, like refrig, so not to subject them to the AC voltage sag during A/C startup.

    When the A/C compressor and air handler blower is running it adds 75 amps load to my 48 volt battery array.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Minimum Battery Bank for XW6048
    RCinFLA wrote: »
    .....
    When the A/C compressor and air handler blower is running it adds 75 amps load to my 48 volt battery array.

    What is your battery array built from ?
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Minimum Battery Bank for XW6048
    mike90045 wrote: »
    What is your battery array built from ?

    1000 AH's , in five parallel strings.
  • Dr. Strangelove
    Dr. Strangelove Solar Expert Posts: 49
    Re: Minimum Battery Bank for XW6048
    RCinFLA wrote: »
    Just a little 4 or 5 Farad cap. That is Farad...

    Help me with the math please:

    62 A / 120 = 0.52 Coulomb per peak

    C = q / V = 0.52 C / 53.2 V = 10K uF.

    Yes, I understand that we're not emptying all the charge out of the capacitor, but where does the derating factor of 500 come in?

    q = C V ==> delta q = C delta V = 0.52

    So for C = 5 F ==> delta V = 0.1 V, is this what you are basing the capacitor size on?

    What about a high Q LC notch filter at 120 Hz, so the inductor can store energy in its H field and reduce the battery ripple and capacitor size?

    Or would something like active PFC to flatten the current profile work better? That would be something for the inverter manufacturer to implement. Does any manufacturer specify DC input ripple power?
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Minimum Battery Bank for XW6048

    There is also internal resistance and inductance in the cap, and it's wire.... But even a modest one, is better than none, to keep from ringng the plates at 120hz
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Minimum Battery Bank for XW6048

    You don't use 53.2 vdc (unless you want that much ripple).

    Yes I used 0.1 to 0.2 vdc for desired maximum ripple voltage. If you go more the batteries will take over current and cap will not help much.

    You have to deliver the peak power with some type of storage element. Inductor would not be a good choice.

    If inverter was three phase grid push it would be much easier on batteries as it would be a much smoother power profile.
  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: Minimum Battery Bank for XW6048
    Altnrgy wrote: »
    Good question. Wire size and cost is the quick answer I came up with. The array is about 150' away from the battery bank. Of course, I was planing on more than one charge controller. Using the Xantrex online Sizing Tool I determined I need 3 of the of MPPT-60's. But to stay under a 2% voltage drop I need a pair of 1/0 copper wires for each charge controller. PV prices have really dropped these last few years so maybe 5% is more tolerable but even if I could tolerate that I'd still need #4 wire.

    Unfortunately they do not have a sizing tool for their MPPT-80 model but if I did the calculations right I can easily get that many watts on two 80's. Paying attention to the maximum and minimum voltage as in allowing for heat to lower the MPPT voltage and the cold to raise the open circuit voltage I can do 9 modules in series That's put's my MPPT voltage somewhere over 225 on the hottest days. #8 wire results in less than 1% voltage drop. So the answer is even though I'm paying more for the controllers I'm making it up in wire cost.

    Thanks, T

    We're working on a new string sizing tool for the MPPT 80 600 SCC. In the meantime, you're on the right track about string size (modules per string), high voltage, low current, and wire size. You might also look into 60-cell modules in the 240 W STC range, as you may be able to do 10 modules in series. Two of these strings in parallel would put you right at 4,800 W STC for a system with a nominal 48 V battery bank.

    Note that charge controllers with just one- or two series- strings per charge controller typically do not require combiner boxes with OCPD's (circuit breakers or fuses) for each string. This is another place to realize some cost- and labor savings.

    See Code Corner #125

    Regards,
    Jim Goodnight
    Global Product Line Manager - XW
    Schneider Electric
  • kakiwaki
    kakiwaki Solar Expert Posts: 28 ✭✭
    Re: Minimum Battery Bank for XW6048

    GUYS GUYS GUYS ...................... THE easiest way to work out the battery required is just take the Inverter watts divided by the required battery voltage and times that by five EG:


    6000 / 48 * 5 = 625 amp/ hour minimum battery required.