Help me build my off grid system... dont know which direction to go.

2»

Comments

  • 2000tranzam
    2000tranzam Registered Users Posts: 15 ✭✭
    Photowhit said:
    Ok so I actually ended up building the setup and it works great... however I did a load test and use a power meter to determine how much KWH usage I actually got. The battery was fully charged to about 53.5 volts give or take,
    It would be good to check your battery spec's, from the link you provided for the battery, which doesn't understand spec's IMHO, since this info is not under Spec's;

    53.5 volts represents about 13.4 volts in a comparable 12 volt system. So perhaps around 50-60% (dumb guess)

    Since Lithium batteries need very little voltage above the existing voltage to charge, I'd assume 14.6 x 4= 58.4 or there abouts to be fully charged. I would ask them for a discharge curve, it may be that it has a very steep knee (which I think Lithium do in general) and it went from 51 to 38 volts with only a minimal discharge. 

    What You Need To Know About LiFePO4 Batteries - Off Grid Ham

    I'd suggest losing the ring posts and buying some power distribution blocks, (you're throwing a ton on money at this, might as well build it right IMO) 

    Hope this helps. 






    From all the information I have gathered... I was under the impression full charge would be about 14.4-14.6 volts per battery, equaling around the 58.4 volts that you had suggested. I had let the battery bank charge for quite a while, and when taking a volt meter to the terminals at one point I saw around 59 volts... but after I let off charging it dropped all the way down to 53.x volts. Does that mean I have an issue with the batteries holding juice? I called the battery company and they advised not to go below 48 volts for the cut off... but that seems ridiculously conservative.   

    Regarding the discharge chart Ill see if I can ask if thats the case... I would guess im only getting 4.3 kwh due to being at 53 volts and not the 58 that I should be (as per 48 volt configuration).

    What is the difference between distribution blocks and ring posts? Dont they distribute power equally?  Thank you for the help... im wondering if there are any people locally in the phoenix area that I could meet with to help, or even possibly pay for consultation to correct the issues (im hoping the inverter/charger I bought is not the reason for my lack of being able to charge fully as it was meant to be used for LiPo batteries as well).
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    I don't use Lithium chemistry batteries, but understand that, some at least, are near 98% efficient. Should get some idea of that from the manufacturers as well, Needing to maintain 58+ volt and see a resting voltage of 53 volts would be around 90% that seems to be wrong to me. I've only read about them so far though...

    It may be that you hold 14.6 volts per battery or 58.4 volts for a 48 volt bank and that the lithium continues to climb as it reaches near parity. If you were to compare resting voltages... So that you would start near 50-60% at then ends near 100%. You might check the current flow with a clamp meter, or if your inverter charger provides the information.

    Power distribution blocks have a single bound connection for each wire. A ring post has a single binder for all the connections.
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    Having a Discharge curve for the battery you are using would be good. 

    You can see from the chart I pulled that 56 volts represents a mostly discharge battery for that chemistry LiFePo4 battery. Though your battery is described as a LiFePO4, I think the characteristics can vary by the anodes and trace catalysts used.
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    It certainly wouldn't be a bad idea to measure voltage for each individual battery when you believe they battery bank is fully charged. This would give you more information and you could see if one battery wasn't even with the rest...
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    So I went out and am reading the manual. Not sure how they are shipped or if or how you 'fully charged' the batteries individually before connecting them in series and parallel. I suspect this has to do with each battery having it's own Battery Management System BMS so they can all start at the same place.


    This pretty much confirms to me that they consider 53 volts to  be 50% charged; So yeah, I don't know the expected charging routine, but checking the voltage of each battery would be in order for me.




    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • Wheelman55
    Wheelman55 Registered Users Posts: 233 ✭✭✭
    Here is a chart from Discover AES LiFePO4 batt showing voltage at different SOC. You can use this for a rough reference. 

    Sorry if it’s hard to read, I’m not able to keep the table format. 

    Table 15. ReCharge Voltage Setting Guidance 
    Remaining 
    Capacity
    Nominal   System Voltage
                        24V    48V
    10-15%. * 24.5V*.  49V*
    15-20%        25V   50V
    20-30%.    25.8V   51.5V
    40-50%.    26V      52V
    80-90%   26.3V    52.5V
    90-100%   27V     54V
    * Not recommended. Inverter may display Low Batt Warning.
    Off-Grid in Terlingua, TX
    5,000 watt array - 14 CS 370 watt modules. HZLA horizontal tracker. Schneider: XW6048NA+, Mini PDP, MPPT 80-600, SCP. 390ah LiFeP04 battery bank - 3 Discover AES 42-48-6650 48 volt 130ah LiFePO4 batteries
  • 2000tranzam
    2000tranzam Registered Users Posts: 15 ✭✭
    Photowhit said:
    So I went out and am reading the manual. Not sure how they are shipped or if or how you 'fully charged' the batteries individually before connecting them in series and parallel. I suspect this has to do with each battery having it's own Battery Management System BMS so they can all start at the same place.


    This pretty much confirms to me that they consider 53 volts to  be 50% charged; So yeah, I don't know the expected charging routine, but checking the voltage of each battery would be in order for me.




    So I just went to put a volt meter to  each of the 8 batteries... 1 of them showed 13.2 volts while all others were at 13.3 volts. Im gonna try to run the charger again and see if I change the settings on the charger, maybe it will allow for the batteries to be charged to a higher voltage?  When I had it charged about a week ago, sitting voltage was about 53.4 or so. Now when I went to check it was about 53.2... how is it possible to have parasitic loss? I had both breakers disengaged so no power was even running to the inverter/charger... what else could drain the batteries if its not even connected?
  • Wheelman55
    Wheelman55 Registered Users Posts: 233 ✭✭✭
    batteries self discharge...maybe your readings are normal?
    Off-Grid in Terlingua, TX
    5,000 watt array - 14 CS 370 watt modules. HZLA horizontal tracker. Schneider: XW6048NA+, Mini PDP, MPPT 80-600, SCP. 390ah LiFeP04 battery bank - 3 Discover AES 42-48-6650 48 volt 130ah LiFePO4 batteries
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    This is a "down the rabbit hole" post... Feel free to skip if not interested "in the details".

    Just out of curiosity, is your meter reading 53.X or 53.XX VDC (one or two+ significant digits). If 0.X VDC reading, then +/- a couple of last digit variation is not out of the ordinary.

    https://www.amazon.com/gp/product/B019CY4FB4

    For example, a "decent" meter has an accuracy of:
    ±(1.0% + 3 digits)

    And a scale/resolution of (guessing that really 599.9 volts max):
    600.0 mVDC max range 0.1 mV 
    6.000 VDC max range 1 mV
    60.00 VDC max range 10 mV
    600.0 VDC max range 0.1 V
    6000 VDC max range 1 V

    So, while we get used to the "exact reading" of the meter--The accuracy you are asking for is a bit much for a less expensive meter.

    Here is a Fluke 76 meter... 0.3% + 1 digit accuracy:

    https://www.manualslib.com/manual/511897/Fluke-76.html?page=42#manual

    But it limits to 40.00 volts (I assume 39.99 volts max)... And if measuring 50.00 volts, you are in the 400.0 volt range, so only 0.1 digit resolution. 

    So, you ask yourself, is my meter accurate (today)? And look for a calibration source:

    https://www.amazon.com/s?k=voltage+calibrator

    Agh... 2-3x or more than the price of the meter itself (at least the "cheap meters" I buy). And do "we" take the meter in once a year for calibration?

    https://en.wikipedia.org/wiki/Segal%27s_law

    Segal's law is an adage that states:

    "A man with a watch knows what time it is. A man with two watches is never sure."[1]
     
    And just to be fully informed:

    Later this was - mistakenly - attributed to Lee Segall of KIXL, then to be misquoted again in Arthur Bloch's book as "Segal's Law". [2]

    And back to "what matters", using Wheelman's data, you are looking at ~6.0 volt range from 100% to 0% for a "48 volt" LiFePO4 battery. And 1% of 55 volts (nominal) = 55.00 +/- 0.50 volts +/- 3 digit accuracy from the Klein meter (first link).

    Rough accuracy of 55.00 volts +/- 0.53 volts (I think) accuracy for the Mid-priced meter (within 1 year of calibration)

    Or for the Fluke 76 meter, 0.3% * 55 volts = 55.0 +/-0.165 volts +/- 1 digit.

    Rough accuracy of 55.0 volts +/- 0.3 volts for the ~$300 Fluke meter model 76.

    The short answer is don't worry--Close enough for government work (and why we suggest something like 20-90% State of Charge range, so that an "inaccurate" meter or other issue (like a single cell or block of parallel cells) do not hit the "hard" voltage limits for over/under charge and cause permanent damage to the cell(s).

    Do watch the DMM's battery... We have had a couple posters & DMM (digital multi-meters) here that had batteries that were going dead (no on screen battery warnings), and the meters did some really bad things... For example, they measured a 12 volt battery at (something) like 10.0 volts, and 120 volts at 144 volts (made up readings). At 12 and 120 Volts, the meter was consistent (change ranges, meter on/off, hours between measurements)... Add two "10.0 volt" batteries in series, got 28.0 volts (or something like that)--Meter was just frustrating consistent but illogical readings. Replace the meter battery(ies) and the world was right again.

    You can use a car battery (known good before~12.7 volts resting) or even an alkaline cell, to confirm that the DMM has got "most of its brains" if things start looking strange.

    There are less expensive (set voltage) calibrators... Here is a thread where folks go down their rabbit hole:

    https://www.eevblog.com/forum/metrology/are-cheap-ad584-units-worth-it/

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭

     When I had it charged about a week ago, sitting voltage was about 53.4 or so. Now when I went to check it was about 53.2... how is it possible to have parasitic loss? 
    I don't now exactly, I don't deal with Lithium batteries.
    Lithium have very little resistance and perhaps the batteries in parallel with the lower voltage battery was trying to bring it up.

    I guess there might also be some losses with an active internal battery management system...

    Wish it was 'my end of the pool' but I just don't know. 
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    Photowhit said:

     When I had it charged about a week ago, sitting voltage was about 53.4 or so. Now when I went to check it was about 53.2... how is it possible to have parasitic loss? 
    I don't now exactly, I don't deal with Lithium batteries.
    Lithium have very little resistance and perhaps the batteries in parallel with the lower voltage battery was trying to bring it up.

    I guess there might also be some losses with an active internal battery management system...

    Wish it was 'my end of the pool' but I just don't know. 
    LiFePo4 do have low self discharge the rate of which depends on temperature along with how high the voltage is above the nominal cell voltage. Seeing a drop from 3.337 VPC (volts per cell) to 3.325 VPC over the course of a week would not be unreasonable, once the plateau voltage is reached the drop will reduce.

    When dealing with lithium batteries it's best to deal at a cellular level and not to relate knowledge acquired from other chemistries, such as lead acid, as their characteristics are significantly different. There are two critical areas to avoid known as the charging and discharging knees, this is when the voltage either spikes or collapses without warning hense the recommendation of using a BMS to reduce charging current at around 3.500 VPC and terminate at 3.600 VPC  or during discharge, disconnect loads around 3.000 VPC. 

    Keeping away from the maximum / minimum values will increase the cycle expectancy as cell drift becomes more pronounced depending largely on where the cells were initially ballanced, top or bottom.

    My suggestion is before using any lithium batteries, take the time to study the particular chemistry one chooses wether drop in or DIY.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.