Numbers Not Adding Up

lzhome
lzhome Registered Users Posts: 39 ✭✭
Hope you can see the attached graphic. This is sizing data for a off-grid system I'm planning and my gut tells me I'm either at too many batteries or not enough PV (4x100w). Would appreciate your comments. Here are usage details:
Daily Watt Hrs: 1553
Daily Ah: 77.65  


TriStar MPPT, 8 x 100w PV, MNPV6 Combiner, 4 x 12v 155ah VMAXTanks AGMs and GoPower 2000w PSW Inverter.

Comments

  • Horsefly
    Horsefly Registered Users Posts: 470 ✭✭✭✭
    I may be a little slow, but I think there are a few things here that are confusing.
    1. It looks like you have some 24VDC loads, and maybe some that are at 120VAC?  If so, you should count up there instantaneous power and daily energy consumption separately. The watt hours at 120VDC would need to be divided by roughly 0.85 to account for the inverter inefficiency. Then you add that number to the watt hours at 24VDC (which doesn't need to go through an inverter) and that gives you the total daily watt hours you will be pulling from your batteries.  I would give you my take on that number, but some of the other stuff doesn't make enough sense yet for me to give you a figure.
    2. You have a 1000W inverter in your load list, you say it is on continuously, but that only means 15 hours per day? And it only pulls 9.6 watts? So you don't actually have anything using any power from the inverter, but you keep it powered up for some reason?
    3. You have a "charging efficiency" of 1.2. What does that mean?
    4. It looks like you are choosing days of autonomy ("reserve days"?) of 1.5. Just to be clear, that means that if you have a really crumby weather day, you only expect to have enough power for 50% of your normal load.
    5. Your depth of discharge calcs are correct, to the degree that the 77.65 Ah is correct, which I don't have much faith in yet.
    6. If you've chosen a 450Ah battery bank at 24V, you will have a charging voltage of around 29V. You will also have some inefficiency between the panels and your batteries (everyone probably has different #'s here, but these are mine): charge controller efficiency (95% for MPPT, lower for PWM), wire efficiency, accounting for wire loss (97%), and panel efficiency, lower than STC (75%). You want something close to a 10% charging rate for each hour of full sun, but between 5% and 13% seems to be what people say is good. So for your 450Ah battery bank you would need:   450Ah * 29V / (0.95 * 0.97 * 0.75)  * 0.10 (10%) = about 1,900W of panels. Note that this would be higher if you use a PWM charge controller.

    Off-grid cabin: 6 x Canadian Solar CSK-280M PV panels, Schneider XW-MPPT60-150 Charge Controller, Schneider CSW4024 Inverter/Charger, Schneider SCP, 8S (25.6V), 230Ah Eve LiFePO4 battery in a custom insulated and heated case.
  • lzhome
    lzhome Registered Users Posts: 39 ✭✭
    Horsefly said:
    ...
    1. Your depth of discharge calcs are correct, to the degree that the 77.65 Ah is correct, which I don't have much faith in yet.
    2. If you've chosen a 450Ah battery bank at 24V, you will have a charging voltage of around 29V. You will also have some inefficiency between the panels and your batteries (everyone probably has different #'s here, but these are mine): charge controller efficiency (95% for MPPT, lower for PWM), wire efficiency, accounting for wire loss (97%), and panel efficiency, lower than STC (75%). You want something close to a 10% charging rate for each hour of full sun, but between 5% and 13% seems to be what people say is good. So for your 450Ah battery bank you would need:   450Ah * 29V / (0.95 * 0.97 * 0.75)  * 0.10 (10%) = about 1,900W of panels. Note that this would be higher if you use a PWM charge controller.

    If we can assume the daily load of 77.65 Ah is correct using a MPPT CC and the efficiency considerations you outlined, are we saying 1,900 w of panels to charge from 75-80% Dod? Just seems like an enormous number of panels for what I would consider a small load requirement. My plan was to go with 800w of panels so I seem to be way off at this point. But I do appreciate your response.
    TriStar MPPT, 8 x 100w PV, MNPV6 Combiner, 4 x 12v 155ah VMAXTanks AGMs and GoPower 2000w PSW Inverter.
  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    edited October 2017 #4
    @lzhome
    If you have a Morningstar MPPT 60 charge controller to charge a 450AH 24 volt battery bank then in my opinion, the best way to configure the array is 3 strings of 3 100 watt panels. A combiner box would make this easier and I would recommend one like this:
    https://www.solar-electric.com/misomnsoarco.html
    This combiner box will enable expansion up to 6 strings of 3 panels. The charge controller and the batteries also will be able to handle the full expansion to 6 strings if necessary.

    The reason for 3 strings of three is the voltage of 2 panels in series is not high enough for an MPPT CC to reliably charge a 24 volt battery and 4 in series is unnecessarily high, whereas 3 panels in series is 54 volts, giving plenty of headroom for the charge controller to down convert the voltage. Besides, you really need the power that the extra panel will produce. 
    900 watts in this configuration, will effectively produce about 30 charging amps or a 6.7% charge rate. That's about 165 amp hours, I know that's more than you plan to use, but believe me, you will need it on partly cloudy days. The general rule of thumb is that an off grid system that is doing daily cycling should have a 10% charge rate, but since you are not cycling too deeply, then 6.7 should give enough charging to handle the occasional cloud and still get the batteries full. But if you need more, just add a string of three panels. The number one reason that batteries die an early death is chronic under charging.

    Rick
    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    In figuring out how much panel you need/want, although we use some rules of thumb, it really comes down to personal preference.  There's nothing "wrong" with using 400w of pv on a 450ah@24v bank.  You could bulk charge with a generator every morning, and let the panels finish charging if the weather's nice.  Depending on location, you may end up doing that in winter anyway.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • Horsefly
    Horsefly Registered Users Posts: 470 ✭✭✭✭
    lzhome said:
    Horsefly said:
    ...
    1. Your depth of discharge calcs are correct, to the degree that the 77.65 Ah is correct, which I don't have much faith in yet.
    2. If you've chosen a 450Ah battery bank at 24V, you will have a charging voltage of around 29V. You will also have some inefficiency between the panels and your batteries (everyone probably has different #'s here, but these are mine): charge controller efficiency (95% for MPPT, lower for PWM), wire efficiency, accounting for wire loss (97%), and panel efficiency, lower than STC (75%). You want something close to a 10% charging rate for each hour of full sun, but between 5% and 13% seems to be what people say is good. So for your 450Ah battery bank you would need:   450Ah * 29V / (0.95 * 0.97 * 0.75)  * 0.10 (10%) = about 1,900W of panels. Note that this would be higher if you use a PWM charge controller.

    If we can assume the daily load of 77.65 Ah is correct using a MPPT CC and the efficiency considerations you outlined, are we saying 1,900 w of panels to charge from 75-80% Dod? Just seems like an enormous number of panels for what I would consider a small load requirement. My plan was to go with 800w of panels so I seem to be way off at this point. But I do appreciate your response.
    You are right that with a lower depth of discharge you can get by with less from the solar. I was just giving you the general answer. I think the point is (and you seem to already know it) that the way you calculated the number of panels you needed was not correct. You can't simply take the Ah and divide it by the full sun hours and the Imp of the panels to get a panel count. I'm still concerned that your daily loads may not be right, but can't tell. If you don't want to explain any of the other issues I brought up that's cool. If you feel confident in it, then it's all good.
    Off-grid cabin: 6 x Canadian Solar CSK-280M PV panels, Schneider XW-MPPT60-150 Charge Controller, Schneider CSW4024 Inverter/Charger, Schneider SCP, 8S (25.6V), 230Ah Eve LiFePO4 battery in a custom insulated and heated case.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Your refer usage seems a bit on the low side... Typically closer to 1,000 WH per day vs 500 WH--Unless this is a chest freezer converted to a chest refrigerator (most AC refrigerator compressors run around 100-120 Watts for 25%-50% duty cycle). Also, I am not sure, but I don't think you have accounted for all of the "losses and deratings" that need to happen with off grid power.

    Anyway... Let us assume that 1,553 WH per day is correct. And, for first cut, all of it is 120 VAC power (85% efficient AC inverter). Sizing the battery bank:
    • 1,553 WH per day * 1/0.85 AC inverter eff * 2 days storage * 1/0.50 max discharge * 1/12 volt battery bank = 608 AH @ 12 volt battery bank
    Then calculating solar array size... 5% rate of charge can work for weekend/seasonal cabins. 10%-13% is recommended for full time off grid:
    • 608 AH * 14.5 volts charging * 1/0.77 panel+controller deratings * 0.05 rate of charge = 572 Watt minimum array
    • 608 AH * 14.5 volts charging * 1/0.77 panel+controller deratings * 0.10 rate of charge = 1,145 Watt array nominal
    • 608 AH * 14.5 volts charging * 1/0.77 panel+controller deratings * 0.13 rate of charge = 1,488 Watt array "cost effective" maximum
    And then there is sizing the array for the amount of sun you receive. (guessing, don't know your location, fixed array):
    http://www.solarelectricityhandbook.com/solar-irradiance.html

    Hattiesburg
    Average Solar Insolation figures

    Measured in kWh/m2/day onto a solar panel set at a 59° angle from vertical:
    (For best year-round performance)
    Jan Feb Mar Apr May Jun
    3.53
     
    4.16
     
    4.80
     
    5.40
     
    5.43
     
    5.24
     
    Jul Aug Sep Oct Nov Dec
    5.21
     
    5.14
     
    5.06
     
    4.99
     
    4.14
     
    3.50
     
    For a lot of regions, 5.5 hours a of sun is a lot--Typically towards the gulf, you have a lot of clouds and marine layer that can reduce sun? Anyway, assume 4.0 hours of sun and use a genset for poor winter sun:
    • 1,553 WH per day * 1/0.52 off grid off grid AC system eff * 1/4.0 hours of sun "break even" = 747 Watt array minimum (4.0 hours of sun minimum)
    If full time off grid, a 747 Watt (~9 months of year without genset) to 1,145 Watt (nominal) to 1,488 Watt (cost effective maximum) array system would supply the planned energy usage.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • lzhome
    lzhome Registered Users Posts: 39 ✭✭
    Horsefly said: I think the point is (and you seem to already know it) that the way you calculated the number of panels you needed was not correct. You can't simply take the Ah and divide it by the full sun hours and the Imp of the panels to get a panel count. I'm still concerned that your daily loads may not be right, but can't tell. If you don't want to explain any of the other issues I brought up that's cool. If you feel confident in it, then it's all good.
    I actually followed a worksheet from a book that divided Ah by Sun hrs to determine PV. And yes, I knew that wasn't working out and was the primary reason I started this thread. I am located in the Florida Panhandle so the 5.5 hrs of sun may not be a reality for every day of the year, especially for my static panels facing South.

     There are several loose ends I need to tie up but thanks to you folks I now have the tools/information to make those decisions. I'm sitting on the fence with several issues, 12v vs 24v, # of panels, and battery bank size. If I could only grow the battery bank as I learn and receive system feedback that would be great; but we know batteries like cloned siblings and not younger/older batteries joining the bank. 

    BTW here is an email quote I received from Trojan which I found very interest: "The T-105 225 and SSIG 06 255 are the same battery. The capacity ratings for the Solar flooded products have been calculated at 86°F (30°C) as many renewable energy applications operate at higher temperatures."
    TriStar MPPT, 8 x 100w PV, MNPV6 Combiner, 4 x 12v 155ah VMAXTanks AGMs and GoPower 2000w PSW Inverter.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    More or less, I suggest of your battery bank is over roughly 800 AH in capacity, I would suggest going up to next bus voltage (800ah@12volts becomes 400ah@24volts. Same energy storage).

    Mostly because in a balanced system design, this keeps battery wiring to s reasonable diameter, and meshes well with 80-90 ampere solar charge controllers (1,000 ah battery bank with 100 amps charging needs 2x solar charge controllers, vs 1x at 500ah and 50 amps).

    But you also need to see which batteries are available, what weight cells you can move around and such.

    Also, what a.c. inverter you want and what features. 12 volt inverters are lower wattage, few features, and good for smaller systems.

    More difficult to find a low wattage 48dcv input inverter.

    Your system is on the edge of 12 or 24 volts. If you want a 3.3kwh per day system and full size fridge, a 24 volt system minimum would be nice.

    If you have a bunch of 12vdc native loads, perhaps 12vdv battery bus would be better.

    Note that many 12 VDC loads do not like 10.5-16.0 volt range that a typical of grid battery system runs at. A major reason o suggest 120 VAC loads and inverter instead (and make the system 1.18x larger to take care of the 85% efficient inverter).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • lzhome
    lzhome Registered Users Posts: 39 ✭✭
    edited October 2017 #10
    Raj174 said:
    @lzhome
    If you have a Morningstar MPPT 60 charge controller to charge a 450AH 24 volt battery bank then in my opinion, the best way to configure the array is 3 strings of 3 100 watt panels. A combiner box would make this easier and I would recommend one like this:
    https://www.solar-electric.com/misomnsoarco.html

    The reason for 3 strings of three is the voltage of 2 panels in series is not high enough for an MPPT CC to reliably charge a 24 volt battery and 4 in series is unnecessarily high, whereas 3 panels in series is 54 volts, giving plenty of headroom for the charge controller to down convert the voltage. Besides, you really need the power that the extra panel will produce. 
    900 watts in this configuration, will effectively produce about 30 charging amps or a 6.7% charge rate. That's about 165 amp hours, I know that's more than you plan to use, but believe me, you will need it on partly cloudy days. The general rule of thumb is that an off grid system that is doing daily cycling should have a 10% charge rate, but since you are not cycling too deeply, then 6.7 should give enough charging to handle the occasional cloud and still get the batteries full. But if you need more, just add a string of three panels. The number one reason that batteries die an early death is chronic under charging.

    Using the string calculator at morningstar.com and I got some different results on 3 strings of 3 100w (900w) panels. Lets call this option "A".

    Minimum Voc (average high temp.)61.47 
    Max. Operating Voc (record low temp.)67.09 
    Max. Voc (record low Temp., morning)72.56<150Max. Vin
    Minimum Vmp (average high temp.)50.67>31.00Vb Max.
     
    Maximum Vmp (record low temp.)56.29 
    Isc (STC)17.25 
    Imp (STC)15.87 

    Option "B", My approach was to use 4 strings of 2 100w (800w) panels providing sufficient volts to charge 24v bank and increasing amps.

    Minimum Voc (average high temp.)40.98 
    Max. Operating Voc (record low temp.)44.73 
    Max. Voc (record low Temp., morning)48.38<150Max. Vin
    Minimum Vmp (average high temp.)33.78>31.00Vb Max.
     
    Maximum Vmp (record low temp.)37.53 
    Isc (STC)23.00 
    Imp (STC)21.16 

    Comments please.

    TriStar MPPT, 8 x 100w PV, MNPV6 Combiner, 4 x 12v 155ah VMAXTanks AGMs and GoPower 2000w PSW Inverter.
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    How far will the panels be from the charge controller?  If fairly close and you use the right wire, you're probably okay with option B.  You might want to check an online voltage drop calculator to be sure though.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • lzhome
    lzhome Registered Users Posts: 39 ✭✭
    edited October 2017 #12
    @Estragon Just under 15 feet of 10 AWG wire from charge controller. I did run the voltage drop calculation and it was pretty small, somewhere around .59 volts.
    TriStar MPPT, 8 x 100w PV, MNPV6 Combiner, 4 x 12v 155ah VMAXTanks AGMs and GoPower 2000w PSW Inverter.
  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    @lzhome
    PV voltage of 2 times the nominal battery voltage is recommended to avoid any possible issues related to low PV voltage with a MPPT charge controller. It would essentially remove the word "probably" from Estragon's comment. In all likelihood two panels in series will be fine with your AGM batteries. If I only had 8 panels, I would go ahead and give the 2 in series strings a try. 

    Rick 
    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • lzhome
    lzhome Registered Users Posts: 39 ✭✭
    @Raj174 Will be using T-105s Flooded LA but probably not any difference. Thanks for your confirmation. 
    TriStar MPPT, 8 x 100w PV, MNPV6 Combiner, 4 x 12v 155ah VMAXTanks AGMs and GoPower 2000w PSW Inverter.
  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    edited October 2017 #15
    If you will be using T-105's then you will very likely not be able to equalize the batteries with 2 panel strings. I have a friend who is using two 100 watt panels in series and he can not achieve EQ at 32 volts with T-105 batteries. Voltage requirements vary with temperature also. He has switched to AGM batteries and charging is ok. No EQ. Of course you can EQ with a generator. 

    Rick

    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • lzhome
    lzhome Registered Users Posts: 39 ✭✭
    Raj174 said:
    If you will be using T-105's then you will very likely not be able to equalize the batteries with 2 panel strings. I have a friend who is using two 100 watt panels in series and he can not achieve EQ at 32 volts with T-105 batteries. Voltage requirements vary with temperature also. He has switched to AGM batteries and charging is ok. No EQ. Of course you can EQ with a generator. 

    Rick

    With the short cable runs and higher temps in Florida panhandle is FLA EQ still expected to be an issue once I expand out to 800w?
    With the expense of batteries, maybe your friends best option may have been to rewire PV configurations. I really need to get this nailed down soon before making battery purchase so anyone's input on this would be greatly appreciated.
    TriStar MPPT, 8 x 100w PV, MNPV6 Combiner, 4 x 12v 155ah VMAXTanks AGMs and GoPower 2000w PSW Inverter.
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    The batteries will presumably also be warm, so EQ voltage doesn't need to be as high as it might be in a cooler climate.  Still, 2 panels in series might have trouble on hot days.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • lzhome
    lzhome Registered Users Posts: 39 ✭✭
    If I'm not off track, shouldn't the batt bank charge at 5%-10% of capacity? So with 450 amh batt bank it looks like I should be producing 22 to 46 amps. What is my compromise between higher voltage to achieve EQ or higher amps to fully charge batteries?
    Kinda related, but what does the MPPT CC do with the "excess" voltage beyond what the batteries need? Anything?
    TriStar MPPT, 8 x 100w PV, MNPV6 Combiner, 4 x 12v 155ah VMAXTanks AGMs and GoPower 2000w PSW Inverter.
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    One of the fundemental mistakes made is undersized array capacity, which leads to cronic undercharging of the battery, given the capacity of 450 Ah the array should be in the region of 1500W, having an array that is too small will limit it's ability to satisfy the demands, which will pull the voltage down to the point where an EQ  cannot be performed, or regular charging for that matter. Having an array which can satisfy the requirements is of paramount importance, this includes any loads that are subsequently connected. The 800 watts PV, is in my opinion simply too low for the battery capacity of 450 Ah, unless of course you use the system on weekends only, then it may  be sufficient.
    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.
  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    @lzhome
    An MPPT controller converts the higher voltage into higher charging amps. But it needs more room/higher PV voltage to do this. It gets nearly 100% of the available power from the panel, whereas a PWM controller, uses battery panel voltage, which is 36 volts for a 24 volt battery, and can only get about 70% of the available power in the panel. The extra 30% is essentially converted to higher amperage by the MPPT controller. The MPPT controller really needs to see PV well above battery panel voltage. PV in at twice the battery voltage is recommended. Once the panel string voltage is determined, based on battery voltage, controller type and sometimes the distance from the array to the controller, then the required array size can be made by adding strings in parallel. By the way, I live in the Florida panhandle also, near Tallahassee.

    Rick
    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.