Pre-wiring for new construction

NBDrancherNBDrancher Registered Users Posts: 13 ✭✭
I'm curious what size conduit I should run and what wire size I need for a 40 foot run from a 3000 watt PV array/combiner box to a Four Star Solar Mini Magnum 4448PAE w/Classic 150 charge controller/48 volt battery bank.  I'm preparing to pour concrete and want to do this right. 

Comments

  • BB.BB. Super Moderators Posts: 26,976 admin
    The National Electric Code would be the first stop (defines insulation/wet vs dry/current/fill factor/etc.). And you need to know the wiring of the Array (series/parallel, do you bring down one cable or one pair for each panel string, will there be a ground wire, etc.).

    The second thing to check for is voltage drop of the run... You may need heavier cable (and larger conduit). Again need array wiring details.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • vtmapsvtmaps Solar Expert Posts: 3,738 ✭✭✭✭
    As Bill points out, the question cannot be answered without knowing how your array is configured.  I will make some guesses...

    With a Classic 150 on a 48 volt system, your string Vmp should be at least 75 volts, and your string Voc should be less than 120 volts.  I guess you have 60 cell panels configured three to a string, with a Vmp of 90 volts and a combined Imp of 33.3 amps.

    That voltage and current through a 40 ft run (80 ft of cable for the round trip) of 8 gauge copper, will give a voltage drop of 1.9%.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Wheelman55Wheelman55 Registered Users Posts: 35 ✭✭
    There are three very useful apps from Southwire: one for "Voltage Drop", one for "Conduit Fill", and one for "Box Fill".  The apps use NEC code. 

    The voltage drop app will ask you questions and then give you wire size. For instance, you'll need to tell the app how many volts and amps will run through this wire(s), acceptable voltage drop (use 2 or 3%), and length of wire run. 

    Once you have wire size, you then determine how many wires need to be run in the conduit. When you know wire size and number of wires, use the conduit fill app to size the conduit. 

    Do this, then consult with the experts here to confirm your results. 
    Building Off-Grid in Terlingua, TX
  • NBDrancherNBDrancher Registered Users Posts: 13 ✭✭
    Thank you all for responses and direction.  The array will be wired in series/parallel-I believe we are using 18 of the 60-cell Kyocera KU265-GX panels with Vmp of 31.0 and a Imp of 8.55 individually.  I'm guessing #3 wire should handle this.in a 1" conduit.
  • vtmapsvtmaps Solar Expert Posts: 3,738 ✭✭✭✭
    Thank you all for responses and direction.  The array will be wired in series/parallel-I believe we are using 18 of the 60-cell Kyocera KU265-GX panels with Vmp of 31.0 and a Imp of 8.55 individually.  I'm guessing #3 wire should handle this.in a 1" conduit.
    31 volts X 8.55 amps = 265 watts.
    18 solar panels X 265 watts = 4771 watt array. 

    By the way, I did my earlier calculations based on a 3000 watt array because you wrote in the original post:
    I'm curious what size conduit I should run and what wire size I need for a 40 foot run from a 3000 watt PV array/combiner box to a Four Star Solar Mini Magnum 4448PAE w/Classic 150 charge controller/48 volt battery bank.
    You still haven't answered the question about how the panels are configured.  I will assume that you have chosen the most sensible configuration for 18 panels, which is six strings with three panels per string (string voltage = 93).

    In that case, a 40 ft run of #6 copper will have a voltage drop of less than 2%.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • NBDrancherNBDrancher Registered Users Posts: 13 ✭✭
    Yes, I believe we will install the panels in 6 strings @ 3 panels-per-string.  Sorry about my original post; it was based on my 1st calculation of requirements for our ranch.  After much hand wringing and thinking I decided we had better go for more power.
     
    Thank you so much for taking the time to provide this information. 
  • BB.BB. Super Moderators Posts: 26,976 admin
    While you are there--Perhaps you may want to lay more conduit? Expansion, possibly sending AC power back out to a shed? Adding low voltage communications (Internet, antenna lead, generator feed+control, etc.)?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • NBDrancherNBDrancher Registered Users Posts: 13 ✭✭
    Indeed Bill, and thanks for pointing that out.  I have 3 or 4 more conduits planned.  I am assuming I will install the array combiner box at the panels themselves. 
  • NBDrancherNBDrancher Registered Users Posts: 13 ✭✭
    vtmaps said:
    Thank you all for responses and direction.  The array will be wired in series/parallel-I believe we are using 18 of the 60-cell Kyocera KU265-GX panels with Vmp of 31.0 and a Imp of 8.55 individually.  I'm guessing #3 wire should handle this.in a 1" conduit.
    31 volts X 8.55 amps = 265 watts.
    18 solar panels X 265 watts = 4771 watt array. 

    By the way, I did my earlier calculations based on a 3000 watt array because you wrote in the original post:
    I'm curious what size conduit I should run and what wire size I need for a 40 foot run from a 3000 watt PV array/combiner box to a Four Star Solar Mini Magnum 4448PAE w/Classic 150 charge controller/48 volt battery bank.
    You still haven't answered the question about how the panels are configured.  I will assume that you have chosen the most sensible configuration for 18 panels, which is six strings with three panels per string (string voltage = 93).

    In that case, a 40 ft run of #6 copper will have a voltage drop of less than 2%.

    --vtMaps
    I'm a little confused in this regard:  If the cumulative voltage from the 3 array strings equals 93 volts, will the 48 volt inverter be over-driven?  Would I be better off with 24 panels; two separate arrays of six-strings with two panels and  voltage number closer to 60 volts?  I know my current system @ 24VDC is being charged by 31 Vmp panels at 4.5 Imp.  
  • BB.BB. Super Moderators Posts: 26,976 admin
    The Classic charge controller, for all the world, acts like a variable transformer... It takes "high voltage/low current" from the solar array and "down converts" (with ~95% efficiency) to "low voltage/high current" needed for the battery bank.

    For most areas, the maximum Vmp-array is around 100 VDC.

    If you had a "cheap" PWM charge controller, your optimum Vmp-array would be ~72 VDC... This allows for Vmp to fall with high panel temperatures in the summer, and some voltage drop in the array wiring.

    In the end, it is the battery bank that sets the operating voltage (input voltage range) for your inverter.

    HOWEVER--there is always a however in engineering land--The operating voltage range of a 48 VDC flooded cell lead acid battery is around 42 volts to 59 volts for a room temperature battery bank under normal discharging/charging conditions.

    Depending on battery construction and battery bank temperatures, charging (and equalization) voltages can reach 60-62 VDC (cold battery banks need higher charging voltages, equalization requires higher charging voltages).

    And, many AC inverter mfg. seem to set their "cutoff voltages" to ~15/30/60 Volts (12/24/48 volt banks). And, especially for our friends in the great white north, they have problems with inverters shutting down in very cold weather (and during equalization).

    AGM batteries do not need (and should not be subjected) to elevated voltages for equalization. Some industrial batteries (forklift, traction) seem to need elevated charging voltages as a matter of course.

    So--The summary:

    Your MPPT charge controller will make efficient use of the elevated Vmp-array charging voltage.

    Your actual battery bank voltage will be set by Load, state of charge,  Absorb, and Equalization voltage set points will be affected by temperature, charging phase, and battery construction. And 60 VDC, especially for cold climates and industrial FLA batteries, will be a real problem for folks with 60 VDC cutoff inverters (i.e., may have to shut down inverter during some charging phases).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • vtmapsvtmaps Solar Expert Posts: 3,738 ✭✭✭✭
    I'm a little confused in this regard:  If the cumulative voltage from the 3 array strings equals 93 volts, will the 48 volt inverter be over-driven? 
    Bill answered your question by explaining that the MPPT controller does the voltage conversion.  I will throw a little more math at the question.

    For starters let's pretend the controller is 100% efficient... then power in = power out. 

    In your case, Power in = 93 volts X  8.55 amps X 6 strings = 4771 watts

    Power out depends on the battery voltage at the moment.
    If the battery is at 58 volts, then 4771 watts ÷ 58 volts = 82.3 amps out of the controller.  Of course the controller is not 100% efficient, so there will be less than 82.3 amps.    The lower the battery voltage, the higher the current (amps) will be.   That's why Bill referred to the MPPT as like a transformer.
     I know my current system @ 24VDC is being charged by 31 Vmp panels at 4.5 Imp.  
    If that works for you, I can't argue with success, but many folks have ruined batteries with a setup like that.  Many 24 volt batteries require 29.5 volts for absorb and 31 volts for equalization.  When panels get hot their voltage drops.  If the panel Vmp is 31 volts at standard temperature, then it is several volts lower when hot.  If your battery needs 29.5 volts and the panel's Vmp is 28 volts... there is a problem.  And that does not take into consideration that the controller loses some of the voltage also.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • NBDrancherNBDrancher Registered Users Posts: 13 ✭✭
    BB. said:
    The Classic charge controller, for all the world, acts like a variable transformer... It takes "high voltage/low current" from the solar array and "down converts" (with ~95% efficiency) to "low voltage/high current" needed for the battery bank.

    For most areas, the maximum Vmp-array is around 100 VDC.

    If you had a "cheap" PWM charge controller, your optimum Vmp-array would be ~72 VDC... This allows for Vmp to fall with high panel temperatures in the summer, and some voltage drop in the array wiring.

    In the end, it is the battery bank that sets the operating voltage (input voltage range) for your inverter.

    HOWEVER--there is always a however in engineering land--The operating voltage range of a 48 VDC flooded cell lead acid battery is around 42 volts to 59 volts for a room temperature battery bank under normal discharging/charging conditions.

    Depending on battery construction and battery bank temperatures, charging (and equalization) voltages can reach 60-62 VDC (cold battery banks need higher charging voltages, equalization requires higher charging voltages).

    And, many AC inverter mfg. seem to set their "cutoff voltages" to ~15/30/60 Volts (12/24/48 volt banks). And, especially for our friends in the great white north, they have problems with inverters shutting down in very cold weather (and during equalization).

    AGM batteries do not need (and should not be subjected) to elevated voltages for equalization. Some industrial batteries (forklift, traction) seem to need elevated charging voltages as a matter of course.

    So--The summary:

    Your MPPT charge controller will make efficient use of the elevated Vmp-array charging voltage.

    Your actual battery bank voltage will be set by Load, state of charge,  Absorb, and Equalization voltage set points will be affected by temperature, charging phase, and battery construction. And 60 VDC, especially for cold climates and industrial FLA batteries, will be a real problem for folks with 60 VDC cutoff inverters (i.e., may have to shut down inverter during some charging phases).

    -Bill

    Bill-
    I have to apologize for missing the obvious when it comes to the rate of charge to/thru the charge controller, not the inverter as I inadvertently communicated. 
    We do reach single-digits in the winter here and I have realized poor charging rates when the existing battery bank has been depleted overnight.  My new installation will be inside a well insulated building, batteries in a plywood box vented to the outside. 

    another Bill-

  • NBDrancherNBDrancher Registered Users Posts: 13 ✭✭
    vtmaps said:
    I'm a little confused in this regard:  If the cumulative voltage from the 3 array strings equals 93 volts, will the 48 volt inverter be over-driven? 
    Bill answered your question by explaining that the MPPT controller does the voltage conversion.  I will throw a little more math at the question.

    For starters let's pretend the controller is 100% efficient... then power in = power out. 

    In your case, Power in = 93 volts X  8.55 amps X 6 strings = 4771 watts

    Power out depends on the battery voltage at the moment.
    If the battery is at 58 volts, then 4771 watts ÷ 58 volts = 82.3 amps out of the controller.  Of course the controller is not 100% efficient, so there will be less than 82.3 amps.    The lower the battery voltage, the higher the current (amps) will be.   That's why Bill referred to the MPPT as like a transformer.
     I know my current system @ 24VDC is being charged by 31 Vmp panels at 4.5 Imp.  
    If that works for you, I can't argue with success, but many folks have ruined batteries with a setup like that.  Many 24 volt batteries require 29.5 volts for absorb and 31 volts for equalization.  When panels get hot their voltage drops.  If the panel Vmp is 31 volts at standard temperature, then it is several volts lower when hot.  If your battery needs 29.5 volts and the panel's Vmp is 28 volts... there is a problem.  And that does not take into consideration that the controller loses some of the voltage also.

    --vtMaps
    Thanks, I'm impressed and relieved with your answers on these points.  Since my ultimate goal was a system with improved efficiency, reliable pure sine-wave delivery and to be able to operate the basics I think we should be in good shape.  I have ruined a few of the T-105 style 6vdc batteries in the past because our array was too small and I over-taxed the battery bank.  

    As both you and Bill have alluded to-a balanced system must be achieved or as close to it as possible.  
Sign In or Register to comment.