Off grid backyard trailer solar

tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
I am planning on installing an off grid solar system for my backyard "office" trailer.  It is about 8' x 14' and I think I can get 4 Canadian Solar 305 watt panels on the roof.  I know nothing about solar systems, so I have some newby questions.

I have measured my power consumption and it is about 2.5-3.5 killowatts per day in the winter.  The window air conditioner can add another 6 killowatts for at total of 10 killowatts on the worst 115 F days in Tucson, AZ (with a 640 watt max window a/c, but I could go down to a 450 watt a/c.)

So, with 4 panels of 305 watts that would be a total of 1220 watts per hour incoming x 6 hours average peak sun hours = 7320 watts total. The solar panels would face south with no shading, and the trailer could be turned to true solar south if necessary or significantly beneficial.

My questions are:

What would be the best battery bank configuration. (FLA or SLA, 6 volt or 12 volt, indoor or outdoor location.) I have plenty of floor space for the batteries inside the trailer, or I could build an insulated  and vented storage box for FLA batteries which would store nicely under the trailer, so it would be shaded but still exposed to hot summer temperatures.  I could leave the window a/c on set at about 80 degrees 24/7, which would keep the battery temperature moderate (assuming that 7320 watts of solar input would support this constant setting.)

I have read about panel mounted micro inverters but have also read they they are not reliable..  Would micro inverters eliminate the need for a central string inverter and are they worth the extra cost? 

In looking at conventional inverters, I see that they seem to have only a 2 year warranty.  Are they designed to last only two years?  I have seen youtube videos of name brand inverters failing shortly after purchase. Others have said that their inverters last forever with no repairs needed.

A solar charge controller stops the charge to the batteries when the batteries are at 100% SOC.  So, what happens to the additional power from the solar panels when the batteries don't need a charge?  Do I need to install a "dump" load.

If supplemental charging of the batteries is needed from a generator, what is the best way to connect the generator to the batteries.  Can a separate charger be connected directly from the generator to the battery bank, or should the charging be done through another system component, like the inverter?  What size and type of charger is needed for FLA or SLA batteries?  Would an RV type converter/charger  be better?

I have read conflicting calculations for the ideal solar panel angle. Some angles are based on the angle from zenith (straight up and down - which I assume would indicate the complimentary angle) and others are based on the angle of elevation from the horizontal surface (which would be the flat roof of my trailer.)  Tucson is at about 32 degrees latitude.  I am most concerned about maximum output from the panels in the summer to accommodate the window a/c, and that seems to be 6"-8" elevation over the length of the 65" Canadian Solar panel.  Should the panels be mounted in a fixed "best summer output angle" or should they be adjustable?

Thanks for any help you can provide.












«13

Comments

  • EstragonEstragon Posts: 2,962Registered Users ✭✭✭✭✭
    4 x 305w panels would likely produce ~ 900w in normal operating temps, maybe a bit less if mounted flat close to the roof with little airflow.  

    There are pros and cons to flooded vs AGM (sealed).  In this application, they may lean to AGM.  They could live in the trailer, whereas having flooded with associated gassing may be best kept outside occupied space.  Cooler is better for batteries.

    3.5kwh/day likely needs something like (eg, just for illustration) a bank of 8x6v L16 size batteries, for about 400 amp-hours at 48v for ~19000 watt-hours.  For long life you don't want to draw down more than 50%, so 8500wh usable - enough to last a couple of days at 3.5kw before starting a generator.  In summer, at 10kwh/day, you'd be running a generator daily rain or shine.

    Assuming panels produce 900w, and most months have 6hrs full sun equivalent, that gives you ~5400 wh/day.  Enough to support loads at 3500 wh/day, but may take a few sunny days to catch up charge batteries after a couple of rainy days (which is okay).

    If batteries are full, and there are no other loads, pv simply won't produce more power than needed to float charge batteries.  A dump load isn't required. 

    For charging, an inverter with integrated transfer switch and charger is probably simplest, and is likely cheaper than quality separate inverter and charger.  Inverter/chargers such as Outback/Schneider etc sold by our hosts generally have a life expectancy of ~10years.

    Some can be series stacked (120/240vac - useful if air conditioning is 240v), and/or parallel stacked (for higher current - useful if you need 10kw at 120vac).

    A rule of thunb for tilt is latitude plus 15° from horizontal in winter, latitude minus 15° (flatter) in summer.  Flatter panels in summer not only give a better angle to the sun, they also produce longer/better in early (east sun) and later (west) hours.  


    They should be pointed roughly south, but it isn't really critical to make them exactly south.

    To support a 10kw/day load is going to run significant money for panels, inverters, batteries, generator, and related wire/boxes/breakers, etc.  If this is a backyard of a grid supplied location, trenching or whatever needed to get grid power to the trailer is likely to be much less expensive.
    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
  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    Thanks for the response.  Am I correct in thinking that if I am consuming 450 watts of energy while the a/c is running during the 6 peak sunlight hours, that the net input to the batteries will be 450 watts per hour during that 6 hour period, resulting in a 2700 watt net input to the batteries during that period?  Also, will there be any additional energy coming from the solar panels in the summer beyond 6 hours, since the sun rises earlier in the summer and sets later in the evening during the summer?

    I am trying to make the trailer self sufficient without grid power.

    Please help me with the math.  If I consume 3500 watts of energy in one day, and I allow 50% DOD, then I need 7000 watts of battery capacity.  If 7000 watts / 6 volts = 1167 AH of battery capacity and If a Trojan T105 is 225 AH, why would I not need 5.18 or 6 of the T105 batteries?  I accept that the generator will have to run a long time during days with no sun. I would not want to add capacity beyond one sunny day's energy production, since we have very few cloudy days in Tucson.

    The number of solar panels is constrained to the footprint of the flat trailer roof, which is approximately 8' x 14'.


  • mcgivormcgivor Posts: 2,334Solar Expert ✭✭✭✭✭
    So your loads are 3.5Kwh per day without including A/C which would be an additional 6Kwh, correct? These loads are extremely high 3.5Kwh is roughly what a modest offgrid home would use, may we ask what the loads are because reducing loads will be way less expensive than building to accommodate them. Anything 100% offgrid is extremely expensive in comparison to grid power costs, if you have grid I would highly suggest using it, if the idea is to offset some of the costs associated that may be more realistic, then again if money is no object, it's your money who am I to judge. 
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 

  • EstragonEstragon Posts: 2,962Registered Users ✭✭✭✭✭
    Very roughly yes, if pv output is 900w, and loads are 450w, there will be 450w available to charge batteries.  

    There are a number of potential complications though.  For example, the a/c unit likely has a "power factor" of less than one.  Residential billing generally ignores this, but providing your own power means you can't.  If the unit power factor was .7 for example, you need to produce 450w÷.7(pf)÷.85(inverter & wire loss) = ~750w.  Using something like a "kill-a-watt" device to measure loads is a good way to get a better grip on actual load requirements.

    The question about early/late charging depends on factors like panel tilt, season, haze/smoke (or lack of), etc.  Some of these are accounted for by the calculator at: http://pvwatts.nrel.gov
    You may want to try various tilts, azimuths, etc for your location to get a better sense of what to expect in terms of average production.

    For the battery bank math, you also need to consider wiring for a given system voltage.  At 3.5kwh+, 48v is indicated.  To get 48v with 6v batteries, capacity is added in series strings of 8 batteries.  A single string would be [email protected]=10800 watt-hours.  At 50% max, this should cover your 3500wh load for one rainy day.  Keep in mind that the battery starts bearing the load late in day 1, all through day 2, and on day 3 until the sun gets strong enough to supply loads and begin charging.

    The panel limitation just means more generator time vs pv.  Using a tracker would help, but basically it is what it is.
    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
  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    Regarding battery type, I have read that AGM batteries charge 5 times faster than FLA batteries.  Is that true?  If so, how long would it take to charge AGM batteries in the 10800 watt hour bank you described above, assuming that they are discharged 50%?
  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    I have used a kill a watt meter to come up with 3.5 kw for a winter daily load.  It sounds like a 1220 watt solar panel array will only yield 900 watts of energy, and that 900 watts will be further degraded to 765 watts (900 x .85), so I would be approaching only one half of the rated 1220 watts of the solar panel coming out of the batteries via the inverter  actually available for loads.  I would have thought that a properly designed system would not result in about a 50% loss of energy, but I guess this reflects my ignorance on this topic.  :*
  • EstragonEstragon Posts: 2,962Registered Users ✭✭✭✭✭
    edited November 3 #8
    Some AGM batteries are designed to charge faster (assuming, of course, there is current available to support the higher charge rate).  A car battery, for example, is designed to deliver high current relative to its size and capacity, and be recharged quickly.  This is accomplished by, among other things, increasing the surface area for contact between plates and electrolyte mats.  In such a design, there are more, thinner plates closely spaced vs a similar sized deep cycle design.  Although good at what they're designed for, these high current designs tend not to do well with repeated deep cycling.  

    All that said, there are some deep cycle AGMs capable of accepting/delivering higher current rates than flooded.  Not especially helpful in your application though, as the current is limited by space on the roof for panels.

    AGMs also tend to be more efficient than flooded cycling at high (~90%-100% ish) states of charge.  Again, not super helpful in this case, where design cycling will be to lower states of charge where both Flooded and AGM are pretty efficient.

    The inverter efficiency number is about average.  They can get into the 90s if loaded at the ideal part of their efficiency curve, but can be very inefficient, for example, with a largish inverter running a small load like a couple of LED light.

    The 1200 to 900w "loss" isn't really a loss, but a matter of how panels are rated/marketed.  The nameplate wattage is Standard Test Condition rating.  STC spec is with the panel exposed briefly to a light at 1000w/m2 at 25°f ambient temp.  There is often a second spec, NOCT listed for the panel operating after having warmed up sitting in full sun.  NOCT is more indicative of real-world conditions in most locations, and tends to be ~75% of STC.  

    At higher temps, panel current remains roughly the same, but voltage drops off, so overall wattage does too.  In cold, dry (typically high altitude) locations, a panel might regularly output more than STC, but not many of us live in such places.

    IIRC, the kill-a-watt has a pf or kva measuring ability.   Assuming the A/C is 120v, you might want to check this to get a better sense of actual power (VoltAmps) needed.  My number was purely a guess for example purposes.
    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
  • BB.BB. Posts: 28,088Super Moderators, Administrators admin
    edited November 5 #9
    If you are using AGM or Li Ion batteries, you can use 0.61 as the derating for Solar panel to 120 VAC off grid solar (vs ~0.51 end to end with lead acid batteries)... For example:

    0.81 solar panel derating * 0.95 solar charger eff * 0.80 flooded cell lead acid battery eff * 0.85 typical AC inverter eff = 0.52 end to end eff...

    Of course if you only draw energy from "solar panels" (during day, solar power>load power), then you can save the 0.80 losses...

    And these numbers are somewhat conservative... Most people will at least "match" the predicted performance (remembering that solar panel harvest can vary by 10% (to as much as 20%) year over year... Getting more exact estimates is hardly worth the trouble as the variability of hours of sun per day is much greater. And you have the issue of what happens when you some days (or weeks) of bad weather... On a very dark and cloudy day, you can have 5% or less of "average" daily harvest. You have to account for that too (size of battery bank for stored energy, turning off "optional loads", using a backup genset, or simply going without power)...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    My question about the AGM batteries being able to charge 5 times faster than FLA batteries had to do with the necessity of using a generator to supplement the solar panel charging of the batteries.  If it is true that I can get the AGM to 100% SOC quickly, then that would reduce the run time (and gasoline consumption) of the generator. I expect to run a generator on any cloudy days, but I don't think that this would be more than a few weeks of total cloudy weather in southern AZ.  So, a battery bank that can charge quickly is a big consideration for me, but I don't think $1000 lithium batteries are a wise investment for this current trailer experiment. 

    I assume that when my 1220 watts of solar panels are generating only 600 watts of energy available for loads during the day, then I am consuming from the battery approximately what solar is putting in to the battery bank, and I am not losing battery bank SOC. I was hoping that I would have some residual energy going into the battery bank from the solar panels during daylight hours but it appears that this won't happen. 

    The ultimate goal here is to equip the trailer with a solar/generator system that it can operate completely off grid - eventually in a remote location far from utility power.  So, supplemental grid power is not a consideration for me in this effort. My backyard is just a test area for this purpose, so grid power will not enter into the equation.

    I believe that I can reduce the winter (and summer) power consumption by dividing the 8' x 14' trailer in half with an insulated partition wall(with an insulated door) down the center (perpendicular to the 14' dimension) to cut the cooled and heated area down to an area about 8' x 9'.  The remainder of the trailer area would be for storage and a bathroom, which would not require constant hvac control. I expect that this would cut the duty cycle of the A/C (and any supplemental heat) to about 1/3, or about 8 hours per day, probably less. Having the window A/C facing east with the solar panels all along the 14' side all facing south I expect the panels would provide some shading for the trailer roof and the remaining unshaded roof area could be covered with some shade cloth. The trailer walls have 1.5" of styrofoam in the walls and 3.5" of styrofoam in the roof, but the floor is made of 2.5" of  wood (1.5" 2x6, plus 2 layers of 1/2" plywood), but the floor could also be insulated.  The trailer has no windows, but does have 3 roof vents.  Two of the vents would be in the 1/2  of the 8' x 9' hvac conditioned space, but I have used foam plugs in those vents to contain heat or cooling and that does help.  Essentially, I am envisioning a big 8' x 9' x 8' insulated cooler. 

    I believe that I could supplement electric power consumption with some use of propane for heating water and cooking, but I really don't think that those are significant contributors to the 3.5 kw daily consumption I measured.  About 1.5 kw of that number is a 10 cu ft. apartment refrigerator.  A dorm fridge uses about 1 kw, so not much savings in downsizing.  An efficient 12 volt chest type fridge is not out of the question, but at about $1,000 for a good one with a not very good warranty makes it my last choice. It appears that reducing power consumption would get me closer to a viable solar power system.  
  • mike95490mike95490 Posts: 7,970Solar Expert ✭✭✭✭
    tucsonjwt said:
    Regarding battery type, I have read that AGM batteries charge 5 times faster than FLA batteries.  Is that true?  If so, how long would it take to charge AGM batteries in the 10800 watt hour bank you described above, assuming that they are discharged 50%?
    AGM can be CHARGED faster.  Instead of limited to C10 rate, you could possibly double it and charge at C5 - but only if you have the solar to produce that many amps.  That would be double the array size for a FLA battery .

    AGM has to have 105% of the consumed power, returned as recharge.   FLA is less efficient and needs 120% returned to the battery , at a slower rate than AGM can accept
    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 ,

  • EstragonEstragon Posts: 2,962Registered Users ✭✭✭✭✭
    Using an AGM with a somewhat faster charge rate like C/5 could work well if you planned to routinely charge with generator in the morning, and use solar to carry loads and finish charging to the extent current is available.

    Once the bank is ~85% full, charging reaches a target voltage and is held there.  Current slowly tapers as it approaches full.  At C/5 (~2kw in the example), going from 50% to 85-90% might take a couple of hours.  Going from 90 to 100% might take another couple even with generator current available, so better to let solar finish if possible.  It's important to not let L.A. batteries sit at low charge for long, but it's not critical they get all the way to 100% daily.

    If the fridge is the type with condenser coil on the back, you may be able to add insulation.  Many use the metal skin instead for heat transfer though, which shouldn't be covered.  I built a small heavily insulated chest fridge/freezer box which uses ~300w/day as a freezer or ~150w as a fridge.
    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
  • mcgivormcgivor Posts: 2,334Solar Expert ✭✭✭✭✭
    edited November 4 #13
    Personally I'm wondering if the OP is thinking of LFP rather than AGM, with reference to 5 times faster charging, although  LFP' has the  ability to accept more current, there has to be current to accept in the first place, so despite being more efficient than both LA chemistries, the PV would need to be increased to match the batteries.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 

  • EstragonEstragon Posts: 2,962Registered Users ✭✭✭✭✭
    OP did mention lithium, which could work well, but suggested "cost not worth the investment for this trailer experiment".
    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
  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    Being a beginner, I don't think and additional $2,000 to $4,000 investment in LiFePo batteries is a wise choice, because I am likely to do something wrong with the system at some point and damage the batteries.  I would rather damage $1,000 of FLA. :) I have read that FLA batteries are more forgiving of improper charging than AGMs, but AGMs use a two stage charging process rather than a 3 stage process for FLAs.  I am leaning toward cheap FLA golf cart batteries unless AGMs can be charged much more quickly than FLAs, so I would save enough on generator fuel to eventually offset the higher cost of AGMs. The five times faster charge for AGM comes from information elsewhere on the internet, which we all know is always 100% correct. :)

    I am open to replacing the existing 10 cu. ft. apartment fridge with a small chest freezer converted to be a refrigerator with a probe thermometer and still running on AC.  I see videos on YouTube showing a 24 hour power consumption of about 300 watts AC.  My current refrigerator consumes about 1500 watts over a 24 hour period.

    Two hours of generator charging each morning to get to 85% of charge and letting solar provide the remaining 15% would be OK with me. How much input would be required from the generator during that 2 hour period?  Would a 2000 watt generator running on an eco setting of 400 watts be sufficient during this two hour period?  Put another way, at a 50% SOC, what would be the appropriate charging amperage input for FLA vs. AGM, and is the amperage setting of this charging expressed  as 120 volt AC or DC voltage? Comparing  FLA vs. AGM batteries, how much faster could the AGM batteries be charged given a fixed charging amperage rate for both types of batteries. Can the generator be wired through the charge controller to charge the batteries?  Can the controller be programmed to allow a higher charge amperage and a different charging profile for AGM vs. FLA? 




  • EstragonEstragon Posts: 2,962Registered Users ✭✭✭✭✭
    Charging current is generally DC bank capacity / rate.  For example a C/10 rate on a 225ah bank would be 22.5adc.  A C/5 rate would be 45a.  To put 45a into a 48v bank (ignoring losses and charger pf) would require 45x48=2160w.  If the generator was something like a Honda eu2000, it would fault overloaded.

    The generator puts out AC current.  This can either supply AC loads, or be converted to charge batteries via a charger.  A charge controller controls DC power from panels, and (depending on controller type) bucks DC string voltage to battery charging voltage and limits current.  The (AC) generator can't be wired through the (DC) charge controller.

    Most quality charge controllers can be programmed to various battery types and charging regimes
    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
  • littleharbor2littleharbor2 Posts: 1,072Solar Expert ✭✭✭✭
    tucsonjwt  I have read that FLA batteries are more forgiving of improper charging than AGMs, but AGMs use a two stage charging process rather than a 3 stage process for FLAs. 
    Not true, AGM batteries charge the same as FLA batteries. Bulk, Absorb and Float. They have different voltage limitations and generally don't get an Equalization like FLA's. Some do allow a semi equalization charge. This varies from brand to brand.

    2.1 Kw Suntech 175 mono, Classic 200, Trace SW 4024 ( 15 years old  but brand new out of sealed factory box Jan. 2015), Bogart Tri-metric, 700 ah @24 volt AGM battery bank. Plenty of Baja Sea of Cortez sunshine.

  • mike95490mike95490 Posts: 7,970Solar Expert ✭✭✭✭
    To charge off of a generator, you will need an AC power battery charger in the 40-80 amp range, with power factor correction.

    With good power factor and a 90% efficient charger, you could get from a 1800w generator,  90 charging amps @ 14VDC, so smaller charger will allow the auto-throttle to activate.
    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 ,

  • mcgivormcgivor Posts: 2,334Solar Expert ✭✭✭✭✭
    Some battery  chargers have a poor power factor which can limit the amount of charging current to less than expected by doing a simple calculation using generator output  wattage versus input wattage to the battery. Also consider that charging current drops when the absorption stage begins, which makes generator charging less efficient as the battery nears full charge, the most efficient use would be the bulk phase. Charging  at maximum current limitations will stress the batteries more than a moderate current, the same goes for dischargeing, just some things to keep in mind.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 

  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    So far I have learned that the watt rating of a solar panel will be about half that rating after the power travels through inherent panel inefficiency, voltage drop along the wires, power loss at the controller, power loss at the batteries, and power loss at the inverter, and final power delivery to the 120 volt ac outlet. Do I have that correct?




  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    Am I better off with a 12 volt battery bank being charged by the 400 watt eco mode of a 2000 watt generator for many hours during the day? Would the long, slow charge be better for the batteries and less wasteful of generator fuel?
  • mcgivormcgivor Posts: 2,334Solar Expert ✭✭✭✭✭
    edited November 5 #22
    tucsonjwt said:
    Am I better off with a 12 volt battery bank being charged by the 400 watt eco mode of a 2000 watt generator for many hours during the day? Would the long, slow charge be better for the batteries and less wasteful of generator fuel?


    For every transition there are inherent losses, gasoline to rotation loss, alternator loss, transformer 120 AC-14VDC loss, wiring loss, battery efficiency loss, they are laws of physics, no way to avoid them. Charging with a generator is very inefficient considering every dollar spent on fuel ~70% is wasted out it exhaust and lost in cooling. With PV the losses are less but there is an upfront cost, this will however be recovered over time, unlike the lost energy from an internal combustion engine. 

    Designing a ballanced system is, in the long run, the most cost effective because it requires less generator support, underestimated load demand is a pitfall most have experienced in the learning curve. Building a marginal system is a false economy, IMHO. Air conditioning is a challenge in itself, I pondered the idea but the demand was simply too great using 400Ah LA, 1500W PV at 24V nominal and a base load of 1.5 to 2 Kwh  pre day, due to demands on the batteries. With LFP now, the option became available mostly due to battery efficiency, I installed a mini split inverter A/C unit, the most efficient available to me, I'm in the testing phase but due to cool weather I've not needed it, there may be a need to increase PV and or battery capacity, neither of which is no problem.

    As I mentioned in post #4 conservation is cheaper than building around inefficiencies, to answer the 12V question, no use the highest nominal voltage possible if 10Kwh per day are the target loads because you'll need about 30Kwh or more in storage capacity to reasonably support those loads with any margin for reliability and minimal generator use. Naturally the array would need to be sized to adequately charge the batteries, that goes with saying.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 

  • BB.BB. Posts: 28,088Super Moderators, Administrators admin
    Sort of correct.... The wire losses tend to be relatively low (1-3% or so)--We generally suggest extra heavy cabling for off grid solar power systems. Both to reduce losses, and to keep cables+terminations from running hot (typically 1.25x maximum continuous current--I.e. a 15 amp load * 1.25 = 18.75 Amp minimum branch circuit/breaker rating...

    Solar panels are affected by panel temperatures... More or less, Vmp falls by x0.81 when they go from room temperature to typical operating temperatures (you put anything out in bright sunlight, it gets hot). Realistically, solar panels will only output near Vmp*Imp=Pmp at freezing and sub-freezing temperatures (panels are cooled to ~room temperature under full sun).

    Others like Flooded Cell Lead Acid batteries we use 80% average (sort of) worst case efficiency... It does depend on how you charge/discharge them... FLA batteries are very efficient below ~80% state of charge, but approach 0% efficiency as you approach 100% state of charge (charging energy turns into heat and Hydrogen+Oxygen gases).

    Regarding what is most efficient for charging... More or less, gasoline gensets are more efficient when loaded >~50% of rated power...

    So, an 1,800 Watt genset would charge at 12 volt battery bank (as an example) at:
    • 1,800 Watts * 0.50 output * 0.80 typical AC to DC battery charger eff * 1/14.5 volts charging = ~49.7 Amps charging current (into 12 volt battery bank)
    The Honda EU (and other inverter-generators) can be more efficient at less than 50% of rated output because the engine can throttle down and use less fuel (more or less, a gasoline genset will use roughly the same fuel flow between 50% and 0% rated output).

    Diesel gensets do run more efficiently at full and part throttle (the Diesel cycle is more efficient overall, and diesel fuel has more BTU per gallon of fuel). However, Diesel engines tend to not like running at less than ~30% to 60% of rated power (newer diesel may run "better" at lower power levels). At lower power levels, diesels can build up carbon internally, glaze cylinder walls, and wet stack...

    So, the charging current for a battery bank depends both on the type and size of cells, and the size/type of genset... A good place to start is 5% to 13% rate of charge for solar panel charging, with 10%+ rate of charge for full time off grid for typical deep cycle flooded cell lead acid batteries (some mfg. say 10% rate of charge minimum).

    If you charge at >~13% rate of charge, a remote battery temperature is recommend to protect the battery bank (hot batteries need to reduce charging voltage setpoint).

    For gensets, typically 10% to 20% (as high as 25%) rate of charge is OK for FLA batteries... Typically, start the genset in the morning at 50% state of charge and run until you get to ~80% state of charge (if you are having bad weather/heavy loads, end at 80% and run the genset the next morning again)--And let the solar panels recharge the last amount.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • mcgivormcgivor Posts: 2,334Solar Expert ✭✭✭✭✭
    This diagram is for a diesel engine but demonstrates where the energy is lost, on the right is heat recovery which increases useful energy from 36% to 49.6%, gasoline will be slightly less efficient.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 

  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    I am now thinking that I can get my winter usage down to about 1.5-2 kw per day.  So, that leaves running a 450 window air conditioner in the summer as the challenge.  I guess that a/c will consume about an average of 400 watts per hour and a likely 30% duty cycle over 12 hours, or about an additional 1.4 kw per day.  So, given sizing for a 3 kw to 3.5 kw average daily summer demand, and supplementing daily with a generator for a bulk charge, what would be a good size charge controller, inverter/charger, and battery bank for this demand? This still is based on 4 solar panels of 305 watts per panel for 1220 watts total (e.g. Canadian Solar CSK-305MS Monocrystalline).  These panels have 32.7 VMP per panel.
  • mcgivormcgivor Posts: 2,334Solar Expert ✭✭✭✭✭
    What nominal voltage do you intend to use? The controller will be able to support more PV wattage the higher the voltage used, so an important part of the calculation in determining the current rating.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 

  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    I don't know what nominal voltage I should use.  I was looking for some suggestions. Thanks
  • mcgivormcgivor Posts: 2,334Solar Expert ✭✭✭✭✭
    edited November 9 #28
    This is the best time to choose a higher nominal voltage now rather than late when expansion is needed. Higher voltage lowers the ballance of system costs, batteries can be in the desirable series configuration to mention a few reasons, but definitely not 12V for the capacity you are looking at, 48V is best 24V ok, one minor downside of 48V is if a small inverter is needed ~300W they are not as common in that voltage.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 

  • tucsonjwttucsonjwt Posts: 34Registered Users ✭✭
    I am unclear about the charge rate for an AGM battery bank.  Using the example of 4 Full River 12 volt AGM 115 AH batteries wired in series of 2 batteries (24 volts) and 2 parallel strings (2 strings of 24 volts), for a total battery bank of 230 AH at 24 volts, what would be the appropriate charging amps for this battery bank?
  • mcgivormcgivor Posts: 2,334Solar Expert ✭✭✭✭✭
    The charging rate depends on useage, part time vs full time, the manufacturer will have a maximum charging rate. There are figures used as guidelines of 5% to 13% of battery capacity for each respectively, so in the above example 230 Ah × 13% = 29.9 A,  so 30A at 24VDC would be a starting point for full time offgrid.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 

  • BB.BB. Posts: 28,088Super Moderators, Administrators admin
    From our host's site (NAWS Northern Arizona Wind & Sun):

    https://www.solar-electric.com/lib/wind-sun/Fullriver_battery_charging_instructions.pdf

    Take the batteries C/20 (20 hour capacity), and multiply by 0.15 to 0.35 for rate of charge:
    • i.e. 115 AH per string * 0.20 rate of charge (nominal) = 23 amps per string nominal charging current during bulk
    • 2x parallel strings * 23 amps = 46 amps nominal charging current.
    Once the battery reaches 2.45 volts per cell, hold charging voltage until battery charging current drops below ~0.012 to 0.02 C (approximately 8 hours maximum charging on absorb).
    • 2.45 Volts per cell absorb * 12 cells in series for a 24 volt battery bank = 29.4 volts absorb stage (at ~77F/25C)
    Hold float charging voltage for a maximum of 8 hours:
    • 2.275 volts per cell float * 12 cells = 27.3 Volts float (8 hours max)
    By the way, I suggest that a single string of batteries is better. And 2-3 parallel strings maximum... Do you have any issues about using a single string of 6 volt @ 220 to 250 Amp*hours?

    https://www.solar-electric.com/search/?q=fullriver

    The one advantage that people have stated--If you have two parallel strings and a "dead battery"--You can take out the one dead battery/string and run on the other string (1/2 capacity) until you get the replacement battery.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
«13
Sign In or Register to comment.