Solar system design help for remote research station on the Equator

mactadpolemactadpole Registered Users Posts: 2


First post here. I must start by saying we are a group of biologists/ecologists who have been able to figure out how to get a very remote research site functioning with clean water, composting toilets, and a minimal hydropower system. But we are not electricians. I have been doing a lot of research as we need to increase our power availability. I really need some help to figure out the details. We are located on the coast of Ecuador basically at 0 degrees latitude. In the dry season we have a lot of overcast days but it is still bright and during the wet season it may rain for a day or two and then be full sun for a day or two or several.

Being an NGO I have managed to obtain a donation of a few components to begin building our system. To start, here is what we have been given:

AIMS Power SCC60AMPPT, 60 Amp Solar Charge Controller 12/24/36/48V

AIMS Power 1500 Watt 24 Volt Pure Sine Power Inverter

2 - bayite DC 6.5-100V 0-100A LCD Display Digital Current Voltage Power Energy Meter

Based on my research I believe we should build a 24V system. We want to be able to power some lights, wifi router, charge a couple laptops (1 or 2 at a time), USB devices (phones/tablets), charge a mix of batteries (AAA, AA, cameras, drone, etc.), and run a 24V fridge/freezer (eventually). We plan to purchase the rest of our items in Ecuador or go to Colombia to get it. I’ve found a company here where we can get most everything. Here is my current hardware plan with eventual upgrades in parentheses. We will not have the 24V fridge/freezer to start.

2 - 350W/24V/9.04A pollicrystaline solar panels (upgrade to 4 total)

4 - 150Ah/12V SBB Gel batteries (upgrade to 6 or 8 total)

(eventually add another 24V 1500-2000 Watt inverter)

My plan was to wire our first two panels in series for [email protected] (700W) and then eventually (or if we have enough $ initially) add two more panels in series and parallel with first set for [email protected] (1400W). With the 4 batteries (maybe 6 to start) the plan was to wire in series/parallel to create 24V/300Ah battery bank (24V 450Ah w/6). Eventually add pairs of batteries to increase battery bank size 150Ah per pair. My plan was to also hook up a battery charger on a timer from our 120V/300 Watt hydro generator to charge the battery bank at night if needed. I would have another timer on the hydro system to run a water heating element in a barrel during the day. It would be nice to at least have some warmer water in the shower.

Hopefully I haven’t committed any gross negligence in my research and design of this system so far. I know you will let me know if I did! But I sincerely appreciate any and all help/recommendations.

Thank You,



  • mike95490mike95490 Solar Expert Posts: 8,067 ✭✭✭✭✭
    My system design parameters indicate that if you start to parallel batteries, you need to instead, wire them in series for a higher voltage system.   If your plans will eventually include a fridge, a 48V system is better.

    Batteries.  Gel batteries are a no-no for solar power.  They are designed for slow recharge, and with solar, you do not have time.
    If you want sealed batteries, use AGM batteries.  If you want cheap batteries, flooded 6v Golf Cart (GC2) batteries are great
    in-between, you can sometimes get sealed golf cart batteries (case size = GC2 )   8 cart batteries in series will give you 48V at about 200ah, or 4800watt hours (that's using 50% of the bank capacity)
    Flooded batteries survive parallel operation better than AGM batteries will, if you require a 24V system for the fridge (is it avaib in 48V?)

    48V bank will need about 80 - 100V from the PV array or hydro generator.  

    Does your hydro run 24-7 ?  daily downtime for de-gunking intake screens ?  What's its output ?

    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

    gen: ,

  • mcgivormcgivor Solar Expert Posts: 2,500 ✭✭✭✭✭
    Your ultimate plan is approaching what a normal small offgrid home system may look like, the difficulty in building piece by piece is the adding of batteries, which results in mismatched batteries in the end. As the loads are currently minimal, it may be better to calculate all loads, inverter self consumption  and how long they are expected to run without being too conservative to establish an Ah baseline, factor in efficiency losses etcetera then buy a batteries to suit current loads and consider them sacrificial with regards to the final system. This would be the learning system, many loose the first set of batteries in the learning curve, so keeping potential losses to a minimum is a good thing, once ready to expand recalculate all loads again to establish the battery requiments then the PV to keep them happy, it always begins with loads to build a ballenced system. Just my thoughts and opinions.

    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 
    900W  3 × 300W No name brand Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah FLA 24V nominal as a backup system. 
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergencies and welding.
  • EstragonEstragon Registered Users Posts: 3,127 ✭✭✭✭✭
    Does the hydro still produce in the dry season? If so, at what rate? Do you have a generator?

    Can you supply a link for the 24v fridge/freezer you're considering?

    Pv will still produce with lightly overcast conditions, but at a reduced rate. According to Equador gets a fairly steady 4+ hours of full sun equivalent on average throughout the year, but averages hide things such as extended overcast periods like you describe for the dry season. The concern would be reduced output for a week or two wouldn't cover loads and charging, resulting in a chronic deficit and premature battery problems.

    4 x 350w panels might output ~1kw in full sun and warmish weather, so 4000ish watt-hours. A fridge freezer might take say 1500ish watt-hours. Assuming other loads take about the same, total loads would be about 3000 watt-hours, so in this example, there should be enough sun to recharge daily on average. If, however, you had a run of overcast with say 50% output, only 2000 watt-hours produced minus 3000 watt-hours loads equals a 1000wh daily deficit. If continued for more than a few days, the result will be dead batteries. If hydro can't make up the deficit in the dry season, a solution might be either more panels to take advantage of what light there is, or a generator (if quality fuel is reasonably available).

    As Mcgivor noted, getting a fairly accurate sense of loads (total watt-hours usage daily) is important in getting a system that works well at reasonable cost. My numbers above are guesswork to get started with.
    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
  • Dave AngeliniDave Angelini Solar Expert Posts: 4,198 ✭✭✭✭✭
    If you are serious about this, how will you monitor the system? Below are examples from Schneider Electric and Outback power systems. They can be monitored anywhere in the world and work well if you use equipment all from their brand. Good Luck!

    "we go where power lines don't" Sierra Mountains near Mariposa/Yosemite CA
    E-mail [email protected]

  • BB.BB. Super Moderators, Administrators Posts: 28,315 admin

    If you could go a step up from Aims to another brand of charge controller and inverter--Would probably be a good idea. If you have a lot of little crawling things (bug/animals), put screening around the electronics (electronics' vents) to keep out the bugs from your systems.

    Adding a refrigerator is what usually makes an off grid power system go from small (1,000 WH per day) to a medium size system (~3,300 WH per day).

    If you have the option, see if you can find an "inverter" refrigerator... These types have a VFD (variable frequency drive) that greatly reduces the starting current (from ~500-600 Watts for a typical 120/240 VAC fridge compressor to less than 150 Watts). Note that frost free refrigerators do have heater inside to melt the ice from condenser that take ~500-600 Watts--So you still need a good size inverter to run them. Using a chest freezer with a thermostat that can run a refrigerator temperatures can save quite a bit of energy overall (typical refrigerator is ~1,000 to 2,000 WH per day, a chest freezer "refrigerator" can be down to 250 WH per day--Warm climate may be higher). Chest refrigerators are a pain, water collects on sides/bottom/need bottom drain, And using baskets that you have to move around to get at food at the bottom...

    Any way, some suggestions. For your system, you need (at least) a 1,200-1,500 Watt AC inverter to run the fridge minimum. Do not "over size" the inverter (inverters take energy just turned on--roughly 6 watts for a small one, large ones can take 20-40 Watts or more... 40 watts * 24 hours per day ~ 960 Watt*Hours per day--Almost as much as the fridge (if you leave inverter running 24 hours per day).

    When mounting the array, put a 5 degree tilt from vertical in it... Will help to "self clean" of dust and leaves when it rains.

    Need to figure out how much sun/location, but taking a guess:

    Average Solar Insolation figures

    Measured in kWh/m2/day onto a solar panel set at a 90° angle:
    (For best year-round performance)


    Sounds like you have a good amount of "Marine Layer"--So "not great" amount of bright sunny days. Using a rules of thumb design for your place might look like this:

    Battery bank... 2 days of storage, 50% maximum discharge, 3,300 WH (3.3 kWH per day) of average usage (basically an energy efficient full size refrigerator (1-2 kWH per day), LED lighting, laptop computer (0.5-1.0 kWH per day), some small water pumping, etc.:
    • 3,300 WH per day * 1/0.85 AC inverter eff * 2 days of storage * 1/0.50 maximum discharge (longer battery life) * 1/24 volt battery bank = 647 AH @ 24 volt battery bank
    For this size system, 24 volts is a good fit (roughly, if your battery bank is >800 AH, go to the next higher voltage... 12 volts @ 800 AH = 24 volts @ 400 AH...). For this size battery bank a 2,000 to 3,235 Watt inverter is about the "maximum" suggested (500 Watts per 100 AH @ 24 volts...). And a maximum of ~ 3,235 Watt array typical).

    The sizing of the solar array... Two calculations. One based on battery bank capacity and rate of charge 5% to 13% is typical for solar. 5% for weekend/seasonal usage. 10%+ is better for full time use/off grid operation.
    • 647 AH * 29 volts charging * 1/0.77 panel+controller deratings * 0.05 rate of charge = 1,218 Watt array minimum
    • 647 AH * 29 volts charging * 1/0.77 panel+controller deratings * 0.10 rate of charge = 2,437 Watt array nominal
    • 647 AH * 29 volts charging * 1/0.77 panel+controller deratings * 0.13 rate of charge = 3,168 Watt array "cost effective" maximum
    And then there is usage based on amount of sun... You don't plan on using 100% of predicted power per day (some days less sun, use less power, or run genset), or plan on using 65-75% of predicted output for base load. Away from the equator, we calculate on winter sun+genset use--Usually gives "lots of extra" energy in summer. In the equator, you have similar sun year round... You have to decide how much solar array you need for "comfortable" operation (you don't want to drive yourself nuts every day managing loads+battery capacity+genset--Especially if you are not always there--Others will not be as diligent):
    • 3,300 WH per day * 1/0.52 off grid ac system eff * 3.80 Hours of sun per day (minimum for several months) = 1,670 Watt array minimum (break even months, some genset usage)
    • 1,670 Watt array * 1/0.65 "base load" usage = 2,569 Watt array for "less genset" / management issues with 3.3 kWH per day base load
    So, I would be suggesting an array of 2,437-2,569 Watts to 3,168 Watts would be "nice" for your setting. Of course, the larger array gives you some extra power for electric hot water (or just a solar thermal collector--They are much cheaper than solar electric water heating--but more plumbing issues).

    Your thoughts?

    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • mactadpolemactadpole Registered Users Posts: 2
    Thanks so much for the feedback thus far! Given me a lot to think about. I provide responses in order of replies.

    mike95490 The fridge/freezers available here are only available in 12/24V. I think we're pretty much locked into a 24 volt system at this point. We are a non-profit in a developing country with limited funds and since we were given the AIMS equipment that is part of the reason we are going that route for better or worse. Seems to be a lot of debate on AGM vs. Gel in the solar setting. In the end, the gels are much cheaper here and if we take care of them they should last quite a while it seems. They hydro runs 24/7 at 120 volt/300 Watt. We actually don't have much problem with it getting clogged since we created an intake system with a grate and screening that overflows most of the time, and has a 10.5cm tube that feeds a reservoir we built to maintain constant flow in the dry season. 

    mcgivorEstragon My initial calculations were between 1500-2000 WH per day without fridge. Man, I hate to think about sacrificing batteries at their cost here. Seems it would just be wiser to invest initially in more batteries and panels, and to begin by hooking the inverter up to the CC to be able to set the low voltage cutoff high in order to not discharge them to low. My thinking was to set it at a voltage equivalent to a maximum of 30% discharge. We always have 2-3 people on site and we have been living off just the hydrosystem for a while so we can monitor the solar system closely at the beginning. We have been able to maintain the hydrosystem working throughout the dry season and we limit our max watts out of it at 250 watts. We can actually charge quite a few AA batteries (~8) and 2-3 USB devices plus led lights at one time with no problem on it. We do have a gas generator that we can run to top up batteries if necessary. Here is a link to the fridge we can get here - Shocked to find it on Amazon! It's specs are 72 [email protected] and an average daily consumption of 550WH. 

    Dave Angelini We always have 2-3 people on site who will be able to monitor the system. We will be setting up a long-range (~8km) wifi system but could never afford components from either of these manufacturers. The import tax alone would be insane.

    BB. Unfortunately we just don't have the $ to invest in other components to replace the AIMS stuff. It is brand new and was brought in from the US by a donor. Excellent suggestion on screening the electronics. And good suggestion on the 5 degree tilt, easy enough. We are actually pretty close to Manta so here are our average figures:
    Better than Ibarra! I can probably convince our ExecDir to splurge now for 4 panels and 6 batteries up front. Especially with all the advice on here. I had calculated for 1 day autonomy on solar alone with the hope that charging off the hydro system at night would top things up sufficiently. We'll be heating a tank of water using an immersion heater during the day from they hydro.

    From my research it is typically not advised to wire the inverter to the DC load of the CC but I like the idea of controlling battery discharge that way to begin with. Once I'm convinced we're not going to toast the batteries I will wire the inverter directly to batteries.

    Well, is there anything else I should be worrying about? I know more panels and batteries is recommended across the board but we just don't have the money for more than the max I proposed. I know it's also going to cost us no less than about $1500 for fuses, wiring, etc.

    Thanks again,


  • BB.BB. Super Moderators, Administrators Posts: 28,315 admin
    In general, wire the batteries to a common Bus Point. This is a good starting point for understanding wiring issues:

    Once you have the common bus point, generally, you run the + lead(s) to each major load/source (AC inverter, charge controller, etc.). Each + connection should have a fuse/breaker (rated for load/wire size--Remember breakers protect wiring, not your loads--more or less).

    Generally, AC inverters take a lot more current than the "load connections" on a typical solar charge controller can supply/manage. Almost always, the DC input for the inverter should be wired to the battery bus directly (through a DC breaker on the + lead from the bus).

    For lead acid batteries, it actually is better to charge/discharge (on average) 25% per day... It is difficult to have enough hours of sun in a day to recharge a lead acid battery bank from 50% to 100% state of charge in one solar day (it takes 8-10+ hours, roughly, to recharge a lead acid battery fully)... Mostly you have from 9-3 or so of "full sun". If you use 50% for a day, then have a few days to "recover/recharge", that can work.

    I did not include the Hydro power--But if you have stable/reliable hydro power, that can help a whole bunch (average power/voltage/current, how many hours per day, seasonal?).

    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • mcgivormcgivor Solar Expert Posts: 2,500 ✭✭✭✭✭
    @mactadpole said
    From my research it is typically not advised to wire the inverter to the DC load of the CC but I like the idea of controlling battery discharge that way to begin with. Once I'm convinced we're not going to toast the batteries I will wire the inverter directly to batteries.

    The inverter has a low battery cut out, LBCO, but it is fixed at 19V which means the battery is pretty much dead, not a good situation, the charge controller on the other hand has an adjustable setpoint for LBCO for the load terminals of 21-25V. What I would do is use the charge controller to drive a relay, 24V DC coil, from the load terminals with a normally closed, NC, contact, one side of the contact connected to ground with a  series resistor of 5000 ohms. The other side would be connected to the live leg of the inverter downstream of a residual current circuit breaker, RCBO, or ground fault circuit  interrupter, GFCI. This would allow the charge controller  load output to drop the relay when in low voltage  creating  a ground leakage greater than the trip value of the ground leakage device, much the same way as the test button, thereby disconnecting all loads downstream of the inverter's output. It would however not shut down the inverter itself, that would rely on its own LBCO should the fault go unnoticed. Once battery voltage is normal a manual reset of the ground leakage device would be required, not perfect but rather a MacGyver solution.

    As the rating of the load terminals, they are rated at 60A, DC loads could be connected directly if low enough of a current but be aware, with regards to the DC refrigerator, make sure it can accept ~29V which is needed due to charging voltages, if not a buck converter, or voltage regulator, of some description will be required. 
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 
    900W  3 × 300W No name brand Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah FLA 24V nominal as a backup system. 
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergencies and welding.
  • mcgivormcgivor Solar Expert Posts: 2,500 ✭✭✭✭✭
    edited July 2018 #10
    Obviously what you are referring to is hystersis, or a delay in transition, it is difficult to accomplish when programming is unavailable, the setpoints would  need to be massaged somewhat over time to achieve approximately the same results, got lemons make lemonade  
    Edit.  This was a response to a comment made by @MrM1 which disapeared or is being edited 
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery bank 
    900W  3 × 300W No name brand Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah FLA 24V nominal as a backup system. 
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergencies and welding.
  • BB.BB. Super Moderators, Administrators Posts: 28,315 admin

    It looks like MrM1 did pull down his post for some reason (if accidental, I can restore if you wish MrM1).

    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • MrM1MrM1 Registered Users Posts: 311 ✭✭✭
    edited July 2018 #12
    No ... McGivor beat me to the punch with a much better answer.  I deleted it as he was adding his comment to me.  If what I had to say made any sense or added anything to the discussion, it would be fine to put it back.  (or I can).  Often in my vague attempts to Help here,  I am also trying to learn, by putting my musings out there to be fact checked by the group.
    REC TwinPeak 2 285W 3S-3P 2.6kW-STC / 1.9kW-NMOT Array / MN Solar Classic 150 / 2017 Conext SW 4024 Inverter latest firmware / OB PSX-240 Autotransfomer for load balancing / Trojan L16H-AC 435Ah bank 4S connected to Inverter with 7' of 4/0 cable / 24 volt system / Grid-Assist or Backup Solar Generator System Powering 3200Whs Daily / System went Online Oct 2018 / System, Pics and Discussion
  • BB.BB. Super Moderators, Administrators Posts: 28,315 admin
    No problem MrM1.

    Have fun,
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
Sign In or Register to comment.