Building a mid size system

mountainman · April 2020

@BB. I have read about your 3.3 kwh per day systems. But I can't seem to find it anywhere.
Im considering building one.
On 24 volts with GC or possibly l16 batts.
Any help would be appreciated. Thanks.

BB. · April 2020

For other new readers here--It is not My System... It is a process that helps do a paper design/scaling for a system that can run a very efficient cabin/small home (lights, fridge, well pump, washer, TV, computer, etc.) off grid. I highly suggest doing a few paper designs and figure out if A) it will provide the energy you need, and 2) making a list of equipment and see if it meets your $$$ point (solar is expensive). And does give some incentive to bo back and look at more conservation to reduce loads and keep the overall system costs down.

I will be a bit more wordy here--To explain "why" certain assumptions/fudge factors where used. Note that most fudge factors here are a bit on the "conservative side"--Because many depend on exactly how the components are used and the system is operated (AC inverters have "less good" efficiency when lightly loaded, FLA batteries are very efficient when operated between 50-80% state of charge, but much less efficient when cycled in the 80-90-100% State of Charge. The basic efficiency factors I use here are:

Solar panels = 81% derating (hot panels in warm/hot weather drop output power from "marketing numbers")
AC inverter = 85% efficient (yes, they can be upwards of 95% efficient--But if you conserve energy and run with lower loads, efficiency does fall)
Flooded Cell Lead Acid battery = 80% efficiency (can easily be >90% efficiency, but depends on cycle depth and age)
Solar Charge Controller + panel deratings = 77% MPPT controllers can be 95% efficient. Solar Panels can be >81% of rating... 0.95*0.81=0.77 derating. PWM controllers work "differently", but a well designed PWM system is about the same "derating" as a well designed MPPT system in warm/hot climates. So, use same 77% for both as starting point.
Base Loads (such as fridge and lighting that must run every day) = 50% to 65% of "predicted" solar output. The "Sun" and weather is highly variable, and so are personal loads--Assuming base loads are 50-65% of "predicted" average output means less genset usage and still can use your "base" or "critical" loads without too much heartburn.
Battery Storage = Plan for 2 days of storage... Not only to "ride through" bad weather, but FLA batteries are not the "fastest/most efficient" charging storage chemistry. If you cycle every day and want deeply cycle at times (bad weather, more loads) and may have some high surge devices (well pump, refrigerator compressor), this "up sizing" of the battery bank (plus the 50% planned discharge) helps ensure that your system will supply loads and surge current through normal operation (state of charge, age, etc.).
Battery max depth of charge = 50% -- I use this number as a planned maximum depth of discharge. Battery will have longer cycle life and charge back to >75% state of charge more quickly (less sulfation/longer life). Yes, you can discharge a FLA deep cycle battery to 20% when needed (emergency, genset breakdown, low on fuel, etc.). And as long as you recharge quickly (in hours or next day), sulfation will not take much life from your battery bank. Battery is there to be used... And if you look at the cycle life curves... You see that a 50% discharge lasts (for example 1,200 Cycles) and a 25% discharge will last ~2,400 cycles... A 2x larger / more expensive bank lasts (very roughly) 2x as long... So cost per stored AH/kWH is about the same.
Battery bank AH capacity = 800 AH max... For many reasons (bus current, size of copper wire, available solar charge controller Amp output, length of wiring)--I suggest using ~800 AH as the "suggested maximum" break point between changing to a higher battery bus voltage. I.e., 12 volts @ 800 AH vs 24 volts @ 400 AH --- Both configurations store the exact same amount of energy but the 24 volt bus is 1/2 the current of the same power at 12 volts (Power=Voltage*Current). Typically a "small system" is 12 volts. A medium size system works well at 24 volts. A larger system typically 48 VDC. But also look at the AC inverters... Some functions/options are generally more available in the higher voltage/larger AC inverters (more $$$, more features).
All numbers above and below are "estimates" for solar... In most cases anything within 10% (or better) is pretty much "exact" (i.e., 100 AH vs 110 AH battery bank--Those are the "same").
AC inverters: Suggest PSW/TSW (Pure/True Sinewave) over MSW (Modified Square/Sine Wave) inverters... MSW inverters are "hard" on induction motors (like well pumps and refrigeration compressors). And can also be hard on SOME AC Power Supplies/Loads (can reduce life of electronics, cause to run hot, cause mechanical noise).
Using "golf cart" batteries for your first battery bank can save you lots of money (cheap, pretty rugged/forgiving).
Paralleling Batteries = I like to suggest 1-3 parallel strings as "optimum". If you need >3 parallel strings, finding larger AH batteries (instead of 6 volt 200 AH, 2 volt @ 600 AH cells, etc.) can be helpful/better choices (avoid "large/heavy" batteries unless you have a way to unload 1,000-2,000+ lbs batteries to where they will be installed.

So, the above help get the paper design "over the target". Your needs/equipment may dictate other choices... But the Rules of thumbs above are good starting points and suggest to double check reasons for "violating" them.

Here is the process based on (I hope) Mountainman's energy needs... As always, feel free to change my assumptions to better match your actual location/etc.

3.3 kWH per day, a full size energy star refrigerator (~1 to 1.5 kWH per day), Leaving another 1.5 to 2 kWH per day for the "rest of the home". And will assume that during bad weather, "optional loads" will be reduced/not used, and/or a genset started (when/if needed) during poor weather/high load needs. This is "mid size" solar power system.

First, design the battery bank. Assume 2 days of stored energy (for a couple days of stormy weather, etc.), and 50% maximum discharge. This seems to be a good "optimum" for a Flooded Cell Lead Acid battery bank. At some point, many folks are thinking about AGM or LiFePO4 type banks--Which can change the FLA model (more expensive batteries such as AGM or Li Ion, then different assumptions can reduce battery AH size and costs).

3,300 WH per day * 1/0.85 AC inverter eff * 2 days storage * 1/0.50 max discharge * 1/24 Volt battery bus = 647 AH @ 24 volt battery bank (or 48 volts @ 324 AH would be OK too)

Next, two design checkpoints. First, make sure have enough charging current for the battery bank. And second, to size the solar array for the loads/hours of sun of the location.

First, the charging current. 5% to 10% to 13% to 20% suggested. 5% can work for an emergency/summer/backup FLA off grid system. 10%+ suggested for full time off grid. 13%+ is nice (less/no genset usage, panels are "cheap" and batteries are "expensive"--Keep battery happy). Rate of charge is xx% * 20 hour discharge rate (100 AH * 0.05 rate of charge = 5 amps).

647 AH * 29.0 volts charging * 1/0.77 panel+charge controller derating * 0.05 rate of charge = 1,218 Watt array minimum
647 AH * 29.0 volts charging * 1/0.77 panel+charge controller derating * 0.10 rate of charge = 2,427 Watt array nominal
647 AH * 29.0 volts charging * 1/0.77 panel+charge controller derating * 0.13 rate of charge = 3,168 Watt array "typical cost effective maximum"

And there is sizing the array based on your loads. Florence South Carolina, fixed array, facing south, array best "average harvest" over the year:
http://www.solarelectricityhandbook.com/solar-irradiance.html

Florence
Average Solar Insolation figures

Measured in kWh/m2/day onto a solar panel set at a 56° angle from vertical:
(For best year-round performance)

Jan	Feb	Mar	Apr	May	Jun
3.59	3.96	4.84	5.46	5.36	5.19
Jul	Aug	Sep	Oct	Nov	Dec
5.12	4.96	4.83	4.82	4.08	3.67

If you use a genset, then you can drop the bottom three months and use (typically February) as your "break even" month. For a sunny region like this, 3.67 Hours of sun ain't bad... Typically I suggest that for areas with 3-2 hours of sun per day during "winter" (or less), it is a choice between more panels (for winter) or more fuel for the genset...

o 3,300 WH per day * 1/0.52 offgrid FLA battery with AC inverter = 1,729 Watt array "December "break even")

Note: FLA batteries "last longer" if hit with a "nominal" charging current like 10-13% all day long, vs a (for example) 20% peak at noon. Also, FLA batteries need longer absorb times the deeper they are discharged (2 hours absorb for shallow cycling, 6+ hours for 50% or deeper cycling).

Using a 2 axis tracking array can extend the time of harvest (earlier in morning, later in evening) to give you bank "more time on charge" during the day. And an alternative to tracking is to split your array... 1/2 facing (roughly) South West and the other 1/2 facing South East. Saves the costs and mechanical complexity of a "real" tracking array. (does require more panels than a tracking array--But panels are relatively cheap these days).

And since the whole reason for having off grid solar power is to consume that power... Some estimates of what such a system can support.

Assuming charging during day and loads at night--All power supplied by battery bank (worst case FLA bank--Other chemistries can supply higher surge current):

o 647 AH * 24 volts * 0.85 AC inverter eff * 1/20 hour discharge rate = 560 Watt load, 5 hours a night, 2x nights to 50% max discharge
o 647 AH * 24 volts * 0.85 AC inverter eff * 1/8 hour discharge rate = 1,650 Watt load, Less than 4 hours to 50% discharge
o 647 AH * 24 volts * 0.85 AC inverter eff * 1/5 hour discharge rate = 2,6430 Watt load, For minutes to an hour (heavy, short term loading)
o 647 AH * 24 volts * 0.85 AC inverter eff * 1/2.5 hour discharge rate = 5,280 Watt load starting surge (well pump, etc.)

From the above... You can justify an AC inverter from 1,200 to 1,500 Watts minimum (suggested to start/run refrigerator+other loads). To a maximum of (5,280 Watt Surge load / 2 = ) 2,640 Watt AC inverter (most quality inverters can support 2x surge load).

For solar charge controllers... PWM + smaller solar panels can be cost effective for smaller systems (say 400 Watt or smaller arrays very close to charge controller/battery bank/battery shed). For larger systems MPPT controllers offer more features and easier/cheaper wiring--Especially when longer Array to Charge controller distances are needed.

I will stop here---Obviously no hardware has been picked/suggested at this point... Look at the overall design and see if it will supply the power your need.

-Bill

BB. · April 2020

One "mistake above"... I assumed December was your "worst sun" month... But January was a bit less... I am not going to bother to edit everything above to correct--Close enough for our discussion. But I should have used January 3.59 Hours of sun per day...

Just to add a bit of detail... Choice between 1,729 Watt array (December "break even"--May need genset at times) to 2,427 Watt array (base on 10% rate of charge)... If you choose 2,427 Watt array (not bad choice for full time off grid), then your harvests look like this:

2,427 Watt array (10% rate of charge) * 0.52 off grid system eff * 3.67 hours of sun per day (average December) = 4,632 Watt*Hours per day
2.427 Watt array * 0.52 off grid system eff * 4.8 hours of "minimum average "summer" sun = 6,058 WH per day "summer" typical

And if much of your load is during the day (refrigerator as an example), you save the "80% battery charging losses" by going from array to charge controller to AC inverter--So even a bit more "free energy".

Another example of "how you use the system" affects the "fudge factors/efficiency" numbers.

-Bill

mountainman · April 2020

Thanks @BB. For the detailed information response.
I've been at this location for 10 years and it is usually very sunny.
Being in the middle of a 300+ acre field which is open to south sun is a plus also.
Snow is almost non existent. In January daytime extreme lows are 30f.
For the 2500 watt array would it be better to use a series parellel combination of 10-60 cell 250 watt panels or say 8 320 watt 72 cell watt panels? For 24 volt charging.
So basically for Batteries. 12- 2 volt 650ah
Or 8- 6 volt l16 330 ah.
With GC batteries it would take 12.
4 series And 3 in parellel. (Not the best option.)
What size charge controller would be needed?

BB. · April 2020

You are very welcome MountainMan. I am continuing with the editorial here. For you and and others that follow. We have not had a single "3.3 kWH per day" system design with all of the details.

Before we get too far... 3.3 kWatt hours per day or ~100 kWH per month is my suggested as a minimum size system that gives you a "near normal" modern electrical life (fridge, lights, washing machine, solar friendly water pumping, laptop, TV, cell phone, etc.). Albeit with lots of conservation and watching others not to exceed your daily energy budget.

Any comments from others here (I have not heard a lot from you folks that are truly off grid about the 3.3 kWH per day system--Is it "viable" or not for a conservation minded lifestyle?)?

Are you OK with this system energy budget/capabilities?

Next, the battery bank... You have it nailed. Besides the issue of series/parallel wiring (more connections, more batteries/cells that can go bad over time)--And the positive of "cheaper" Golf Cart batteries for the first months to 3 years or so (as always, hope for the best, plan for the worse)... That is 3p x 4s GC batteries or 36 cells to check water levels monthly. Whereas a single string of ~6xx AH 2 volt batteries is only 12 cells to check. The L16 batteries have had a good track record here (from others).

One thing to watch for... There are some "2 volt" L16 batteries (Trojan) that are actually 3 cells in parallel for a 2 volt overall voltage. As far as batteries go, not an issue. However, that does give you 3 caps (cells) per 2 volt "cell" to check.

And for wiring parallel battery banks, if you do end up with series/parallel, this website has useful wiring configurations to help ensure the batteries share charging/discharging current:

http://www.smartgauge.co.uk/batt_con.html

About the solar panels... Either would be fine. HOWEVER, if you ever go to a 48 volt battery bus, the 72 cell (and larger) panels with Vmp ~ 36 volts and higher gets into a bit of a dilemma. For typical larger MPPT charge controllers, their maximum input voltage is around 145 to 150 VDC. And in very cold climates that works out to a Vmp-array of ~72 volts to 100 VDC. You really want >72 volts Vmp-array to run an MPPT controller in its "MPPT Range", and you want higher Varray voltage to keep current and wire gauge size down.

With Vmp of ~36 volts, that leaves 3x 36 volts = 72 volts--right on the bottom edge of suggested Varray, and 3x is 108 volts, which runs into Voc-array-cold--So you have to check the minimum temperature/maximum Vinput for the MPPT charge controller.

with Vmp~30 volts... You can run 3x 30 volt panels in series for Vmp-array~90 volts, which will gives you a bit higher Varray operating voltage and headroom for very cold climates.

Midnite solar has the very nice Classic controller which uses a nice Array Configuration tool that allows us to compare panel configuration against panel specs against ambient temperatures. Also, if Vinput is exceeded, it will shut down and "be safe" up to Vmax+Vbatt (such as 150 VDC + 24 VDC battery bus = 174 volts, or 150 VDC + 48 VDC = 198 volts).

Note, there are "high voltage" controllers available too these days... where Vmp-input can run upwards of 400+ VDC (and Vmax~600 VDC). Really nice for larger arrays and long wiring runs, but also not cheap.

Here is an example (you need to plug in your panel/array data, I am just using generic numbers here for 325 Watt panels).

https://www.solar-electric.com/rec-solar-rec325np-monocrystalline-panel.html

8x 325 Watt panels
Vmp: 34.4 Volts
Voc: 40.7 Volts
Imp: 9.46 Amps
Isc: 10.28 Amps

Using Midnites' Classic string sizing tool:

http://www.midnitesolar.com/sizingTool/index.php

Using the 2s x 4p configuration (good for best price--Array close to home/battery shed/charge controller with Vin max 150 VDC):

Using above specs, 20F minimum and 100F maximum ambient temperatures.

The results look good for a Classic 150 -- And you are pretty much right at the limit (cannot add more panels to this unit, unless you go with a 48 volt battery bank). And if you ever did go higher bus voltage, you can put 3x in series (assuming 20F minimum temperature).

About the only "change" is the number of parallel array connections. Each one needs its own series protection fuse or breaker (when 3 or more parallel strings, you need series protection fuses). Note: Fuses are nice because you can turn off each string (for debugging) under load. Touch Safe fuse holders cannot be used to turn off operating current under load (they can arc and catch fire if flipped open but not removed).

Roughly, the charger will output to your battery bank (derated panel output in summer, probably will not see any higher except in cold/clear weather around solar noon):

8x 325 Watt array * 0.77 panel+controller derating * 1/29.0 volts charging = 69 Amps derated charging current
69 Amp charging / ~647 AH = 0.107 = 10.7% rate of charge (good for full time off grid)

The other array 10x 250 Watt panels would be fine too... 2s x 5p works fine. 3s x 4p, erc. configurations would work too.

Generally, for MPPT charge controllers, roughly 2x battery bus voltage (i.e., 2x 29 volts = 58 volts) is where the controllers are "most efficient". However, higher voltages make sense for wiring (higher voltage, lower current, smaller AWG wiring) and for long wire runs from array to charge controller (again, higher array voltage, lower current, lower voltage drop).

MPPT charge controllers can safely and accurately to the Controller's maximum output current rating (over paneling is fine). I suggest that 1/0.77 is a good "cost effective" max for over panelling (1/0.77= 1.3x larger array than controller rating).

With the 8x 325 Watt array, any MPPT charge controller > 70 Amps would be fine. Do note that output current is affected by battery bus voltage--So you will get >70 amps when the battery is discharged/heavier DC Bus loads at lower Vbus voltage). Example 25 volt bus:

8x 325 Watt array * 0.77 panel derating * 1/25 volt battery bus = 80 Amps (when battery bus voltage is "low")

-Bill

petertearai · April 2020

Ive got 400 ah battery at 24 volts 2200 watts of panels . Outback fm80 and 2kw fxr inverter .
It work's seamlessly . Off grid . 2 fridge freezers , lights tv microwave , toaster , vacuum , washing machine . Cooking gas , hot water gass , heating wood fire . Temperate climate , so only fans in summer. Water gravity feed. Big header tank so only need to pump water during day, and then only if sunny
So yes very doable . Regrets..... not going 48 volts and slightly bigger inverter .however it works well .

Photowhit · April 2020

I'm going to add in a couple 'pulls' from what Bill has said;

Fudge factor based on other things like system use ...
and
January is actually your worst month...

It's important to look at how you will use your system, In South Carolina, you may find you want to run an air conditioner in the summer and have limits to what your system can provide, so summer might be your worst months since your need will be greater.

For others in more northern latitudes, you may find Winter much more harsh in having limited charging hours. If this is taken into account during the design stages, you may find angling your array somewhat higher will help you catch more sun in the winter when the sun is lower on the horizon.

Also while you are considering a 24 volt system, and I understand your fears working with higher voltages. Most of us who have lower voltage system of this size regret not having gone with a higher 48 volt system. You might still consider it, Your array voltage will still need to be higher for the most part above the 48 volt number. With a 3000 watt array you will need 2 charge controllers vs just 1 for a 48 volt system.

Inverters for 24 vs 48 volt systems will tend to cost about the same, particularly for UL house rated inverter (UL 1741) so there would be significant savings. Though it should be looked at carefully as some designs are difficult. Particularly in colder climates, configuring an array with 72 cell panels can be hard to find a sweet spot with strings of 2 to low a voltage for proper charging and strings of 3 too high a VOC for many charge controllers. Strings of 3 - 60 cell panels usually works out fine. 3-72 cell panels is usually fine in warmer climates.

mountainman · April 2020

Thanks @Photowhit and @petertearai.
I've researched 100 year record night lows for my area 9f in 1918 and 11f in 2018
Doable with bills generic 325 panels 3 in series without maxing voc at 103 volt vmp. If I go with a 48 system
Wouldn't 58 charging x2 require 116 vmp?
I understand that without loads sizing a system is next to impossible.

Is there there someone in my area who uses A 14000 btu lg inverter ac.
That I could get a estimate for what my daily usage might be.

I can see advantages Going to 48 volts
Quality Controllers are expensive. (1vs2)
And the capacity for a larger array because here at 34 latt. July temps often exceed 100. So ac is a must.
And if I were to use the 8 l16 batteries no parellel connection with 48 vs 24 with 2p 2s

So far we've covered everything except inverters. First 1 that would run 24/7
For Refrigerator. 1500-2000 watts???
What of when i want to use the microwave
Or my s/o turns on the hair dryer
When the fridge is in defrost mode?

#2 What size larger inverter would be needed
Only on for running a solar friendly water pump to a gravity tank.
#3 would I need an additional inverter for summer for a 14000 btu mini split?

mike95490 · April 2020

I run a 6kw inverter, mainly because of a water pump (1/2 hp) 2 fridges and a freezer. It's idle power is about 30W, and my household loads range from 200 - 350W unless the pump is on. So, some of the large-enough-to-do-it-all inverters have pretty low idle losses, and you don't have to worry much about load management.

mountainman · April 2020

30 watts is low. my 300 takes 8 watts.
For some reason I can't see signatures.
Which inverter do you use?

mike95490 · April 2020

posting screen shot of data sheet. Sometimes, it's worth paying for the big stuff. I dithered for weeks, 2 small inverters or one honker. finally dug into specs and decided big boy gear is cheaper & simpler in long run. And solved minor imbalance worries, many inverters deal poorly with L1 - L2 imbalance

BB. · April 2020

Regarding seeing signatures, as far as I know, users should see signatures when on a "desktop" computer. And on a mobile computer, signatures are suppressed.

On a mobile computer (at least my Android devices running FireFox browser), in the upper right corner are three stacked dots. Click on that, and 1/2 way down the menu, click on "request desktop site" to see the "desktop" theme.

THEN hit REFRESH.

-Bill

mountainman · April 2020

I used to be able to see SIG's on my Android
But Not anymore.

BB. · April 2020

Still works for me, but I forgot to add "hit refresh" after clicking on request desktop...

The "new" forum software (installed a few years ago), does not show signatures on Mobile themes. The old VBB forum, you could configure to see or not see signatures.

-Bill

Dave Angelini · April 2020

Maybe he needs an Android update?
The question I ask my clients is how long are they going to do this? If it is a long time then an Outback or Schneider system all on the same network so one can see graphs and not just numbers. I mainly use the one below since 2005. 3 different models over the years and only in 48vdc. The same smart design went in to cool it and keep it alive. There is no air across the components like most. It is cooled by a tunnel that goes up the middle. A grant from the US military for design and initial XW build. Must be 100's of thousands of these by now.

Image: https://us.v-cdn.net/6024911/uploads/editor/tc/v78d31aco4lq.png

petertearai · April 2020

Indeed selecting desk top shows signature . I tried it , but on the phone, i don't like it ,. Not used to it i suppose .so back to mobile version .

mountainman · April 2020

Seems that there are diy small systems such as mine.
And then there's big systems
1kwh per day cost me $1.15 per watt.
Im sure going to 48 volts and a 5kwh will be more than 5x the cost.
@mike95490 For a large system as yours what is the approximate cost per watt.

mountainman · April 2020

For a 5 kwh per day useage. Approximately how much would the System cost?
Anyone.

BB. · April 2020

What size battery bank? What size solar array? What size AC inverter? Do you want an AC inverter or AC Inverter-Charger? What backup (genset/charger)? Any vendors you like or not (Schneider, Midnite, Outback, Magnum, etc. or EBay vendors?)?

The "quality" choice---Easily 2:1 or greater price difference. FLA, AGM, Li Ion battery chemistry? If FLA, cells or "Fork Lift" type battery bank?

What sort of battery life are you looking for? 3-5 year? 15+ year?

Going with Roof/Ground/Mfg. Rack/wood frame DIY/tracking array mounting?

How much do you want to pay up front for system?

Do you want a highly networked system with remote/Internet control/monitoring?

Self install or consulting/turnkey install?

Even if purchasing from our Host in Arizona) is not practical for you--They are not a bad place to start your equipment search (and pricing--Note, I believe that NAWS uses list price for website, and then offers discounted prices on some things if you create an account an log in). Always include shipping and insurance charges... Large/bulky items like high wattage solar panels can be expensive to ship (sometimes the almost the same price for shipping 1 panel vs a whole pallet by truck--And shipping small quantities can exceed the cost of the panel itself).

Then go through the same pricing through EBay/Other Vendor(s) and see what the difference is.

80% of the system price will probably be the major components (panels, charge controller, inverter, battery bank, possibly solar panel mounting hardware, backup genset system costs). The pricing part should not take long--It is all the other issues (as above) that take more time, and can drive system costs higher or lower -- Depending on those choices.

Use NAWS pricing for a good quality system as your baseline... Then figure out what you are willing to adjusg "specific requirements" to better meet your needs and cost point.

Places like NAWS--You can call them up, give them your requirements, and they can create a complete kit and give you the costs.

Or, you can go with an installer and use their experience to design/cost/install (as desired) for your system (and their labor/overhead).

Not trying to be snarky here... but you can get a large spread depending on those choices.

Decade ago, GT Solar was $10 per Watt installed. Today it is probably in the range of $2.50 to $5.00 per Watt. Add a battery bank cost--And that should give you a decent "floor" for an installed system cost (GT Solar is/was high volume business--Off Grid solar, lower volume, higher complexity, generally out "in the middle of nowhere").

Costco Battery cost: 6 volts * 200 AH = 1,200 WH.
$90 per battery / 1,200 WH - $0.075 per WH (3-5 year life)

Rolls / Surrette Battery L16 6 volt @ 390 AH $317 each; 6v * 390 AH = 2,340 WH
$317 / 2,340 WH = $0.135 per WH (8+ year life?)

Concorde L16: 6 volts * 405 AH = 2,430 WH
$673 / 2,430 WH = $0.277 per WH (5-7 year life?)

Almost a 4:1 ratio in costs just for "Lead Acid" batteries (prices for Rolls & Concorde are list prices from NAWS--I did not research shipping costs)...

Solar panels around $0.50 to $1.00 per Watt (no shipping cost included)

5,000 Watts * $1.00 per Watt = $5,000

A 980 WH @ 24 volt or 490 AH @ 48 volts = 23,520 WH battery bank (just scaled from 3.3 kW to 5.0 kW system)
23,520 WH * $0.135 per WH = $3,175 for FLA "Rolls" battery bank

Of course, the Rolls L16 does not "fit" into a 490 AH @ 48volt string--So there will be some "adjustments" different battery voltage/AH, or other vendors, etc.... But just for argument... The cost of the solar array and the battery bank are "on the same scale"... Pricing for both can be cut by 1/2 by shopping (different brand/model at the same website) and/or looking to less expensive brands and sources...

-Bill

littleharbor2 · April 2020

The Rolls S6 L16-SC is a 487 ah. battery.
https://rollsbattery.com/battery/s6-l16-sc/?pdf=8465

Estragon · April 2020

In post #8 you asked about charging 48v nominal at ~ 58v. Not sure where the 2x =116v comes from. You'd want string Vmp when hot to be low-mid 60s volts minimum. 103vmp would be fine. My strings run mid 90s hot and charge the 48v system fine.

BB. · April 2020

The x2 is from the old rule of thumb "set Vmp-array to ~2x Vbatt" voltage for "optimum charging" point (lower and higher Varray voltages tend to be "slightly less" efficient than the ~2x "optimum").

But for Vbatt~48 volt bus (~59 volts charging) with Vmp-array>~108 volts or so, you start having to worry about very low ambient temperatures in winter (well below 0F?) and Voc-cold exceeding the ~145 to 150 Vpanel max for the higher end MPPT charge controllers.

-Bill

mountainman · April 2020

Guess I was thinking charging voltage needed to be double for mppt. But I understand now vmp
needs to be ~ 1.5 charge voltage.

BB. · April 2020

Well, for "proper" MPPT function, you need something like 1.3x Vbatt minimum... But since Vmp-array falls when panels get hot, it is better to be higher. If you run your Vmp-array-hot close to Vbatt+a couple of volts for battery drop, the MPPT controller behaves much more like a PWM controller (no "MPPT" "help" for higher output power from higher voltage array).

This all based on "derating" panels because the cell Vmp/Imp/Voc/Isc/etc. are rated at a cell temperature of ~25C/75F (Lab/Marketing test conditions). And we all know that in warm weather, on a sunny day, the panels/cells are >>>25C/75F in temperature.

A bit more detail about PTC/Temperature deratings (PTC assumes ~20C increase due to panels under full sun).

https://newenglandcleanenergy.com/energymiser/2015/12/01/stc-vs-ptc-why-solar-panel-testing-matters/

And why Solar Panels + Controller pairs are a "pain"... Matching solar panel IV curves vs local temperature min/max, and Vpanel range for XYZ solar controller (and PWM vs MPPT)--The matching is not something to be taken lightly.

-Bill

mountainman · April 2020

I can get these 250 watt locally for $60
I was hoping to run a medium 1500- 2000 watt array. 150 volt cc.
with 2 panels in series and 8 GC 200 AH on a 24 volt system.
for washer a 18 cuft refrigerator
(340 wh per year)
micro tv dvd and 4 12 watt led lights.
With the hopes that later I could rewire the panels to 3 in series.
And rewire batteries for a 48 volt system.
And as for the travel trailer a 12 volt dc system.
I would use 2 panels 500 w with 2 GC 200 ah batteries.

mountainman · April 2020

I ran the panel voc numbers and 3 in series is good to-99f. Using 0 f as my low temperature
So with solar as every thing else you get what you pay for.
you can cheap out and get a troublesome system that doesn't last very long.
Or buy quality for long term use.

Dave Angelini · April 2020

A major indication in solar panel quality is the product warranty period and cost is the other. Really high quality panels like sunpower, LG, panasonic and one other have 25 year product and power warranties. It is an indication of redundancy built-in and that cost money. To me it is just a much lower chance that I will have to help a client troubleshoot an array. The product warranty also tells that wind, snow and both have been well thought out.

Horsefly · April 2020

I'm a little late posting in this thread, but I do have one data point for what might be considered a medium sized system.

3 years ago we installed a system at our cabin, designed for 3.3kWH / day. We were pretty careful to document all our material costs right down to the PVC conduit elbows and paint for the equipment room. We had no labor costs, as my brother-in-law and I did all the work. The total cost was $9,754.

In case the signature doesn't show up, the main components we used were:

6 x 280W Canadian Solar mono panels
Schneider MPPT60-150 charge controller
Schneider SW4024 Inverter / Charger
Schneider SCP control panel
4 VMax XTR12-155 batteries (design was for no extra days of autonomy - long story)
Lots of wire - Array to charge controller was just under 200 ft

@mountainman - I can provide more detail if you need it.

softdown · April 2020

Been focused on refrigeration as of late. This is my current recommendation for off grid applications:
Refrigerator - has no freezer area. There are some available. Fridges have 1" of insulation and that doesn't work very well for areas kept at 0F.

I have been playing with the conversion of an upright commercial freezer to a refrigerator with an external (after market) temperature controller that keeps it at 40F. I bought a simple temp controller this time since the last electrical wonder gave me the heebie jeebies before I gave up on it.

Chest freezer kept in the coolest area possible. Cold air falls out when doors open - not a problem with a chest.

The refrigerator/freezer is the power monster with off grid. Just due to battery wear and tear we can normally look forward to spending ~ $350/year or $1/day for cool and cold food.

mountainman · April 2020

@Horsefly yes if you would elaborate on what he powers with the system. And why would he not want reserve?

Dave Angelini · April 2020

I believe Steve has a part-time cabin. Quite a bit different than a full time home where you need a reserve or like to run a generator.

Building a mid size system

Comments

Florence
Average Solar Insolation figures

Categories

Building a mid size system

Comments

FlorenceAverage Solar Insolation figures

Categories

Florence
Average Solar Insolation figures