Help: Lots of A/C and not enough panels/batteries?

kojmania
kojmania Registered Users Posts: 15
BB. wrote: »
This is sort of how I do it--Say you are near Tulsa OK, using flooded cell batteries and an inverter (0.52 derating). The controller is either a well designed PWM system or MPPT with fixed solar panels pointing south at latitude:



Say that you would like to use solar panels only for ~9 months of the year, and for the 3 months of winter you will use a backup genset. That gives us the 4th month as January and 64 kWH per month usable AC power (out the inverter outlet) per 1,000 watts of solar panels:
  • 64 kWhr per month * 1/30 days = 2.1 kWHrs or 2,100 Watt*Hours per day base load
  • 2,100 WH / 12 volts = 175 AH @ 12 volts
You may have seasonal loads--Let's say that 6 months of the year you need fans/small AC unit--roughly 80 kWhrs per month per 1,000 watts of panels:
  • (80kWH-64kWH) * 1/30 days = 0.533 KWH or 533 WH per day extra power (summer)
If you get a small Sanyo mini-split AC system, set to low power:
  • 533 WH excess per summer day / 300 watts Sanyo on low = 1.77 hours per 1kW of solar panels
Anyway--I just tried something different to recognize that many places have different winter/summer numbers... Tulsa seems to have reasonably consistent power over the year (on average)... So, you don't get a huge winter/summer difference as somebody who lives else where.

I try not to get too generic in laying out a solution--it can create lots of confusion... Generally, it works better if you describe your needs and we try to work out possible answers based on your requirements.

And remember that these are all estimates based on ~20 year weather history--Some places have very consistent weather and other have highly variable weather from year to year--So any estimate we work out while generally be really good if we get within 10% of your actual power available/usage.

I try to be a bit on the conservative side when making estimates so as not to surprise you (in a bad way) when you have to run the genset much more than your original plans.

-Bill

Hi Bill,

I have to say, you've been very very thorough with your explanation. My dad and I were recently trying to do a move to solar energy in Ghana, and while I am no expert, he knows I love renewable energy, and we were trying to avoid the high cost of hiring a consultant in these parts, so I did some basic snooping around to figure out how much we'd need, and I'm realizing from your posts that I grossly underestimated the system. The bad news is he's already spent a few grand getting a system made up of 24*2V 500AH batteries, an inverter, PWM controller, and 8*250W panels. My understanding was that you just look at how much gets consumed over a 24hr period, divide that by the number of hrs of sunlight, which we estimated to be about 10, and then u know what size panels to get. We're consuming abt 20KW a day because we have a small business and use a handful of AC's. This past week has been hell, feeling like the system was inadequate. Now I'm realizing a handful of things:

1. The wattage produced from the panels is in DC, and we had failed to factor in the derate value (0.52) for our off-grid system, so we actually only generate abt 1.04 KW AC for every 2KW DC power produced

2. Based on the PV Watts website, looking at this date in the past, there are realistically only about 5 hours in the day when the panels even generate over 1KW AC in an hr, so even if we did have 10hrs of brightest possible sunlight, we'd only generate abt 13KW AC tops!

3. Our battery capacity per your explanations, would be 4,608 KWH when fully charged, and according to one post above, we should be trying to not use it at more than half its capacity, so realistically, we'd wanna use abt 2,304 KWH per day.

Based on just this alone, (and please tell me if I'm off in these calculations) I'm realizing that to even power one air conditioner that consumes 2000 W, we'd use all our battery power within an hr, maximum 2hrs, and have more or less no energy left to power our building after that!

Needless to say, he's not too pleased, my pops, but I havent yet mentioned all this as Im just stumbling on it now. Can you advise if I'm making a mistake, and what would you think is the best way to maximize use of our system now? I know our month with the least amt of sunlight is June, and even then, we get 129 KWH for the month (per the PV Watt site again).

HELP!!
«1

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    Kojmania,

    I have moved your post from KWH to AH Battery to its own thread so we can discuss your needs/questions. Makes it a bit less confusing regarding the original questions from the other poster.

    Yep, you pretty much have it down with the back of the envelope calculations. Of course, there are many more details to designing/installing the system (and the added costs)...

    Batteries are the "heart" of your off-grid system design... If you don't do that correctly, you will find that the batteries will not have a very long life (months or a couple of years vs 5-10 years--depending on quality of batteries too).

    Also, many of things you asked about are "rules of thumb"--basically good design practices that get you "close" to the answers without a lot more detailed questions and designs.

    For example, 50% discharge is the maximum we use use to give you good battery life... Many people here aim for around 25% maximum discharge (75% state of charge) as the batteries will last even longer--Yes, the batteries will probably last a bit more than 2x longer, but they also purchased 2x the amount of batteries to do that... So, on a cost basis, the two systems will cost almost the same amount of money to run (one system may run 4 years on a bank of batteries, and the other 8+ years and a bank twice as large).

    I guess, what I am coming down to is--What is it you need from your system?

    Off-Grid solar PV power is very expensive--In the US where we have (relatively) inexpensive solar hardware and power--An off-grid system will cost around ~10x the cost of utility power. So, on this forum, we recommend that folks work very hard on conservation first (insulation, double pane windows, energy efficient appliances and lighting, and also look at energy efficient Air Conditioning units too...

    The reason we look at conservation first is because it almost always costs less to conserve a kWH than to generator a kWH... And, for people living in hot climate with Air Conditioning, conservation has a second advantage. You are generating less heat in the building, so you have to spend less on Air Conditioning to move the heat outside (100 watt filament lamp replaced with 23 watt CFL--compact florescent lamp--aka "twisty bulb"). So, now instead of heating the room with a 100 watt reading lamp, you are only heating the room with a 23 watt lamp--And that saves 73 watts that the A/C would have had to remove (roughly, the AC would use 1/2 of 73 watts or ~36 watts to move that heat outside).

    For many people, computers and other electronics are actually quite heavy power consumers. Looking for more efficient computers, refrigerators, and such (lap tops, new monitors, turning off laser printers when not needed, setting computers to "sleep" when not being used, etc.) all helps.

    So, stoves, motors, lamps, etc. that put heat into the occupied space should be efficient and/or even moved/vented directly outside to prevent heat buildup in the building.

    Next, while "renewable energy" is a nice concept--it still requires a lot of money and resources (those heavy battery banks) to function. So, don't feel like conservation is not a worthwhile goal--In my humble opinion, conservation is "greener" than Solar Power.

    Once you have addressed conservation--next you need to define the needs of your alternative power system...

    Is it to save money, have power during power failures, avoid fuel (and fuel transportation costs) or something else...

    Many people have a need for reliable power in a region where power blackouts are common events during the day/evening. It is not unusual to setup an AC battery charger + battery bank + inverter to be the equivalent of a whole house/building UPS.

    If this is your need--Setting up such a system then adding solar panels to help charge the battery bank (and perhaps reduce generator usage) is frequently a good compromise. You only power the "critical" loads with the battery bank (keep the home/business operating) and you add solar panels as your money/needs dictate.

    Anyway--I will stop here for now. As you have seen, most people tend to overestimate the amount of power a solar PV system can supply and underestimate your loads.

    Assuming you have 220 VAC 50/60 Hz power--Perhaps you can obtain a Kill-a-Watt meter (220 VAC version here) and start measuring some of your loads and figure out the conservation angle first.

    Once you figure out your loads and what the system needs to supply, then we can work out on paper what your design may look like.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Bill,

    I totally agree with you on the conservation being more of an importance. I'll be sure to start looking for more ways to do that here. We did start a few practices such as installing rain gutters on all the roofs and collecting the water for things like watering the plants, washing dust off feet etc. We also installed the double pane tinted windows, which are becoming more common here. Every room has shades as well, to try and prevent too much sunlight from heating the room, and we already use CFL's and low-consumption ceiling fans. I think the concrete used throughout the building is one of the reasons a lot of heat gets trapped indoors, which is why we started using AC's so much.

    This past weekend, we took the air conditioners off the system altogether, and they are now running on the national grid. We've kept lights, fans, and the 9 or so installed sockets on solar power. It seems to be working quite well thus far.

    I guess our overarching goal was to reduce how much money we spend on electricity in a month, by getting as much of our equipment on the solar power as possible. Our second goal was to not have to rely on the national grid because power-cuts are frequent, and the fuel for the generator is even more expensive than paying for the grid.

    Looking at how much power is generated through our 2kw (DC) panels, and it looks like we are currently operating at 1.4KW (AC) per hour, we're probably relying on the batteries a bit already. But now I'm trying to figure out if I'm calculating in the right way, the amount of time it takes to charge our batteries fully (there doesnt seem to be a way to check the level so i have to settle for estimates), and then how much time it'll take to discharge (if we were running solely off batteries). This will help to estimate how to use power so we don't keep draining the batteries. How would you approach this part of our complex equation?

    And thank you, I'm feeling much more comfortable talking to someone who gets this.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Help: Lots of A/C and not enough panels/batteries?

    battery state of charge can be determined with 'at rest voltage' readings and if your battery is an fla then specific gravity readings. we've had many threads on that. the rough power used/generated may be monitored with a battery monitor for a rough idea on the soc as well once it is calibrated correctly.
    it is a big undertaking to go solar, but rewarding too when you see outages and you then have power without noise. i'm waiting for them to invent a renewable wallet to help and i'm sure you and many others can relate to that.:p
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    Take a read through a couple of Battery FAQs:

    Deep Cycle Battery FAQ
    www.batteryfaq.org

    On average, you need to budget about 80% efficiency for battery charging/discharging and 85% efficiency for the inverter losses.

    So, if you use 1.4 kWH of AC power 5 hours per day, then, roughly wattage wise, you would need:
    • (1.4 kWH * 5 hours per day) / (0.80*0.85) = 10.3 kWH of DC power to charge...
    Unfortinatuley, each of the small losses in your system adds up to large losses when all is taken into account...

    Roughly, you will lose 1/2 of the power (Solar Panel STC Ratings to Inverter AC output) for an off-grid system... You may do a little batter, or you may do a little worse (lots of variables)--but I would not expect that you will do better (conservative estimates--fewer surprises).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Again Bill,

    Thanks for the battery resources. Read through some of it, and now think I may have found a good solution to our problem. We're running all the lighting, fans, and most of the electrical equipment on the solar power, and the AC's are on the national grid. But recently, 2 nights last week, the power cut out at midnight. I suspect the batteries are not fully charged. But there's no indicator for the battery charge level, and unitl I buy one, I'm trying to estimate how long it takes to charge. Can you tell me if I'm doing it right:

    Using 1,400 W AC from 8am to 8pm, 12hrs= 16, 800 W AC during the day.

    At night, 6 lights, say, 100 W if all equipment is off. 100W x 12hrs = 1200 W AC.

    Total daily use = 18,000 W AC = 11,448 W DC.

    Panels produce 7,059 W AC per day (4490 W DC), batteries supply the rest, which is 10,941 W AC (or 6958 W DC) per the PV Watts site for this day in 1978.

    Battery capacity: 48V x 12000 Ah= 576,000 Wh DC (since there are 24 batteries, each is 2V, 500 Ah)

    Safe usage value: 50% of 576,000 = 288,000 Wh DC

    To charge battery: 576,000 Wh/4490W = 128 days

    Is this correct? Could it really take 128 days to fully charge all 24 batteries? I know this is approximate based on solar radiation in the past, but I want to make sure I'm looking at this right. Thanks in advance.

    -Kojmania
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    Kojmania,
    kojmania wrote: »
    Using 1,400 W AC from 8am to 8pm, 12hrs= 16, 800 W AC during the day.

    At night, 6 lights, say, 100 W if all equipment is off. 100W x 12hrs = 1200 W AC.

    Total daily use = 18,000 W AC = 11,448 W DC.
    Depending on where you measure the power... The inverter is ~85% efficient and flooded cell batteries are somewhere around 80% efficient.
    • 18,000 Watts AC / (0.85*0.80 efficiency for DC conversion/storage) = 26,471 Watt*Hours DC from charging sources
    So, you actually need more power/energy on the DC side to run your AC appliances--not less.
    Panels produce 7,059 W AC per day (4490 W DC), batteries supply the rest, which is 10,941 W AC (or 6958 W DC) per the PV Watts site for this day in 1978.
    PV Watts is an average--It is easy for you to get +/- 10-20% or more differences from predicted vs your specific day (average weather vs your weather that day).
    Battery capacity: 48V x 12000 Ah= 576,000 Wh DC (since there are 24 batteries, each is 2V, 500 Ah)

    Safe usage value: 50% of 576,000 = 288,000 Wh DC
    With Series connected batteries, the voltage adds, but the AH rating remains the same. With Parallel connected batteries, the voltage remains the same and the AH add..

    In your case, you have:
    • 48 volts * 500 AH =24,000 DC Watt*Hours in your battery bank.
    • 24,000 DC WH * 0.50 max discharge = 12,000 DC Watt*Hours recommended max discharge
    To charge battery: 576,000 Wh/4490W = 128 days
    • 24,000 WH battery bank / 4,490 WH per day from panels = 5.3 "full sun days" to recharge a 100% discharged bank
    I am not sure of the 4,490 WH from panels is coming from--just using your numbers.

    It would not hurt to recharge your bank with solar+grid power---Especially
    right now because your batteries are so severely depleted.

    Also, just in general, be real careful about what loads you choose to run from your solar/battery system. Batteries are expensive and need to be replaced every 3-8 years (+/-). And it is real easy to kill a bank within months (or faster) if not properly charged/maintained and/or is over discharged.

    Solar off grid power is very expensive (probably at least as expensive as running a genset for power)... So, conserve its resources only for important loads.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?
    So, you actually need more power/energy on the DC side to run your AC appliances--not less.

    I suspected this, but wasn't sure what i was doing wrong, so thanks for clarifying.
    With Series connected batteries, the voltage adds, but the AH rating remains the same. With Parallel connected batteries, the voltage remains the same and the AH add..

    I didn't realize this. When would one use parallel connected batteries in a solar setup?
    It would not hurt to recharge your bank with solar+grid power---Especially
    right now because your batteries are so severely depleted.

    Unfortunately, we don't have that capability right now. (Unless every grid does and I don't realize it). But I hear the government is trying to pass a policy soon that will allow us to start putting the systems on the grid, and be able to sell back power. So fingers crossed that it happens soon.
    Also, just in general, be real careful about what loads you choose to run from your solar/battery system. Batteries are expensive and need to be replaced every 3-8 years (+/-). And it is real easy to kill a bank within months (or faster) if not properly charged/maintained and/or is over discharged.

    For the time being, we don't really have any appliances with motors. It's predominantly lights, fans, TV, and on occasion someone will plug in a kettle for a few mins, but then thats it. We've tried with the most important things running, and the inverter shows that its putting out a max of about 4 amps usually (@ 220V, which is 880W AC), and it is able to provide this consistently for a whole day. What I should probably look into next is getting a few more panels so that our system is not as dependent on the batteries as it currently is to make up the shortage in what we're using versus what the panels are providing.

    -Kojmania
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    Kojmania,
    kojmania wrote: »
    I didn't realize this. When would one use parallel connected batteries in a solar setup?
    As an example... You have a 12 volt truck battery and need more storage capacity--You would connect a second battery in parallel with the first (+ to +; - to -). This gives you more Amp*Hours but keeps the voltage the same.

    Adding batteries in series is done to get the correct operating voltage. But you cannot add more capacity to a battery bank without adding more strings in parallel.

    Remember that Power = Voltage * Current

    So, adding too batteries in series (which doubles voltage) or two batteries in parallel (which doubles current / Amp*Hour capacity) still doubles the available power/energy storage (2xP=2xV*Current=V*2xI note: * and x both mean times--just trying to clarify the relationship between voltage, current, and power).
    Unfortunately, we don't have that capability right now. (Unless every grid does and I don't realize it). But I hear the government is trying to pass a policy soon that will allow us to start putting the systems on the grid, and be able to sell back power. So fingers crossed that it happens soon.
    This may be a bit of confusion over terms... I am typing about connecting an AC battery charger to your utility power and recharge your battery bank from the "grid" (utility power). Or, connect the battery charger to your genset.
    For the time being, we don't really have any appliances with motors. It's predominantly lights, fans, TV, and on occasion someone will plug in a kettle for a few mins, but then thats it. We've tried with the most important things running, and the inverter shows that its putting out a max of about 4 amps usually (@ 220V, which is 880W AC), and it is able to provide this consistently for a whole day. What I should probably look into next is getting a few more panels so that our system is not as dependent on the batteries as it currently is to make up the shortage in what we're using versus what the panels are providing.

    Power usage is power usage... Does not matter if electronics/computers or air conditioner motor.

    Things like electric kettles and other heating devices would really be better to use a fuel powered stove (propane, natural gas, petrol, diesel, kerosene, etc. -- whatever is handy and "safe" for your home/business) instead of solar power.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Bill,
    As an example... You have a 12 volt truck battery and need more storage capacity--You would connect a second battery in parallel with the first (+ to +; - to -). This gives you more Amp*Hours but keeps the voltage the same.

    Adding batteries in series is done to get the correct operating voltage. But you cannot add more capacity to a battery bank without adding more strings in parallel.

    Remember that Power = Voltage * Current

    So, adding too batteries in series (which doubles voltage) or two batteries in parallel (which doubles current / Amp*Hour capacity) still doubles the available power/energy storage (2xP=2xV*Current=V*2xI note: * and x both mean times--just trying to clarify the relationship between voltage, current, and power).

    This makes a lot of sense, and I think I am getting it. So let's say we want to power our off-site office with our even smaller system (2*100W panels with 2 batteries-each is 12V, 100 AH), and we do a few calculations and realize that to power 2 laptops, 1 standing fan and a handful of lights, we'll be overusing the system batteries. Would it be theoretically possible to buy more 12V batteries locally and then connect them in parallel? Would the batteries have to have the same ratings, or could we use, say a 2V 500 AH battery with the already existing 12V 100 AH ones? Or is this not a good way to increase the system's capacity?
    This may be a bit of confusion over terms... I am typing about connecting an AC battery charger to your utility power and recharge your battery bank from the "grid" (utility power). Or, connect the battery charger to your genset.

    You mean, like these?: http://www.cclcomponents.com/range.asp?ID=45
    Things like electric kettles and other heating devices would really be better to use a fuel powered stove (propane, natural gas, petrol, diesel, kerosene, etc. -- whatever is handy and "safe" for your home/business) instead of solar power.

    Noted. Thanks.

    -Kojmania
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?
    kojmania wrote: »
    This makes a lot of sense, and I think I am getting it. So let's say we want to power our off-site office with our even smaller system (2*100W panels with 2 batteries-each is 12V, 100 AH), and we do a few calculations and realize that to power 2 laptops, 1 standing fan and a handful of lights, we'll be overusing the system batteries. Would it be theoretically possible to buy more 12V batteries locally and then connect them in parallel? Would the batteries have to have the same ratings, or could we use, say a 2V 500 AH battery with the already existing 12V 100 AH ones? Or is this not a good way to increase the system's capacity?
    Yes, you can do it--It is not a great thing to do--but it does work.

    First, batteries are actually very sensitive to voltage levels... For a 12 volt battery, 0.1 volt difference is (very roughly) a 10% difference in charge level. Charging a 12 volt battery at 13.8 volts will take a long time to recharge. Charging at ~14.4 volts will recharge quickly. Recharging at 15 volts will eventually damage the battery if done all the time. "Storing" a battery at ~13.7 volts on a charger (Float) will keep a battery charged without overcharge/uncharging damage.

    Adding small batteries (different brand/models/types) in parallel together can be an issue... Different batteries have slightly different voltages--can charge/discharge differently (some will share/carry more load, others will carry less).

    In the end, using a few large Amp*Hour (capacity) batteries in parallel (or even one string) is usually better than placing a bunch of small capacity batteries in parallel (fewer cells to check water levels on, fewer parallel electrical connections where batteries do not share well, some battery failures such as a shorted cell can discharge the batteries around them, etc.).

    Yes. For long term charging--You want battery chargers that do ~14.4 volts (or ~58-59 volts for 48 volt bank) fast charging, and float around ~13.7 volts (~54.8 volts) for long term standby (setting on AC charge).

    You are very welcome.

    By the way, I am very impressed with your English--It is better than mine (and English is my only language :blush:;)).

    Some links for you to read:

    Battery Connections
    www.batteryfaq.org
    Deep Cycle Battery FAQ

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Bill,
    By the way, I am very impressed with your English--It is better than mine (and English is my only language ).

    Haha. Thanks a lot. Don't know if my primary school English teachers would agree, but I appreciate that!
    Adding small batteries (different brand/models/types) in parallel together can be an issue... Different batteries have slightly different voltages--can charge/discharge differently (some will share/carry more load, others will carry less).

    Ok, so we've decided to actually try and get more panels for both systems. We did an audit of the daytime and nighttime usage and realized that in fact, the battery capacities are enough in both systems to last longer than the whole night. Thus, if we can provide more power in the daytime, we ensure that the system can stay on all day, and there will be sufficient power left over to charge the current battery sizes (if the sun works with us, of course). But now we just want to make sure the charge controllers and inverters can handle the new loads:

    The smaller system has a charge controller rating of 20A, 48V and a 600W inverter, 2*12V, 100AH batteries, and 2*100W mono solar panels. We're thinking to add 4*100W panels to make a total of 6*100W.

    The larger system has a charge controller of 48V, 50A, a 5000W inverter, 24*2V, 500AH batteries and 8*250W mono solar panels. Here, we're thinking to add 4*250W solar panels.

    From previous estimations I did, i believe the systems should work fine with these upgrades, but would you agree?
    Some links for you to read:

    Battery Connections
    www.batteryfaq.org
    Deep Cycle Battery FAQ

    Thanks a lot. These sites were very helpful and I've already referred to them a few times.

    Cheers,

    -Kojmania
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    kojmania,

    Not to avoid doing any work... But could you show your math here? I.e.;

    XXX watts of load * Y hours = XXX*Y Watt*Hours of load

    Remembering that you have loads and inverter losses (80-85% efficient), etc.

    And for your panels:

    AAAA Watts of panels * B Hours of sun per day * C system derating = ABC Watt*Hours per day.

    More or less, "Prove" to us (and yourself) that your new system paper design will meet your needs.

    It will be a lot more accurate than me guessing from 1/2 a world away.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Bill,

    So I've created a pdf sheet showing my calculations from excel. Please let me know if you can see it (attached). Attachment not found.

    I only attached the one for the smaller system. Please let me know if my estimates look right. As you'll see, I like to work with Direct Current where possible because it allows you to work more easily with the batteries.

    What do you think?
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Help: Lots of A/C and not enough panels/batteries?
    kojmania wrote: »
    .....I like to work with Direct Current where possible because it allows you to work more easily with the batteries.

    What do you think?


    I'd like to convert it to AC asap, because the 48V 400A fuses are real expensive. A 30A 240V ckt bkr is <$10
    "high voltage" AC is much easiser to handle, and parts for it are much more common and affordable.
    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 ,

  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    I am a bit confused about how you calculate the amount of power generated.

    Let's take the last problem. You need 2,455 Watt*Hours per day for the "small system". Assume you have:
    • 0.77 solar panel + charger eff * 0.80 battery eff * 0.85 inverter eff = 0.52 end to end derating (solar panel to AC power).
    Assume you have 5+ hours of sun per day (fixed array) for most of the year:
    • 2,455 WH per day load * 1/0.52 system eff * 1/0.85 inverter eff * 1/5 hours of sun = 1,110.9 watts minimum solar panels
    • Note: The above line is in error--the 1/0.85 is already included in the 1/0.52 derating number. Corrected equation:
    • 2,455 WH per day load * 1/0.52 system eff * 1/5 hours of sun = 944 watts minimum solar panels
    Assume 24 volt battery bank. Minimum of 50% of daily capacity to a maximum of 50% and 3 days of no sun:
    • 2,455 WH * 1/24 volt battery * 1/0.85 invert eff * 1/0.50 max discharge = 240 AH @ 24 volts minimum
    • 2,455 WH * 1/24 volt battery * 1/0.85 invert eff * 1/0.50 max discharge * 3 days =722 AH 2 24 volts for high side estimate of bank capacity
    So, from looking at your calculations--I think you have made some different assumptions (6+ hours of sun, perhaps less derating factors)...

    There is something else to double check--Make sure we have enough solar panels to properly charge your battery bank. Normally, we recommend a range of 5% to 13% of your bank's Amp*Hour rated capacity (20 Hour Rating).

    For example, say you choose 800 AH of battery bank (fairly large battery bank for your planned loads):
    • 800 AH * 29 volts charging * 1/0.77 panel+charger derating * 0.05 rate of charge = 1,506 Watts solar panel minimum
    • 800 AH * 29 volts charging * 1/0.77 panel+charger derating * 0.13 rate of charge = 3,917 Watts solar panel maximum
    So, in your case, a large battery bank will need even more panels to ensure there is sufficient amount of solar power to properly charge the bank.

    Doing the same calculations for a "small" bank (50% max discharge or 1 day of storage) 240 AH rounded up to 300 AH:
    • 300 AH * 29 volts charging * 1/0.77 panel+charger derating * 0.05 rate of charge = 565 Watts solar panel minimum
    • 300 AH * 29 volts charging * 1/0.77 panel+charger derating * 0.13 rate of charge = 1,469 Watts solar panel maximum
    So, you will need to overlay how much solar panel you need for power (1,110.9 watts minimum--correct to 994 watts minimum) against the size of battery bank and how much solar power you need to keep the batteries "happy".

    For example, if you chose an 800 AH battery bank, the range of panels is from 1,506 watts minimum--More than the 1,110.9 994 watts minimum that the load itself would need. Too large of battery banks tend to be a pain--you need more solar panels or have to charge the bank often with a genset/utility power to keep the bank operating properly.

    The whole attempt here is to balance the needs of the loads against the size of solar panels and battery bank to give you an "optimum" usage of your resources (also known as "money").

    My guesstimate is probably a bit conservative--But in real life, it will be difficult to get 100% usage out of your system. You have limited storage and some days you may have more sun, and on others a bit less. Also, some days you will use more power and other days less...

    I would would not recommend a "production system" (where you must use power to be in business) to plan on 100% useful power (from the above numbers). Your solar panels + batteries should be a bit more than your needed loads (allow for bad weather, batteries wearing out, dirty solar panels, etc.) and/or an alternative power source (genset, grid power).

    You may choose to build out your system with a full complement of batteries and inverter--And a minimum amount of solar panels. Run the system for a while (monitoring power usage)--Then later, when you have more accurate numbers, decide how many more solar panels you will wish to purchase (offset utility power / generator fuel costs).

    Am I making sense?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Bill,

    This was a bit tricky, but I think I get the basic argument: First, you calculate the total load you have, then you design the system against that, but within the system itself, there are certain standard factors to bear in mind such as the fact that battery charging varies depending on the size of the bank and number of panels. But coming down to the nitty gritty of it, there are some areas I was confused abt:
    0.77 solar panel + charger eff * 0.80 battery eff * 0.85 inverter eff = 0.52 end to end derating (solar panel to AC power).
    Here, I'm a bit confused about where the 0.77 came from. Is that the panel efficiency? How did you get it? And for the derate factor, our manufacturer actually provided it as 0.85. Does this seem plausible to you or should we do our own calculations to find out? Any ideas how we could even go about doing that?
    Assume you have 5+ hours of sun per day (fixed array) for most of the year:
    2,455 WH per day load * 1/0.52 system eff * 1/0.85 inverter eff * 1/5 hours of sun = 1,110.9 watts minimum solar panels
    So this is assuming that for each of those 5hrs, our system is actually generating 1,110.9 WH AC per hour, right? How does one take into account the varying light intensities and solar radiation per hr? That's where PV Watts comes in, right?
    Assume 24 volt battery bank. Minimum of 50% of daily capacity to a maximum of 50% and 3 days of no sun:
    2,455 WH * 1/24 volt battery * 1/0.85 invert eff * 1/0.50 max discharge = 240 AH @ 24 volts minimum
    2,455 WH * 1/24 volt battery * 1/0.85 invert eff * 1/0.50 max discharge * 3 days =722 AH 2 24 volts for high side estimate of bank capacity
    So for our example, we actually need to add more batteries as well? Isn't it a good idea to try and get enough panels to cover the daytime use (as you've done above, and then enough batteries to cover primarily the nighttime use? In that case:
    895 WH (our nighttime load) * 1/24 volt battery * 1/0.85 inverter eff * 1/0.50 max discharge= 88 AH @ 24 volts minimum
    895 WH * 1/24 volt battery * 1/0.85 inverter eff * 1/0.50 max discharge * 3 days= 263 AH @ 24 volts for high side estimate of bank capacity. Is this a good thing to do, or does it merely complicate things for the user in the event that even with the max battery storage, any bad weather during the daytime would give them considerably less power than the 3-day ideal?
    So, from looking at your calculations--I think you have made some different assumptions (6+ hours of sun, perhaps less derating factors)...
    Yes. According to the PV Watts site, June is our month with the least solar radiation, but even then, different values are obtained across a 12-hour period. Check out:http://rredc.nrel.gov/solar/calculators/PVWATTS/version1/International/pvwattsv1_intl.cgi.

    If you Output the hourly performance data, you'll see that power is generated across an almost 12-hr period. I input the values for a 600 Watt DC system and a derate factor of 0.85.

    So to recap with our new derate value, let's assume we have 10+ hours of sun per day:
    2,455 WH per day load * 1/0.85 system eff * 1/0.85 inverter eff * 1/10 hours of sun = 340 watts minimum solar panels. Does this look right?
    For example, say you choose 800 AH of battery bank (fairly large battery bank for your planned loads):
    800 AH * 29 volts charging * 1/0.77 panel+charger derating * 0.05 rate of charge = 1,506 Watts solar panel minimum
    800 AH * 29 volts charging * 1/0.77 panel+charger derating * 0.13 rate of charge = 3,917 Watts solar panel maximum
    Assuming we just used the new derate factor in this calculation:

    800 AH * 29 volts charging * 1/0.85 panel+charger derating * 0.05 rate of charge = 1,364 Watts solar panel minimum
    800 AH * 29 volts charging * 1/0.85 panel+charger derating * 0.13 rate of charge = 3,548 Watts solar panel maximum

    But for our purposes, maybe we could try doing an average of the minimum and maximum battery bank capacity (allows us to save cost without undervaluing the battery bank too much is my thought here):

    (722 AH +240 AH) / 2 = 481 AH.
    Now to calculate the panel wattage needed for charging:
    481 AH * 29 volts charging * 1/0.85 panel+charger derating * 0.05 rate of charge = 821 Watts solar panel minimum
    481 AH * 29 volts charging * 1/0.85 panel+charger derating * 0.13 rate of charge = 2,133 Watts solar panel maximum

    So worst case scenario, we could increase our projected number of panels from 600W to 900W, and maybe leave the battery bank at its current size? This of course depends on whether you think it's a good idea to continue to project the battery use as a nighttime factor.

    By the way, what determined the rate of charge you used above (5 to 13%)?
    You may choose to build out your system with a full complement of batteries and inverter--And a minimum amount of solar panels. Run the system for a while (monitoring power usage)--Then later, when you have more accurate numbers, decide how many more solar panels you will wish to purchase (offset utility power / generator fuel costs).
    The larger system has been in operation for some time, and we notice that after a few days of use, the system often shuts off after sunset, so we realize that we're doing something slightly wrong, hence the desire to upgrade. But what I'll do next is apply your explanations here to that system as well to determine what the ideal system would look like.

    Last question though: What's the best way to size the inverter and charge controller? Does one look at the max draw from the load and then choose an amperage for the inverter that is maybe 10% more? I am worrying about upgrading our current controllers and inverters (costs that I'm really crossing fingers not to have!)

    Thanks, Bill. You definitely are making sense to me. I think my document was still quite simplistic, but it's a work in progress, so bear with me.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    Kojmania,
    kojmania wrote: »
    Here, I'm a bit confused about where the 0.77 came from. Is that the panel efficiency? How did you get it? And for the derate factor, our manufacturer actually provided it as 0.85. Does this seem plausible to you or should we do our own calculations to find out? Any ideas how we could even go about doing that?
    The 0.77 derating is from the PV Watts website Derating Page (and seems to agree pretty well with my GT system experience. Basically, for warm climates, you can derate the PV solar panels to ~81%, and assume around 5% losses for the charge controller:
    • 0.81*-0.95 = 0.77 solar panel + charge controller derating
    So this is assuming that for each of those 5hrs, our system is actually generating 1,110.9 WH AC per hour, right? How does one take into account the varying light intensities and solar radiation per hr? That's where PV Watts comes in, right?
    The "5 hours" or "6 hours" of "full noontime sun equivalent" assumes a "bell curve" where you may produce varying power for 12+ hours per day (sun rise to sun set, + average cloud/weather conditions day by day, converted to 1,000 Watts/sq.meter of Noontime sun equivalent).

    So, PV Watts takes the "24 hours of measured 20 year average sunlight" and integrates that into a "number" that is based on fixed array, or tracking array, etc... And giving us an equivalent number of hours per day we can use in our math (tracking arrays keep "square" to the sun through the day, so you don't get "cosine error" -- i.e., Power of Sun * Cosine angle to sun = effective power on panel, etc.).

    The "temperature" derating is (mostly) the reason you have that 81% solar panel derating (the lab conditions dramatically underestimate the real cell temperatures under sunlight--Basically a quick flash of light on panel = Watt output--and cells do not get time to heat up like a real installed panel would).

    STC (standard test conditions) vs PTC ratings:
    California and many other states require solar panel manufacturers to submit operating specifications for their products. These products are then further tested by an independent laboratory. At least in California, the agency approved to do this is PVUsa. The results of PVUsa tests are dubbed a panel’s PTC rating (PVUsa Test Conditions).
    Therefore every panel will have both an STC rating (standard test conditions) and PTC rating. The PTC rating involves more extreme, real world test conditions, so you’ll find it to be lower than the STC rating and more indicative of how a solar panel will hold up. So the higher PTC rating, the better.
    Also, it is the PTC rating that the state uses to calculate system rebates (i.e., expected performance). So while the STC rating for a panel may be 175 Watts, if the PTC rating is only 160 Watts, then that will be the benchmark for your cash rebates — an important tip when figuring expenses and payback period for your solar system.
    PV Watts (as far as I know) does not take into account that hot panels produce less power vs very cold panels (because Vmp falls as the panels get hot).
    So for our example, we actually need to add more batteries as well? Isn't it a good idea to try and get enough panels to cover the daytime use (as you've done above, and then enough batteries to cover primarily the nighttime use? In that case:
    895 WH (our nighttime load) * 1/24 volt battery * 1/0.85 inverter eff * 1/0.50 max discharge= 88 AH @ 24 volts minimum
    895 WH * 1/24 volt battery * 1/0.85 inverter eff * 1/0.50 max discharge * 3 days= 263 AH @ 24 volts for high side estimate of bank capacity. Is this a good thing to do, or does it merely complicate things for the user in the event that even with the max battery storage, any bad weather during the daytime would give them considerably less power than the 3-day ideal?
    Yes, for your needs--You could size the panels to support daytime use and size the batteries for night-time use (assume that most of the power used by the business is during the day). And since you are not charging the battery bank, you don't have the 80% battery efficiency to worry about...

    But--There is a risk to this... Are you willing (able) to shut down the business if the sun is not shining (or transfer over to utility power).

    The way I did the calculations, assumes that you are off-grid and want to minimize generator usage. If you can avoid using the solar+battery when it is cloudy--you can make different assumptions (I am trying to be conservative in the recommendations and assuming worst case operation--all power generated during day, all power consumed at night).
    Yes. According to the PV Watts site, June is our month with the least solar radiation, but even then, different values are obtained across a 12-hour period. Check out:http://rredc.nrel.gov/solar/calculators/PVWATTS/version1/International/pvwattsv1_intl.cgi.
    You know your conditions better than I... You learn how to do the calculations and make appropriate assumptions for your setup--And you won't need me any more (I feel so abandoned :cry:;)).
    If you Output the hourly performance data, you'll see that power is generated across an almost 12-hr period. I input the values for a 600 Watt DC system and a derate factor of 0.85.
    • 0.77 = solar panel * charge controller derating
    • 0.80 = Flooded Cell battery losses (new batteries may be closer to 90%, older batteries closer to 80%).
    • 0.85 = Inverter Losses
    So, if you are powering mostly AC loads:
    • System Derating = 0.77 * 0.80 * 0.85 = 0.52 end to end system derating
    So to recap with our new derate value, let's assume we have 10+ hours of sun per day:
    First, no place that I know of has more than 8 hours of sun (full noontime sun equivalent) with a fixed array... You may have 10-12 hours of sun per day, but if you integrate under the bell shaped curve--you will get 6 hours (or whatever is appropriate for your area, array angle, season, etc.).
    2,455 WH per day load * 1/0.85 system eff * 1/0.85 inverter eff * 1/10 hours of sun = 340 watts minimum solar panels. Does this look right?
    Personally, I think that you are too optimistic... If I understand everything you told me--Assuming that you are running your business loads in daylight (and very light loads at night):
    • 6 hours of "full noontime equivalent) sun
    • 0.654 = 0.81 panel derating * 0.95 charge controller * 0.85 inverter
    • 2,455 WH per day * 1/0.654 daytime eff * 1/6 hours of sun = 626 watt panel minimum
    But, I would be using 0.52 instead of 0.654 derating because I have to assume that some days the sun will not be shining and you will still want to run from the battery bank:
    • 2,455 WH per day * 1/0.52 anytime eff * 1/6 hours of sun = 787 watt panel minimum
    Assuming we just used the new derate factor in this calculation:

    End of part 1
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    Part 2:
    800 AH * 29 volts charging * 1/0.85 panel+charger derating * 0.05 rate of charge = 1,364 Watts solar panel minimum
    800 AH * 29 volts charging * 1/0.85 panel+charger derating * 0.13 rate of charge = 3,548 Watts solar panel maximum
    I would use 0.77 derating vs 0.85 as in real life, the panel is ~81% efficient and the charge controller is around 95% efficient.

    But for our purposes, maybe we could try doing an average of the minimum and maximum battery bank capacity (allows us to save cost without undervaluing the battery bank too much is my thought here):
    (722 AH +240 AH) / 2 = 481 AH.
    Now to calculate the panel wattage needed for charging:
    481 AH * 29 volts charging * 1/0.85 panel+charger derating * 0.05 rate of charge = 821 Watts solar panel minimum
    481 AH * 29 volts charging * 1/0.85 panel+charger derating * 0.13 rate of charge = 2,133 Watts solar panel maximum
    Perfectly OK to choose something between 240 and 722 AH... 481 is a good a choice as any (minimizes $$$ costs up front, not too large of bank, so less solar panels needed too).
    So worst case scenario, we could increase our projected number of panels from 600W to 900W, and maybe leave the battery bank at its current size? This of course depends on whether you think it's a good idea to continue to project the battery use as a nighttime factor.
    Part of the issue with "unbalancing" the battery/panel/load is you start to have to look at other factors... For example, a good rule of thumb for a flooded cell battery is to limit surge current to C/2.5 (battery bank AH rating * 0.40 = maximum surge current). If you have a "small" battery bank, the amount of surge power that you can draw (with your inverter/starting motors, etc.) is reduced.

    If you draw more than 40% of the battery bank's AH rating (motor starting), the battery may drop its output voltage so much that the inverter will cutoff/your other DC loads may "brown out".
    By the way, what determined the rate of charge you used above (5 to 13%)?
    Basically, there are several battery manufacturers that recommend a 5% minimum rate of charge for equalizing their battery bank. Also, as batteries age, they have self discharge... If you go down to 1% rate of charge--basically, the solar panels are just "floating" the battery bank (making up for self discharge) and you will get very little useful power from the battery bank.

    13% or ( ~C/8 ) is the maximum charge rate for most flooded cell batteries. Above that rate, you could run the risk of overheating them.

    In reality, you probably could go as high of charging current as 30%--because once the batteries are >~80% charged--the charge controller will be cutting back on charging current to maintain ~14.5 volts (for a 12 volt bank)--and any extra charging ability (AC battery charger, Solar panels + charger, etc.) is sort of wasted at that point. (solar panels are expensive--so we try to not recommend too many of them).
    The larger system has been in operation for some time, and we notice that after a few days of use, the system often shuts off after sunset, so we realize that we're doing something slightly wrong, hence the desire to upgrade. But what I'll do next is apply your explanations here to that system as well to determine what the ideal system would look like.
    Basically, it sounds like, what is usually called, "deficit charging"... Basically, you start with a battery that is 100% charged... You take it down 20% and recharge it 10% ever day (100% to 80% to 90% to 70% to 80% to 60% etc.). So at the end of 5-10 days, you are at 20% taking it back town to 0% backup to 10% and back down to 0%) at the end...

    There are a few major problems with doing this.
    • Taking a battery below 20% state of charge can kill one or more cells in a battery bank (reverse charging)... When a cell is reversed charged (flipping voltage polarity because it started out as a weak cell)--it is usually destroyed.
    • Keeping a battery below ~75% state of charge for days, weeks, or months--The batteries will begin to sulfate (the lead sulfates harden or crystallize). These sulfates do not return to solution on recharging and permanently lose battery capacity. Once the sulphation is severe enough--your battery's storage capacity is no longer enough to take you through the night.
    • The deeper the cycling, the less total cycles the battery will survive. So, a battery that is cycled down to 75% state of charge every night may last 2,000 cycles. A battery that is cycled down to 20% state of charge may last 500 cycles (need to check manufacturer's specifications for actual numbers).
    Last question though: What's the best way to size the inverter and charge controller? Does one look at the max draw from the load and then choose an amperage for the inverter that is maybe 10% more? I am worrying about upgrading our current controllers and inverters (costs that I'm really crossing fingers not to have!)
    For the charge controller--You can estimate the Solar Array Power * 0.77 for charge controller' maximum output power... For example a 60 amp MPPT charge controller on a 12 volt battery bank charging at 14.5 Volts can support a maximum array (cost effectively--0.77 is my humble guestimate):
    • Array Size = 14.5 volts * 60 amps rated output * 1/0.77 derating = 1,130 Watt Array
    You can choose more or less wattage. The controller will work fine either way. But, the above would be my "optimum" price/performance recommendation--Others probably have their own (some may use 0.85, others in very cold climates may choose 1.0 as a derating factor--personal choices).
    Thanks, Bill. You definitely are making sense to me. I think my document was still quite simplistic, but it's a work in progress, so bear with me.
    Kojmania, you are very welcome--You are definitely jumping in with both feet into the whole Solar PV / Off-Grid thing.

    The first time or two -- It is really confusing. After you have built your first couple systems and measured their output power vs your loads--Your next iterations will be much easier and probably better meet your price/performance goals.

    Remember, in the end, if your measured numbers agree with predicted numbers within 10%--That is probably more luck than skill... There are just a lot of variables and variability in the sun/batteries/loads/etc. and getting closer is really not practical.

    -Bill

    PS: Regarding your inverter sizing question... I probably cannot give you much useful information. Every inverter and load is different. And also depends on wiring, battery capacity, and such.

    As an example... In the USA we have "Energy Star" rated refrigerator/freezers. They typically draw around 120 Watts to run the compressor. But they are also "frost free" which means that they have 500 watt heating elements to defrost the evaporator (cooling coils) and run the ice maker (melt a bit of ice so the ice maker can eject the ice).

    That ends up needing a 1,500+ watt AC Inverter because the inverter needs to handle the total loads (120 watts running, interior fridge light, maybe 600 watts starting, and a 500 watt defrost cycle at the same time).

    And, if you measure surge current--most DMM (digital multi-meters) are not fast enough to record the initial surge current (for example, an old TV set or large computer system, that could be 45+ amps inrush or more for a few cycles).

    Also, turning 5 computers one at a time is not near the surge load as turning on all 5 at the same time (say you turn on a power strip).

    Sorry I do not have a simple formula that you can plug in and get the "right answer".

    A couple Inverter FAQs (both are "good reads"):

    All About Inverters
    Choosing an inverter for water pumping

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Bill,

    Thanks for some really in-depth explanations.... wow......
    The "5 hours" or "6 hours" of "full noontime sun equivalent" assumes a "bell curve" where you may produce varying power for 12+ hours per day (sun rise to sun set, + average cloud/weather conditions day by day, converted to 1,000 Watts/sq.meter of Noontime sun equivalent).

    So, PV Watts takes the "24 hours of measured 20 year average sunlight" and integrates that into a "number" that is based on fixed array, or tracking array, etc... And giving us an equivalent number of hours per day we can use in our math

    I think I understand in theory what you mean, and to test it out, I used the hourly performance data for one of the PV Watt Hourly Output readings. The figures for the day were:
    Year, "Month", "Day", "Hour", "AC Power (W)"
    1991, 1, 1, 01:00, 0
    1991, 1, 1, 02:00, 0
    1991, 1, 1, 03:00, 0
    1991, 1, 1, 04:00, 0
    1991, 1, 1, 05:00, 0
    1991, 1, 1, 06:00, 0
    1991, 1, 1, 07:00, 0
    1991, 1, 1, 08:00, 287
    1991, 1, 1, 09:00, 686
    1991, 1, 1, 10:00, 1031
    1991, 1, 1, 11:00, 1256
    1991, 1, 1, 12:00, 1372
    1991, 1, 1, 13:00, 1290
    1991, 1, 1, 14:00, 1108
    1991, 1, 1, 15:00, 1026
    1991, 1, 1, 16:00, 681
    1991, 1, 1, 17:00, 262
    1991, 1, 1, 18:00, 2
    1991, 1, 1, 19:00, 0
    1991, 1, 1, 20:00, 0
    1991, 1, 1, 21:00, 0
    1991, 1, 1, 22:00, 0
    1991, 1, 1, 23:00, 0
    1991, 1, 1, 24:00, 0

    So I added the power and got 9001 W AC for that day. The hour with the max output was 12 noon, with 1372 W. Dividing 9001 by 1372 gave me ~6.5. So that means on this day, there were approx 6.5 hours of full noontime sunlight, correct? But when you refer to a "number" above, is that the PV Watts column called "Solar Radiation"? I hadn't understood how that column would be important in calculations, until now. So thanks.
    The way I did the calculations, assumes that you are off-grid and want to minimize generator usage. If you can avoid using the solar+battery when it is cloudy--you can make different assumptions (I am trying to be conservative in the recommendations and assuming worst case operation--all power generated during day, all power consumed at night).
    Great clarification for me! And one of my friend's dads who is a mentor to me just said to me this weekend that he would be interested in having solar used in his home and asked yours truly for some projections. So this is something I'll bear in mind cos for residential use, you're totally right, the power gets stored up during sunlight, and used at night, so the ratio of his panels to batteries will vary significantly from some of our current projections. Esp depending on whether he wants back-up power or simultaneous power. Will figure all that out soon.
    You know your conditions better than I... You learn how to do the calculations and make appropriate assumptions for your setup--And you won't need me any more (I feel so abandoned ).
    Actually, you were right here. I hadnt yet understood the bell curve argument. And no, Bill, I'm pretty sure we just became partners for life! :) You've been truly invaluable! Thank you.
    Part of the issue with "unbalancing" the battery/panel/load is you start to have to look at other factors... For example, a good rule of thumb for a flooded cell battery is to limit surge current to C/2.5 (battery bank AH rating * 0.40 = maximum surge current). If you have a "small" battery bank, the amount of surge power that you can draw (with your inverter/starting motors, etc.) is reduced.

    If you draw more than 40% of the battery bank's AH rating (motor starting), the battery may drop its output voltage so much that the inverter will cutoff/your other DC loads may "brown out".
    Very good to know! What's "brown-out" by the way? Hopefully not synonymous with "blown"!
    And to make sure I get it, if we had a battery bank of 500 AH, then 0.40*500 = 200. This means if initial current is greater than 200 amps, we'll have problems, right? Either the system shuts off or we're actually damaging our battery with the huge draw, correct?
    13% or ( ~C/8 ) is the maximum charge rate for most flooded cell batteries. Above that rate, you could run the risk of overheating them.
    Also good to know. And what does "C" stand for?
    Basically, it sounds like, what is usually called, "deficit charging"... Basically, you start with a battery that is 100% charged... You take it down 20% and recharge it 10% ever day (100% to 80% to 90% to 70% to 80% to 60% etc.). So at the end of 5-10 days, you are at 20% taking it back town to 0% backup to 10% and back down to 0%) at the end...
    You're right, and thats why we wanna try and get panels to add to the system now. And if I've understood some of your previous explanations, then to avoid deficit charging for our system, we should have sized our panels based on the battery bank like you did here:
    800 AH * 29 volts charging * 1/0.77 panel+charger derating * 0.05 rate of charge = 1,506 Watts solar panel minimum
    800 AH * 29 volts charging * 1/0.77 panel+charger derating * 0.13 rate of charge = 3,917 Watts solar panel maximum
    In our case, 500 AH * 55 volts charging * 1/0.77 panel+charger derating * 0.05 rate of charge = 1,785 Watts solar panel minimum
    500 AH * 55 volts charging * 1/0.77 panel+charger derating * 0.13 rate of charge = 4,643 Watts solar panel maximum.
    2 questions about this: a) Our 2000 W panels therefore do fall into the correct range, don't they? And wouldn't that mean we shouldn't be having the charging deficit problem?
    b) What determines the charging voltage? I was reading through some of the battery FAQ sites, but they mostly make reference to 12V battery charging, and even then they give a range of what the voltage may be, depending on whether you're in the float, absorption or bulk phase. How did you thus decide on 29? I picked 55 kind of randomly (since our battery bank is 48V, 500AH)
    For the charge controller--You can estimate the Solar Array Power * 0.77 for charge controller' maximum output power... For example a 60 amp MPPT charge controller on a 12 volt battery bank charging at 14.5 Volts can support a maximum array (cost effectively--0.77 is my humble guestimate):
    Array Size = 14.5 volts * 60 amps rated output * 1/0.77 derating = 1,130 Watt Array
    So in our case, rating is 48V, 50A; 55 volts * 60 amps rated output * 1/0.77 derating = 3,571 Watt Array. So we could, in actual fact, add an additional 1,500 W worth of panels, and exceed neither the maximum panel size for charging the battery (4643 W above) nor the max output from the charge controller. We only wanted to add 1000 W, so allowing for some slightly "off" estimations, we should still be in good shape. Would you use this same approach for the inverter? i.e: output voltage * rated wattage * 1/0.52 array to inverter derating? eg a 5000 W rated inverter outputting 220V: 220V * 5000W * 1/0.52 = 2115384.6. LOL. My gosh, this is clearly wrong, but it was worth a shot :)
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    Good day Kojmania,
    kojmania wrote: »
    So I added the power and got 9001 W AC for that day. The hour with the max output was 12 noon, with 1372 W. Dividing 9001 by 1372 gave me ~6.5. So that means on this day, there were approx 6.5 hours of full noontime sunlight, correct?
    Basically correct... The "variable" is that we are talking about 1,000 Watt/sq.meter of solar irradiance.

    The noon-time sun for your location may be a bit above or below the "standard" as this is just an "average" day.
    But when you refer to a "number" above, is that the PV Watts column called "Solar Radiation"? I hadn't understood how that column would be important in calculations, until now. So thanks.

    Yep--The single "6.5" hours of sun is just the "integral" (from Calculus--area under the bell power curve) averaged over that 1 month / 1 year "average" set of samples.

    Great clarification for me! And one of my friend's dads who is a mentor to me just said to me this weekend that he would be interested in having solar used in his home and asked yours truly for some projections. So this is something I'll bear in mind cos for residential use, you're totally right, the power gets stored up during sunlight, and used at night, so the ratio of his panels to batteries will vary significantly from some of our current projections. Esp depending on whether he wants back-up power or simultaneous power. Will figure all that out soon.
    Details do matter--but many times in solar pv--if you get within 20% of predicted numbers (from PV Watts or equivalent)--That is about the limit of the accuracy. Worrying about the 20% losses of battery for night power use vs daytime power use--is almost lost in the "noise" of the daily variations in real daily solar collection.

    Also, in real life--you will never pull 100% of the available power during the 24 hour period from the PV Array/Battery system...

    It is not a bad idea to have 50% to 100% (i.e., double your daily estimated power usage) for the system design.

    Most people tend to add more loads to their systems over time (computers, radios, vacuuming, etc.) and designing a system to support 100% of your daily loads today will make you crazy as you try to get everyone to conserve power because a little bit of clouds are predicted for tomorrow.
    Very good to know! What's "brown-out" by the way? Hopefully not synonymous with "blown"!
    Brown-out is probably a US term... It is when the utility has so many loads (air conditioning/electric heating during weather exteremes, power grid failure, etc.) and the 120/240 VAC of the power grid falls to ~50-80 volts or so... Your filament lamps are dimming or looking "brown" instead of white.

    In the olden days when this used to happen (50 years ago)--the utility just let the power sit at that low voltage until they could address the problem... Many people had their refrigerator motors burnout from the low voltage. We used to have to run around the home and unplug appliances/pull fuses until the power was restored.

    Today, the utility will simply turn off the power instead of letting the grid brown out if they cannot regulate voltage (usually--on rare occasions, I have seen brownouts when a power pole was hit in a traffic accident).
    And to make sure I get it, if we had a battery bank of 500 AH, then 0.40*500 = 200. This means if initial current is greater than 200 amps, we'll have problems, right? Either the system shuts off or we're actually damaging our battery with the huge draw, correct?
    Surge current may not damage the batteries--but they simply will not be able to supply full voltage (like your car lights dimming when you start the engine). In some cases, it can cause problems (see brown-out from above--motors that don't start but stay stalled and overheat--can happen to both DC and AC motors).
    Also good to know. And what does "C" stand for?
    C= Battery / Bank Capacity in Amp*Hours...

    Typically batteries may be rated in 5/10/20/100 Hour ratings--Or, for example if you have a 500 AH battery bank at the 20 hour rating:
    • 500 AH / 20 Hour rating = 25 amp current for 20 hours
    Of course, that is 100% to 0% state of charge--Never run your battery bank to 0% state of charge--it is too easy to ruin one or more batteries in your battery bank.

    You're right, and thats why we wanna try and get panels to add to the system now. And if I've understood some of your previous explanations, then to avoid deficit charging for our system, we should have sized our panels based on the battery bank like you did here:
    • In our case, 500 AH * 55 volts charging * 1/0.77 panel+charger derating * 0.05 rate of charge = 1,785 Watts solar panel minimum
    • 500 AH * 55 volts charging * 1/0.77 panel+charger derating * 0.13 rate of charge = 4,643 Watts solar panel maximum.
    That is correct.
    2 questions about this: a) Our 2000 W panels therefore do fall into the correct range, don't they? And wouldn't that mean we shouldn't be having the charging deficit problem?
    Probably--based on the assumed daily load and daily power available.

    In any case, monitoring your battery bank state of charge during use/charging and how much current/power you consume/generate (AH or Watt*Hours) and what the battery "says" (specific gravity, resting voltage, battery monitor reading, etc.) will tell you if the battery is "happy".
    b) What determines the charging voltage? I was reading through some of the battery FAQ sites, but they mostly make reference to 12V battery charging, and even then they give a range of what the voltage may be, depending on whether you're in the float, absorption or bulk phase. How did you thus decide on 29? I picked 55 kind of randomly (since our battery bank is 48V, 500AH)
    Battery charging voltage depends on:
    • state of charge (near dead batteries will take lots of current at less than "charging voltage").
    • amount of charging current (more current, higher battery voltage)
    • Bulk/Absorb charging voltage (~13.8-14.5 volts) is the recommend "fast charging" voltage range. If you "charge" below 13.8 volts, the battery will probably never reach 100% charge. Charging at 14.5 volts charges the bank about as fast as it can safely take the energy)
    • Float charging voltage (~13.7 volts) is the "optimum" voltage to keep on a battery bank in storage... Higher voltage can damage the positive plates and lower voltage can damage the negative plates (over time).
    • B/A/F charging voltages are dependent on temperature (defined for 77F/25C). Hot batteries need a lower charging voltage. Cold batteries need a higher set of charging voltages.
    • For tropical climates, a battery retailer may use slightly less strong acid mix (electrolyte) in the battery cell... This will drop the charging/output voltage requirements a tiny bit.
    • Different battery types/manufacturers/etc. use different compounding agents in the battery plates (fork lift batteries, car batteries, UPS batteries, solar RE batteries, AGM, sealed, etc.) and different specific gravity electrolyte to optimize for the specific design/battery usage. Always refer to your battery manual/vendor information for your recommended charge current/voltage settings.
    For the numbers I used above referenced a 12 volt battery bank--Simply multiply by 4x for a 48 volt battery bank. Some battery specs. will be based on Per Cell voltage/ratings (around 2.x volts per cell).
    So in our case, rating is 48V, 50A; 55 volts * 60 amps rated output * 1/0.77 derating = 3,571 Watt Array. So we could, in actual fact, add an additional 1,500 W worth of panels, and exceed neither the maximum panel size for charging the battery (4643 W above) nor the max output from the charge controller. We only wanted to add 1000 W, so allowing for some slightly "off" estimations, we should still be in good shape.
    Yep--For a MPPT type charge controller, you can exceed the controller's input wattage rating... They will only output their rated current (maximum) regardless of how much input power they have available (with the overall ratings of the MPPT charge controller).
    Would you use this same approach for the inverter? i.e: output voltage * rated wattage * 1/0.52 array to inverter derating? eg a 5000 W rated inverter outputting 220V: 220V * 5000W * 1/0.52 = 2115384.6. LOL. My gosh, this is clearly wrong, but it was worth a shot :)
    The 0.52 is all of the losses combined into one number (panel, charge controller, battery bank, inverter)... So:
    • 1,000 watts of solar panels * 0.52 system derating * 6.5 hour of sun per day = 3,380 Watt*Hours
    So, you have 3.38kWH of estimated available power from your 230VAC outlet over night.... If you ran the inverter only during the day--then you would get 1/0.80 or 1.25 more available power (not cycling the battery bank day/night).

    Also, remember that inverters have other losses... A large inverter may lose 60 watts with no loads... Leave that on 24 hours per day:
    • 60 watts * 24 hours = 1,440 WH idling losses per day...
    If you only ran the inverter 8 hours a day to power your loads (and turned of the DC input "at night"):
    • 60 watts * 8 hours = 480 Watt*Hour losses per day from inverter
    So, you can get almost another 1kW of power per day from your large inverter if you turn it off when not in use.

    Hope that all makes sense. Solar is frequently a game of "small" numbers that add up to "big" losses. That is why we push conservation so heavily...

    I.e., use the large inverter during the day to run your business... Run a small inverter (that may have 6 watts of idling losses) at night to run a few lights, laptop computer, TV/Radio.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Bill,

    A few quick ones:
    It is not a bad idea to have 50% to 100% (i.e., double your daily estimated power usage) for the system design.
    Hmm... This should be interesting because when I took his readings, he's really only interested in using his system as a back-up one, and in their area, power usually goes out for not more than 4 hours at a time, so I thought I'd use 8 hours as his anticipated time during which power is required.
    In the olden days when this used to happen (50 years ago)--the utility just let the power sit at that low voltage until they could address the problem... Many people had their refrigerator motors burnout from the low voltage. We used to have to run around the home and unplug appliances/pull fuses until the power was restored.
    Oh wow. Hollywood never showed us this (;)that I remember)
    The 0.52 is all of the losses combined into one number (panel, charge controller, battery bank, inverter)
    This makes sense, but then it confuses me a bit when we did this earlier:
    Assume you have 5+ hours of sun per day (fixed array) for most of the year:
    2,455 WH per day load * 1/0.52 system eff * 1/0.85 inverter eff * 1/5 hours of sun = 1,110.9 watts minimum solar panels
    If the inverter's efficiency has already been factored in to get 0.52 as the end to end derating, how come we multiplied it again in determining the minimum panel wattage required here? Had meant to ask that earlier.

    Lastly, attached is the approach I am taking for sizing the system I told you about. This time, I used your approach first, and then "proofed" it with my earlier approach, using the PV Watts site recorded values. What do you think?
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    You are correct, I goofed and included the inverter loses twice.

    I will have to check your numbers later today.

    -Bill

    Update: Take out my 0.52*0.85 adding additional inverter losses goof and all looks well... -Bill

    Update 2: Fixed original post with error in formula.:blush:
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Bill,

    Oh, no worries. At worst we would have simply had more power to use...

    Please note then that my estimate in the pdf will be slightly off as well since I did use the inverter losses twice there.

    Thanks!

    -Kojmania
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Hi Bill,

    Wondering how you're doing. Been a while.... How's my sensay this week?

    Cheers,
    Kojmania
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    Good Day Kojmania,

    Life goes on. :roll:

    How is it going in your part of the world? Are you going to look into becoming a solar installer yet?

    Best Regards,
    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Haha. The good news is "Yes"!

    The "other" news is that while some aspects may seem straightforward, I'm sure you know better than anyone that there's a lot of nitty gritty stuff that it helps to know as well, so I'm trying to find a course locally that might give some exposure. But we're limited this side of the world. Do you have any recommendations?

    And is it safe to assume I can proceed with my most recent estimates?:
    I will have to check your numbers later today.

    -Bill

    Talk soon.

    Sorry for the long delay. I teach at a local school, and there has been more work the past 2 weeks than usual. Happy Thanksgiving!

    -Kojmania
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    I have heard rumors of people with "lives" out there. ;)

    The numbers look fine withe 0.85 extra derating factor removed.

    Try looking for courses geared towards electricians.

    What seems to cause the most problem is that battery banks are typically 12-48 volt systems and electricians on your side of the world probably work with 230 VAC.

    Basically, on the DC side of the system, the current is about 10x that on the AC side (24 vdc vs 230 vac). On occasion, the electrician forgets to use the P=V*I equation to figure out the low voltage DC side currents and voltage drops.

    Higher currents, lower voltage drops (a 12 volt battery system will cutout around 10.5 volts--1.5 volts of headroom--A 230 volt system will probably operate down to 205 volts or so--25 volts of drop is OK).

    DC current is also "harder" to break with switches and fuses. DC current will sustain an arc much better than AC--So any switches/fuses/breakers are usually physically larger when rated for DC, or the DC rated voltage is much lower than its AC rating.

    Working on large battery banks--you have to be very careful as storage batteries can output 1,000's to 10,000's of amperes or more (lots more than a home AC system can supply into a dead short). So proper safety precautions are critical so that nobody is injured or fires started on the DC side of the system (proper fusing/wiring/switches).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Wow. Funny I should just be reading this... Last night, I got a call from my dad's office saying the batteries were sparking and there was a burning smell. You won't believe someone needed a bolt (probably for a car or something), and removed 2 of the bolts holding the wires to the battery terminals, and the poor contact caused the 2 terminals to spark and blacken the wiring cable, and started to melt the plastic covering on the battery!

    It was annoying, but they've promised to pay closer attention, but its amazing how quickly an investment can wash down the drain just cos s'one thinks "Oh, look they have so many, they shouldn't miss 2 of them."

    Unbelievable, right?
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: Help: Lots of A/C and not enough panels/batteries?

    It always ticked me off that people locked up everything (tools, parts, equipment, etc.) at work--Until you have a couple of instances where everything was working fine/in place until somebody decided that something would not be missed.

    Then--We had to pop ceiling tiles around parts cages for when we were working weekends on time critical projects and the guy with the keys could not arrange for access...:roll::cry:;):grr:blush:

    Bill "It never ends" B.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • kojmania
    kojmania Registered Users Posts: 15
    Re: Help: Lots of A/C and not enough panels/batteries?

    Ugghh!!! Tell me about it!

    By the way BB, I have one more question for you. Based on the sizing we've been doing, I wanna try to upgrade slightly the 2KW system. Looks like my panels can go up by 500 and that should be fine. But the batteries are 2V, 500 AH, 24 of them. Now i realize we need a minimum of 700 AH. Can you help me figure out how many more batteries to get and how I can connect them (in parallel, I believe) to obtain the 48V, 700 AH I need?

    Thank you....

    -Koj