Battery Bank Sizing

The friendliest place on the web for anyone with an RV or an interest in RVing!
If you have answers, please help by responding to the unanswered posts.
Alpena Jeff said:
Please explain this statement.


Simply put, the main issues arise from cable,connection and internal impedance, if we eliminate cabling and terminal contact
variances ( very difficult for DIY as you don't have the equipment, but achievable ) we are left with internal impedance ( apparent resistance of the battery ). this is quite variable, certainly more so with AGM

Now, a battery with a higher impedance will, over time, result in less charge current to that battery, thus resulting in a lower average
SOC ( state of charge ), this has the effect of aging the battery and increasing its internal impedance. over many cycles the effect becomes more pronounced. when the battery pack is idle ( minimal load ) other batteries in the string of higher SOC will discharge into the lower SOC battery and self equalize the voltages, but, this effectively makes the whole set equal to the worst battery in the pack.

this just gets worse as we add more batteries to the string, the net result is shorter lifespan and capacity due to all the unequal current flows.

with FLA these effects can be mitigated somewhat by periodically charging with an equalizing charge for an extended period, but as we know, equalizing causes more plate corrosion and reduces life.. !! we only equalize to reduce sulphation and electrolyte stratification. with agm, you can't equalize, so you have to hold at float for at least 24 hours to get the SOC to balance somewhat.


other issues with parallel FLA include cell shorts, imagine what happens when one cell shorts in a parallel bank.. the other batteries discharge into the failed battery and that's going to give you a very bad hair day with fire, smoke and hot acid !!!

now, with eight agm's having lower and more variable internal resistance, the owner has no way in hell of ever balancing the charge currents, I know this from experience actually measuring charge and discharge currents on parallel strings, it just ain't gonna happen..
so again,  the net effect is much shorter life and reduced capacity. all at a 200% premium over FLA.


 
Thank you for the replies.  Question if the panels I get put out a VMPP  of 36 volts I wouldn't
need to run the in series (theoretically) correct? Could they each be independently run to
the charge controller? Won't that reduce the shading problem to only affect whichever panel
is being shaded.  Also the panels seem to have 3 bypass diodes so it seems that partial shading
only takes out some of the panel not all of it.  Does this make any sense or am I thinking about
it incorrectly.
Here are panels that I was looking at.

http://www.amerescosolar.com/sites/default/files/Hyundai%20305W%20Solar%20Panel_0.pdf

https://www.solar-electric.com/lib/wind-sun/AXITEC_AXI_power_72c_310W-330W.pdf
 
Solarman, if not 8 Group 31 AGM's how would you solve the battery problem within the constraints of a typical RV battery bay and budget? 

As I see it 8 6V GC2 AGM's in series parallel might help, but you are still faced with similar issues, just on a smaller scale at 4 parallel banks, 4V or 2V industrial deep cycles might be nice, but then you get sticker shock on the price, and may face other physical limitations.  Flooded deep cycles might work, but often there is a limitation on access when squeezing in more batteries than a coach originally came equipped with...
 
Butchiiii:


Sounds like your thinking is much like mine.  Here is a link to my project: http://www.rvforum.net/SMF_forum/index.php/topic,106669.0.html


I used panels made from 5" ingots rather than the more standard 6" ones, which made the panels closer to 32" wide.  This still allows me to walk on the roof of our 30' TT.


AFAIK, yes, shading will only take out the shaded panel if they are all wired in parallel, and if they have bypass diodes (mine do), then it only takes out the sections of the panel that are shaded.  Whether 36v is optimal or not may well depend on the controller you use (my panels are rated at 39Vmpp).  But you are probably in the ballpark.


I deliberately put more panel on the roof than I needed to account for periodic shading, and cloudy days.  Panel cost for me was relatively minor compared to the total cost of the project, with shipping and controller being the 'big ticket' items.  4 panels or 6 really didn't make that much difference, so I went for 6.  I have zero regrets.  My system is panel heavy, bit I don't mind.  My batteries are charged by 1 PM the next day most of the time.  If I am wasting sunlight, well, sunlight is still free. :)  I also have room for growth in battery capacity if I ever go to lithium.  And I still get 15 to 25 amps in the rain.


My 1200 watt system is overkill for us.  This winter in Arizona and SoCal we were without shore power almost exclusively for 3 months, but we did not spare ourselves any conveniences.  Microwave, waffle iron, electric toaster, hot air corn popper, computers and laser printer -- none of them taxed the system at all.  I even ran the fridge one afternoon on 110 from the inverter because where we were was SO quiet that the noise of the flame in the fridge bothered me. :)  Like others have suggested, 2kw of solar may be far more than you will use.


Frank.
 
Kevin Means said:
Solarman, initially, there was indeed a fair amount of "guesswork" as I tried to size our battery bank and solar array to meet our needs. So much so that I hired a guy named Bob Sheerer (AKA HandyBob) to help me. A Killowatt meter was used to determine our A/C usage, and a Trimetric RV 2030 was installed to measure our typical DC consumption. In the end, our system works very well.

handybob.. omg, i'm sorry you had to do that.

As long as we have good sunlight, we go to bed with about a 90% SOC and our house batteries are never below 75% SOC in the morning. During the winter months, our batteries are fully charged by about 1:00 PM if we tilt the panels, 3:30 PM if we don't.

for off grid we use 5 days autonomy, rv probably better to design for 3 to 4 days..
system should be designed to use no more than 15% capacity each day to allow for cloudy days.
this assumes FLA. for agm you can go lower, so one more day with same capacity or reduce capacity

We've got a 2800 watt inverter (85% efficient) that runs 24 hours a day to power our 22 CF res fridge, and those two devices alone consume between 180 and 200 amps per day (depending on how often the compressor runs.) The coach has two macerator toilets, a 40" LED TV that runs maybe two hours a night, LED lighting throughout and we charge our phones overnight. The microwave oven and TVs have A/C cutoff switches to eliminate parasitic draws when they're not being used, and we typically use only our catalytic LP heater to heat the coach vs. the Aqua Hot furnace.

that's good, however, handybob did not teach you that amps per day is wrong, consumption is always in watt hours.


I think we were pretty thorough in determining our actual usage, but I'm open to constuctive criticism. Our goals were to prevent our batteries from discharging below 75% SOC overnight, and to reduce generator run time as much as possible. In the end, we had to increase our battery bank's AH capacity by about 30%, and install 960 watts of solar to achieve that. With those goals in mind, what would you have done differently to determine actual usage when sizing a solar array and battery bank?

you did quite well, however your first mistake was to stick with 12 volts and parallel so many batteries when there is a simpler and lower cost solution. secondly for 960 w and 12 V you will need an 80 amp mppt controller and that's expensive.
moving to 24 or even 48 volts would halve the controller and cabling costs and you would not need to parallel.



It's been my experience that most RVers don't monitor their battery's SOC very well, usually because they have no way of doing so. They just want their solar panels to recharge their batteries, and they don't want to have to think about it. Getting them to measure their actual usage, for the purpose of determining how much solar and AH capacity they'll need, can be a lesson in futility.

Kev

quite true, it is very easy to make costly mistakes with solar, i have an uphill battle on my hands right now.. LOL

 
butchiiii said:
Thank you for the replies.  Question if the panels I get put out a VMPP  of 36 volts I wouldn't
need to run the in series (theoretically) correct? Could they each be independently run to
the charge controller? Won't that reduce the shading problem to only affect whichever panel
is being shaded.  Also the panels seem to have 3 bypass diodes so it seems that partial shading
only takes out some of the panel not all of it.  Does this make any sense or am I thinking about
it incorrectly.
Here are panels that I was looking at.

http://www.amerescosolar.com/sites/default/files/Hyundai%20305W%20Solar%20Panel_0.pdf

https://www.solar-electric.com/lib/wind-sun/AXITEC_AXI_power_72c_310W-330W.pdf

both those panels are good..

let's say you had room for 6 at 300 watts each for 1800 W
then you could connect them as a 2s3p array, that is, 2 in series and three sets in parallel
and feed a 150 Volt MPPT controller into a 24 or 48 Volt battery.
for 24 V you would need a 1800/24=75 amp controller
for 48 V you would need a 1800/48=37 amp controller

in practice you will not get 1800 w so a lower cost 35 amp controller would suffice for 48V

1800 W is unsuitable for 12 Volts as you would need multiple expensive controllers, heavy gauge wiring etc..



 
Isaac-1 said:
Solarman, if not 8 Group 31 AGM's how would you solve the battery problem within the constraints of a typical RV battery bay and budget? 

As I see it 8 6V GC2 AGM's in series parallel might help, but you are still faced with similar issues, just on a smaller scale at 4 parallel banks, 4V or 2V industrial deep cycles might be nice, but then you get sticker shock on the price, and may face other physical limitations.  Flooded deep cycles might work, but often there is a limitation on access when squeezing in more batteries than a coach originally came equipped with...


the stated cost is $300 each.. so eight of them is $2400 and he has a parallel setup that will need to be replaced sooner than he would like. also, i assume he paid for heavy interconnect cabling as he is stuck at 12 volts

from his statement his capacity is 840 Amp/hr at 12V
so the capacity of this bank is  12*840 = 10080 watt hours

his 960 watt panels is outside normal design practice and should really be 24 or 48 Volts, i'll explain in
the tutorial i'm going to post shortly in a new thread

with 10080 Whr  then at 24 Volts would be 10080 / 24 = 420 A/hr
and at 48 Volts would be 10080 / 48 = 210 A/hr

if we went real cheap and used 6 Volt golf cart batteries then a good fit is 48 volts
so he could have used eight of those at $85 each for a total of  85*8 = $680

ok, so this is FLA and not AGM, so lets do the math for agm

again going cheap we could use Duracell agm at $180 ea so that's 180*8 = $1440

the lower spec batteries will most likely last longer than the higher spec'd units
as they will not suffer any of the parallel issues.


the biggest problem i see is the dumb asses that build coaches, they buy and build on a commercial chassis and because
it's 12 volts, they automatically assume that solar is going to be 12v too.. and it's not the case.


 
solarman said:
both those panels are good..

let's say you had room for 6 at 300 watts each for 1800 W
then you could connect them as a 2s3p array, that is, 2 in series and three sets in parallel
and feed a 150 Volt MPPT controller into a 24 or 48 Volt battery.
for 24 V you would need a 1800/24=75 amp controller
for 48 V you would need a 1800/48=37 amp controller

in practice you will not get 1800 w so a lower cost 35 amp controller would suffice for 48V

1800 W is unsuitable for 12 Volts as you would need multiple expensive controllers, heavy gauge wiring etc..


Solarman that makes sense. At the battery output to feed the RV what is used to step the power back down to 12v?
 
butchiiii said:
Solarman that makes sense. At the battery output to feed the RV what is used to step the power back down to 12v?


you need a 48V to 12V buck converter, they are used in the golf cart world and cost about $50 for a 360 Watt unit
that may be sufficient for your needs, larger units are available.
you will also need a means to charge the battery from the mains 120 V and also an inverter to power 120 V appliances,
this is available as a combined inverter/charger/transfer switch, I have one in my own rv.

you throw away or disconnect the POS WFCO 12V converter in the RV




 
Solarman that DC-DC converter may be fine for keeping the control electronics working on the water heater, fridge, etc, but how do you hand cranking the generator from the house bank if it is running at 48VDC?
 
This is rather long, but I feel I need to address a few issues. I didn't have to go to Bob, I elected to. He's an opinionated guy, no doubt, but he's got an extensive background in off-grid solar applications - including RV solar. I don't agree with all his opinions on specific products, but so far, his technical info has been right on.

When I was looking for someone with a strong background in RV solar, I kept encountering his name. I also found a lot of people who claimed to be experts, but their background and experience was dubious at best. They were usually short on actual RV solar experience, and long on the tongue.

There's nothing wrong with using five days as your energy target, but with adequate sunlight, we routinely go a week or more without running our generator at all, and our batteries are fully charged every day. Our holding tanks always end up being our limiting factor.

Designing a system that would only consume 15% of a battery bank's capacity would mean that our battery bank would have to have a capacity of more than 1600 AH, which is about double our current capacity. Our fridge and inverter alone consume nearly 200 amps per day - nearly 25% of our AH capacity. A battery bank with that much capacity would be pretty impractical and in my experience, totally unnecessary. Did I misunderstand what you were saying?

You're right. Bob uses both amp hours and watts when making specific points. When discussing battery capacity I like to use amps, but it's easy enough to convert from amps to watts, or vice versa

Thank you. Our system does work well. However our battery bay does not have the physical height to install, say 6 volt GC batteries with 220 AH capacity. Group 31s barely fit. Assuming that one type, or size of battery will work for everyone's needs is a mistake.

Our last AGM battery bank lasted seven years. I replaced it in January, because it seemed to be taking longer to charge. Seven years boondocking on RV batteries ain't bad in my book. I'm quite sure we'll get more life out of current bank, because they won't be drawn down as far. Why would you say they would need to be replaced sooner than later? How long would you expect flooded cell batteries to last in an RV that was primarily used for boondocking? 

If there is an imbalance in our eight AGM battery setup, it's unnoticeable. The system actually works quite well. It's only two more batteries than came with the coach from the factory. Your assumption that a 960 watt array would need an 80 amp controller is exactly what I assumed, which is why we have an Outback FM-80 controller. That assumption, however, was wrong.

The highest output we've seen in direct sunlight has been about 55 amps. We routinely see 50 amps in the winter months, as long as the panels are tilted. The fact is, a good 60 amp controller would have worked fine. I'm curious as to how many amps your 960 watt array puts back into your battery bank. We've got 8 gauge wire between the panels and the junction box, 4 gauge wire between the junction box and the controller, and 2 gauge wire between the controller and the batteries, so wire length and diameter isn't a factor.

We have six panels and every two of them are wired in series. The three sets of two panels are then wired in parallel. That was Outback's recommendation, because according to their tech rep, the FM-80 is most efficient at converting volts to amps with an input voltage of 40 to 44 volts. That's the output of two of our panels when wired in series.

Lots of RVers have talked about converting their RVs to a 24 or 48 volt system, but few actually end up doing it. There are good reasons for doing it and good reasons not to. I'd be interested in hearing from those who have.

Kev
 
Hello SolarMan,

solarman said:
Now, a battery with a higher impedance will, over time, result in less charge current to that battery, thus resulting in a lower average SOC ( state of charge ), this has the effect of aging the battery and increasing its internal impedance. over many cycles the effect becomes more pronounced. when the battery pack is idle ( minimal load ) other batteries in the string of higher SOC will discharge into the lower SOC battery and self equalize the voltages, but, this effectively makes the whole set equal to the worst battery in the pack.

I'm sorry, but you do mean "resistance" instead of "impedance" above, correct? AFAIK, "impedance" is for AC.

Or do you use "impedance" when referring to the batteries for some special reason, perhaps because on other parts of an RV electrical system ("after" the inverter, and internally in a buck converter, for example) there will be AC and therefore impedance, and you mean to handle some complex interaction of these with the DC part?

(Not trying to nitpick, just trying to learn as you really sound knowledgeable in these matters).

Cheers,
--
  Vall.
 
Hi Kev,

Kevin Means said:
Lots of RVers have talked about converting their RVs to a 24 or 48 volt system, but few actually end up doing it. There are good reasons for doing it and good reasons not to. I'd be interested in hearing from those who have.

If you haven't already, you might be interested in reading this: http://rvnerds.com/2016/01/19/why-does-a-48-volt-battery-bank-make-sense/

This guy has done it primarily because he changed his battery bank to lithium, in fact, a repurposed Chevy Volt battery (!), and these are more easily usable in 48V than on 12V contexts; but he discusses the general whys and wherefores of 48 vs 12V at length.

The above is his main post on the matter, but the rest of his blog is very informative too, there are many other articles on the matter.

Cheers,
--
  Vall.
 
hogwash, battery size and panel watts are dependent on each other

I like to keep my hogs healthy, so I wash them regularly.  ;)

I'll stand by my statement but I think you misinterpreted it. They are indeed complimentary, but both need be sized to meet the daily power demand. The battery bank is sized to store a days energy needs (plus some extra for cloudy days) and the panel is sized to replenish that daily usage based on a typical day's sunshine.  If your power demand is only 1000 watt-hours per day, you don't need a huge battery bank whether your panel wattage is 2000 or 200.  You also need to have some idea of the solar energy available daily in your area - a relatively small panel array can provide enough energy to fill a large battery bank IF there are plenty of hours of sunshine each day.
 
Isaac-1 said:
Solarman that DC-DC converter may be fine for keeping the control electronics working on the water heater, fridge, etc, but how do you hand cranking the generator from the house bank if it is running at 48VDC?

you connect it to the coach starting battery,  that's how I would do it. that way if you have an issue with the house side, you can
still start and use the genny.


 
VallAndMo said:
Hello SolarMan,

I'm sorry, but you do mean "resistance" instead of "impedance" above, correct? AFAIK, "impedance" is for AC.

Or do you use "impedance" when referring to the batteries for some special reason, perhaps because on other parts of an RV electrical system ("after" the inverter, and internally in a buck converter, for example) there will be AC and therefore impedance, and you mean to handle some complex interaction of these with the DC part?

(Not trying to nitpick, just trying to learn as you really sound knowledgeable in these matters).

Cheers,
--
  Vall.

no special reason,

Impedance is just the sum of the DC component, the AC component is reactance.

If a circuit has only DC applied to it, then the inductive reactance is zero and the capacitive reactance is infinite
so we can use the impedance term in a DC circuit if we recognize that the AC frequency is zero.

written in plain text without special characters:

L = inductance, C = capacitance, F = frequency and "PI" = 3.14159... then:

XL = 2 PI FC  and  XC = 1/2 PI FC

in the equations above, if we set  F = 0 then the impedance of both the inductor and capacitor are zero and you are left with DC resistance.

in this context they are one and the same.. :)



 
There's nothing wrong with using five days as your energy target, but with adequate sunlight, we routinely go a week or more without running our generator at all, and our batteries are fully charged every day. Our holding tanks always end up being our limiting factor.

Designing a system that would only consume 15% of a battery bank's capacity would mean that our battery bank would have to have a capacity of more than 1600 AH, which is about double our current capacity. Our fridge and inverter alone consume nearly 200 amps per day - nearly 25% of our AH capacity. A battery bank with that much capacity would be pretty impractical and in my experience, totally unnecessary. Did I misunderstand what you were saying?

the 5 days is what we would normally spec, for an rv i did say 3 would be better.. in your case you are at 2 1/2 days approx..
so that's good..

Thank you. Our system does work well. However our battery bay does not have the physical height to install, say 6 volt GC batteries with 220 AH capacity. Group 31s barely fit. Assuming that one type, or size of battery will work for everyone's needs is a mistake.

this was not a mistake, I suggested a lower cost solution in response to another post as an example.. obviously space
contraints have to be considered in an RV, my own included..


Our last AGM battery bank lasted seven years. I replaced it in January, because it seemed to be taking longer to charge. Seven years boondocking on RV batteries ain't bad in my book. I'm quite sure we'll get more life out of current bank, because they won't be drawn down as far. Why would you say they would need to be replaced sooner than later? How long would you expect flooded cell batteries to last in an RV that was primarily used for boondocking?

your batteries started to destroy each other the moment you used them, i have outlined this already
a flooded set in series of the same type would outlive a parallel configuration by a considerable margin, proven fact.



If there is an imbalance in our eight AGM battery setup, it's unnoticeable. The system actually works quite well. It's only two more batteries than came with the coach from the factory. Your assumption that a 960 watt array would need an 80 amp controller is exactly what I assumed, which is why we have an Outback FM-80 controller. That assumption, however, was wrong.

you didn't notice because you have no way of checking it, coach builders don't care what they do, they are only interested in cost, so they gave
you the best solution for them and the worst one for you, this is a sad fact of the RV industry..

the controller selection btw was correct, one always designs with headroom in mind


The highest output we've seen in direct sunlight has been about 55 amps. We routinely see 50 amps in the winter months, as long as the panels are tilted. The fact is, a good 60 amp controller would have worked fine. I'm curious as to how many amps your 960 watt array puts back into your battery bank. We've got 8 gauge wire between the panels and the junction box, 4 gauge wire between the junction box and the controller, and 2 gauge wire between the controller and the batteries, so wire length and diameter isn't a factor.

the highest I have seen is 905 watts on a cool sunny day in summer at high elevation in Colorado, that's not suprising as panels
output is higher when cold and these are mono types. any other time the best is no greater than 800 watts


even though we have the same wattage, here is the difference in cabling:

I have 10 awg from panels to cc and 10 awg from cc to battery with less than 1% loss and 4 awg from battery pack to the inverter.



anyway, let's just agree to disagree on the finer points and leave it at that ?

 
Thanks Vall. Yeah, I've actually visited that site before, and I agree with rvnerd's and Solarman's logic for a 48 volt system, but I keep coming back to the actual need, cost effectiveness and practicality of making the swap. Yes, the wiring for solar panels is more expensive for a 12 volt system, because of the need for heavier gauge wire, but if done correctly, it's a one-time expense that doesn't exactly break the bank.

Our Magnum 2812 inverter/charger with remote unit is a pretty pricey unit ($2000.00) but the 125 amp charger portion obviously couldn't be used anymore. Kind of hard to swallow that while having to buy a new charger. How much are 48 volt FL batteries with a similar AH capacity? That would probably be a good time to go lithium. My math says I'd only need about 600 AH capacity in Li batteries, but that's still another $6000.00 (minimum).

Then we're looking at mods to enable our AGS work, because it senses low voltage based on 12 volts, not 24 or 48, and it won't work with 12 volt Li batteries either. I'm sure it's technically feasible to overcome those issues, but at what cost? Anyway, I'm sure you understand my concern.

Solarman's comments did prompt me to investigate a different method of wiring our eight batteries together, and sure enough, there is a better way, and it will ensure equal charging and discharging of all eight batteries. I appreciate that.

Kev
 
Kevin:


I didn't have to go to Bob, I elected to. He's an opinionated guy, no doubt,


Yeah, that is common in the solar industry, it seems.


Last winter I tried a place in Quartzsite that specializes in solar. I never got past the outgoing phone message which was something like "If we don't answer it is because this is our busy time of the year and we have customers backed out the door. If you want to leave a message we'll get back to you eventually" or something similar displaying the same 'attitude'. Needless to say, I didn't bother them with my business.  ;D


That is one of the things that I have appreciated about you, Kevin. You help without belittling.

Solarman's comments did prompt me to investigate a different method of wiring our eight batteries together, and sure enough, there is a better way, and it will ensure equal charging and discharging of all eight batteries. I appreciate that.

Do share that with us, please.

Frank.
 
Thanks Frank. By the way, where are you guys? Solarman.. I agree to respectfully disagree on a few points, and offer this as a handshake over the Forum.

Kev
 
Back
Top Bottom