Choosing the Right Batteries

There seems to be a lot of confusion about how to choose the right batteries. It’s easy to see why – with so many different battery technologies, prices, and specs (both claimed and real), you’d need to immerse yourself in the data to make an informed decision.

Well, I did that for you. It’s what I do.

So we can get through this, and I can get to the important point quickly, I’m going to stay on the fairway here. If you want to get into the weeds, I’m happy to do that with you – just reach out to solar@offgridforever.net and let me know what’s on your mind.

Background

Three major chemistries currently available: flooded lead acid (FLA), absorbed glass mat (AGM), and lithium (LiFePo4, Li-ion, et cetera). Note that this is not an in-depth treatment of all the options – these are general classes and not every class is represented, but these are the basic options I would consider right now for renewable energy.

Flooded Lead Acid

OLD OLD OLD technology. The FLA battery was invented in 1859.

  • Brutally simple
  • Expensive to ship
  • Laborious to maintain, have to keep (unfrozen) distilled water on hand and check/fill weekly
  • Needs equalization
  • Cheap (but probably not cost effective any longer)
  • Shortest life expectancy of the three
  • Heavy as shit
  • Off-gasses hydrogen, needs venting to outdoors
  • Spillable
  • Can freeze when discharged or when it’s really freaking cold out

Absorbed Glass Mat

  • Still the same basic technology as a FLA, but with significant improvements in how it’s applied
  • Electrolyte paste soaked into a fiberglass matt
  • Non-spillable, can be mounted in any position
  • Barely any off-gassing, if any. I install these in enclosed spaces without issue
  • Also heavy as shit
  • I haven’t found a need to equalize, but I also don’t run below 70% SOC very often
  • Does not tend to freeze. At least, I’ve never seen one freeze
  • Can ship non-hazmat and arrives charged since non-spillable
  • Higher life expectancy than FLA. 6 years is typical, but you can push further if you don’t need high performance out of them
  • Very cost-effective

Lithium

  • Most expensive to get into of the three types described here
  • Longest life expectancy, often with warranty up to 10 years
  • Non-spillable, non-hazmat, can be mounted in any position
  • Zero maintenance
  • Can be finicky about voltages. Make sure to get one with a built-in BMS and it will basically take care of itself, though
  • Have a battery-powered drill? Love it? Yeah, lithium batteries, baby.
  • Most cost effective barring any new technology coming out
  • Light as shit. 31 pounds for 100 usable Ah vs. 73 pounds for 50 usable Ah from an AGM. If you’re oscar mike with your power system, this can really matter. If you’re stationary, maybe it doesn’t matter to you at all.

Let’s do some MATH

Math is the only way to figure this out, so let’s build three renewable energy systems and see which battery system wins. This will depend, of course, on what we consider winning, so let’s define that first.

Winning, to me, is best energy value over time for dollar spent, assuming you’ve got the $$$. Your mileage may vary, but this is based on extensive experience living off-grid, both mobile and stationary. All bets are off if you’re grid tied. We can talk about that another time.

Let’s say you need 15KWh usable energy from your system and have got to be able to pull at least 4000W continuous from your batteries with a surge of 6000W. This should be enough to power most small homesteads and even larger RVs minus the soul-sucking AC units they come with. Unplug that shit.

We’ll use the 20-hour rate (since most of us function on a 24-hour day, that makes sense, right?) and a depth-of-discharge that will eek the best life out of the battery type in question. We’ll also take into account how that battery charges and the variation in sun hours throughout the year in common US-latitudes. If you’re equatorial, good for you. Let’s talk about how that changes things some time.

Battery Bank Sizing and Costing for Various Battery Chemistries:

AGM

My choice is the Universal Power Group 12100. I love these batteries for the right application (non-mobile).

  • 20-hour rating: 100 amp hours
  • 12V
  • operate between -20 and 45 degrees C
  • 64 pounds
  • $169

I don’t tend to discharge these below 70%, but you can run them much lower if you don’t mind shortening their life. I had a set for 6 years that still worked pretty well. Definitely could have kept using them, but I sold them as perfectly good trolling motor batteries and got half my money back out of them to buy new ones. Opinions vary, but that’s a great deal in my book.

Calculating at 50% DOD, we’ll need 30KWh of battery bank capacity to get our 15KWh usable.

12V X 100Ah = 1200Wh per battery. 30000Wh needed/1200Wh per battery = 25 batteries. Since any sane person running a 4000+W system is going to build with 48 volts (I’m not necessarily sane, but let’s pretend for the moment), let’s go with 24 batteries. We’re a tiny bit short on bank size, but it saves us buying 3 extra batteries.

24 batteries X $169 = $4056. Tax and delivery not included.

* Note that delivery can be by UPS or FedEx (or by me), as these are not HazMat, and no upcharges.

So, $4056 for the batteries. Any old charger will work, let’s use a Midnite Solar Classic 150 and use a “panel factor” of 1 for the AGMS. This is a variable I have made up – panel factor – that means how you need to adjust your array wattage to deal with battery losses and resistance to charging. AGMs fall in-between FLA and Lithiums, so we’re giving them a “1” to representy them as the basis for comparison.

If we think we’ll be using 15KWh per day, we’ll need to have about 25% more panel than will supply that on our lowest average sun-hour day, often right around the winter solstice. Here where I am (where we have a solar score of 94) we have an average of 3.4 sun-hours per day on December 21st. 15KWh/3.4 = 4400W * 1.25 = 5500 watt array.

Let’s say that our array will cost us $1/watt, so $5500. We’ll use $1 per watt for all systems, so if it’s off it doesn’t matter. The comparison will still hold.

  • Batteries: $4056
  • Panels: $5500
  • Charger: $678
  • Total: $10,234
  • Expected lifespan: 6 years * 365 days per year * 15KWh / day = 32,850KWh
  • 31.15 cents per KWh average lifetime equipment cost
  • No labor upcharge, as these batteries don’t need filling.

FLA

My choice is none, but we’ll go with what people seem to gravitate toward so we can do a real-life comparison. We’ll use the venerable Trojan T-105, designed for golf carts and used by early renewable energy peeps because there was nothing else but submarine batteries available. We’ll need two charge controllers, since we won’t be able to get all the power in through one due to the high “panel factor” for FLAs. You can use this same basic info for L16s. They are just bigger and heavier. LOL.

  • 225 amp hours at the 20-hour rate
  • 6V
  • 62 pounds (without electrolyte)
  • $186
  • Must ship by truck freight – or I can pick up/deliver (not free), but they are not UPS or FedEx-able.
  • Do not let these discharge too much when it’s cold – they can definitely freeze and then they are junk. Since they need to be vented to the outside, this can be a stumbling block.

If you want these to last, you’ll need to have distilled water on hand and keep them full with at least weekly checks, more often if you are running them hard or before and after equalization. I would recommend never going below 60 SOC or 40% DOD for maximum lifespan.

Calculating at 40% DOD, we’ll need 38KWh of battery bank capacity to get our 15KWh usable.

6V X 225Ah = 1350Wh per battery. 38000Wh needed/1350Wh per battery = 28 batteries. Unfortunately, 28 is not divisible by 8 (sets of 8 batteries needed to get 48V) so we’ll have to add 6 batteries to reach 32 and we’ll have four banks of 8 batteries. I’d go down to 24, but we’d fall very short of needed capacity. Also, capacity will decrease somewhat quickly over the years with these batteries, so the extra capacity is recommended.

32 batteries X $186 = $5952. Tax and delivery not included.

Because these self-discharge pretty quickly and are much slower to take a charge, I’m giving these a panel factor of 1.25. That means we’ll need to upsize the array by 25%. On our 3.4 sun-hour day, that will be 15KWh * 1.25 = 18.75KWh / 3.4 hours to get our charge = 5500 watt array * 1.25 again for panel realities = 6875 watt array.

  • Batteries: $5952
  • Panels: $6875
  • Charger: $678 * 2 to handle the wattage of the array
  • Total: $14,183
  • Expected lifespan: 4 years * 365 days per year * 15KWh / day = 21,900KWh
  • 64.76 cents per KWh average lifetime equipment cost
  • Whatever you value your labor at, or labor costs when you are away for more than one week, to maintain these batteries.
  • Whatever distilled water costs, as well as the time and vehicle to go get it or the energy to make it.

Lithium

My choice is the Battle Born 12V. I love these batteries for the right application.

  • 20-hour rating: 100 amp hours
  • 12V
  • operate between 25 and ? (hotter than you want to be) degrees F. These will be damaged if you charge them at low temps!!! The BMS (battery management system) in a good quality lithium should control this and protect your batteries.
  • 31 pounds
  • $899

These can be discharged to 0%. Really. Well, this is what they’d have you believe. The reality is that you can discharge them 100% of their rated capacity, which is 100Ah. The actual capacity of the battery is more, but the BMS will stop the discharge after 100Ah. Let’s plan on not using more than 80% so we don’t have our little energy system go down while we’re binging on NetFlix or making bread in the Kitchenaid or whatever we’re doing.

Calculating at 80% DOD, we’ll need a 18.75KWh battery bank capacity to get our 15KWh usable.

12V X 80Ah = 960Wh per battery. 18750Wh needed/960Wh per battery = 20 batteries. That works out nicely for a 48V, 24V, or 12V system and we’ll have plenty of reserve if we need it.

20 batteries X $899 = $17,980. Tax, not included, but delivery is free.

I recommend a lithium-capable charger for lithium batteries, not just a lead-acid charger with some settings tweaked or disabled. Let’s go big with this Victron that will handle whatever is needed.

I’m giving lithiums a panel factor of 0.8, but it’s likely closer to 0.75 or even 0.7. These things charge FAST and hold that charge incredibly well. So, 15KWh /3.4 = 4400W * 0.8 = 3600 watt array.

  • Batteries: $17,980
  • Panels: $3,600
  • Charger: $834
  • Total: $22,414
  • Expected lifespan: 10 years (well, that’s the warranty!) * 365 days per year * 15KWh / day = 54,750KWh
  • 40.93 cents per KWh average lifetime equipment cost
  • No labor upcharge, as these batteries don’t need maintenance.

Conclusion

I invite you to reach your own conclusions based on your needs, or reach out to me and we can walk through every option in detail for your particular situation. My choice for stationary is AGMs, for mobile systems of any real size is lithiums.

Thanks for reading! I hope this has been helpful to you and I look forward to hearing your thoughts!

Winslow

solar@offgridforever.net

Leave a Reply