Latest / Logosofia no Dia a Dia / LiFePO4 Batteries: The Workhorse Chemistry for DIY Projects
Transcript
- [Prompt] custom: Let's talk about one of the most useful batteries to know about for anybody working with 12 to 24 V | Hosts: corn, herman [Response] Corn: Daniel sent us this one — and it's exactly the kind of question that separates a project you build once from a project you rebuild every eighteen months. He's asking about LiFePO₄ batteries for twelve to twenty-four volt hobbyist and project use. What capacities make sense, how to actually connect them with XT60 adapters, how to charge them properly, and whether to buy pre-built or go with bare cells. Herman: You've probably killed a few lithium-ion packs by over-discharging or leaving them fully charged. The chemistry just punishes that. LiFePO₄ doesn't. Corn: That's the hook right there. With solar generators, portable power stations, and DIY off-grid projects absolutely everywhere now, the battery chemistry you pick is the single biggest determinant of whether your power supply lasts two years or ten. Herman: Most people don't realize they're making that decision by default every time they grab whatever's cheapest on Amazon. LiFePO₄ — lithium iron phosphate — is the workhorse chemistry, and it earns that reputation in ways that are genuinely surprising if you've only ever dealt with standard lithium-ion. Corn: Where do we start? The chemistry itself, I think — because LiFePO₄ gets lumped in with "lithium-ion" as if it's all the same thing, and it's really not. Herman: LiFePO₄ is a specific sub-chemistry of lithium-ion. The stuff in your phone, your laptop, most power tool batteries — that's lithium cobalt oxide. Most electric vehicles use NMC, nickel manganese cobalt. LiFePO₄ replaces the cobalt with iron and phosphate, and that swap changes basically everything about how the battery behaves. Corn: Cobalt is expensive, ethically messy to source, and thermally — let's say "excitable. Herman: That's one way to put it. Cobalt oxide cathodes can enter thermal runaway at around one hundred fifty degrees Celsius. LiFePO₄ doesn't break down until well over two hundred seventy degrees. You basically can't make a LiFePO₄ cell catch fire by overcharging it. The phosphate-oxygen bond is just too stable. Corn: Which is why you see LiFePO₄ in home battery walls and server rack backups where "don't burn the house down" is a hard requirement. Herman: The thing that really matters for hobbyist projects is the voltage curve. A standard lithium-ion cell has a nominal voltage of three point seven volts, swinging from four point two volts at full charge down to about three volts when nearly empty. LiFePO₄ has a nominal voltage of three point two volts per cell, and here's the key — it stays near three point two volts from about ninety percent state of charge all the way down to twenty percent. Corn: That flat discharge curve is a gift for voltage regulation. If you're building a twelve-volt system, four LiFePO₄ cells in series give you twelve point eight volts nominal, and it barely budges. Your downstream electronics aren't constantly compensating for a sagging input. Herman: Compare that to lead-acid, where a "twelve volt" battery drops from twelve point seven volts fully charged to eleven point five under load. Or standard lithium-ion, where a three-cell series pack swings from twelve point six volts down to nine. Regulating that is a headache. LiFePO₄ just sits there at twelve point eight, hour after hour, until it's nearly drained. Corn: Then there's the cycle life, which is where the "workhorse" label really comes from. Herman: A good LiFePO₄ cell is typically rated for two thousand to five thousand cycles to eighty percent capacity retention. A standard lithium-ion pack — think the one in your cordless drill — is doing well to hit five hundred cycles before it's noticeably degraded. Lead-acid deep-cycle batteries, if you're careful about depth of discharge, maybe three hundred to five hundred. Corn: You're looking at roughly an order of magnitude more cycles. If you cycle a LiFePO₄ pack once a day, you're talking five to ten years before it hits eighty percent capacity. The same daily cycling kills a lead-acid battery in eighteen months. Herman: That's before we talk about usable capacity. With lead-acid, you really shouldn't discharge below fifty percent if you want any kind of lifespan — the Peukert effect and sulfation punish deep discharges hard. So a hundred amp-hour lead-acid battery only gives you about fifty amp-hours of actual usable energy. LiFePO₄ happily runs at eighty percent depth of discharge, so that same hundred amp-hours delivers eighty usable amp-hours. You'd need a one hundred sixty amp-hour lead-acid battery to match it, and it would weigh twice as much. Corn: Which brings us to the practical side of Daniel's question. With that chemistry foundation, let's get concrete — what sizes do these batteries actually come in, and how do you hook them up to your project? Herman: The capacity ladder is worth understanding because it's not just "bigger is better" — it's about matching the battery to what you're actually powering. For twelve-volt projects, you're building a four-cell series pack — four S in the shorthand — and the sweet spot is the hundred amp-hour size. Corn: Why a hundred amp-hours specifically? Herman: It's where weight and usable energy cross in a way that makes sense for most builds. A twelve-volt hundred-amp-hour LiFePO₄ pack weighs about thirteen kilograms — call it twenty-eight pounds — and stores roughly one point two eight kilowatt-hours of energy. That's enough to run a fifty-watt LED light strip for about twenty hours, or a twelve-volt camping fridge for two to three days. And it's still light enough that one person can lug it around without regretting every life choice. Corn: Whereas the lead-acid equivalent to get the same usable energy would be a one hundred sixty amp-hour deep-cycle, which weighs closer to forty-five kilograms and takes up twice the footprint. Herman: Costs about the same upfront, which people don't expect. A decent lead-acid deep-cycle in that size runs two hundred fifty to three hundred fifty dollars. The LiFePO₄ is five hundred to nine hundred pre-built, but it lasts five to ten times longer. The math favors lithium hard once you look past the sticker price. Corn: What about the smaller capacities? Not every project needs a car-battery-sized brick. Herman: The common sizes cascade down from a hundred amp-hours: fifty, twenty, ten, five. A twenty-amp-hour pack is great for portable ham radio setups or powering a Raspberry Pi cluster in the field. A ten-amp-hour pack can run LED lighting for a weekend campsite. The five-amp-hour sizes are basically what you'd use for small sensor nodes or backup power for a single board computer. Corn: On the other end, two hundred amp-hours exists, but at that point you're dealing with something that weighs twenty-five kilos and you're probably better off paralleling two hundred-amp-hour packs instead — easier to move, easier to replace a single unit if something fails. Herman: For twenty-four-volt projects, you jump to eight cells in series — eight S — giving you twenty-five point six volts nominal. The common capacities there are fifty and a hundred amp-hours. Twenty-four-volt systems are less common in pure hobbyist builds, but they're standard in marine and RV setups, and small off-grid solar. Higher voltage means lower current for the same power, so your wiring can be thinner and your losses lower. Corn: Daniel also asked about connectors, and this is where a lot of people trip up — they buy a battery and then realize they have no clean way to plug it into anything. Herman: The XT60 connector is the de facto standard for hobbyist power connections, and for good reason. It's rated for sixty amps continuous with a brief surge tolerance up to about eighty amps. It's polarized — you can't plug it in backwards — and it locks securely with a physical detent. You're not going to accidentally yank it loose. Corn: They're soldered, not crimped. That matters because a bad crimp on a high-current connector creates resistance, which creates heat, which creates — well, problems. Herman: If soldering XT60s isn't your thing, pre-crimped XT60 pigtails are widely available from Adafruit, SparkFun, and across the hobbyist supply chain. The connector was originally designed for RC drone and plane batteries, but it's been adopted across the entire maker ecosystem because it's just a solid design. Corn: The specific adapter Daniel mentioned — XT60 to two-pin — that's the bridge between the hobbyist battery world and the screw-terminal world that most project hardware lives in. Herman: The XT60 side plugs into your battery. The two-pin side is typically a five-point-five millimeter by two-point-one millimeter barrel jack — the standard DC barrel connector you see on everything from LED drivers to small inverters. Some versions terminate in bare wire leads instead, which gives you more flexibility. Adafruit product number five thousand twenty-four is the canonical example — it's an XT60 to two-pin screw terminal adapter, so you land your wires under screws and you're done. Corn: That adapter is the thing that lets you go from "I have a battery" to "my project is running on battery power" in about thirty seconds. Herman: Here's the part people skip that they absolutely should not skip: put a fuse or a circuit breaker between the battery and the adapter. XT60 connectors handle sixty amps, but if that adapter cable gets pinched or the barrel jack shorts internally, your LiFePO₄ battery can dump hundreds of amps into a dead short almost instantly. The internal resistance on these cells is extremely low — sometimes under ten milliohms. A short across a fully charged twelve-volt hundred-amp-hour pack could theoretically push over a thousand amps for a split second before the BMS trips. That's enough to weld the connector contacts together before the protection kicks in. Corn: Which is the kind of failure where you don't get a second chance. A twenty-dollar inline fuse holder with a properly sized blade fuse is the cheapest insurance you'll ever buy. Herman: A practical setup looks like this: battery with an XT60 output, connected to an XT60-to-two-pin adapter, with an inline fuse on the positive lead between the battery and the adapter. Then the two-pin side goes to your charge controller, your inverter, your LED driver — whatever the load is. Clean, modular, and if something goes wrong, the fuse blows instead of the wiring. Corn: The BMS is the silent partner in all of this. Every LiFePO₄ battery — pre-built or DIY — has a battery management system that's watching cell voltages, pack current, and temperature. It's the thing that prevents overcharge, over-discharge, and short circuits at the pack level. The fuse is your last line of defense; the BMS is your first. Herman: The BMS is so critical that I'd say it's the single component you should never cheap out on. A bad BMS will let one cell drift high while the others lag, and over time that imbalance kills the pack. A good BMS — something from Daly, JBD, or Overkill Solar — does active balancing, low-temperature charge cutoff, and over-current protection. The difference in cost between a junk BMS and a quality one is maybe thirty dollars. The difference in outcome is a battery that lasts ten years versus one that dies in eighteen months. Herman: One thing I want to nail down about the XT60 to two-pin adapter, because I've seen it trip people up — there are actually two common variants and they're not interchangeable. The barrel jack version, five-point-five by two-point-one millimeter, is rated for about five amps max. If you try to pull twenty amps through one of those, the barrel connector itself becomes a heating element. Corn: Which is a lesson usually learned by smell. Herman: The other variant terminates in bare wire or ring terminals, and that's the one you want for anything drawing more than five amps. You land those directly onto the screw terminals of a charge controller or fuse block, and the current path is limited by your wiring gauge, not the connector. Corn: The barrel jack version is for low-current stuff — LED strips, small pumps, microcontroller projects. The bare-wire version is for everything else. And Daniel, if you're listening, Adafruit's five thousand twenty-four is the screw-terminal flavor, which is the more versatile of the two. Herman: Let me give you a real-world setup that ties all of this together. A two-hundred-watt solar panel feeding an MPPT charge controller set to a LiFePO₄ profile. The controller output goes to a twelve-volt hundred-amp-hour battery. The battery's XT60 output connects to an XT60-to-ring-terminal adapter, with a thirty-amp inline fuse on the positive lead. The ring terminals land on a DC fuse block, and from there you branch out to a twelve-volt camping fridge, some LED lighting, and a USB charging panel. Corn: That's the kind of setup where you can run a fridge for a weekend, lights for a week, and the whole thing fits in a plastic ammo crate. Herman: If you built the same system with lead-acid, you'd need a hundred fifty amp-hours of battery to get the same usable capacity, the whole thing would weigh twice as much, and you'd be replacing the battery every two years instead of every ten. Corn: The weight difference is worth underscoring. A hundred-amp-hour LiFePO₄ at thirteen kilos is something you can mount on a wall bracket or tuck under a desk. The lead-acid equivalent at forty-five kilos needs a floor and a reinforced shelf and a plan for what happens if it tips. Herman: Lead-acid off-gasses hydrogen when charging. Not a lot, but enough that you need ventilation. LiFePO₄ is sealed, no off-gassing, no maintenance. You install it and you basically ignore it for a decade. Corn: That's the workhorse pitch in one sentence. Now that you know what to buy and how to connect it, the next question is how to keep it alive — and whether you should build it yourself or buy it ready-made. Herman: This is where the "just use whatever charger you have lying around" instinct will absolutely ruin your day. LiFePO₄ needs a charger with a dedicated LiFePO₄ profile. The absorption voltage for a twelve-volt pack — four cells in series — is fourteen point four to fourteen point six volts. That's three point six to three point sixty-five volts per cell. Corn: A standard lithium-ion charger is targeting four point two volts per cell. So if you plug a four-cell LiFePO₄ pack into a charger meant for standard Li-ion, it's going to push sixteen point eight volts into cells that are designed to max out at fourteen point six. Herman: Best case, the BMS sees the overvoltage and disconnects the pack before damage occurs. Worst case, you don't have a BMS or it fails to trip, and you're overcharging iron phosphate cells to voltages they were never designed to handle. You'll get gas generation, electrolyte decomposition, and permanent capacity loss. You won't get a fire — the chemistry is too stable for that — but you will get a dead battery. Corn: The charger has to explicitly say LiFePO₄ on it, or have a selectable profile. Herman: After the absorption phase, the charger drops to float voltage — thirteen point six to thirteen point eight volts for a twelve-volt pack. That's the maintenance voltage that keeps the battery topped off without over-stressing it. And this is where LiFePO₄ diverges from lead-acid in a way that changes how you should think about charging habits entirely. Corn: Lead-acid punishes you for not charging to full. The sulfation starts the moment you leave it sitting at partial charge. Herman: LiFePO₄ doesn't sulfate — there's no lead and no sulfuric acid electrolyte to form sulfate crystals. Partial state of charge operation is not just tolerated, it's actually better for the cells. Cycling between twenty and eighty percent state of charge can push cycle life past five thousand cycles. You only need to charge to full occasionally for the BMS to recalibrate its state-of-charge tracking and do cell balancing. Corn: The old instinct of "plug it in and let it hit a hundred percent every time" is actively shortening the lifespan of a LiFePO₄ pack. The battery wants to live in the middle of its range. Herman: A hundred-amp-hour pack charging at zero point five C — that's fifty amps — goes from empty to full in about two hours. But if you're only topping up from forty to eighty percent, you're looking at under an hour. With solar, you just set your charge controller to stop at fourteen point two volts instead of fourteen point six, and you're operating in the longevity sweet spot without thinking about it. Corn: What about temperature? I've seen people charge these in unheated garages in winter and then wonder why the pack stopped working. Herman: That's the one hard rule with LiFePO₄: do not charge below zero degrees Celsius. Thirty-two Fahrenheit. The internal resistance spikes, and instead of intercalating into the anode, lithium ions plate onto the surface as metallic lithium. That plating is permanent, it reduces capacity, and in extreme cases it can grow dendrites that puncture the separator and cause an internal short. Corn: The battery isn't going to explode, but you're slowly destroying it from the inside every time you charge it cold. Herman: This is where a quality BMS earns its keep. Units from Daly, JBD, and Overkill Solar include low-temperature charge cutoff — they sense pack temperature via a thermistor and simply refuse to allow charging current below zero. Discharge is fine down to about minus twenty Celsius, so you can still use the battery in cold weather. You just can't charge it until it warms up. Corn: Which for a solar setup means either bringing the battery inside, or buying one with built-in heating pads — some of the Battle Born and Renogy units have those now. Herman: Which brings us to the pre-built versus DIY decision. Pre-built batteries from Dakota Lithium, Battle Born, Renogy — these come in sealed IP sixty-five rated enclosures with integrated BMS, often with Anderson PowerPole or screw terminal outputs, and they carry warranties. A hundred-amp-hour twelve-volt pre-built runs roughly five hundred to nine hundred dollars. Battle Born's hundred-amp-hour unit is around nine hundred fifty dollars with a ten-year warranty. Corn: On the DIY side? Herman: Prismatic LiFePO₄ cells — the big rectangular blocks — are commonly available in three-point-two-volt sizes at a hundred, two hundred, and two hundred eighty amp-hours. To build a twelve-volt hundred-amp-hour pack, you buy four one-hundred-amp-hour cells, which currently run about two hundred to three hundred dollars total. Add a quality BMS for fifty to a hundred dollars, plus bus bars, a compression fixture, and an enclosure. You're looking at maybe three hundred fifty dollars all-in versus nine hundred fifty for the Battle Born. Corn: You save about six hundred dollars. What's the catch? Herman: If you don't torque the bus bars evenly, you get hot spots. If you don't compress the cells properly, they swell slightly with each cycle and delaminate internally over time. If you wire the balance leads wrong, the BMS can't do its job. And if you skip the compression fixture entirely — which a lot of first-time builders do — you're taking a battery that could last ten years and turning it into one that lasts three. Corn: The DIY route makes sense if you're comfortable with the electrical work and you're willing to invest the time to do it right. But for a first LiFePO₄ project, the pre-built removes the most common failure points. Herman: The warranty matters. If a Battle Born battery fails in year seven, you get a replacement. If your DIY pack fails in year two because you bought a cheap BMS that didn't balance properly, you're buying new cells. Corn: Let's put hard numbers on the workhorse reputation. A single hundred-amp-hour LiFePO₄ cycled daily to eighty percent depth of discharge — that's three thousand six hundred fifty cycles over ten years — still has eighty percent of its original capacity at the end. A lead-acid deep-cycle doing the same daily cycling is dead in about five hundred cycles, which is under two years. Herman: Total cost of ownership flips the upfront price argument completely. Nine hundred fifty dollars for a Battle Born that lasts a decade is ninety-five dollars per year. A three-hundred-dollar lead-acid that lasts two years is a hundred fifty dollars per year. And that's before you account for the fact that you need a bigger lead-acid battery to get the same usable energy. Corn: The expensive battery is actually the cheap one. It just takes a spreadsheet to see it. Corn: All of that technical detail is useful, but let's boil it down to three things you can do this week to get started with LiFePO₄. Herman: First: for a first project, buy a pre-built twelve-volt hundred-amp-hour battery with an integrated BMS and an XT60 pigtail. The pre-built removes the two most common failure points in DIY builds — bad BMS wiring and unbalanced cells that never got a proper top-balance before assembly. You unbox it, you connect it, it works. Corn: If it doesn't work, you call someone and they send you another one. That's not a small thing when you're trying to get a project off the ground and you don't want to spend a weekend debugging a cell balance issue. Herman: Second: use a charger or charge controller that explicitly has a LiFePO₄ profile. This is non-negotiable. A standard lithium-ion charger pushes four point two volts per cell, which will overcharge a LiFePO₄ pack and either trigger the BMS cutoff or damage the cells. The charger needs to say LiFePO₄ on it, or it needs a selectable chemistry profile. If you're using solar, set your MPPT controller to the LiFePO₄ preset and verify the absorption voltage is fourteen point four to fourteen point six volts. Corn: I'd add that this is the mistake I see most often in forum posts — someone buys a nice LiFePO₄ pack and then plugs it into the charger from their old lead-acid system or a generic lithium charger and then can't figure out why it won't charge past eighty percent or why the BMS keeps tripping. Herman: The BMS tripping is doing its job. It's the battery telling you the charger is wrong. Listen to it. Corn: Third: if you do go the DIY route and build from prismatic cells, the BMS is not the place to save twenty dollars. Get something from Daly, JBD, or Overkill Solar with low-temperature charge cutoff and active balancing. Herman: Active balancing matters more than people realize. Passive balancing just burns off excess voltage from the highest cell through a resistor — it's slow and it wastes energy as heat. Active balancing shuffles charge from high cells to low cells, which means your pack stays balanced even if you never charge to full. For a battery that's going to spend most of its life between twenty and eighty percent state of charge, active balancing is the difference between cells that drift apart over months and cells that stay within millivolts of each other for years. Corn: Low-temperature cutoff is the thing that keeps you from accidentally destroying the pack the first time you leave it in a cold garage overnight and the charge controller kicks on at dawn. Herman: Those features together add maybe thirty to fifty dollars to the BMS cost. Replacing four prismatic cells because one drifted low and reversed under load costs two hundred to three hundred dollars. The math is not complicated. Corn: While we're giving rules of thumb: size your battery so your daily depth of discharge stays under eighty percent. Ideally under fifty percent if you want to maximize cycle life. A hundred-amp-hour battery should not regularly discharge more than fifty amp-hours per day. Herman: That fifty-percent target sounds conservative, but it's where the cycle life curve really flattens out. If you cycle between thirty and eighty percent state of charge — so fifty percent depth of discharge — you're looking at well over five thousand cycles. At eighty percent depth of discharge you might get three thousand. Both numbers are excellent compared to lead-acid, but the difference is several years of additional service life. Corn: The nice thing is you don't need to obsess over hitting exactly fifty percent. The battery doesn't care if you go to fifty-five or forty-five. It's a guideline, not a curfew. The point is just don't run it to the BMS cutoff every single day and call it normal operation. Herman: If you find yourself regularly hitting the low-voltage cutoff, you didn't size the battery correctly for your load. Buy a bigger battery or reduce your consumption. Running a LiFePO₄ pack until the BMS disconnects it is like driving your car until the engine seizes and calling that your refueling strategy. Corn: That's an image I'm going to enjoy. So to recap: buy pre-built for your first project, use a LiFePO₄-specific charger, and if you build your own, spend the money on a real BMS. Plus the sizing rule: fifty percent daily depth of discharge as your design target. None of this is complicated, but all of it is the difference between a battery that outlasts your enthusiasm for the project and one that becomes the reason you abandon it. Herman: One open question I keep coming back to — with cell prices now around eighty to a hundred dollars per kilowatt-hour, we're approaching the point where LiFePO₄ is cheaper than lead-acid on a per-cycle basis. So will we actually see these replace lead-acid in automotive starting batteries? Corn: The C-rate limitation is the barrier, right? Starting an engine pulls hundreds of amps for a few seconds. Most LiFePO₄ cells are rated for one C continuous — that's a hundred amps from a hundred-amp-hour pack — and maybe three C for a brief surge. Herman: A car starter can draw three hundred to five hundred amps for a couple of seconds on a cold morning. You'd need a LiFePO₄ pack rated for that surge, or you'd need to oversize the battery just to handle the peak current. There are high-discharge LiFePO₄ cells that can do ten C or more, but they're more expensive and trade off some energy density to get there. The standard one-C prismatic cell that makes so much sense for solar and hobbyist builds is not the right tool for cranking an engine. Corn: For now, the starting battery stays lead-acid, and the house battery goes lithium. That's already the standard in RVs and boats — a dual-battery system where each chemistry does what it's best at. Herman: I think that split is going to hold for a while. But the broader landscape is shifting. Sodium-ion batteries are emerging as a real alternative — cheaper raw materials, no lithium, no cobalt, no nickel, and they can be discharged to zero volts without damage, which is something even LiFePO₄ can't do. Corn: Sodium is basically table salt. The supply chain isn't geopolitically fraught the way lithium is. Herman: And sodium-ion cells have better low-temperature performance — they can charge below freezing without the lithium plating risk. The trade-off right now is lower energy density and a shorter cycle life than LiFePO₄. We're seeing maybe three thousand cycles versus five thousand plus for iron phosphate. But the cost curve is steep, and several Chinese manufacturers are already shipping sodium-ion cells for stationary storage. Corn: The question is whether sodium-ion eats LiFePO₄'s lunch in the next five years. Herman: I think for hobbyist and small off-grid projects, LiFePO₄ is safe for at least the next three to five years. The ecosystem is just too established — chargers, BMS units, enclosures, connectors, the whole supply chain is built around three-point-two-volt iron phosphate cells. Sodium-ion operates at a different voltage range, so none of that infrastructure transfers directly. And cycle life still matters more than upfront cost for something you're going to install and forget about. Corn: The workhorse doesn't get replaced by the first cheaper option that shows up. It gets replaced when the cheaper option is also more capable, and we're not there yet. Herman: Next time you're speccing a power supply for a twelve-volt project, ask yourself: do I want to replace this battery in two years, or in ten? That's the real question Daniel's prompt gets at, and the answer is almost always LiFePO₄. Corn: Now: Hilbert's daily fun fact. Hilbert: In the early nineteen hundreds, a Scottish carpenter on the Isle of Lewis built a mechanical tide-predicting computer out of discarded clock parts and fishing line. It used a series of pulleys and cams to sum ten harmonic constituents and could forecast tides for any date within a nineteen-year cycle. The device was lost in a barn fire in nineteen thirty-seven, and no photographs of it survive. Herman: A tide computer made of fishing line. Corn: That's somehow the most Outer Hebrides sentence I've ever heard. Herman: This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop, and thanks to Daniel for the question. Corn: If you want to send us your own weird prompt, email the show at show at my weird prompts dot com. We'll be back next time.