DIY LiFePO4, the build begins

Four 280 amp hour LiFePO4 cells ready to become a 12-volt battery

In the last couple of months I’ve installed two of the three main types of lithium iron phosphate (LiFePO4) batteries. In March I installed Mastervolt’s system integrated MLi batteries on Have Another Day and just last week I finished up the installation of Battle Born’s 8D drop-in batteries on another boat. With two out of three types covered I figured it was time to get my hands dirty with the third type, a do-it-yourself build of a 12-volt LiFePO4 battery.

For my DIY LiFePO4 battery, I picked four, 3.2-volt, 280-amp hour cells direct from China via Aliexpress. The batteries took about two months to arrive, which isn’t surprising for heavy items with the current international shipping challenges. The seller says these are brand new cells and they look it, though I’m not sure I’d know if they weren’t. I selected an Overkill Solar 4S, 12-volt, 120-amp BMS paired with my batteries. Although the BMS was back-ordered, it arrived a few weeks later and well ahead of the batteries.

The batteries’ specs aren’t overly clear but they appear to be able to sustain 1C continual load or 280 amps. I don’t have a specific application for this battery yet and felt that 120 amps was likely to be enough for any use I dream up for the batteries, so I stuck with the 120-amp BMS.



Top balancing underway with all four cells wired in parallel

The first step of my build is to get all four batteries balanced at the same voltage. In the picture above all four batteries are connected in parallel and a 3.65-volt charge is being applied. As you can see on the multi-meter, the batteries are about 0.3 volts below that charge so it will take a while for them to reach top balance with all four fully charged and balanced with each other. Once that’s done it will be time to move on to connecting the BMS and physically connecting the four cells into a single battery.

I’m a rookie, so I’m taking it slow and being as careful as I can. If any of you who have done this before have words of wisdom to share, I’m all ears. Otherwise, I’ll just be here waiting for my meter to edge up to 3.65 volts.

Ben Stein

Ben Stein

Publisher of Panbo.com, passionate marine electronics enthusiast, 100-ton USCG master.

30 Responses

  1. Ben Stein Ben Stein says:

    Top balancing takes a long time. I have a 10-amp bench power supply and I’m trying to top balance 1,120 amps of 3.2-volt cells. In about four hours I’ve brought them up .0012 volts, they need to come up about .35 volts.

    -Ben S.

  2. 2,5 A per battery, sure, that is going to take a long time. Lithium cells have a pronounced hockey-stick charge – voltage diagram, so you will find that they remain stable in voltage until they are _almost_ full then the voltage starts rising quickly.

    > I’m trying to top balance 1,120 amps of 3.2-volt cells

    Ben, please don’t be casual with units when batteries and electricity are involved. It is hard enough for readers to keep their heads around this without you writing “amps” when you mean “amp-hours”. You meant to write “1,120 Ah of 3.2 V cells”. “Yes but it is hard” is IMO no excuse when you are a journalist writing about this.

    • Ben Stein Ben Stein says:

      Kees,

      Mea culpa, I did indeed mislabel the unit of capacity on the batteries. Wired in parallel, the four 3.2-volt, 280 amp-hour cells yield a total capacity of 1,120 amp-hours.

      Also, indeed as you mention, given the very stable voltage of LiFePO4 chemistry batteries I don’t expect much fluctuation in voltage until the cells get very close to 100 % state of charge. I’m at about 18 hours of charging right now, which means I’ve probably replaced about 180 amp-hours of energy, perhaps another day or two to go.

      -Ben S.

      • Joel says:

        But as a 12 volt system it will be 280 aH.

        Joel

        • Ben Stein Ben Stein says:

          Yes, you’re correct. The 1,120 Ah figure is only relevant while top-balancing the cells in parallel. So, once this step is done I can charge at 12v nominal (higher actual voltage from the charger) which will cut charge times by four.

          -Ben S.

  3. Joel says:

    Ben, I was thinking of buying a kit similar to yours here: http://lithiumbatterykit.com/.

    Is there any way to judge the quality of the cells?

    • Ben Stein Ben Stein says:

      I don’t know of a way to assess the quality of the cells. I think you could contact the seller and ask if they have any reports on the cells. But, overall, you’re likely placing your trust in the reputation of the seller. In this case, I know nothing about the seller, so you may have a leap of faith involved.

      You’re definitely paying for the top balancing, US warehousing, and kitting the seller has done. My total cost for the 280 Ah batteries and BMS was around $600. There are additional components included from this seller that might add another $100 or so. But, I also have all the tools, wire, and connectors needed on hand which makes the build easier for me.

      The one question I’m left with is how the seller can ship these batteries top balanced. My procedure for top balancing will have me connecting to the BMS fully charged. My understanding of U.S. shipping regulations limit batteries to 30% SOC for shipping.

      -Ben S.

  4. Brian Stannard says:

    I wouldn’t purchase cells that were top balanced by the supplier. LiFePo4 should not sit fully charged. If the supplier top balances how long do they sit on a shelf fully charged – months?

    As far as supplier I purchased Calb cells from one of the main distributors in China. Much less expensive that buying from elsewhere.

  5. Andreas says:

    I had in mind to do exactly what you are doing but I’m stuck on something. How do I charge these with an alternator? When they are fully charged, the BMS will disconnect them and the resulting spike will damage the alternator. Sterling makes an alternator protection device. That might work. I’d prefer a BMS that could signal an impending disconnect…

    • Ben Stein Ben Stein says:

      Andreas,

      First, at least with regards to the BMS I’m using (from Overkill Solar), the BMS doesn’t disconnect when the battery reaches full charge. The cells taper to accepting nearly zero amperage, but there’s not a disconnect event. The only time there’s a disconnect event is when a cell or the battery as a whole violates the over-voltage thresholds. In my normal charging thus far, with the bench power supply set to 14.4v I haven’t seen this event. I did force it by setting the power supply to 14.8v with a fully charged battery. That quickly resulted in the BMS shutting down charging — as it should.

      I think (but I’m still learning here as well so everyone should feel free to chime in) you have several options to avoid damaging your alternator. First, and possibly the safest option because it changes the charging system as little as possible, would be to maintain a lead-acid battery for engine start and then use a DC to DC converter from the lead-acid battery to the LiFePO4. That way your alternator is only connected to a battery that’s not ever going to disconnect. The second option is indeed an alternator saver like you mentioned. I don’t believe the Overkill BMS has an external trigger to shut down the alternator, though it would be nice.

      -Ben S.

      • Richard West says:

        I tried this setup with the alternator charging the start battery and a DC-DC converter charging the LiFePO4 batteries. I used two Victron Orion TR 30 amp DC to DC converters. The problem was that the converters get REALLY HOT and drop their output to less than 20 amps each.

        I’ve since changed the system so that I’m charging the LiFePO4 bank from the alternator (with a Balmar regulator) and am using a Sterling alternator protection device. I am now using one DC-DC converter to change the start battery. This is working out very well and performed flawlessly on a recent five week cruise.

        My before and after system diagrams can be found here: https://twoatsea.com/auckland-boat-projects-3/

        • Ben Stein Ben Stein says:

          This makes a lot of sense when considering the flow of energy through the boat. I’ve been hesitant to go the route you describe only because I like the idea of disturbing the engine 12v system as little as possible. I’m using a Victron Orion TR on my RV to charge from engine start to house (though currently, they are both lead-acid). I haven’t noticed the converter getting hot but it lives in a pretty well-ventialted area that’s seeing a lot of airflow when the RV is moving along at 60 mph.

          -Ben S.

        • Val Vechnyak says:

          Rich, question for you. How big is your Balmar alternator and what AMPs do you see coming out of it?
          I just had a chance to use my brand new DIY 200ah Lifepo4 bank for over a week. As far as charging it I only see 20amps at the start of the engine which drops to sustained 15A. This is going into the battery. I dont know how much is coming from the alternator. I havent had a chance to see if the internally regulated 60amp alternator heats up or the DC-DC Orion drops it because it gets too hot.

          • Richard West says:

            My alternator is 100 amps but I’ve never seen that much. When it’s cold, I can get almost 90 amps for a short time (maybe a minute or two). Then it drops so that I get about 65 or 70 amps from it.

            I have also had to de-rate the alternator using the “belt manager” setting in my Balmar regulator. I have it set to B1. Without this setting, the alternator gets too hot. (I have a temperature sensor on the alternator so the regulator will cut the power if the alternator gets too hot, but the temp climbs fast then the power remains at about 50% for a long time. With the belt manager set to B1, I get longer charging periods at 60 or so amps without overheating the alternator.)

            My Balmar regulator (MC-614) must be an older version and doesn’t have a lithium battery setting. I’m using the “halogen” setting with some adjustments and that’s working well.

  6. Andreas says:

    I was hoping to build four 280AHr battery packs, and wire them in parallel.. that would give me 1120 Ahrs. That’s 13.44kWh 🙂 I could even run a small A/C unit off my inverter for a few hours 🙂
    To recharge that battery bank, I actually considered not using the alternator at all, and just going with shore power/genset and solar charging only. Not as nice as having the alternator charge while we cruise, but very usable. I’ll figure it all out eventually.. I’m also still learning about having several LIFEPO4 batteries in parallel.. each one has 4 cells and a BMS.. what happens when one of the prallel packs experiences a problem and disconnects? I almost need a BMS to keep track of the battery pack.
    Maybe I’m overthinking this.. So much to learn 🙂

  7. Val Vechnyak says:

    Andreas,

    Good question you brought up in your first post. This great article talks about two different alternator damage situations and ways to protect them.

    https://livsailing.com/2019/11/04/protecting-your-alternator-when-switching-to-lithium/

    Still, personally, I would resort to only DC to DC charger simply because I can never be comfortable pushing 80, 120, etc. amps that big alternators put out through the wires on my sailboat.
    My needs are pretty conservative. Even with all possible systems turned on i only use 5-6amp hour. My battery is 200ah.
    With the big battery bank like yours my solution may not work for you.

  8. Richard R says:

    Will Prowse tested these batteries and was impressed.

    https://www.youtube.com/watch?v=3U4ZfQ_IToI

  9. lawrence says:

    how can one take advantage of fast LFP charging to minimize engine run time if they are charging into a LA battery and trying to charge the LFE with a DC-to-DC converter only capable of 50A or so?

    I want a minimum of 250A of steady charging current into my LFP bank from my twin 150A alternators.

    Really don’t understand this.

  10. Christopher says:

    Mastervolt and Victron allow the connection of DC-DC chargers in parallel 5x50A = 250A

  11. lawrence says:

    Thanks Christopher, but 5 converters in parallel? Ouch?

    I am installing the 800Ah Blue Lithium bank to directly be charged by the twin 150A alternators (thru a battery selector switch). The other side of the switch will optionally connect to a 100Ah AGM reserve battery that will be kept charged on a DC to DC converter off the lithium bank. See any issues?

    • Bill says:

      Yes there is an issue. You need a controller on the output of your alternator to prevent it from burning up and actually catch on fire. Your LithiumS are no doubt rated at one C which means they could very well call for 810 amps if you let your Lithiums go all the way down to 10% SOC. This will burn up your 80 amp alternator.

  12. lawrence says:

    Bill,

    I would like to make a point tangential to your concern.

    It is a common misconception that lithium batteries have a higher practical charge acceptance rate than quality AGM batteries such as TPPL AGMs which we have standardized on.

    There are no limitations on charge current on these AGM batteries, whereas marine lithium batteries, by comparison, are severely restricted by the mfgs in regard to recommended charge rates.

    Just look at the specs on the Lithium Blues above. Or even the more robust and heavy duty well tested MasterVolt line.

    People often confuse the longer time it takes to get the last 10-15% into an AGM battery, with it having an inferior Bulk charge acceptance rate.

    I can assure you that our present 860Ah AGM bank, with no limitations on how big a charging source it can be connected to, has the ability to take 300A in Bulk charging mode. And in a much safer manner than lithium. You don’t have to safe guard these AGMs from very large charging sources as you do with lithiums.

    I find it ironic that you can hook up a 300A alternator source to a 50A AGM without issue, but you would typically need 6 lithiums rated at 100Ah each to stay within the mfg’s recommended charging specs.

    Having said that we have temperature sensors on our alternator sources that keeps them protected, yet reduces there full charging performance very little.

    They have been in service for 10 years now and are still going strong. They won’t care WHAT is putting a load on them, only what their internal temperature is.

    I find it ironic that you can hook up a 300A alternator source to a 50A AGM without issue, but you would need typically need 6 lithiums rated at 100Ah each to stay within the mfg’s recommended charging specs.

    Where quality lithium batteries shine is there high current output at low states of charge, hence our interest in them – if they actually live up to that performance promise.

    Weight savings will be appreciated too.

    Best,
    Lawrence

    • Ben Stein Ben Stein says:

      Lawrence,

      I have to respond to several of the assertions you’ve made. I don’t think what you’ve described tells the whole story.

      First, you note that there are no limits on charge current for AGM batteries. That’s not what my experience has shown and it’s not what I’ve seen recommended by battery manufacturers. For example Rolls (http://support.rollsbattery.com/en/support/solutions/articles/4345-agm-charging) states:

      We recommend a charge current of 20% of the 20 hr rate for both Bulk & Absorption charge phases on AGM & GEL VRLA models.

      [10% min, 20% recommended, 30% max]

      That 30% maximum is 0.3C. Quality LiFePO4 batteries are typically rated for 1.0C or 100% of rated Ah capacity. Inexpensive batteries might be at 0.5C but 0.3C for a LiFePO4 battery would be lower than nearly any specs I’ve seen.

      Additionally, I think you’ve blurred the length of Bulk and Absorption cycles when you describe the time it takes to get the last 10-15% of charge. That additional time is because of the lower charge acceptance as the battery approaches 100% SOC. And, it’s a double-edged sword for AGMs, SLA, and FLA batteries because without achieving 100% SOC you’re also not clearing sulfation from the plates and risking lasting damage to the batteries.

      Lastly, I think you’re comparing different charge acceptance behavior and deciding it’s indicative of the superiority of one chemistry over another. It’s true that lead chemistries typically protect the alternator because of their higher internal resistance and lower charge acceptance. It’s a happy circumstance of their charge characteristics, but it’s also part of why they take longer to charge. LiFePO4 and other lithium chemistries don’t have these limits on charge acceptance and as a result, the alternator needs to be protected from killing itself delivering all the power the battery will accept. I think that situation will improve over time. The vast majority of alternators in the world were designed and built long before batteries with differing charge-acceptance behavior existed. So, we will need to adapt.

      -Ben S.

  13. Lawrence says:

    Ben,

    From the MasterVolt 12/5500 400Ah battery manual:

    “Recommended charge current ≤ 120 A”

    That’s 0.3C, no? And this is a high quality lithium battery. And the most expensive I have seen.

    I have never seen a lithium battery with a recommended charging current of 1.0C
    Do you have an example?

    From my Enersys AGM manual:

    “there is no need to limit the inrush current, allowing the battery to be rapidly charged.”

    And I note there is no caveat whatsoever on charge current limits in their specs, as with the MasterVolt battery.

    Also, I am referring to BULK charge acceptance rates when comparing lithium charging rates to our AGMs. The slow Absorption and Float times for the last 10-15% are not relevant when the goal is to alternator charge in the briefest amount of engine run time, such as one does at anchor to make the day’s amp hours for house loads.

    In our case that is around 250Ah. This is easily achieved in one hour as you watch a sustained 250A for an hour go into the batteries, and then you shut down the engine.

    In regards to holding sulfation at bay, 10 years ago the chief technical officer at Enersys told me that a full charge every 7-10 days takes care of that. And I can report it has. Its easy to do on windless passages when the engine runs 8 hours or more. All too common in cruising. In fact, happens several times a week. And of course any time with shore power in a marina does the same.

    And I find this direct quote from the same MasterVolt 12/5500 400Ah battery manual interesting (page 21):

    “To prevent damage the batteries must be charged to the full 100% regularly. It is a common misconception that Lithium-Ion batteries should not be fully charged.”

    Not sure I agree with that as my Tesla manual differs, but it sure contradicts a big perceived advantage of Li over AGM if MasterVolt is right.

    With regard to your last paragraph I am confused by your statement of lithium not having charge acceptance rates issues as with our AGMs.
    From the recommended charge limits appearing in the manuals I have read, I can only conclude its the opposite.

    I think one difference perhaps in our individual perspectives is that I am alternator charging centric, and with our SOP of Bulk charging only, the generalized issues you are referring to simply don’t manifest. I am not trying to make the case for AGM vs Lithium, just trying to stay objective about the application realities of the choice.

    Cheers,
    Lawrence

    • Ben Stein Ben Stein says:

      Lawrence,

      Thanks for the interesting discussion. This is new technology and the dynamics have changed. Lead-acid batteries don’t have a BMS involved,which means the parameters are only about maximizing cell life. In LiFePO4 batteries it’s difficult to know if the limits and recommendations are made to preserve cell life, to protect the BMS, or both.

      First, you’ve correctly pointed out that I didn’t pay enough attention to the recommended versus maximum charge current. For example, a Mastervolt MLi 12/5500 ultra battery does indeed list a 120 amp (0.3C) recommended charge current. But, its maximum charge current is 400 amps (1C). Compare to the Rolls battery specs at 0.2C recommended and 0.3C maximum.

      But, it also seems like Mastervolt is more conservative in how they rate their batteries. Battleborn doesn’t list a recommended and maximum, instead, they just list a single number of 0.5C (https://1t1pye1e13di20waq11old70-wpengine.netdna-ssl.com/wp-content/uploads/2019/01/BB10012-Spec-Sheet.pdf – second page of the spec sheet in an FAQ), Lion Energy also only has one number, but it’s 100 amps for a 100 Ah battery or 1C (https://lionenergy.com/collections/batteries/products/lion-safari-ut-1300). Lithionics shows a recommended charge current of 50 amps on a 130 Ah battery (0.38C) (https://lithionicsbattery.com/product/12v-125ah-g31-battery/) and a maximum of 100 amps (0.77C).

      Mastervolt’s charge recommendation is the lowest of any of the majors, but at 1C for maximum they’re in range with the others. I’m going to reach out and ask them the difference to the battery between recommended and maximum charge current. Does charging at maximum reduce cycle life? Produce more heat? Shorten BMS life? I’m curious to know the difference.

      As for the 100% charge recommendation, first of all that was a good catch. It’s in a weird section of the manual focused on MasterBus events. I haven’t seen any other LiFePO4 manufacturer make that statement or recommendation. And, as you mention, it flies in the face of much of what we’ve heard about ideal use of these batteries. It’s often recommended that the absolute longest life comes from a maximum charge of about 80%.

      Again, I thank you for the interesting discussion and challenging me when I’m wrong.

      -Ben S.

  14. Richard West says:

    The recommendation to charge periodically to 100% might be to allow the BMS to balance the individual battery cells. Balancing works best when the charge current is small, as it is when the battery is almost at 100%. Also, some BMS units don’t even start to balance until a threshold voltage is reached.

    • Ben Stein Ben Stein says:

      Richard,

      Thanks for chiming in. Not sure why it didn’t click for me before. I’m sure you’re right that the 100% recommendation is about cell balancing.

      -Ben S.

  15. Jeff Shukis says:

    For a lithium battery pack, often the maximum charge current is set by the limitations of the BMS, not the cells themselves. For example, I have a 48V, 300AH pack powering an electric runabout. If you look a the battery cell specifications, the maximum charge current is 2C or 600 Amps, but the BMS specs say 200 Amps maximum. So while the cells can handle 600A, the pack as a whole is good for only 200A based on my choice of BMS. Had a spent a bit more, the maximum charge rate could have been higher.

  16. Bottom Balance only on new installs.
    I’m an electric vehicle tech/ emobility specialist.

    Check my channel out, it might help on a future build.

    https://youtube.com/channel/UCbOo4Qc1XOL2juUvUiOp28g

    • Brian Stannard says:

      For a marine LiFePo4 pack – assuming cells with separate BMS – top balance, not bottom balance. Electric vehicles are very different in use than boats. On a boat it is rare to get near the bottom, unlike a vehicle. It is most important that the cells in a marine system are in balance at the top.

Join the conversation

Your email address will not be published.