Charging LiFePo4 (LFP) to 3.4V and 3.5V with and without Absorption. What a difference!

[@loganv0410]

A Note at 13:45 - "What am I doing here?" You are educating me. I appreciate that. Fantastic work. Please keep it up!

[@OffGridGarageAustralia]

Thank you very much. Happy to share all my data here on the channel. More to come.

[@richardeccles4997]

One of the most usefull videos I have ever seen on this subject. Actual evidence for the charge voltages instead of just opinion!

[@OffGridGarageAustralia]

Thank you Richard!

[@markstemmett5296]

Found your channel today, at Microcare we have been manufacturing solar equipment for 30 years. I really think you are doing great work for the industry, Thank you and well done

[@pchris6662]

This video should be pinned and bookmarked by everyone just getting started in LiFePO4 battery charging. It’s so helpful to see those curves, and work through the differences between long absorption and faster charge times and how it does and doesn’t affect total charge effectiveness. Thanks so much for taking all that time to document these results! 🙏

[@fritsriep3830]

Very good information. At my home I am using a 12V LiFePO4 6ah battery as a DC backup for my home network. I have a simple power supply to both run the network equipment and float charge the battery and had been trying to determine what would be the best voltage setting to keep the battery fully charged but also not damage the battery. There are many articles which state that you should never float charge an LiFePO4 battery. However, a PowerSonic article stated that you can use 13.8V (equivalent to a cell voltage of 3.45V). Your testing shows that I should be fine by setting at 3.4V per cell (13.6V for the battery), so that's what I am now using. Excellent info! Thanks

[@junkerzn7312]

Excellent testing! Next you need to run the same test at 3.375V, 3.35V, 3.325V, and 3.20V. You will find that the behavior begins to change at these lower voltages, even though they are still above the cell's nominal operating voltage. (And this is why 'Float' voltages are typically set between 3.35V and 3.375V... doing the least damage to the battery while still being able to support loads without having to drain the battery, which would cycle it unnecessarily). You could even test 3.25V and 3.20V, but you will find that these voltages hit diminishing returns very quickly and just don't charge the cell up sufficiently. For a 4-cell 12V battery: 3.20V is 12.8V

[@OffGridGarageAustralia]

Thanks, Matt. As always great advice and input. More testing to come, for sure.

[@sameershinde4594]

This is one of the best video on KZbin explaining important of charge voltage. thank you very much.

[@garycampbell1439]

I really find your videos informative, and I appreciate the time and dedication you take, great to see you following the evidence instead of following the crowd, thanks Andy. You make it easier for us to follow the evidence as well👍👍

[@OffGridGarageAustralia]

I appreciate that! Thank so much Gary!

[@StultusRex]

probably one of the most informative or educational videos on youtube that i've ever watched. ty very much for your time and effort.

[@MrEroshan]

My charge controller charges my LiFePO to 3.45. No absorption. Never discharge below 2.75. Going on 5 years off-grid use. No degradation noticed.

[@mikeholt7881]

Yep, I've told this Guy many times....

[@TheGalifrey]

Exactly what mine is set to, absorption is pointless on LiFePO4

[@BradCagle]

Yup this is it!

[@Rajesh_Singh301]

What's your float voltage set at?

[@woodknack1]

@@Rajesh_Singh301 Ya, float voltage confuses me. I have a Growatt 3000 and it says if set on lithium battery setting (USE2) that the top charge setting and float setting will be set at the same voltage. My batteries are Ampere Time batteries (i have two 12 volts hooked together to get 24v 100amp hour) and they are telling me I should be setting my charger to charge the battery up to 29.2 volts which seems high to me. Even after its fully charged my battery drops down and seems to settle at 26.6v.

[@KevIsOffGrid]

All good Andy. It took me around a year of being off grid to learn what you have detailed and proved here. And it is always nice to see the proofs. A bit like your boat man example, I'm in the UK, the weather is often intermittent; I change the MPPT settings based on weather, need - when it is that week of sun and low power needs being able to set a 3.35-3.4v/cell charge makes me feel I'm protecting the battery, and increasing the lifespan of the battery. I'm sure I've read that avoiding higher voltages can increase the "at 80% capacity" cycles past 5000 even as high as 10,000. For a solar system that's 25 years on the battery and panels, so worth knowing if you can lower the charge voltage and still have the system power what it needs to.

[@OffGridGarageAustralia]

Yeah, right, that's another good example. I guess I'm a bit too spoiled with all the sunshine here. Lowering the voltage will make a difference in cell life over time and if you can do this as often as possible it just makes sense. That's far more important then the compress question in my opinion.

[@marcass100]

@@OffGridGarageAustralia What about having a slightly oversized battery and leaving charging discharging voltages in a middle of a curve?

[@chuckziska988]

@@marcass100 Andy is brilliant. everything is a trade off. If you can afford the $ and space, huge is good. I think Andy is trying to find the best compromise by using what you have in the best manner. Riding that edge without getting sliced. Good service from not gluttonous amounts of resources. Each individual must decide, more WH or more $$ ??? For their system in the long run. Accountants probably call it resource efficiency. But now what does your crystal ball say for future prices of solar and batteries? Interesting.

[@phillipchen7749]

Thanks for these videos, when i first started using this chemistry battery/cells it was almost impossible to find the correct charging parameters. As we say in Jamaica, respect brother.

[@OffGridGarageAustralia]

Thank you very much, Philip 😊

[@rilosvideos877]

This is a very good and valuable test! It shows a LFP cell can already be overcharged at 3.4V if you charge too long and with too little current. If you charge only to 3.4 and let it absorb its probably less stress for the cell. If you charge to 3.5 with a cutoff current high enough (about 0.02C) the cell will be fine but if you keep charging until 0.00xC you will most likely overcharge and damage the cell already a bit. A Voc (open circuit voltage) of 3.5 is already massive overcharging for one cell! You are more on the safe side with 3.4 or even 3.35V with absorption.

[@gumpster6]

This sort of stuff is fascinating to me even though I don't have any of this equipment and likely never will. You are doing things at a level of detail I don't think anyone else is on KZbin. Great stuff!

[@Raphael_Hofmann]

You might want to do a test with different chrage currents...charging with a high current to 3.4V without a CV absorption afterwards, might only give you 60% or 70% SOC. I appreciate your videos a lot! Thanks!

[@diyEVguy]

Good point!

[@OffGridGarageAustralia]

Thanks Raphael. I'll put this on my list for a future experiment.

[@honumoorea873]

I got my charge to 54.4 bulk mode, when it reach that... it goes in absorbtion with 54V... This way, on low Amp charge it will get near 90% SOC in the bulk part and in case of high amps (where apparent voltage is high) the bulk part will stop early but the absorbtion will end it nicely.

[@boscodog4358]

@@honumoorea873 the voltage that you charge to , what is that per cell in your battery bank? Sorry for the simple question 😅 I am very new at this.

[@honumoorea873]

@@boscodog4358 each LiFePo4 cell is 3.2V nominal, 54.4V is for a 16S pack (16*3.4V)

[@FutureSystem738]

Great work Andy. We’ve just been away in our caravan for a full month, where I typically charge to about 3.48v with no absorption time, using a Morningstar MPPT solar controller. We use the battery for everything if solar conditions are good- lighting, Sat Tv, fridge, microwave, air fryer, induction cooking, A/C reverse cycle heating, (a 2.5kW Panasonic inverter split system), electric blanket on a cold night, and even hot water for washing up and showers. I only use LP gas for cooking and hot water if solar conditions are not great, or if the conventional oven is needed. (I bought my 4 x 400 Ah cells, and assembled my system in 2015.) As it’s in a caravan, and we are not using it to live in full time, it hasn’t done a huge number of full cycles, probably equivalent to a hundred and fifty or so. I did a discharge test on our last day away this trip, and found no evidence of degradation whatsoever even taking it down to just 10% SoC according to the Victron meter (360 Ah out of a 400Ah battery, of 4x Winston cells.) Battery voltage was still great on all four cells, and nicely balanced.

[@OffGridGarageAustralia]

Great setup. Thanks for sharing, Dave. That's a fairly beefy installation you have there. How much solar have you got connected to this battery?

[@FutureSystem738]

@@OffGridGarageAustralia Andy I’ve got about 900w of solar on the roof, plus another 350w of portable panels on an 8m lead. The portable panels are a great addition if there is any shade around. In good weather I can get a consistent 60A into the battery.

[@carl2591]

very intresting and lots of time to do the test.. One thing I think we forget sometime is most of the time you are not discharging to 2.5 per cell.. many people try to have enough capacity to power everything to where the discharge is only to 80 or 70% at lowest.. When are not having to recharge from 0 to 100% all the time the time to charge become less in theory.. one your 3.4 v test and 4 hrs abs time if you are only down to 3.2 or 3.0 v then recharge and abs time is lower.. This test has opened a lot of eyes i am sure on how to keep your cell going longer with less deteration.. THANKS mate..

[@mortendiysolarshinybilplei4646]

Hope it comes a new clip soon;) I enjoy too see your clips not only for solar stuff but your seens of humor and your german english ;)👍 Every time I see a new post from you I put on the coffee and chocolate, big blanket and lay down on the sofa. I really enjoy and learn from you sir 👍😉 Best regards Morten from Norway 🇳🇴

[@quasimojo7399]

Brilliant information. Thank you for putting your time and effort into making this video.

[@TheComputerGuyDR]

Great great info. This is EXACTLY what I've been curious about since you started messing with your Victron all those months ago! Can't wait to see more testing.

[@jasondevine6014]

Thanks for your hard work on this. So the question becomes when in a pack with small cell differences how to change charge current based on the highest cell voltage. I look forward to your video on pack settings and why. I think we will see a new generation of charge controllers that interface with individual cell monitoring to really take advantage of this.

[@OffGridGarageAustralia]

Thanks Jason. That would be ideal, if we have charge controllers who know about cell voltages and can charge them individually and correctly. You could almost get rid of the BMS...

[@Corvus52]

There is another thing to consider when charging batteries in parallel. I was charging to a higher voltage 3.6, but the problem was that not all the cells reached the same voltage. Some cells would be at 3.45. I changed my charge profile to 3.45v and now the min/max cell voltages are very close. I’m glad to hear I should still get most of my capacity at 3.45.

[@JoseMVV]

Is not necessary to go to 3,6 on LFP

[@baygon12]

Hi Andy, I have learned more from watching your videos than from any other source of information. I am French and so my English is not really fluent, thanks to your accent and pronunciation I understand everything! Unfortunately, I discovered your channel after ordering batteries and BMS... too bad for the discount but now I think I know how to set up my hybrid inverter. Thanks again.

[@SpeakerKevin]

You can set the cutoff current in the Victron and it will stop charging at that point before the 2 hour absorption takes place. It's called tail current, and you can find it when you're in Expert Mode.

[@OffGridGarageAustralia]

Yes, I know about the tail current. But this would also kick in if the solar is low. It would turn off charging even the cells are not absorbed yet. So I turned off tail current for the moment. It's obviously different with the tester as it 24/7 delivers the same power.

[@lornebonnell2007]

@@OffGridGarageAustralia (not tested) But I would think the tail current measurement would also consider the input power (solar), because how can you have output without input?

[@aureltanasan1413]

Not in my experience. If solar is low the voltage drops and the charger switches to bulk. The cycle starts again until enough solar can sustain the full absorb cycle. This can happen several times a day, depending on solar availability and present loads during absorb cycle.

[@Pim57]

I had the same idea that charging with 3.4 would eventually charge the cell full. I started testing but never finished, so I am very happy that you took the time and effort! Great video and explanation!!!

[@User1462uuw8w]

Is it normal, that in few hours after charge ends at 28,2, voltege goes down to 26,6 - or even to 26,4v What does it mean?

[@muddy11111]

What an effort, the best video so far, now we are learning. My guess is having absorption time on a battery will be better long term as it gives time for the cells to top balance. It would be an interesting experiment if you have a month to spare.... Updating charge control settings to 3.4 with absorption.

[@OffGridGarageAustralia]

Thank you very much! I have changed my Absorption voltage in my SCC from 3.5V to 3.45V now with some absorption time of 10-20min). Further testing necessary though... Solar is so inconsistent and bad for such testing. hence I;m so glad to have this cell and tester now. I'm doing a test and charge to 3.35V with absorption right now. It takes such a loooooong time... ;)

[@dmadsen78]

You should put the Victron shunt in line and see if it matches what your new battery tester says

[@OffGridGarageAustralia]

I've got another shunt I can test against the tester. But then again, I can calibrate the tester to whatever I want. So which one is correct?

[@charleslivingston2256]

What matters is the time from starting charging, not how long it is in absorb. It looks like 20A at 3.5 volts takes about 4.5 hours to start absorb and it's at about 98% at that time. 20A at 3.4 volts entered absorb a little more than 3 hours after starting (and isn't almost full), but by 4.5 hours after starting, it looks to be at about 97% - almost the same. In addition, if charging from panels at MPP (instead of a power source), there is a given amount of power available. If you increase the voltage from 3.4V to 3.5V, the charging current will drop about 3%. As others have said, with multiple cells in the battery, the higher voltage will accentuate the small differences between the cells. The lower voltage will fill more of the capacity of the lowest cell when the highest cell is full (and increasing resistance to charging current)

[@kevineaton9558]

Reference the comment about the boat and keeping the battery charged for the radio etc… it is really just to make sure the beer is always cold. :) ❄️🍺

[@OffGridGarageAustralia]

Hahaha, love that. 100% correct! 🍺

[@calvinflager4457]

I appreciate you doing test to answer questions using equipment not available to many of us. thanks again.

[@cuanblane43]

Great video! The reason Victron have the 2 hour absorption step pre configured in their solar controllers, is related to their LFP batteries and the BMU built into their batteries. The BMU in their batteries top balances the batteries when the cells reach >3.5V, the charging parameters in the MPPT then maintain constant voltage for 2 hours. They do this to ensure their batteries stay balanced.

[@wouterdevos8305]

Thank you. So maybe charge to 3.5 with absorption, say once a month?

[@garyenwards1608]

From what ive been reading most LIPO4 need to reach 14.4V on a 12V battery for the BMS to balance. What length of time I dont know. I have a 200amp Ampere Time

[@Wolfie6020]

Thank you for taking the time to do this. The information is extremely helpful. I used to review flight simulator software and would sometimes leave them running all night on a Pacific crossing to Australia from the USA in a 747 to see how well it modelled the long flights on autopilot. This multi day testing of yours reminded me of it.

[@OffGridGarageAustralia]

Thanks for your feedback. I had all these battery questions myself and could not find answers, so started testing myself. It's good these information help others as well.

[@MrNitrify]

Thanks Andy, great video. Interesting results 👍

[@OffGridGarageAustralia]

Thanks, Thomas 👍

[@Tro1086]

Awesome video, really helps explain and demonstrate what we have all been mucking around in our heads. Really appreciate you taking all the time to run these tests and show us the graphs, helps a ton!!!

[@OffGridGarageAustralia]

Thanks for your feedback. Glad this helps a lot of people.

[@devmich99]

For Victron Lithiums absorption is not to squeeze the last bit of amps into the cells but to give the balancer(s) time to balance the cells.

[@OffGridGarageAustralia]

Well, that's what I thought too. But as along as the charge is supplying power to the cells (which it does during absorption), the smaller balance current of the BMS has no chance to have an effect. With different BMSs on the market these days, we see balance currents of 20mA to 200mA in general. If the battery is still getting charged with 0.5A during absorption, the balancing does nothing. I have found the balancing is most effective during float charge as the SCC will take over all the load and leave the battery alone. This is the only time the balance current can effectively work on these cells.

[@devmich99]

@@OffGridGarageAustralia Not really true for Victron Lithiums. The balancers start balancing at 13.8V and independent of the charge state or charge current.Each balancer takes about 250mA. But yes, it may be different for your BMS.

[@SiriusSolar]

@@OffGridGarageAustralia some top balancers activate above x voltage while others activate only while x amount higher than the other cells and over x voltage. Either way balancing is being achieved whether you're charging or not if the above parameters are met. Easy enough to tell when the resistors are active because they feel warm or you can look at them with a thermal camera.

[@danielardelian2]

@@OffGridGarageAustralia The balancer only has to "burn" the cell IMBALANCES (DIFFERENCES in Ah). The charging current is not that important. Let's say you have a 2-cell battery with 100Ah capacity and 1 Ah imbalance between the two cells. A 200 mA balancer will need 1 Ah / 0.2 A = 5 hours to re-balance those cells. Notice that the charging current is not needed in the equation. If you have a 0.5A charging current, one cell will receive 0.5 A * 5 h = 2.5 Ah and the other cell will receive 0.3 A * 5 h = 1.5 Ah over the 5 hour balance process....and the difference is exactly equal to the 1 Ah imbalance. The purpose of the balancer is to "burn" the imbalance...it is NOT to BYPASS the charging current over some of the cells.

[@uliberndt4255]

Greetings from Germany, ;-) ....it´s very interesting to listen to your adventures in Solar/Pv/ Battery -Land.

[@OffGridGarageAustralia]

Hello there! Thank you, Uli and all the best back to Germany!

[@abelteixeira9544]

Hi Andy, great idea you have, tanks that you dedicate your money, your time and share it with us. Very helpful for me and for sure, to all here that try understand witch are the best parameters for our applications. Lifepo4 are the best. Ps: I'm glad you appreciate my country. Maybe you was only on south but if you can you should visit north Porto, Minho and Trás-os-Montes. Beautiful part of us. Again many thanks for your educational videos.

[@OffGridGarageAustralia]

Thanks, Abel 👍

[@deerfootnz]

My batteries BMS will only start balancing at 3.5v. So absorption time at 3.5v is for balancing, not storing extra energy.

[@joesmith1142]

This is what Battleborn suggests as well on their videos. They suggest charging until the batteries reach 14.2 and then holding them in absorbtion for 15-20 minutes per battery for balancing.

[@ZookeeperJohnG]

Great video, thanks for the efforts! I have changed my plans to charge to the top and will no longer do that.

[@OffGridGarageAustralia]

How far did you charge?

[@Scott-ig1zd]

Brings back memories of testing control valves with the Fisher Flowscanner. The graphs look just like a double acting piston actuator.

[@trevortrevortsr2]

Interesting good to see you showing precise data - Its charging and discharging into the "heals" ( the part of the graph that starts to rise or fall sharply ) that prematurely shortens the life of the pack - years ago I read from Battery Doctor the first 3 charges on my new pack I let the cells go as high as 3.6 but after that it was the 5th, 10th and 50th then ever 50 or 3 months - at all other times I charge to 3.4v and discharge to 3v - it seemed to make sense

[@OffGridGarageAustralia]

Thanks Trev. Yeah, higher voltage, more stress, higher degradation. That's what it seems to come down to.

[@danielardelian2]

@@OffGridGarageAustralia Based on what I've read, LiFePo4 can withstand up to 4.2 volts before it starts to degrade. It's the charging to HIGH SOC that is harmful, not the voltage. But also messing with voltages above 3.40 volts increases the risk of high SOC charging / overcharging.

[@trevortrevortsr2]

@@danielardelian2 I higher voltage will allow more headroom to raise the current charge rate - In terms of Danger Lifepo4 can be over volted up to 4.2v without hazard however it will stress the cells and shorten the life - my RV solar cuts charge at 14.4 and the alternator at 14.2 through a DC to DC unit - so 3.55 - 3.6 - both units allow a modest charge 20a each - which also gives the alternator an easier time

[@lkeil84]

I also have a mitor yacht I spends 4-5 months at a time on in the Bahamas and Florida. Mostly when I am at anchor, I set the charge controller to 14 volts, or 3.5 volts per cell and absorbtion time to 6 hours. I use about 10-20% per night depending on how much cooking we do out of the 1500 amp bank. One night I did turn on the front AC for better sleeping and woke up at 25%. This ks not normal though as it takes several days to get the batteries back up. My large bank is also there because I can be 3-5 days of little sunshine. When charging we'll, it ussually supplies 50-70 amps for a large portion of the day amounting from 2-7 kw a day from solar. When I move locations, the engines will supply about 200 amps more for 1-8 hours depending on the trip. These have vehicle alternators that I have changed the regulators in to stop at 14.0volts instead of the 14.3 volts they shipped with and with temperature compensation to not kill the prematurely. When I am at dock not using the batteries, I change to 13.5 volts, 3.25 per cell just keeping them on standby. This is where I let them remain until I am under way again, sometimes a month later. I could not tell you how this affects my batteries, but it is what makes me comfortable after watching your and other videos. Thanks for all the good work.

[@Shep5847]

On mine, I use absorption a couple of times a year to make sure that all of my cells are above the flat charge curve, i.e. top balanced equally. I change my Batrium BMS settings and I push all of my cells up to about 3.6v and set the BMS to discharge any cells (@1.7amps) that hit 3.6v before any of the rest. By doing this “maintenance” a couple times a year, I am making sure mine are staying top balanced regularly without having to disassemble the pack. If I were to discharge to 2.5v, I would have a handful of Ah beyond the rated capacity. So on a normal day, I get 100% capacity without having to actually go much below about 2.7v or so. I love it! But I rarely get below 20-25% anyways. Unless I’m doing a capacity test. 😀

[@OffGridGarageAustralia]

That's great and with such a high balance current, it makes sense to do that. almost all BMSs have only 20-200mA so there is no point to absorb and balance at the same time. It just does not work. The Batrium is good in so many ways, I should really look into it...

[@Shep5847]

@@OffGridGarageAustralia The Batrium can also interface with a Cerbo and send its data to the Victron system and replace the victron shunt entirely. I actually unplugged my 2k amp SmartShunt a few weeks after installing it and now I get more accurate data in the victron interface than with the smart shunt. However, I haven’t been about to get it to work with the RPi running Venos OS yet.

[@PowerPaulAu]

Andy that's brilliant!!! I came so close to buying one of those testers, but you've just done the exact tests I wanted to see (basically what was discussed a couple of months ago). Excellent results too, it's very clear to see the effects. Huge thanks! I had already decided that during summer, when I can normally reach full charge before midday, I will have a lower charge voltage. And during winter, I will bump that up to push in as much charge as possible while I have the short charge events here in Melbourne. So this confirms the hypothesis. Nice work!

[@OffGridGarageAustralia]

Yeah, I remembered your question from a few months back in our chat and on your website. I think it makes sense, right? The lower the voltage, the longer the absorption time to get to the same SOC. You're making a good point here which I didn't think about before. The difference in solar production through different seasons may require a change i settings in the SCC and BMS. Thanks, Paul!

[@fredio54]

Hey Paul! Two things: 1, You'll be proud of me, I clicked 'Read more' this time. 2, I gave my channel a make over to reflect my experience. I hope you like it, it's just for you

[@georgesackinger2002]

A long desired question answered. Love seeing real time testing. Thanks

[@camielkotte]

This is very clear explained by you. I generosity in information is endless.

[@OffGridGarageAustralia]

Thank you very much for that kind feedback.

[@2flight]

You save us all a lot of time and energy!!! Thank you!

[@RhysOrwin]

Awesome video as always. That's put so many questions to bed. Thank you!

[@OffGridGarageAustralia]

Thank you very much.

[@spiro6363]

Hi Andy, once again, a great deal of testing to answer a very important question on how to best to charge a cell, It’s clear that different environments determine what your configuration will be. Cell chemistry and application very useful many thanks for a very informative and detailed test, Awesome.🙏

[@OffGridGarageAustralia]

Thanks again, also for the donation. greatly appreciated!

[@sichambers9011]

Excellent videos Andy and very helpful to me as I live in an RV and off grid so there's no possibility of testing or taking something to the lab bench. Just got my 4s pack together and it's great to get a reasoned and evidence based approach that provides different perspectives and usages and how you would change the charge profile to meet them. A lot more useful and informative than just being given a number. Thank you!

[@john_in_phoenix]

Nice, I am doing the same thing, but with an Eve 280AH cell at 40 amps. Yes, it takes a while.

[@OffGridGarageAustralia]

Nice! Waiting for mine to come in soon...

[@glennsonne9786]

Thank you for the weekend spent doing this charge to 3.4 vs. 3.5. Very enlightening. I will be expanding my absorption at a lower voltage and watch my string. I think 3.4 is easier on the Anode (OR was it cathode) and will make my cells last longer w/ no loss of capacity. Very well done!

[@OffGridGarageAustralia]

Fantastic! Thanks for your feedback. 3.4V is great, let it absorb there and you still get almost 100% capacity with stressing the cells.

[@ぴーちゃんA]

The same question about absorption and capacity. Thank you for this video!!

[@lawrencerubanka7087]

Hi Andy, thanks for the detailed testing. I really enjoy your videos and learn so much. Keep up the great work!

[@OffGridGarageAustralia]

Thanks, will do!

[@tassieflash63]

Brilliant Andy, thanks so much for your groundwork. This Taswegian will shout you a beer for you efforts :-)

[@jmaus2k]

Great testing. So you want MPPT to charge at 3.4V, but BMS to allow up to 3.65V per cell before cutoff and to start balancing at 3.401V

[@MrSummitville]

How much time does this allow for top balancing?

[@OffGridGarageAustralia]

Actually not. Balancing at 3.65V or even 3.5V does not make sense. I'll do a video soon where I share my thoughts. I also thought to balance cells, the voltage needs to be higher.

[@fredio54]

Great video with really good meaningful real world tests, Andy. The best one I've watched yet from your channel. In contrast, in my application I'm using low capacity high current density cells that are kept at "high" voltage while I'm using my vehicles, but are then subject to slow trickle discharge while parked between uses. In the worst car I get about a week before 8AH is depleted by the alarm, in a few others I have 16AH and they can last for months with no degradation (and no alarms). The nice thing is that they'll still dump out hundreds of Amperes to start the car even if there's only 2AH left in them, so you don't have to catch them early, just any time before the end, really. I don't really get a choice of charging voltage or absorption cut off, but fortunately most of them charge between 14V and 14.4V anyway which is 3.5V to 3.6V. I sometimes run a small capacitor balance board on them to keep them even, but one of them has been going for a few years now without that and hasn't deviated from what it should be and always works. The little balance boards are totally incapable of keeping things in check depending on the difference in state of charge and rate of charging, but it does make for a perfectly even discharge curve right to empty. Note, none of these cells have a BMS because I'm drawing several hundred amps from an 8AH pack and dropping up to about 100A back into them, the BMSes that can do that are not at all affordable and have excessive quiescent current requirements. So I run them naked :-)

[@barryevans5232]

Andy, this was extremely helpful. I have seen so much clearly uninformed speculation about charging LiFePO4. It's really good to see some actual controlled testing productng coherent data resuilts. By the way I noticed that you have an EPever SCC on the bench similar to mine which I really like. I would be interested to see your conclusions about it but have not yet seen your video on the subject. I also use a Daly 100A BMS which is great value and works well, but the SOC reading is pretty useless.

[@oluoginni]

Andy, I have learnt more watching your videos than from any other channels. You are the best at what you do an thank for keeping me and everyone one informed, educated and entertained. You are the best. Thank you 👍

[@OffGridGarageAustralia]

I appreciate that! Thanks for your feedback and kind words.

[@jethrotull5847]

I’m creating my own analog of your example. So far I’ve integrated a pzem-017 300v 300a dc meter with a esp-8266. This methodology will ( with the integration of node-red and mttq ) feed a database and dashboard . Initially it will be on a private network. There are still some issues I have to work out, but IOT is the way to go ! Everyone should learn IOT, the real bonus is solving issues, cheaply, effectively and remotely. I expect my highest cost will be a power supply, with a huge kickback of knowledge. Thanks for you videos, im learning a lot !

[@jethrotull5847]

For detail if anyone wants to start a journey of IOT, I suggest getting a D1 mini ( ESP-8266 ), PZEM-017 ( a shunt based dc power meter ) . Program the ESP using Tasmota “sensors.bin” and find online instructions to wire the devices. All devices are on Amazon. Price varies some but around $20 bucks for both. You will be able to remotely monitor dc readings but at some point you will want to set up a MTTQ (free) server on a raspberry pi 3 or 4, and install node-red. After some work, configuration and swearing, you will feel the power ! ESP chips have GPIO pins and data buses that allow connections to countless available sensors, DC,AC,Temp /Humidity, Lux, UV, Ozone, switch relays, Air quality, everything. Anyhow, I find Andy’s channel the best to understand what I want to do. I give great thanks to his “Kung fu”.

[@ddthames]

Good topic. I have a used battery with some cells that tend to run ahead if charged too high. So I have been trying to find the best settings for get a good charge and keep the voltage under control. So thanks for this test as it is helpful.

[@OffGridGarageAustralia]

Thank you. That is actually a good point what you rise here. If you have mismatched cells, charging to a lower voltage could overcome the problem of single cells peaking and potentially triggering the BMS.

[@christianriedel7017]

I loved your logic about living on a boat and charging to a higher voltage, this made everything click for me as I'm living on my boat and building my first battery bank (8X EVE 280s) as we speak and had always planned on following your lower voltage recommended but this made it so much more clear that I should use the higher voltage in my situation. I'm always on anchor never at a dock and the only solar supplement I have is my engine which I don't want to run a bunch of hours up on. Thank you for all you do. ❤

[@OffGridGarageAustralia]

Thanks so much Christian. It's good to hear that all my testing helps others out. That is great!

[@user-jj2bj4mp7c]

Vielen Dank! Mit diesem Video hast Du einen großen Beitrag zu meiner eigenen Projektplanung und meinem persönlichen KnowHow geleistet.

[@asderven]

Just a suggestion for a super long test :D. Charge to 3.4v/3.5v, absorption up to you, give it a rest of 4~6 hours, dishcarge to 2.7v~3V (so as to not damage the cell), give it a rest of 2~4 hours. That would be what a normal day would look like. See over 2~4 cycles like this, whether the charging and discharging is very similar across the parameters. I know very long, tedious test and boring, but it would be good to see wether these rest periods cause us to lose any significant battery capacity or not over time.

[@OffGridGarageAustralia]

Great suggestion, I'll put this on my list. Thank you.

[@asderven]

@@OffGridGarageAustralia glad to help, your channel has given me courage to try and order the batteries/cells myself. I am still trying to figure out the cheapest way to charge and discharge. Only way I can think of is to use a 24V/48V off grid inverter. That will give me logging capabilities, plus I can test higher loads, I can attach smaller devices like shunts and monitor to get multiple results from the same test. I am more interested in testing the 16 cells at once, rather than in piece meal. Your new battery tester is awesome. I reached out to the ali express folks, if they made a 12v, 24v or 48v, or a higher amp setting one, sadly they don't. The highest variable dc power supply I found was 30V 60A or 60V 30A. Charging up those 280ah cells should be fairly quick with that at 0.2C.

[@charlesadamski_197]

Sir that was an awesome test and it explained a problem I had. Thank you so much. . It's about my camper batteries. I keep them charged with a trickle charger. . I couldn"'t understand why my first 2 days of camping plenty of power but whwn it depleted I could run my engine all night and no holding power very long. I see now vehicle alternators are not concerned about absorbtion. They only care about fast hard punch to get your engine running again.

[@OffGridGarageAustralia]

Thanks Charles. Do you have solar as well for these batteries or just the alternator?

[@charlesadamski_197]

@@OffGridGarageAustralia Mostly Solar

[@TheCrozenfox]

Your tests in your videos are great and helpful. I learned a lot. Plus you are a funny guy with lot of humor, it's nice. Keep it up and hello from France 😁

[@TallTom77]

Andy, by far the best explanation ever! Keep it up.

[@OffGridGarageAustralia]

Thank you. I'm charging to 3.45V now for 30min and fall back to 3.35V for floating. Works perfectly and charges the cells to 99.8% without stressing them.

[@unofficialfuture3120]

You mad genius! Thank you for this testing - I was told by the battery manufacturer to charge my new 12v lipo battery to 14.2v or 3.55v per cell to reach 80%! According to your research I should be aiming at 13.6v or 3.4v to be around 80 %.

[@OffGridGarageAustralia]

If you have LiFePO4 cells, it difficult to charge to a certain voltage. It's all about absorption. If you keep the voltage lower you can still charge to 100% without any harm of the battery. It just needs longer to absorb. So 13.6V should be fine and if you still keep the charger connected at this voltage it will increase the capacity without putting much stress on the cells.

[@frankz1125]

I am glad you are using this charging method. It really is better for your batteries. I said this to you in the comments about 7 videos ago. Cheers

[@STEELROCK11]

That's absolutely right for a 1 cell setup like this. But as soon as you got a few ones in serial, with no absorbtion, you'll have one that peak and stop the charge which make the balancing almost impossible

[@SiriusSolar]

Test takes long time. You wondered why you bother? I appreciate very much your testing results because I've always pondered the answer to this question because I cannot adjust my charge voltage. So I am forced to stop at 3.395 and hold at that voltage until the sun goes down. Now I know I must be getting at least 95%. I've noticed I can accomplish top balancing at this voltage but it top balances a bit slower than I would have otherwise if I had gone to a higher voltage. Thank you so so much for your testing I truly appreciate and understand how long it takes to accomplish.

[@hommerdalor6301]

Hello. If we look at the graph 3.5V+absorbsion, we see that it takes 2:20 h to go +0.075V , from 3.35V to 3.425V but only 20min to go +0.075V, from 3.425 to 3.5V. Also, the balance current is only occurring when there is enought voltage difference between the cells, let's say 0.01V, like the 6A balance board with capacitors on it, for 0,01V, it will transfer 1A, only from 0.06V difference and more, it will deliver 6A. Now if your bms can balance 50mA, 0,05A, you have to be near peak best cell voltage, 3.65V to obtain effect, I imagine.

[@OffGridGarageAustralia]

3.395V is perfect, Trevor. There is nothing wrong with it for longevity of your cells. You will be close to 90% with your battery bank and it will still absorb energy until the sun goes down.

[@SiriusSolar]

@@hommerdalor6301 my balancers are probably more finely tuned than most DIY systems. They are Valence u27-12xp lithium magnesium iron. But thier voltage characteristics are pretty much the same as lipo4. My main system runs 65 kwh. 152cell all in series for 516v float. I also run larger battery Banks of the same brand at 48 volts.

[@crappieconnection]

So, most chargers have us charge the cells to 3.65 each. Is this too high for the good of the cell? should I find a charger that charges to 3.5 or 3.6 per cell?

[@Wktr777]

Great info...thanks for sharing ❤❤

[@OffGridGarageAustralia]

Thank you!

[@timmathewson9111]

I must say this information for me is so so valuable, thank you so much for sharing this video with us.

[@lennieadi]

I'm wanting to have the raw cell lifepo4 battery in jamaica but I'm wanting someone there who understands this chemistry and can top balance whole pack without dismantling. I'm start looking for a electronics engineer there closer to the time when I'm ready to emigrate. Keep up the good work Andy. 🇬🇧🇯🇲

[@OffGridGarageAustralia]

Thank you very much Lennie. Can you not do it all yourself?

[@lennieadi]

@@OffGridGarageAustralia hi Andy. I’ll try but I’ve only learned from you And Will Prowse. I’ve no electronic or Math background what’s so ever and will Be in mid to late 60’s when when I make the move So will need a tonne of Confidence to manage a huge battery of over 50 kWh.

[@swolath]

It’s been 3 days. Need more of your fantastic videos. Learning all the time. And now I know you’re Swedish even better. Skol

[@TG-lp9vi]

The other thing to remember is that new battery will have retained its charge but when the battery gets old many half life the equalizing will be a better way to charge also when the cels are in series equalizing/absorption is preferred.

[@mikeg7924]

What a great video. I especially like the boat analogy. Thank you!

[@jws3925]

I can not clearly read the times at the bottom of the screen. I was hoping they were in 60 minute increments but they don't appear (as much as I can make out) to be standardized minutes between vertical lines. So, my question is this; how long did it take to charge the cell from 2.5v to 3.5v (no absorption) VS how long did it take to charge the cell from 2.5v to 3.4v? I think this is very good information to have but I can not make out the times at the bottom of the graph. Hope you can supply this information.

[@OffGridGarageAustralia]

The scale of the graph changes as time goes buy. It makes it a bit hard to read and check for these information. You can download the graphs from my website and zoom in. 2.5V - 3.5V without absorption: 5:30h 2.5V - 3.4V without absorption: 4:50h 2.5V - 3.4V with absorption: 6:30h off-grid-garage.com/battery-data/

[@j.k4825]

Sehr hilfreich, ersparrt mir eine menge Zeit das Ich nicht alles selber testen muss :)

[@OffGridGarageAustralia]

Gerne 😊

[@SpeakerKevin]

I think the lowest voltage you can use and still get a full charge is 3.37 volts (with absorption of course).

[@davebutler3905]

Very interesting experiments... You are asking the important questions and checking our assumptions. We are told high voltage reduces battery life. Would be interesting to see if 3.3v can charge to 100% given anough time.

[@AveRage_Joe]

Awesome video with GREAT information! Thanks for your time in the search!!!💥🤘

[@Josh-b3c]

I have been doing 3.48v per cell with no absorption

[@OffGridGarageAustralia]

And that is totally fine and will give you good results.

[@FutureSystem738]

Yep, what I do on my caravan battery (400 Ah Winston cells).

[@techmaster_sree]

I have no idea how did I miss this video . Just seen this amazing video and trying to figure it out for my 16S LFP Config used in my main off grid setup. using 52.6V float and 55V Absorption (3.43V per cell with 120minute) on my Victron mppt. I think it will be better if changing 54.4V with 120minute absorption. using only 3kw from the pack so may be its better to put 54V for absorption. Thanks for the great video.👍

[@evil17]

Another great vid Andy,I have been curious about this info since I saw a YT vid about how NASA & ESA lower the voltages of their satellites or mission orbiter space craft to catch a comet, as it could be a 10 year mission to catch up to the comet, so they drop the voltage of the batteries & put the space. Craft into a sleep type mode to preserve the batteries over the mission life. Good to see some real data on this subject and lots of good comments too, I’m sure there are a lot of variables that can be considered regarding boats, home, Caravan, camping,etc . If $ can support it, bigger is better in the battery dept. Thanks for all ur hard work & info Andy.

[@mikewasowski1411]

So interesting! I’ve always wondered the same. 👏👍

[@OffGridGarageAustralia]

You and me both! 😂

[@w3bb0y]

A brilliant video. It really shows practically about bulk and absorption. I feel well educated. I've now deduced based on your teaching to charge to between 3.40-3.45v a cell and discharge down to 2.7. hopefully this means I'm reducing my use on the battery resulting in more cycles.

[@OffGridGarageAustralia]

Thank you very much. I too have reduced my settings for Absorption in the SC from 3.5V to 3.45V and allow a bit of absorption time. It's now all about finding the right settings and understand what they actually change.

[@w3bb0y]

@@OffGridGarageAustralia brilliant stuff. Do you plan to expand your videos by talking about float modes? From what I understand after absorb it lets the battery voltage drop to a slightly lower voltage and maintain that setpoint (constant voltage??)

[@MarcusPocus]

DaLy BMS for LiFePo4 cells start to balance at 3.6v/cell (from them spec), so i think that you miss the balancing stage if you charge below it and balancing will be insufficient if you don’t give any stabilization/absorption time for it.. that mean sense??

[@TheGalifrey]

Don't rely on the Daly, use an active balancer

[@jasondevine6014]

I cooked a pack with a Daly. Don't use Daly with mppt charge controllers as they can't handle the failure mode of mppt charge controllers. The discharge FETs fry when they try and turn off with a failed mppt charge controller.

[@TheGalifrey]

@@jasondevine6014 TBH the Daly is cheap compared to the MPPT :D

[@jasondevine6014]

@@TheGalifrey the point is that if/when the mppt fails it will cook your pack if you have a daly. Ask me how I know!

[@TG-lp9vi]

You check battery capacity without discharging. You can measure the internal impedance of the battery instead. The lower the internal impedance the higher is the capacity.

[@cdkipp]

What a great video. We always debate. Now we have answers.

[@OffGridGarageAustralia]

Thank you very much. I had the same questions for a long time and read so many different 'opinions'. I needed to know!

[@alistairwarburton628]

I think it would be wise consider/acknowledge batteries with onboard BMS... Or for that matter naked cells with an external BMS. Some absorb time is required, as far as I understand it, to allow a BMS to achieve cell balance. Assuming your charger will drop from absorb to float at 3.5 VPC, or some tail current, down to circa 3.35/3.4 VPC the ability of the BMS, to balance will be compromised. Ultramax recommend 3.65 absorb, for 6 mins, dropping to 3.45 float. Both of which sound high to me but that is what they said. BTW 80%DOD / 20% SOC will be approximately 3.225 VPC for these batteries with counting being a better option than voltage for detecting that. I also think, just an opinion, that a BMS, in a battery, will probably skew these voltages when compared to a STACK naked cellS.

[@Tezza66]

do you get a diff, result if you polish the terminals. :p also victron lets the battery absorb for 2hrs to let a basic bms balance out the cells. when no active balancer is used. i.e as in when in a pack. not a singular cell

[@OffGridGarageAustralia]

Yeah, polished terminals are always better! I proved that, right? 😂 I balance at 3.35V and it works fine. I don't believe you need to have a high voltage to balance. It's on my list of videos to make soon. I think there is some widespread misunderstanding about that in the community.

[@SpeakerKevin]

Great video, There is nothing better than real world testing.

[@OffGridGarageAustralia]

Thanks, Kev!

[@MindMeetMaker]

Great video thanks for doing the hard work and sharing 👌

[@OffGridGarageAustralia]

No problem 👍

[@PattyDung]

I have measured two different impedances in LiFePO batteries: one is instant, and it is R, the internal resistance. The other impedance is of magnitude ~2R but appears slowly, with negative exponential time constant ~30 sec. Therefore after several minutes of constant current discharge, the voltage drop (or rise when charging) is ~3R times the current. This second impedence I call "slow polarization".

[@OffGridGarageAustralia]

What devices have you used to measure that? I've got an internal resistance tester too and I'm just waiting for more batteries to come in to test them and introduce the tester for that.

[@PattyDung]

@@OffGridGarageAustralia It's actually simpler than you might think. Simply monitor the voltage while you have a constant current drain (or a constant charge current). Record the voltage drop within the first second of applying the current drain. That tells you the R (which is the instant ΔV/I). Then watch the voltage drop further over the next minutes. Ultimately the voltage drops about 3x as much as the instant drop. Also, when you remove the current drain, the open-circuit voltage rises a little instantaneously (because of R), then more rise over the next minutes.

[@arnoldsolof]

My understanding is that for a lifepo4 battery, the main reason for the absorption time is to allow the BMS to balance the cells.

[@ArthurDentZaphodBeeb]

Really good tests. Very informative. Next question is how higher charge voltage effects longevity. Are you significantly affecting battery/cell life at 3.4v vs 3.5v?

[@OffGridGarageAustralia]

That's a long term study, I cannot really simulate. I would be interested in this too. How much cell life do you sacrifice by charging to a higher voltage?

[@na88egt]

I know it’s been said before but you really are doing some good work.

[@JR24-w3p]

Hi Andy, complete beginner here. Love your videos and your humour. There's a lot of info there to try to get my head around. But in a nutshell are you saying that its best to only charge each cell to 3.4v rather than the 3.65v that seems to a common figure used. I want my batteries (when built) to last as long as possible. Appreciate your thoughts maybe. Thanks