Great video again, Andreas! I have used an AAA size LiFePO4 cell in an outdoor project (temp/hum/pressure sensor), charged by a small solar panel. I'm using that red charger board with TP5000. I'm located in Norway, and it has seen temperatures down to -10 or lower - without any apparent problem. The LiFePO4 cell still seems to work well after 3 years. (It may have lost some capacity, I have not tried to measure.)
@@TheChopain Good idea! But as the battery works well also at cold temp, I will continue like this. I forgot to note that I replaced resistors on the charger module to reduce the charging current. I charge at around 1C for the AAA cell.
@@lambdaprog The device wakes from sleep each 15 minutes, using design features derived from Kevin Darrah's Trigboard - so zero consumption while sleeping. Wake time is around 2-3 seconds including Wifi transmit. Using an 11x15 cm solar panel it *almost* gets me through the winter: I have had to recharge the battery once each December. It helps of course that I am located as far south in Norway as you can be.
"when the student is ready the teacher appears"- Loa Tzu. I was looking for LifePO4 battery pack for electric bicycle project and your video appears magically. Thanks
I replaced my li-ion (Mn) pack on my bike with LiFePO4, and it works great, but the big downside, as mentioned in this video, is that you really need a BMS or other tool to track the amount of energy that goes in and out of the battery to get a reliable reading of how full the cells are. The bike itself has 4 bars, and it's usually stuck at the 3rd one for 85-90% of the pack's capacity, and then stays at 2 for a bit, and after that the voltage gets drained so quickly you'll only get like 2-3 minutes before the BMS cuts it off
To avoid confusion: *LiFePo are Li-ion batteries* . "Li-ion vs. LiFePo" is only a colloquial distinction, but won't help you when looking for technical or scientific information. It is pretty much the same as the other often used colloquial distinction between "Lipo and Li-ion", labels which were wrongly assigned to pouch cells vs. cylindrical cells (the contents are pretty much the same). Li-ion is the name of all cells that use Lithium to store energy, which LiFePo do as well. The difference is that the cathode material contains iron-phosphate instead of the usual cobalt/nickel/manganese oxides. The rest is pretty much the same.
@@AndreasSpiess That's the thing. There isn't one "other" type either. Have you ever seen something like IMR, INR, ICR, IFR, LMO, NMC, NCA, NCO, LCO or LFP written on cylindrical cells or in their datasheet? Those signify different chemistries (some of those shorthands are the same, just different by manufacturer). Some use aluminium (NCA) or no cobalt (LMO), but the most used chemistries are LCO (lithium cobalt oxide, also ICR) and NMC (nickel manganese cobalt oxide, also INR). LCO are those that gave li-ion batteries the bad rap for needing so much cobalt. This chemistry is what is used in phones, laptops and pouch cells (the latter colloquially being called "li-po"), but has peaked and is being replaced by NMC. NMC uses a mix of the 3 mentioned metals. Maybe you've seen "NMC 111" or "622" or 811? That's the ratio of the metals (yeah even within a chemistry there's further division). The cobalt content is reduced a lot in these. That's what is used in many car batteries. If you've seen a maximum voltage of 4.35V on a cell it was most likely this type (has to be built for it; no any NMC can do that). Of course there are no clear cut lines where which type will be used (everyone is trying to use less cobalt), and for low current use (
Thank you for your explanation! After your comment I also looked a little into it. My problem is: I do not want to use 50% of the video's time to correctly name two types of batteries. Most of the viewers understood what I was talking about. And this counts for me. And I fear that, if I would use the correct names, most people would not understand what I am talking about :-( One possibility for the future for me is that I talk about 4.2 and 3.6 volt batteries because this is what matters for an electronics guy. One thing a lot of people like on this channel is that complicated things are made easy. This one seems to be one of the tougher problems to solve for me.
@@AndreasSpiess Yes, I fully understand that. My comment wasn't so much about you changing your content, but more for those people that go to the comments to maybe find useful information. Video content ist mostly colloquial and calling them "li-ion" and "life" is fine. Maybe just a single sentence at some point that "life" are "li-ion" as well, just sufficiently different from the "usual li-ion" batteries to get their own name.
@@superdau Thank goodness someone reacts to this LiPo/Li-ion miscalling. As a battery systems engineer, it always makes me wanna correct people :) Of course it's not only the anode/cathode composition in terms of what elements either, the ratio also important.
I love my LiFePos. No regulators needed to drive my ESP stuff, quite safe handling, ideal for my lower power / long life sensors. I can run my doorsensor for several months with one charge from LifePos. I use Li Ion only if there is no way around it ...
@@rickhuntraslam2925 Smart phone makers want you to buy a new phone every 5 years :-). The marketing is capacity-centric rather than robustness. But I would really love to see a cell phone with a LFP battery in it. I would happily take slightly reduced capacity for it. -Matt
@@junkerzn7312 LFP has a very high price still currently a lipo battery 100Ah is less than 100 bucks in alibaba, a battery of that same capacity in Lifepo4 costs about 350 to 600 dollars, So I prefer to buy 3 lipo batteries for applications and change them every 5 years( 3 x 5=15 years) than a single lifepo4 battery that lasts 8- 10 years
Great informative video and thanks for sharing. Let's hope to see some ready made ESP32 development boards like the Lolin range with built in Lifepo4 support
This was very useful, thank-you. I'm looking to replace the Ni-Cad cells in my military Clansman radio and have been given some LIFePo4 battery packs. Your information has given me some ideas for producing a safe and efficient project.
Hi Andreas, Thanks for a good talk on batteries, very relevant for makers. I have noticed that discharge at low temperatures does not affect these batteries much. I bought some camping lanterns and measured their light output while running them flat in an environmental chamber. The temperatures were 20, 0 , -20 DegC. The internal resistance rises slightly at cold temperatures, which resulted in reduced peak output but extended the discharge time as a consequence. The average energy yield appeared to stay the same. It was very interesting testing. Kind regards, South Africa
Andreas, you are amazing. First the video on 3D printing cases exactly when I needed it. And now this very week-end I was experimenting with different battery set-ups and this video surfaces. I'm beginning to think you have mystical powers ;)
Thank you! For me, the improved safety of LFP cells is a big reason I prefer them wherever they make sense. Also - I've had enough tech damaged by leaking 1.5v single-use cells that I'm experimenting with using pairs of LFP cells and spacers in AA and AAA sizes for that kind of application, especially in devices that don't seem to work well on NiMH rechargeables.
@@AndreasSpiess the problems I have with AA LFP in common electronics are the usual inaccurate battery level and no undervoltage protection. It would be nice to have a protection circuit in the "connector", otherwise I noticed it is very easy to over discharge them.
Great video. BTW, LifePO4 batteries are fully discharged at 2.5 volts, not 3.0. There’s not a lot of energy left between those voltages, but to be accurate your graph should have shown that.
Great overview and summary, Andreas! Even Tesla will change/has changed their battery packs and there are already Tesla models released (initially only for China but now also in Europe) that run with LiFePo4 batteries. And even more interesting, they are operated with a heat pump to have a most efficient heating of the batteries in cold environments. So with your video you are again leading edge! Thank you for sharing!
@Juerg: Other viewers also confirmed that Tesla seems to change chemistry. I always asked myself what happens if 500 e-cars are in a Garage and one starts to burn... It seems easier to extinguish burning gas.
@@AndreasSpiess It's already reality! in Germany the Tesla3 models manufactured in China already run on LiFepo4 cells with a special monitoring procedure of the BMS to switch off the battery in case the temperature increases significantly above the environmental temperature. I understand this as a fire protection measure.
Grüezi aus DE lieber Andreas! Ich habe zwar persönlich ausserhalb von PC-Service bisher wenig mit Elektronik zu tun, deine Videos machen aber Lust drauf, mal selbst den Lötkolben in die Hand zu nehmen. Vielen Dank und weiter so!
I have used LTO batteries in my 6.6kwh Powerwall, I know the chemistry isn't ideal for your projects, but it has the greatest change and discharge rate of 6 minutes, also withstands the coldest of all temperatures on charging cycle, extremely safer then Li-ion and Lifepo4, and will have over 20,000 full charge cycles at 80% still left in capacity , I have pulled over 2000kwh out of it , I have even has 2 cells go into reverse voltage before I installed a BMS which seems to have done no harm to them, I think this LTO battery bank will out live me and my grandchildren.!.
I have a few cells in the lab but so far did not use them because, as you said: Their voltage is not ideal. In addition I di not find an "TP4056" style board for this chemistry :-(
Thanks for the excellent update. I think I'm still best served by the unprotected li-pos (sold for RC use), as most of my applications fly on model planes and thus weight is critical, but they also tend to place large pulse loads on the battery. I may very well move towards lifepo4 anywhere else I can though. Their safety is a compelling feature.
I have made a LiFePO4 battery from individual cells; which is nearly three years old, there is no need to fear this battery technology, as long as you have designed it to work within manufacturers guide lines and you have a good Battery Management System (BMS). BMS is a major topic in its own right, with active and passive devices being hotly debated.
I believe the most important difference between the lifepo and others is that need for regulator. Although you mentioned the need for the component in case of higher voltages I think you missed one crucial aspect and that is the efficiency and idle current of the regulator. If you need a regulator in your battery operated project it is crucial how much quiscent current it is speced and what current it needs when loaded, this needs to be taken care of at design and requires testing of the overall system. For lower currents an ldo suffice but for higher loads a switching supply is needed with low efficiency in non-loaded cases giving you terrible headaches. Its a huge benefit when you can connect directly a battery to your mcu/system and not having to deal with optimization and or alternative designs. For me it would be a nobrainer to change to lifepos once they will be widely available. I don't know why the battery operated system designer are not pushing this technology when you can really save 1-3$ component on your board and increase lifespan.
It doesn't seem quite safe enough for some applications. For example, an ESP32 has a maximum Vdd of 3.6V, and the LiFePO4 is charged at 3.65. But if you have an LDO, it needs an exceptionally low dropout for any 3v3 parts (though the esp32 itself is OK down to 1.8, the modules often have a higher limit - perhaps the memory eeprom needs more).
For 5v projects I would also suggest looking at the real new kid on the block, Lithium Titanate (LTO) their working voltage is (unless you are putting extreme loads on small cells) around 2.6-2.4 volts, so with 2 of them in series, my 5v stuff that I tested it on work absolutely fine, and they appear to be as safe as LiFePO4 (Some claim they are even safer). Slightly lower energy density though, but they claim 10 000-20 000 cycles on it. (havent been able to test that myself yet ;) )
I have a few of them in my lab. But as you write, the voltage does not really fit. Too high for 3.3 and too low for 5V (2 cells). It seems they are better in cold conditions. So maybe this is an application. Otherwise I do not see a big advantage for small projects.
@@AndreasSpiess I've had some luck with 5v projects with 2 of them in series (large cells so that they can run long without recharging). Currently they are absolutely not cost effective, but I figured I'd give them a try to determine if they live up to the claims. Maybe I just got lucky with parts that aren't all that sensitive voltage wise :)
I really would like to see a easy solar charge controller for LTO. I'm thinking about installing a project in an extremely remote location where the only way to access is by helicopter. If I could have the LTO long life span and stability it would be perfect.
I bought a 105AH LiFePO4 Drift battery from Fogstar in the UK. BMS is on board and there's a Bluetooth connection to an App so the battery tells you how it's feeling. It replaces two 104AH lead acid standby cyclic batteries. I power my amateur radio and TV activities with it including power amplifiers. The lead acid batteries would decline in charge really quick but the Fogstar Drift does the whole day's operation no trouble! Also my back loves it. My battery pack is a quarter of the mass it was. Yes it and the fast charger by Victron energy are expensive. So are all the sulphated and now useless lead acid leasure batteries and inflated LiPo battery packs. 73 M0YDH
Also a big thank you from Switzerland (also if for once the infos were not new to me). Nice summary and overview though! In the table at the end of the video (13:00) there are two errors: IMHO the discharge current of LiFePo4s may be even higher than the one of LiIon (because of the lower ESR) and the charge time could also be much lower than the one of Li-Ions... Keep up your great work and best regards from the Kanton Bern! :-)
Nice video. At around 2:10 you use 'amps PER hour' instead of 'amp-hours' for capacity twice. For the rest, I have learned a lot. Thanks. p.s. Tesla is switching from Li-Ion to LiFePo4.
I'm pretty sure this is because Tesla decided that at a certain point we can have much safer batteries in cars without too much of a tradeoff in energy density (I still think that EV's need a much lighter and faster charging and longer lasting battery chemistry to overtake the ICE in every metric aside from pollution after manufacture).
@@plemli I think you are right, but it is not absolutely clear. He could be saying 'amps per hours' or 'amperehours'. Closed Caption also doesn't know what he says. Maybe it is just his Swiss accent.
As a long term user of lifepo4 batteries. Charging is not a problem, get boost or buck converters with current limiting set em to 4 volts per cell and the current limiting to something appropriate and just leave it connected. 4 volts won't hurt them and the'll just settle at that and the current draw will drop to near zero. You'll need to watch the low volts cut off. The discharge curve is fairly flat but not as flat as your graph shows. I can estimate mine to within around 10% The'll rapidly fall from say 4 volts to about 3.4 volts and then gradually discharge to about 3.2 volts and then fall away quickly. Almost all the capacity falls betwen 3.2 volts and 3.4 volts They are affectively fully charged at 3.85 volts but you want a bit more to drive the current in. You can make a convincing 12 volt battery by using 4 cells, You cant with standard lithium ion.
Thank you for sharing your experience. I would not charge them higher than 3.65 volts because this is what the specs say. As you write, there is not much more energy stored if you go higher.
Very good video, now I know a lot more about 2 battery tecnologies, for me I loved to know that lifepo4 does not flame cause this is an issue when using li-ion, on the other hand I always want to know the battery capacity, but connnect a lifepo4 to my project without a need of regulator is a great thing, so to choose what use will depend of the project kind, Thank you Andreas for this valuable information, bye!!!!
The claim to fame for LiFePO4 (also known as LFP) batteries is cycle life, a flat discharge curve, low internal resistance, large voltage tolerances, and a nice failure mode. All at the cost of some capacity verses NCA or NMC. (LFP has slightly less capacity due to the lower voltage). I really love the chemistry. * Cycle life is roughly 4 x vs NCA or NMC * The discharge curve is very flat, which is especially nice when you are pulling high currents. * Low internal resistance means the battery has almost no chance of overheating. * Large voltage tolerances. A cell is typically charged up to 3.65V (and settles down to 3.2-3.3V). State of charge depends on the C-rate but you basically get it fully charged if you stop at 3.65V, then let it settle. Electrolyte break-down voltage is way up at 4.2V, so there is a massive tolerance between fully-charged and break-down. * Failure mode is simply a slow loss of capacity over time. No instant failures, no sharp failure curve. So much so that used LFP batteries are often resold as lower-capacity B-grade cells. However, you still have to be careful about bloating if using prismatic cells. Poorly manufactured batteries and batteries that are abused in their first life can wind up building up gasses inside the cell, bloating the cell. Cylindrical LFP cells are more resilient to this effect, but are harder to deal with (e.g. do you go with four big prismatics or 64+ cylindrical cells? Most people go with the aluminum-cased prismatics. * Slightly lower capacity than NCA and NMC lithium ion chemistries. But if you don't care for the application, all of the above positive points trumps the slightly lower capacity. A well-maintained LFP battery will easily last 15 years. Even longer. -Matt
Very informative video, as always. I'm planning a PCB with a CN3058 chip onboard and a small PV panel. This charger chip has a TEMP pin, where you can connect an NTC thermistor, disabling this way the charging at subzero temperature.
Super caps have a very low capacity per size. They are very good for high currents and many charge cycles. All things we do not need for our small projects. They have other uses, though.
indeed LiFePO4 are annoying when you try to measure the capacity because the battery have a really stable 3.3v nominal voltage. but that stability is a bliss for an mcu! 3.3v directly connected without a buck converter/regulator? heck yeah!!
Page 3:30 remember the internal resistance of LiFePO4 cell? It has the lowest among all candidates. It is least appreciate in low current applications. However is important to (very) high rate charge and discharge applications such as high power tool and electric motor cars. LFP is more tolerance to high temperature c/d cycles than its counter parts.
I looked at the use case "small projects" where the maximum charge/discharge rate is not so important. So I did not investigate a lot into this matter.
I kept many broken laptop battery packs (18650 cells), so I saved money each time opening those for 50-80% recycling as it's just 1 cell or the circuit board which killed the pack and maybe caused extra cells to go under, but not all.
Hi Chris, Yes I agree that FePos are cool 😎... I used them a long time in my model planes... If you are flying around 10 minutes with the motor with normal performance this is a perfect choice... You can charge them wit 6c and discharge with 6c nearly infinite times... But you should keep them warm +30°C... The weight is higher but you can even discharge them with effective 30c... I think 200 cycles are possible... Later I used a 120c premium LiPo and after 30 cycles with 30c you see the performance dying... Currently I am using LiPos due to the better power/weight ratio... At the 6s 5000mAh level the difference is significant! So I strongly disagree with your comparison of charging and discharging rates! The reason why you do not find any charging boards is because you do not need them... Take a stabilized voltage supply with max 3.6V and be happy...
@@AndreasSpiess I was thinking of using a current generator (0.1A) switched on and off with an arduino to check the tension every minute, but I may buy a correct charger...
Since LiFePO4 batteries don't leak (as well as not catching fire), I've been replacing pairs of AA and AAA batteries with a LiFePO4 and a blank in all our devices. I lost a $2000 USD device to leaking alkaline batteries, so I don't ever want to go through that again. So far, they have been perfect. With a few spares (kept charged monthly), a quick swap keeps devices in service. Also, Tesla is switching to LiFePO4 for some of their vehicles.
LiFePO4 batteries as well as any other battery always leak. The leakage is determined by how well manufacturer manufactures the cells and how likely it is for the battery to leak through building pressure.
Also other viewers mentioned that Tesla is changing the chemistry. I can imagine having hundreds of Li-Ion batteries in a parking garage could be dangerous if one starts to burn...
@@richardlighthouse5328 Some chemistries, like alkalines, use the case as one of the electrodes. As a result, they are consumed over time, leading to leakage. That is not the case with LiFePO4 batteries.
The video does make it clear what was meant. I was offering a tip, I didn’t mean to criticize. I suspect that it might come up in a conversation where it’s not clear, so it’s worth knowing the US usage. I’m sorry if I offended.
A video on how to integrate the TP BMS on a larger battery pack would be appreciated. I’m working on (future project) 16s 7p pack for a Honda e-cub and the standard multi BMS leave me wondering if I’m missing a trick. Shout out to Great Scott who provided a link to your channel
One great usage of LiFe is the direct replacement of Pb battery of motorbikes or cars, a 4 elements LiFe battery can replace a 12V 6 elements Pb battery, with about 14.4V at maximum charge given by alternators.
Thank you Andreas, great information. Sadly, I am halfway through designing something, when halfway through your video it turns out that I can't find flat packs for lifepo4.. strange, your video is 3 years ago but I still can not find flat packs. Is it true?
There is also a great material advantage for LiFePO4, they don't use cobalt, which outside being expensive to mine has significant additional environmental issues as the countries where cobalt is extracted (outside of Austalia, Canada and the US) have very poor environmental track records and also tend to use child labour under health hazardous circumstances.
I have been using a LiFePO4 32700 battery for a humidity sensor for a while. It lasts half a year on a charge and I love the idea that it won't spontaneously combust and burn down the house. Of course, that's only because its a 12 million MAH Chinese cell and has nothing to do with tuning the ESPs deep sleep and WiFi settings ;-)
I think they currently are a niche product. For small projects they are hard to use because of the voltage. I did not find standard chargers or protection circuits for them.
Is there an error on charging time? At 1:32 you present a chart in which Fast-Charge Time is faster for LiFePO by about 2x but at 12:50 the chart shows Charging time is longer for LiFePO. Have I missed something? Yes, I do understand that for "our projects" charging time is relatively unimportant.
Yes, I wasn't convinced by that. My experience is Li-ion/Li-po can do >1C charge but you have to monitor temperatures to he safe. I charge LiFe cells at 5C with no special precautions.
You are right. There are several Li-Ion technologies. Some are optimized for extremely high C rates. The LiFePO4 batteries I have are not particularly optimized for in that respect. This is what I meant in 12:50. But always the datasheet counts.
There was a test tittled 'LEO life tests of ABSL batteries using SONY 18650HC' that managed to use those lion cells for 75000 cycles, I understood that this due to not discharging the cells too deep, meaning that even lion cells are entirely viable if you just rather shutdown your device than deep discharge them and don't overcharge them.
@@AndreasSpiess availability for both cells and chargers, which thankfully is being leveled by LiFePo4 cells and chargers being more available. It also comes with the implication that more than likely the LiFePo cells can also do the same thing with even less loss of capacity (the Sony cells had lost 30% after 75k cycles). So if you need to leave a sensor in the wild for years, how deep the battery cycles go can mean the life or death of the cell, so less maintenance is required.
Nice overview ! Btw, never trust LiIon packs.... sooner or later they'll all inflate, it's just a matter of time. But so do many other battery technologies. I've been working with lead/acid for years, eventually they too will inflate. If you're lucky they don't burst.
LFP can be used and charged from near 0% to 100% with no degradation. To maintain a reasonable cycle life, LiIon are best kept between 20% to 80% only going outside that range when maximum usage is needed. This considerably reduces the claimed capacity advantages of LIon.
So you said the BMS will protect the cell even when charging. Does that mean I don't need a charger as long as I have BMS for all my LiFePo4? All I need is a voltage regulator for that charging voltage?
I use my lab power supply where I enter the maximum voltage of the battery and the maximum current I want to charge the battery. If you use a charger with a higher voltage it will reduce its voltage to the max. current. As soon as the battery voltage is at the maximum the BMS cuts off. I would not use this method with a higher voltage for very high currents, though.
I'm impressed. I will be glad when drone batteries last longer (flight time) and do not have to be monitored and maintained when not used for extended periods....1 month of none use.
Interesting video. I wonder if you could understand how some apparatus don't need charge in 10 years of operation. In Italy our wireless toll system is a small low power battery operated transceiver that communicates with the two vertical antenna system on each side in the toll gate lane. There is no issue working in temperatures from 0 to 60°C. No damage to the battery.
@@AndreasSpiess yes they do have, if a remember correctly, a one cell li-ion battery about the AA size. There is also a buzzer than makes two short sounds when the communication is ok. I'm not convinced that they are RFiD because the receiver is typically placed under the center mirror in the car and that's usually over 1,5 meters from the reciever on the toll lane.
@@santopino2546 The Norwegian toll road devices are RFID without battery, same location in the car. I believe it is a SAW (surface acoustic wave) device.
@@opsahle2 i suppose, at this point, in our device, that the battery is only for the buzzer. Some kind of low self discharge battery. I've just searched the web and read about some RFID with a 5m distance communication, that incredibile.
You can also see electric cars switching from lion to lifepo4 due to the availability of iron over nickel. Because the batteries are cheaper it makes sense for lower end models where 300 plus miles of range are not needed.
LFP cells will become more predominant mid next year onwards as a patent on LFP expires then. This was highlighted in a video by "The Limiting Factor" titled "A Brief History of LFP". LFP is desirable because there is no Nickel or Cobalt; both in very short supply. Lithium is the only constraint and that's mostly a processing constraint. This matters greatly in vehicles where there are a lot of cells used.
Almost all hobby chargers support LiFePO4. For desk chargers the cheap LiitoKala single and double bay ate OK. The single bay Opus analysing chargers are great for testing. The 4-bay Opus also handles LiFePO4 however you need to access the internal switch to change it for LiIon, LiPo and LiFePO4.
I have 2 lifePO4 12v 8Ah I bought the first of this month. I bought a charge you have to set type of battery number of cells, voltage, etc. I tried to see the charger work. When I hook either battery up and try to start the charge I get a cell warning on both and no charge. My other charger gives me a cell balance problem. What do you think? I just got one of the Miady 20ah batteries today. The charger is at my office so haven’t had a chance to try it with either charger.
The chemistry has to be the same on the charger. Otherwise, it assumes wrong voltages. My charger, for example, does not support LiFePo4, only standard Li-Ion batteries.
I would love to see a video about coulomb meters / gas gauges / fuel gauges for batteries and how to use them in projects. I've only came across MAX17043 that wasn't super expenisve, but is only for a single cell.
These MAX17043 just measure the voltage and would not work with LiFePO4. Coulomb meters measure the energy transferred in and out of the battery. Maybe I will make once a video about it. However I do not see a big need for them for small projects because usually we just need a signal to recharge the batteries and we fully charge them before the next deployment.
FYI: I am going on 7 years for my GoldenMotor LiFePO4 36 Volt, 12 Ah battery Pack on my EBike. I have absolutely no idea how many charge cycles it has gone through :-) I have also resurrected A123 LiFePO4 cells from the dead (i.e. no detectable voltage).
Hi, I have a questions.as for the max output voltage of LiFePo4 batteries are different than Li-ion let say 18650 batteries. How do i connect to brushless motor ESCs that has a rated battery voltage ( can accept wide range from 2s to 6s ): Example 6s of 18650 can produce 25.2V max, while 32650 LiFePo4 is 21.6V max. How do Electronics components do accept the voltage actually? Does it matter across different chemistry. Is the ESC can handle it okay by its own even the 6s voltage of normal lipo or liion is not the same with 6s of 32650 batteries? As ESCs usually connected to li ion or lipo. Or i actually need a separate component to add in between to jump up the voltage to match the desired 6s of 25.2v of esc. 5s of 18650 li ion= 21V 6s of 32650 lifepo = 21.6V Does the esc runs well when connected to a lifepo4 pack 6s2p with a bms then just connected directly to the ESC with regular XT60 connector for example. Thank You.
Your esc does not care about the chemistry, only the voltage. Check the specified input voltage of your esc and decide how many batteries with the chosen chemistry you need. The connector also does not matter.
My LiLyGO ESP32 SIM800L module doesn't power using LiFePo4 cell (at 3.6V) when I connect through the back built-in connector.. There's also no V-in pin.. Is it safe to connect that 3.6V to 3.3V pin directly to power?
My solar pole light used LiFePO4 batteries in 18650 size. The one that came with it said 2000mah and I was unable to find a replacement with that high a capacity anywhere. I was able to fit in a larger cell that had I think 3600mha.
Hello Andreas, I have been following you for years; thank you for all your support and knowledge, This is off topic but my be interesting, have you heard about the thread protocol? it is supposedly supported by the ESP 32 chips. it's and new open and supported mash network by many companies, But I can't find any information about it.
Thread is similar to Zigbee and will be supported by the (not yet available) ESP32-H2. It is too new and I will cover it when it is ready. BTW: Thread is a stupid name for a new protocol because this name is used for completely different purposes.
so it is ok if i put a solar panel (5.5v 1.8w) -> TP5000 board -> 14500 lifepo4 -> ESP32 connected on the same pin of the battery while it is charged by the TP5000?
Guten Morgen Herr Spiess, I am looking for a LiFePO4 charger, but I can't really find one. I have 4 LiFePO4 cells in parallel and I want to charge at the same time. They're 14505 in size, have a capacity of 600mAh and have a standard discharge rate of 0.2C; to my understanding that means, that I need to charge them each with 120mA for a total of 420mAh at 3.65V. Right? Devices such as the TP5000 say in the datasheet that they support AT LEAST 1A charging current, which is more than double what I apparently need. How would you solve this problem? Thank you very much in advance!
You should be able to change a resistor Rs (you find the datasheet with google) to reduce the max. charging current. On most boards it has R100 (0.1 ohms)
I am looking at Protection Board you linked... specs say Over discharge voltage: 2.1 ± 0.05V... As far as I know, that is way too low... Discharging to 2.1 V will harm batteries and propably to loose some capacity. I was using some LiFePo4 for years, but only with low voltage protection at 2.8V.
I also have such a board and I do not use it. Everything below 2.5 is a no-go for me. Below 2.9 volts there is anyway only a small amount of energy left.
Can i replace a 3.7v 18650 li-ion with an 3.2v 32650 that has a bms on it ? In the case of a USB fan that i have, the internal battery is connected to a 18650. I also have a few LED lanterns that use 18650 that charge via usb charger. Ive also seen some garden solar lights using 3.2v lifepo4... wonder if direct replace would work.
@@AndreasSpiess I'll have to buy a few and play around i guess. In the case of my USB fan and LED lantern. Normally they fully charged and run for many hours then turn off when the voltage drops below 2.5v i guess. i'm hoping it will work the same as the lifepo4 that keeps the voltage more steadily around 3v.. then only let the Fan or LED lantern switch off once it's below 2.5v Maybe the last 10min or 20min before battery is dead. Charging them up again via 5V power input, the BMS should keep battery's safe. only providing 3.3v till full. then protect it from being overcharged and over voltage. Thanks for the feedback
Great video again, Andreas! I have used an AAA size LiFePO4 cell in an outdoor project (temp/hum/pressure sensor), charged by a small solar panel. I'm using that red charger board with TP5000. I'm located in Norway, and it has seen temperatures down to -10 or lower - without any apparent problem. The LiFePO4 cell still seems to work well after 3 years. (It may have lost some capacity, I have not tried to measure.)
Thank you for sharing your experience. Low temperatures created a lot of discussions in past videos.
For a temp sensor with a solar panel you can automatically disable the charging when temperature drops bellow 0 by using a transistor.
@@TheChopain Good idea! But as the battery works well also at cold temp, I will continue like this. I forgot to note that I replaced resistors on the charger module to reduce the charging current. I charge at around 1C for the AAA cell.
Solar... in Norway? Explains the AAA size.
@@lambdaprog The device wakes from sleep each 15 minutes, using design features derived from Kevin Darrah's Trigboard - so zero consumption while sleeping. Wake time is around 2-3 seconds including Wifi transmit. Using an 11x15 cm solar panel it *almost* gets me through the winter: I have had to recharge the battery once each December. It helps of course that I am located as far south in Norway as you can be.
"when the student is ready the teacher appears"- Loa Tzu. I was looking for LifePO4 battery pack for electric bicycle project and your video appears magically. Thanks
I concentrate more on the low power applications. Bicycles are a bit higher power…
@@AndreasSpiess Perhaps.. but we found the video "useful" and "interesting"!
I replaced my li-ion (Mn) pack on my bike with LiFePO4, and it works great, but the big downside, as mentioned in this video, is that you really need a BMS or other tool to track the amount of energy that goes in and out of the battery to get a reliable reading of how full the cells are. The bike itself has 4 bars, and it's usually stuck at the 3rd one for 85-90% of the pack's capacity, and then stays at 2 for a bit, and after that the voltage gets drained so quickly you'll only get like 2-3 minutes before the BMS cuts it off
To avoid confusion: *LiFePo are Li-ion batteries* . "Li-ion vs. LiFePo" is only a colloquial distinction, but won't help you when looking for technical or scientific information. It is pretty much the same as the other often used colloquial distinction between "Lipo and Li-ion", labels which were wrongly assigned to pouch cells vs. cylindrical cells (the contents are pretty much the same).
Li-ion is the name of all cells that use Lithium to store energy, which LiFePo do as well. The difference is that the cathode material contains iron-phosphate instead of the usual cobalt/nickel/manganese oxides. The rest is pretty much the same.
What is then the right name for what I named Li-Ion and pretty much everybody understood? I am no Chemist ;-)
@@AndreasSpiess
That's the thing. There isn't one "other" type either. Have you ever seen something like IMR, INR, ICR, IFR, LMO, NMC, NCA, NCO, LCO or LFP written on cylindrical cells or in their datasheet? Those signify different chemistries (some of those shorthands are the same, just different by manufacturer). Some use aluminium (NCA) or no cobalt (LMO), but the most used chemistries are LCO (lithium cobalt oxide, also ICR) and NMC (nickel manganese cobalt oxide, also INR).
LCO are those that gave li-ion batteries the bad rap for needing so much cobalt. This chemistry is what is used in phones, laptops and pouch cells (the latter colloquially being called "li-po"), but has peaked and is being replaced by NMC.
NMC uses a mix of the 3 mentioned metals. Maybe you've seen "NMC 111" or "622" or 811? That's the ratio of the metals (yeah even within a chemistry there's further division). The cobalt content is reduced a lot in these. That's what is used in many car batteries. If you've seen a maximum voltage of 4.35V on a cell it was most likely this type (has to be built for it; no any NMC can do that).
Of course there are no clear cut lines where which type will be used (everyone is trying to use less cobalt), and for low current use (
Thank you for your explanation! After your comment I also looked a little into it.
My problem is: I do not want to use 50% of the video's time to correctly name two types of batteries. Most of the viewers understood what I was talking about. And this counts for me. And I fear that, if I would use the correct names, most people would not understand what I am talking about :-(
One possibility for the future for me is that I talk about 4.2 and 3.6 volt batteries because this is what matters for an electronics guy.
One thing a lot of people like on this channel is that complicated things are made easy. This one seems to be one of the tougher problems to solve for me.
@@AndreasSpiess
Yes, I fully understand that. My comment wasn't so much about you changing your content, but more for those people that go to the comments to maybe find useful information. Video content ist mostly colloquial and calling them "li-ion" and "life" is fine. Maybe just a single sentence at some point that "life" are "li-ion" as well, just sufficiently different from the "usual li-ion" batteries to get their own name.
@@superdau Thank goodness someone reacts to this LiPo/Li-ion miscalling. As a battery systems engineer, it always makes me wanna correct people :) Of course it's not only the anode/cathode composition in terms of what elements either, the ratio also important.
Nice vid, I'd only capitalize the O for oxygen, it's phosphorus and oxygen, not polonium
hehe radioactive cathode!
You are right!
I love my LiFePos. No regulators needed to drive my ESP stuff, quite safe handling, ideal for my lower power / long life sensors. I can run my doorsensor for several months with one charge from LifePos. I use Li Ion only if there is no way around it ...
I am on the same trajectory ;-) But I often use these TTGO LoRa boards because they are so handy. And they require Li-Ion.
But we still using Lipo batteries in our smartphones 😎🤞
@@rickhuntraslam2925 Smart phone makers want you to buy a new phone every 5 years :-). The marketing is capacity-centric rather than robustness. But I would really love to see a cell phone with a LFP battery in it. I would happily take slightly reduced capacity for it.
-Matt
@@junkerzn7312 LFP has a very high price still currently a lipo battery 100Ah is less than 100 bucks in alibaba, a battery of that same capacity in Lifepo4 costs about 350 to 600 dollars, So I prefer to buy 3 lipo batteries for applications and change them every 5 years( 3 x 5=15 years) than a single lifepo4 battery that lasts 8- 10 years
2 days ago I "found" 40 lifpo4 solar light batteries I bought a year ago because I needed them, your timing Andreas is perfect.
Sounds like enough for many projects;-)
Great Video!
Btw: The P is phosphorus, the O is oxygen, so it should be LiFePO₄ (lithium iron phosphate)
Saw that too. I was like, "Polonium! Holy crap!"
Yes, you are right
Great informative video and thanks for sharing. Let's hope to see some ready made ESP32 development boards like the Lolin range with built in Lifepo4 support
I hope so too!
Very good video and a great overview, thanks Andreas!
You are welcome. After 5 years (last LiFePo4 video was in 2016) I thought it is time for such a video ;-)
Magnificent presentation. I really enjoyed it.
Many thanks! A lot of LiFePO videos focus on high power applications. So I thought I do one for low power...
Another great video mate.
I have been using lifepo4 batteries in my espnow projects since your early videos.
A good decision. Particularly if you use deep sleep.
This was very useful, thank-you. I'm looking to replace the Ni-Cad cells in my military Clansman radio and have been given some LIFePo4 battery packs. Your information has given me some ideas for producing a safe and efficient project.
Glad it was useful!
Hi Andreas,
Thanks for a good talk on batteries, very relevant for makers.
I have noticed that discharge at low temperatures does not affect these batteries much.
I bought some camping lanterns and measured their light output while running them flat in an environmental chamber.
The temperatures were 20, 0 , -20 DegC.
The internal resistance rises slightly at cold temperatures, which resulted in reduced peak output but extended the discharge time as a consequence.
The average energy yield appeared to stay the same.
It was very interesting testing.
Kind regards,
South Africa
Thank you; very interesting comment.
I agree, very interesting info. What about Li-Ion batteries? Did you test them, too?
@@AndreasSpiess Sorry, should have mentioned, these torches are fitted with Li-Ion batteries.
@@Uncle-Duncan-Shack So when you said "...does not affect these batteries much" you referred to Li-Ion batteries?
@@opsahle2 Correct
Andreas, you are amazing. First the video on 3D printing cases exactly when I needed it. And now this very week-end I was experimenting with different battery set-ups and this video surfaces. I'm beginning to think you have mystical powers ;)
Andreas has been known to stalk random subscribers before. I'd be careful.
I hear this quite often. What I think is that we all have the same problems to solve ;-)
Thank you! For me, the improved safety of LFP cells is a big reason I prefer them wherever they make sense. Also - I've had enough tech damaged by leaking 1.5v single-use cells that I'm experimenting with using pairs of LFP cells and spacers in AA and AAA sizes for that kind of application, especially in devices that don't seem to work well on NiMH rechargeables.
A good idea. Soshine even delivered a "connector" (spacer) battery with their AA cells.
@@AndreasSpiess the problems I have with AA LFP in common electronics are the usual inaccurate battery level and no undervoltage protection. It would be nice to have a protection circuit in the "connector", otherwise I noticed it is very easy to over discharge them.
Very useful again! I have been using a LiFePo battery in my motorcycle for the last two years, Works fine and much smaller/lighter.
I still have lead acid. I feared to change because my Harley consumes a ton of current (2 cylinders, 1700cc). Maybe I try one next time.
Great video. BTW, LifePO4 batteries are fully discharged at 2.5 volts, not 3.0. There’s not a lot of energy left between those voltages, but to be accurate your graph should have shown that.
You are right. Other sources go even deeper. But, as you wrote, not much to gain.
Great overview and summary, Andreas! Even Tesla will change/has changed their battery packs and there are already Tesla models released (initially only for China but now also in Europe) that run with LiFePo4 batteries. And even more interesting, they are operated with a heat pump to have a most efficient heating of the batteries in cold environments. So with your video you are again leading edge! Thank you for sharing!
LiFePO4 !
= Li-Fe(III)-Phosphate
@Juerg: Other viewers also confirmed that Tesla seems to change chemistry. I always asked myself what happens if 500 e-cars are in a Garage and one starts to burn... It seems easier to extinguish burning gas.
@@AndreasSpiess It's already reality! in Germany the Tesla3 models manufactured in China already run on LiFepo4 cells with a special monitoring procedure of the BMS to switch off the battery in case the temperature increases significantly above the environmental temperature. I understand this as a fire protection measure.
Thanks again for the detailed yet concise video 👍
You are welcome!
Helpful information
Glad you think so!
Very informative, thanks for your efforts and sharing 👍🇬🇧
This will always be my go to for information for a quick update if I need to recap
Glad it was helpful!
Thank you for all your wonderful videos
You are welcome. Glad you like them!
great comparison and summary. many thanks!!!
You are welcome!
Grüezi aus DE lieber Andreas! Ich habe zwar persönlich ausserhalb von PC-Service bisher wenig mit Elektronik zu tun, deine Videos machen aber Lust drauf, mal selbst den Lötkolben in die Hand zu nehmen. Vielen Dank und weiter so!
Sehr schön. Das freut mich! Willkommen im „Klub“
Thanks for sharing this awesome video for small use cases.
My pleasure 😊
I have used LTO batteries in my 6.6kwh Powerwall, I know the chemistry isn't ideal for your projects, but it has the greatest change and discharge rate of 6 minutes, also withstands the coldest of all temperatures on charging cycle, extremely safer then Li-ion and Lifepo4, and will have over 20,000 full charge cycles at 80% still left in capacity , I have pulled over 2000kwh out of it , I have even has 2 cells go into reverse voltage before I installed a BMS which seems to have done no harm to them, I think this LTO battery bank will out live me and my grandchildren.!.
I have a few cells in the lab but so far did not use them because, as you said: Their voltage is not ideal. In addition I di not find an "TP4056" style board for this chemistry :-(
Thanks for the excellent update. I think I'm still best served by the unprotected li-pos (sold for RC use), as most of my applications fly on model planes and thus weight is critical, but they also tend to place large pulse loads on the battery.
I may very well move towards lifepo4 anywhere else I can though. Their safety is a compelling feature.
I agree. For your usage you are best with these specialized batteries.
I have made a LiFePO4 battery from individual cells; which is nearly three years old, there is no need to fear this battery technology, as long as you have designed it to work within manufacturers guide lines and you have a good Battery Management System (BMS). BMS is a major topic in its own right, with active and passive devices being hotly debated.
I agree. But this is probably a topic for another video...
My 11 year old LiFePo4 RC car battery still works great. It's a 3S 9.9V 4500 mAh battery, and can be charged at 15C (67A).
Thank you for sharing your experience. So this must be one of the first commercially available batteries.
Great update video as always 👍🙂
Thanks for sharing your experience with all of us 👍😀
My pleasure!
I believe the most important difference between the lifepo and others is that need for regulator.
Although you mentioned the need for the component in case of higher voltages I think you missed one crucial aspect and that is the efficiency and idle current of the regulator.
If you need a regulator in your battery operated project it is crucial how much quiscent current it is speced and what current it needs when loaded, this needs to be taken care of at design and requires testing of the overall system. For lower currents an ldo suffice but for higher loads a switching supply is needed with low efficiency in non-loaded cases giving you terrible headaches.
Its a huge benefit when you can connect directly a battery to your mcu/system and not having to deal with optimization and or alternative designs.
For me it would be a nobrainer to change to lifepos once they will be widely available. I don't know why the battery operated system designer are not pushing this technology when you can really save 1-3$ component on your board and increase lifespan.
You are right with the quiescent current. It was covered many times on this channel and was not in focus today (because we do not need an LDO)
It doesn't seem quite safe enough for some applications. For example, an ESP32 has a maximum Vdd of 3.6V, and the LiFePO4 is charged at 3.65. But if you have an LDO, it needs an exceptionally low dropout for any 3v3 parts (though the esp32 itself is OK down to 1.8, the modules often have a higher limit - perhaps the memory eeprom needs more).
For 5v projects I would also suggest looking at the real new kid on the block, Lithium Titanate (LTO) their working voltage is (unless you are putting extreme loads on small cells) around 2.6-2.4 volts, so with 2 of them in series, my 5v stuff that I tested it on work absolutely fine, and they appear to be as safe as LiFePO4 (Some claim they are even safer). Slightly lower energy density though, but they claim 10 000-20 000 cycles on it. (havent been able to test that myself yet ;) )
I have a few of them in my lab. But as you write, the voltage does not really fit. Too high for 3.3 and too low for 5V (2 cells). It seems they are better in cold conditions. So maybe this is an application. Otherwise I do not see a big advantage for small projects.
@@AndreasSpiess I've had some luck with 5v projects with 2 of them in series (large cells so that they can run long without recharging). Currently they are absolutely not cost effective, but I figured I'd give them a try to determine if they live up to the claims. Maybe I just got lucky with parts that aren't all that sensitive voltage wise :)
I really would like to see a easy solar charge controller for LTO. I'm thinking about installing a project in an extremely remote location where the only way to access is by helicopter. If I could have the LTO long life span and stability it would be perfect.
Great explanation, Andreas. Congratulations.
Glad you like it!
Excellent video!
I just loved it!
Thank you!
I bought a 105AH LiFePO4 Drift battery from Fogstar in the UK. BMS is on board and there's a Bluetooth connection to an App so the battery tells you how it's feeling. It replaces two 104AH lead acid standby cyclic batteries. I power my amateur radio and TV activities with it including power amplifiers. The lead acid batteries would decline in charge really quick but the Fogstar Drift does the whole day's operation no trouble! Also my back loves it. My battery pack is a quarter of the mass it was. Yes it and the fast charger by Victron energy are expensive. So are all the sulphated and now useless lead acid leasure batteries and inflated LiPo battery packs. 73 M0YDH
I agree. I also built a 7.3kWh LiFePo battery for our remote station (24V). These batteries are extremely good!
Also a big thank you from Switzerland (also if for once the infos were not new to me). Nice summary and overview though!
In the table at the end of the video (13:00) there are two errors: IMHO the discharge current of LiFePo4s may be even higher than the one of LiIon (because of the lower ESR) and the charge time could also be much lower than the one of Li-Ions...
Keep up your great work and best regards from the Kanton Bern! :-)
Thank you for the additional information!
Supper Thanks 😊 now I now i can add lion protection to my charger boards
Glad I could help
Nice video. At around 2:10 you use 'amps PER hour' instead of 'amp-hours' for capacity twice. For the rest, I have learned a lot. Thanks. p.s. Tesla is switching from Li-Ion to LiFePo4.
I'm pretty sure this is because Tesla decided that at a certain point we can have much safer batteries in cars without too much of a tradeoff in energy density (I still think that EV's need a much lighter and faster charging and longer lasting battery chemistry to overtake the ICE in every metric aside from pollution after manufacture).
Andreas says ampere hours, which is entirely correct. Not the meaningless "amps per hour" that a lot of Americans get wrong.
@@plemli I think you are right, but it is not absolutely clear. He could be saying 'amps per hours' or 'amperehours'. Closed Caption also doesn't know what he says. Maybe it is just his Swiss accent.
As a long term user of lifepo4 batteries. Charging is not a problem, get boost or buck converters with current limiting set em to 4 volts per cell and the current limiting to something appropriate and just leave it connected. 4 volts won't hurt them and the'll just settle at that and the current draw will drop to near zero. You'll need to watch the low volts cut off. The discharge curve is fairly flat but not as flat as your graph shows. I can estimate mine to within around 10% The'll rapidly fall from say 4 volts to about 3.4 volts and then gradually discharge to about 3.2 volts and then fall away quickly. Almost all the capacity falls betwen 3.2 volts and 3.4 volts They are affectively fully charged at 3.85 volts but you want a bit more to drive the current in. You can make a convincing 12 volt battery by using 4 cells, You cant with standard lithium ion.
Thank you for sharing your experience. I would not charge them higher than 3.65 volts because this is what the specs say. As you write, there is not much more energy stored if you go higher.
Very good video, now I know a lot more about 2 battery tecnologies, for me I loved to know that lifepo4 does not flame cause this is an issue when using li-ion, on the other hand I always want to know the battery capacity, but connnect a lifepo4 to my project without a need of regulator is a great thing, so to choose what use will depend of the project kind, Thank you Andreas for this valuable information, bye!!!!
Glad you find the content useful for your projects!
Wirklich super erklärt, du bist ein auf den Punkt Kommer! 👌👍
Vielen Dank!
Thank you Andreas very informative.
Glad you enjoyed it!
The claim to fame for LiFePO4 (also known as LFP) batteries is cycle life, a flat discharge curve, low internal resistance, large voltage tolerances, and a nice failure mode. All at the cost of some capacity verses NCA or NMC. (LFP has slightly less capacity due to the lower voltage). I really love the chemistry.
* Cycle life is roughly 4 x vs NCA or NMC
* The discharge curve is very flat, which is especially nice when you are pulling high currents.
* Low internal resistance means the battery has almost no chance of overheating.
* Large voltage tolerances. A cell is typically charged up to 3.65V (and settles down to 3.2-3.3V). State of charge depends on the C-rate but you basically get it fully charged if you stop at 3.65V, then let it settle. Electrolyte break-down voltage is way up at 4.2V, so there is a massive tolerance between fully-charged and break-down.
* Failure mode is simply a slow loss of capacity over time. No instant failures, no sharp failure curve. So much so that used LFP batteries are often resold as lower-capacity B-grade cells. However, you still have to be careful about bloating if using prismatic cells. Poorly manufactured batteries and batteries that are abused in their first life can wind up building up gasses inside the cell, bloating the cell. Cylindrical LFP cells are more resilient to this effect, but are harder to deal with (e.g. do you go with four big prismatics or 64+ cylindrical cells? Most people go with the aluminum-cased prismatics.
* Slightly lower capacity than NCA and NMC lithium ion chemistries. But if you don't care for the application, all of the above positive points trumps the slightly lower capacity. A well-maintained LFP battery will easily last 15 years. Even longer.
-Matt
Thank you for your additional info!
Good job
Thank you!
Very informative video, as always.
I'm planning a PCB with a CN3058 chip onboard and a small PV panel. This charger chip has a TEMP pin, where you can connect an NTC thermistor, disabling this way the charging at subzero temperature.
This is an interesting feature!
Hi Andreas , Andy of 'Off-Grid Garage' in sunny Queensland has lots of interisting videos about LIFEPO4!
Thank you for the link! I will have a look.
Thank you. Needed the information.
You are welcome!
Very helpful video. Thanks!
You are welcome!
I think that supercapacitors are better because charge cycles, no overdischarge, temp amplitude bigger.Thank you for the great video!
Super caps have a very low capacity per size. They are very good for high currents and many charge cycles. All things we do not need for our small projects. They have other uses, though.
@@AndreasSpiess Oops!I really didn't think about the capacity.Thanks!
Fun, the little hand pointer as frm 8:05
Thank you. It has many fans ;-)
indeed LiFePO4 are annoying when you try to measure the capacity because the battery have a really stable 3.3v nominal voltage. but that stability is a bliss for an mcu! 3.3v directly connected without a buck converter/regulator? heck yeah!!
I agree. Usually I am more interested in when to recharge. This signal fortunately can be generated also for LiFePO4 batteries.
Page 3:30 remember the internal resistance of LiFePO4 cell? It has the lowest among all candidates. It is least appreciate in low current applications. However is important to (very) high rate charge and discharge applications such as high power tool and electric motor cars. LFP is more tolerance to high temperature c/d cycles than its counter parts.
I looked at the use case "small projects" where the maximum charge/discharge rate is not so important. So I did not investigate a lot into this matter.
I kept many broken laptop battery packs (18650 cells), so I saved money each time opening those for 50-80% recycling as it's just 1 cell or the circuit board which killed the pack and maybe caused extra cells to go under, but not all.
Good idea! Cheap and good for the environment.
Hi Chris,
Yes I agree that FePos are cool 😎... I used them a long time in my model planes...
If you are flying around 10 minutes with the motor with normal performance this is a perfect choice...
You can charge them wit 6c and discharge with 6c nearly infinite times...
But you should keep them warm +30°C...
The weight is higher but you can even discharge them with effective 30c... I think 200 cycles are possible...
Later I used a 120c premium LiPo and after 30 cycles with 30c you see the performance dying...
Currently I am using LiPos due to the better power/weight ratio... At the 6s 5000mAh level the difference is significant!
So I strongly disagree with your comparison of charging and discharging rates!
The reason why you do not find any charging boards is because you do not need them... Take a stabilized voltage supply with max 3.6V and be happy...
Thank you for adding in some info. Because I focused on small projects I did not investigate too much into charging and discharging rates.
Thanks for the info. I just save a few packs that were ready to be thrown away at my job...
So you can start to play with them…
@@AndreasSpiess I was thinking of using a current generator (0.1A) switched on and off with an arduino to check the tension every minute, but I may buy a correct charger...
Since LiFePO4 batteries don't leak (as well as not catching fire), I've been replacing pairs of AA and AAA batteries with a LiFePO4 and a blank in all our devices. I lost a $2000 USD device to leaking alkaline batteries, so I don't ever want to go through that again. So far, they have been perfect. With a few spares (kept charged monthly), a quick swap keeps devices in service.
Also, Tesla is switching to LiFePO4 for some of their vehicles.
LiFePO4 batteries as well as any other battery always leak. The leakage is determined by how well manufacturer manufactures the cells and how likely it is for the battery to leak through building pressure.
Also other viewers mentioned that Tesla is changing the chemistry. I can imagine having hundreds of Li-Ion batteries in a parking garage could be dangerous if one starts to burn...
@@richardlighthouse5328 Some chemistries, like alkalines, use the case as one of the electrodes. As a result, they are consumed over time, leading to leakage. That is not the case with LiFePO4 batteries.
Isn't it burning and can cause fire, it's better because it's not exploding.
An American-English tip.
Blow-up = explode (fire).
It’s more clear to say “puff-up” or “inflate”
The video does make it clear what was meant. I was offering a tip, I didn’t mean to criticize. I suspect that it might come up in a conversation where it’s not clear, so it’s worth knowing the US usage. I’m sorry if I offended.
@Em Wintle.: No problem. I know my linguistic limitations and I always learn ;-)
Great video, very informative but there was no mention of the cost difference, is there much?
As always: It depends. But for sure it is not relevant for a decision.
A video on how to integrate the TP BMS on a larger battery pack would be appreciated. I’m working on (future project) 16s 7p pack for a Honda e-cub and the standard multi BMS leave me wondering if I’m missing a trick. Shout out to Great Scott who provided a link to your channel
So far I did not cover higher voltage battery packs because this channel is more geared towards low power and microcontrollers...
One great usage of LiFe is the direct replacement of Pb battery of motorbikes or cars, a 4 elements LiFe battery can replace a 12V 6 elements Pb battery, with about 14.4V at maximum charge given by alternators.
You are right. We also use them for out radio transmitters which work best at around 1.6 volts.
Andreas, can you share how you store your lithium batteries in your lab?
Nothing special. But we have a good Fire insurance ;-) And I do not have many and mostly distributed in devices.
Great battery video!
Thank you!
Thank you Andreas, great information. Sadly, I am halfway through designing something, when halfway through your video it turns out that I can't find flat packs for lifepo4.. strange, your video is 3 years ago but I still can not find flat packs. Is it true?
I found a few, but they are pretty big. Small packs are still with the "old" chemistry :-(
@@AndreasSpiess Yes, odd. I have changed to the oldschool lithium for this project
There is also a great material advantage for LiFePO4, they don't use cobalt, which outside being expensive to mine has significant additional environmental issues as the countries where cobalt is extracted (outside of Austalia, Canada and the US) have very poor environmental track records and also tend to use child labour under health hazardous circumstances.
I agree.
I have been using a LiFePO4 32700 battery for a humidity sensor for a while. It lasts half a year on a charge and I love the idea that it won't spontaneously combust and burn down the house. Of course, that's only because its a 12 million MAH Chinese cell and has nothing to do with tuning the ESPs deep sleep and WiFi settings ;-)
More capacity is always good. The Chinese understood that fact fast ;-) and they found a lot of believers.
Great video....How is life cycles determined in lifepo4 batteries?
I do not know. It seems they survive more cycles if the discharge is small. So far, I have never had problems with the lifetime and no experience.
Thanks for your comparing and time. Could you tell me about LTO Lithium Titanate Battery?
I think they currently are a niche product. For small projects they are hard to use because of the voltage. I did not find standard chargers or protection circuits for them.
Looking forward to a LiFePO4 wall battery!
Then go buy some. They are the most common type, unless you are using recycled laptop cells.
Next year I plan to mount solar cells on my roof. And maybe I will add a "Power Wall"...
Is there an error on charging time? At 1:32 you present a chart in which Fast-Charge Time is faster for LiFePO by about 2x but at 12:50 the chart shows Charging time is longer for LiFePO. Have I missed something? Yes, I do understand that for "our projects" charging time is relatively unimportant.
Yes, I wasn't convinced by that. My experience is Li-ion/Li-po can do >1C charge but you have to monitor temperatures to he safe. I charge LiFe cells at 5C with no special precautions.
You are right. There are several Li-Ion technologies. Some are optimized for extremely high C rates. The LiFePO4 batteries I have are not particularly optimized for in that respect. This is what I meant in 12:50. But always the datasheet counts.
Great video, great batteries!
:-)
There was a test tittled 'LEO life tests of ABSL batteries using SONY 18650HC' that managed to use those lion cells for 75000 cycles, I understood that this due to not discharging the cells too deep, meaning that even lion cells are entirely viable if you just rather shutdown your device than deep discharge them and don't overcharge them.
Good Point. But what is then the advantage of their higher capacity? If you do not use it?
@@AndreasSpiess availability for both cells and chargers, which thankfully is being leveled by LiFePo4 cells and chargers being more available.
It also comes with the implication that more than likely the LiFePo cells can also do the same thing with even less loss of capacity (the Sony cells had lost 30% after 75k cycles).
So if you need to leave a sensor in the wild for years, how deep the battery cycles go can mean the life or death of the cell, so less maintenance is required.
Nice overview ! Btw, never trust LiIon packs.... sooner or later they'll all inflate, it's just a matter of time. But so do many other battery technologies. I've been working with lead/acid for years, eventually they too will inflate. If you're lucky they don't burst.
Good to know. I only use lead acid batteries in my rides…
LFP can be used and charged from near 0% to 100% with no degradation. To maintain a reasonable cycle life, LiIon are best kept between 20% to 80% only going outside that range when maximum usage is needed. This considerably reduces the claimed capacity advantages of LIon.
Thank you for the additional information!
So you said the BMS will protect the cell even when charging.
Does that mean I don't need a charger as long as I have BMS for all my LiFePo4? All I need is a voltage regulator for that charging voltage?
I use my lab power supply where I enter the maximum voltage of the battery and the maximum current I want to charge the battery.
If you use a charger with a higher voltage it will reduce its voltage to the max. current. As soon as the battery voltage is at the maximum the BMS cuts off. I would not use this method with a higher voltage for very high currents, though.
I'm impressed. I will be glad when drone batteries last longer (flight time) and do not have to be monitored and maintained when not used for extended periods....1 month of none use.
I am not sure if this chemistry will appear soon in drones because its Wh/kg ratio.
Interesting video.
I wonder if you could understand how some apparatus don't need charge in 10 years of operation.
In Italy our wireless toll system is a small low power battery operated transceiver that communicates with the two vertical antenna system on each side in the toll gate lane.
There is no issue working in temperatures from 0 to 60°C.
No damage to the battery.
Do they have batteries at all? If they are UHF RFIDs, the energy is induced by the "receiver". I once made a video about it
@@AndreasSpiess yes they do have, if a remember correctly, a one cell li-ion battery about the AA size.
There is also a buzzer than makes two short sounds when the communication is ok. I'm not convinced that they are RFiD because the receiver is typically placed under the center mirror in the car and that's usually over 1,5 meters from the reciever on the toll lane.
@@santopino2546 The Norwegian toll road devices are RFID without battery, same location in the car. I believe it is a SAW (surface acoustic wave) device.
@@opsahle2 i suppose, at this point, in our device, that the battery is only for the buzzer.
Some kind of low self discharge battery.
I've just searched the web and read about some RFID with a 5m distance communication, that incredibile.
@@santopino2546 And also while having to account for significant doppler effect due to car speed. Yes, amazing technology!
Can you please do a video on USB PD and designing for variable voltages?
Maybe. It is on my list...
You can also see electric cars switching from lion to lifepo4 due to the availability of iron over nickel. Because the batteries are cheaper it makes sense for lower end models where 300 plus miles of range are not needed.
I agree. And most home batteries also use this technology.
LFP cells will become more predominant mid next year onwards as a patent on LFP expires then. This was highlighted in a video by "The Limiting Factor" titled "A Brief History of LFP". LFP is desirable because there is no Nickel or Cobalt; both in very short supply. Lithium is the only constraint and that's mostly a processing constraint. This matters greatly in vehicles where there are a lot of cells used.
Thank you for the additional information! I will search for the video.
@@AndreasSpiess ua-cam.com/video/F_GtSA4Ig9s/v-deo.html
For you, though you've probably found the video by now, and others who may be interested.
I invested some money into the SkyRc MC3000 charger. It can charge all kinds of cells and it's firmware can be upgraded for supporting new kinds
Good choice. The newer chargers should offer LiFePo. Older did not have it.
Almost all hobby chargers support LiFePO4. For desk chargers the cheap LiitoKala single and double bay ate OK. The single bay Opus analysing chargers are great for testing. The 4-bay Opus also handles LiFePO4 however you need to access the internal switch to change it for LiIon, LiPo and LiFePO4.
I have 2 lifePO4 12v 8Ah I bought the first of this month. I bought a charge you have to set type of battery number of cells, voltage, etc. I tried to see the charger work. When I hook either battery up and try to start the charge I get a cell warning on both and no charge. My other charger gives me a cell balance problem. What do you think? I just got one of the Miady 20ah batteries today. The charger is at my office so haven’t had a chance to try it with either charger.
The chemistry has to be the same on the charger. Otherwise, it assumes wrong voltages. My charger, for example, does not support LiFePo4, only standard Li-Ion batteries.
thanks !!! any advice on using LFP ( lithium iron phosphate ) for multiple aruino boards ???
I do not understand your question :-(
I would love to see a video about coulomb meters / gas gauges / fuel gauges for batteries and how to use them in projects. I've only came across MAX17043 that wasn't super expenisve, but is only for a single cell.
These MAX17043 just measure the voltage and would not work with LiFePO4. Coulomb meters measure the energy transferred in and out of the battery. Maybe I will make once a video about it. However I do not see a big need for them for small projects because usually we just need a signal to recharge the batteries and we fully charge them before the next deployment.
what is the advantage of the golden tape?
It is very heat-resistant
all bms is balancer too? thanks
At least if they support more than one cell they should include a balancer
FYI: I am going on 7 years for my GoldenMotor LiFePO4 36 Volt, 12 Ah battery Pack on my EBike.
I have absolutely no idea how many charge cycles it has gone through :-)
I have also resurrected A123 LiFePO4 cells from the dead (i.e. no detectable voltage).
Thank you for sharing your experience!
At last LiFePO4, winter is coming!
True. At least on the northern hemisphere.
Hi, I have a questions.as for the max output voltage of LiFePo4 batteries are different than Li-ion let say 18650 batteries. How do i connect to brushless motor ESCs that has a rated battery voltage ( can accept wide range from 2s to 6s ):
Example 6s of 18650 can produce 25.2V max, while 32650 LiFePo4 is 21.6V max. How do Electronics components do accept the voltage actually? Does it matter across different chemistry.
Is the ESC can handle it okay by its own even the 6s voltage of normal lipo or liion is not the same with 6s of 32650 batteries? As ESCs usually connected to li ion or lipo.
Or i actually need a separate component to add in between to jump up the voltage to match the desired 6s of 25.2v of esc.
5s of 18650 li ion= 21V
6s of 32650 lifepo = 21.6V
Does the esc runs well when connected to a lifepo4 pack 6s2p with a bms then just connected directly to the ESC with regular XT60 connector for example. Thank You.
Your esc does not care about the chemistry, only the voltage. Check the specified input voltage of your esc and decide how many batteries with the chosen chemistry you need. The connector also does not matter.
My LiLyGO ESP32 SIM800L module doesn't power using LiFePo4 cell (at 3.6V) when I connect through the back built-in connector..
There's also no V-in pin.. Is it safe to connect that 3.6V to 3.3V pin directly to power?
I do not know this board. Particularly I do not know if the SIM800L module needs 5 volts. So you have to consult the diagram.
My solar pole light used LiFePO4 batteries in 18650 size. The one that came with it said 2000mah and I was unable to find a replacement with that high a capacity anywhere. I was able to fit in a larger cell that had I think 3600mha.
My 18650 LiFePO4 batteries only have around 1500mAh (tested). The 3600mAh are probably 4.2V cells.
He's absolutely convinced me...... to wait and see what happens.
We can always wait and see. In battery technology we will see a lot of innovation over the next years, I am sure. It is a huge market.
Great infos and that's gona help in my project ;) Although I'm having a little challenge to add the ultrasonic loop within my LoRa32 V2 using VsCode
Challenges are here to be solved ;-)
@@AndreasSpiess I can't agree more that's why it's called Hardware ;)
Your title is wrong, it is LiFePO4 not LiFePO. Also you pronounce those batteries as lithium (Li) iron (Fe) phosphate (PO4).
You are right.
Hello Andreas, I have been following you for years; thank you for all your support and knowledge, This is off topic but my be interesting, have you heard about the thread protocol? it is supposedly supported by the ESP 32 chips. it's and new open and supported mash network by many companies, But I can't find any information about it.
Thread is similar to Zigbee and will be supported by the (not yet available) ESP32-H2. It is too new and I will cover it when it is ready.
BTW: Thread is a stupid name for a new protocol because this name is used for completely different purposes.
so it is ok if i put a solar panel (5.5v 1.8w) -> TP5000 board -> 14500 lifepo4 -> ESP32 connected on the same pin of the battery while it is charged by the TP5000?
I think so. The 1.8W is probably a little small... If you want, you can have a look at the "power path" I showed in another video.
So did you manage to make yours sytem work with this connection?
Guten Morgen Herr Spiess,
I am looking for a LiFePO4 charger, but I can't really find one. I have 4 LiFePO4 cells in parallel and I want to charge at the same time. They're 14505 in size, have a capacity of 600mAh and have a standard discharge rate of 0.2C; to my understanding that means, that I need to charge them each with 120mA for a total of 420mAh at 3.65V. Right?
Devices such as the TP5000 say in the datasheet that they support AT LEAST 1A charging current, which is more than double what I apparently need.
How would you solve this problem?
Thank you very much in advance!
You should be able to change a resistor Rs (you find the datasheet with google) to reduce the max. charging current. On most boards it has R100 (0.1 ohms)
I am looking at Protection Board you linked... specs say Over discharge voltage: 2.1 ± 0.05V... As far as I know, that is way too low... Discharging to 2.1 V will harm batteries and propably to loose some capacity. I was using some LiFePo4 for years, but only with low voltage protection at 2.8V.
I also have such a board and I do not use it. Everything below 2.5 is a no-go for me. Below 2.9 volts there is anyway only a small amount of energy left.
I used some Li-Ion for LoRa nodes with heltec v3 that have TP4056 but even so one bad cell started arcing itself and i had to dispose of it.
Standard Li-Ion cells should work well with the TP4056. LiFePo4 of course not because they are not charged up to 4.2 volts.
Can i replace a 3.7v 18650 li-ion with an 3.2v 32650 that has a bms on it ? In the case of a USB fan that i have, the internal battery is connected to a 18650. I also have a few LED lanterns that use 18650 that charge via usb charger.
Ive also seen some garden solar lights using 3.2v lifepo4... wonder if direct replace would work.
Standard Li-Ion batteries have a higher voltage (3.7V) and LiFePo4 only 3.2 V. So you have to check if your device works with the lower voltage.
@@AndreasSpiess I'll have to buy a few and play around i guess. In the case of my USB fan and LED lantern. Normally they fully charged and run for many hours then turn off when the voltage drops below 2.5v i guess. i'm hoping it will work the same as the lifepo4 that keeps the voltage more steadily around 3v.. then only let the Fan or LED lantern switch off once it's below 2.5v Maybe the last 10min or 20min before battery is dead.
Charging them up again via 5V power input, the BMS should keep battery's safe. only providing 3.3v till full. then protect it from being overcharged and over voltage.
Thanks for the feedback
Hey I am doing my project on health care sensor devices which battery would be the better to use
You decide according your needs. You should find the info about the battery features in the video. I am no healthcare specialist.