Sodium Cell Charging and Discharging Analysis, Hardware Compatibility. Future or Niche product?

[@universeisundernoobligatio3283]

Good to see this test. Marketing department ordered the heat shrink sleeves with 1500mAh on them, the engineering department was ignored when they specified 1300mAh

[@OffGridGarageAustralia]

Yeah, that seems to be the case. They are still advertised with 1500mAh though. I haven't tested the other three cells yet...

[@MarkusNemesis]

@@OffGridGarageAustralia The datasheet you were looking at said 1300mah 🤔 I suspect they hoped for 1500 in serial production, but got 1300.

[@FutureSystem738]

Fabulous Andy! This video is JUST what the doctor ordered, very interesting. I’ve been wondering about these for quite some time, and previously knew almost nothing! The massive voltage range means big problems, and actually shows how terrific the almost flat voltage curve on LFP really is.

[@OffGridGarageAustralia]

Thanks, Dave. It's quite interesting, ey?

[@FutureSystem738]

@@OffGridGarageAustralia VERY interesting indeed! Thanks again. Initial impressions are that my 36 LFPs are in no danger of being replaced LOL :-)

[@martinmooren9536]

Interesting batteries and tests. Gives the same excitement when you started exploring lifepo4 batteries. I'm already looking forward to the next entertaining videos you'll be making for us! 😊

[@OffGridGarageAustralia]

Thank you. Yeah, very exciting. More to come...

[@ahbushnell1]

I down loaded your charts. Nice. I calculated the power in and out. I then integrated the power and coverted to kW-hr. The round trip efficiency was 79%.

[@David-bs6bv]

Thanks for all that you do and show. I also enjoy the "to do" cheekyness.

[@user-th1fs3if6i]

Excellent video as usual. Someone may have already picked up on this. Sorry I was not going to read nearly 300 responses. I suggest you add an another column to the csv to calculate the joules per 2 second sample. If you do you should see that energy between 3.95V and 2.325V is about 12760J (3.54Wh) While 3.95V to 2 V is 13540J (3.76Wh). That is roughly 90% and 96% of capacity. Inverters do need to be upgraded to properly deal with Sodium-ion but in it's current state the Victron would allow you to get 90% of the energy out of a 16 cell Sodium-Ion battery. If all the claims about Sodium-ion turn out to be true, the wider voltage range will be a very minor problem. The average electronic consumer device already comes with a SMPS capable of 100V to 240V input.

[@OffGridGarageAustralia]

Thank you. I cannot really alter the csv files as they won't be compatible with the EB-Tester software any more. But feel free to download them and add as much info as you want😉

[@user-th1fs3if6i]

@@OffGridGarageAustralia Thanks Andy. Poor wording on my part, I did not expect you to rewrite the code for EB-Tester 😀 I meant, if you take a copy of the spreadsheet ...

[@ThanosSustainable]

CJ is one of the most common "generic" Chinese cell brands, I got lots of Li-Ion cells from that firm. Otherwise, congratulations for the video, I really enjoyed it. Most important advantage I see is the potential to easily calculate SoC by just getting a voltage reading. This used to work fine in li-ion and I really missed it on LFPs. Balancing would also be much easier, for the same reason.

[@OffGridGarageAustralia]

Thanks a lot for your feedback!

[@michaelmedicworldoftanks33fps]

Not many sodium battery tests in the world so nice to see some of that rare testing done on your channel 👍

[@benwilliam1010]

I'm really curious about the discharge curve at 3C discharge and thank you for the great video!

[@Raphael_Hofmann]

Setting up a 0.5C Cycle test would be 6 full cycles per day or 2190 cycles per year. Setting up a 1.0C Cycle test would be 12 full cycles per day or 4380 cycles per year. Setting up Longterm tests for NMC, LFP and Sodium would be fun ☺

[@OffGridGarageAustralia]

Do it. We will all be watching 😄

[@mendohomepower7492]

Thanks for taking these on Andy

[@OffGridGarageAustralia]

No worries. More to come.

[@towatai]

Glückwunsch zu den 80k 🎉🎉🎉

[@OffGridGarageAustralia]

Thank you.

[@todddunn945]

To really evaluate how useful they will be for an inverter you should determine voltage sag as a function of load. Say take a battery at 50% SOC, measure the no load voltage then apply 1C, 2C and 3C loads to see how much the voltage sags. Your preliminary result suggest that most inverters will only be able to use about 60-65% of battery capacity. Usable capacity might be considerably less if there is significant voltage sag with load. I am looking forward to seeing more tests.

[@PavolFilek]

YOu use more cells and Studer or Victron inverter and no problem.

[@madsskov_dk]

Extremely interesting video, Andy. Thanks! The chemistry may be cheaper and better for the environment, so it could be relevant for stationary/solar use! Topics: - How to treat the battery good? 80% cycle? You found no info. If someone knows, it would be good knowledge. - kWh / price and expected development. And compare that to LFP. - If you have contacts with Victron, maybe reach out and learn if they will adapt to this? - if a 50V inverter produce 1kW, that’s 20A at the beginning, but 40A at the end, 25V / 0% SoC. In larger setups, it’s gonna be a lot of amps, and fat cables!

[@matzke089]

Great to watch someone so excited about charging a battery.

[@OffGridGarageAustralia]

I'm always very excited about charging batteries😄

[@gnuorder]

What I was thinking could be done with these batteries is have two banks. One of bank sodium at a high voltage or say 102V or higher, charged more or less directly from solar. The second bank that the inverter runs off would be LiFePO4 at 48v with a DC/DC converter between the two banks. The sodium bank would hold the bulk of the storage while the LiFePO4 bank would keep the voltage for the inverter constant overnight. I wanted to see more testing on the large sodium batteries before I try it.

[@pcmoeller]

Danke Andy, sehr interessant. Ich denke schon, dass wir noch mehr von der Natrium-Technologie hören werden und dass es auch irgendwann entsprechende Hardware dafür geben wird. Vermutlich aber nicht im ESS Bereich, sondern in anderen Bereichen. Zum Thema Überladen und Kurzschließen: Also wenn du es damit in die Nachrichten schaffst, dann würde das deinem Kanal doch noch mal so richtig Aufmerksamkeit verleihen.... kleiner Scherz, Sicherheit geht vor.

[@gordon6029]

Now leave it going, charging and discharging for a 1000 cycles.

[@PavolFilek]

YOu can not simulate normal use in PV, because you can not simulate temps inside cells, charging temps, dischargong temps, charging algos, not all inverters and MPPT use CC-CV. which is one of the worst charging algos. They are better charging algos, that save your battery and you can use it not 15, but 35 years. The is a lot of pdfs, how to build DC-DC BUCK, BOOST, FLYBACK charger, AC-DC charger and how to build BMS with 15 - 35 AMPS balancing curret, so you can use 35 AMPS MPPT or DC-DC charger, without communication with your inverter or BMS and BMS is still able to balance quickly to this 35 A limit.

[@SaveOurSouls-SOS]

use RE-EMF CHARGER

[@kadensmith5586]

⁠@@PavolFilekI’m currently designing a BMS for a sodium ion pack for my electric motorcycle, is there any charging algorithms you would recommend looking into? (You mention CC-CV is one of the worst but what’s the best?) Would you say it would be necessary to write my own algo specific to the cell?

[@PavolFilek]

@@kadensmith5586 I do not curently make SW for BMS, because no sutable parts and oscilloscope, but there is pdf , where is comparison algos like : Six PPC, Standard PPC, PWM, PAM, CC-PC and PC-CV. My work on BMS with 35 AMPS active non isolated and 15 AMPS active isolated balancer is stopped.

[@raystone4673]

Very important point you touched on regarding upper limits of inverter voltage ratings. Not only the max voltage that you can expect it to work at, but the upper voltage you are allowed to have it connected to the battery. At least a decade or more ago, when 'Raylite' lead-acid 48 volt systems were being equalised (for an hour), on a monthly basis at the required 64 Volts, there were inverters with a specified max "allowed" connection voltage of 60 volts being regularly used. I believe internal capacitors or semiconductor ratings would have been the design limit consideration for this.

[@Scrogan]

60V is a common maximum voltage for MOSFETs, while for capacitors I often see 63V maximums. Neither bode well.

[@Roel_Scoot]

@@Scrogan High power MOSFETS can handel 100V easy, some types go even till 1500V. Capacitors can easy have 300V or higher like the electrolytic capacitors in tube amplifiers and radios.✌🙂

[@Scrogan]

@@Roel_Scoot yeah of course there are higher voltage FETs and caps. But the point is, when you’re designing a 48V appliance, you don’t pick 1500V transistors or 300V caps. You get better specs per dollar spent if you only marginally exceed the maximum conditions. For a 48V appliance, choosing 60V FETs is certainly what I’d do, as the spec difference (mainly the trade-offs with gate-charge and on-resistance) between 60V FETs and 100V FETs is quite significant. Going for larger caps isn’t as unlikely, as there’s a better selection of 100V caps than of 63V caps. The cost increase isn’t as bad as it is for higher voltage FETs, the only real trade-off is size. People often recommend over-speccing the voltage ratings of caps anyhow, for longevity. There’s a good selection of 50V caps, but any engineer would judge that as too marginal for 48V systems. Hence, if you’ve got a 48V inverter, you may be able to push it above 50V, but likely not above 60V. If it says “absolute maximum 60V” then obey that. FYI, I’m making this exact decision to design a 48V kW brushless motor driver.

[@SpecialeW]

Hi Andy, I recently saw a video here on KZbin of somebody building a sodium ion battery for a pre built camper system. They used Victron DC -> DC converters to make sure the inverter could keep working even if the battery gets below a certain voltage. I'd like to know if that is the way to go when working with sodium batteries.

[@OffGridGarageAustralia]

That's definitely not a solution but just a workaround.

[@geekmystique]

very interesting indeed.. Making efficient electronics for these is for sure a challenge with this huge factor 2 voltage curve. It is exciting that batteries can be made from ubiquitous materials but as long as the cycle count and storage loss are not resolved this is going to be a super niche product.

[@MR-ub6sq]

To eliminate the significant voltage differences between the battery's full and minimum charge level, you may need a DC-DC converter. I don't know what and what kind, but one word comes to mind: POLOLU. I think that you should pack so many cells in series that at the time of minimum charge you still have the necessary voltage and when the battery is full you do not exceed the input voltage level of the converter. Now, initially, a smaller current is taken from the batteries to the converter, because the overvoltage is converted into the required current. When the battery is empty, the amount of current drawn from the battery increases, because the desired output voltage is approached at the input of the converter. This is challenging but interesting because we need a converter capable of delivering a large enough current. Interesting, but challenging!

[@JanWerber-b4y]

Hello Andy, or as we say moin Andy.... I love your videos man. You do a very good job, very structured and intense I think. Besides the learning effect, there is also a lot of fun. I can confirm your results with my results on a 32140 sodium ion cell. I only use 80% of the cell capacity of these cells. The 12V battery has a voltage range of 10V - 15.2V. That works perfectly. Let's see how long such a battery lasts..... Have fun and keep going on!👍

[@SolAce-nw2hf]

Interesting results. I am already thinking about building a Sodium-Ion solar setup for the unheated garden shed which often goes below 0 celcius. Not sure about the low voltage range, but losing some usable capacity does not seem to bother most people with NMC car batteries. If I remember this right you should not charge above 80% with NMC and not touch the bottom 30% in case of very cold weather. Or am I just trying too hard to see the good side of Sodium-Ion ?

[@DeophobicMind]

FINALLY! a Sodium Battery Test yey! Been waiting for this, Andy! haha The only problem I saw here was the voltage range of the battery, which will require new hardware for most of us if we were to upgrade or "migrate" to sodium. Voltage curve for me is very nice as it's easier to determine the capacity based on voltage alone compared to LiFePO4. We need more data for the life cycle in the near future if this will indeed replace LiFePO4. Do sodium batteries have better shelf life? How does it handle stress usage? So many questions! Looking forward for the next tests to come in the future, Andy. Thank you very much! If we are look at the bigger picture, lithium reserves are starting to deplete (or so I've heard and read) thus making it harder to mine and everything, and ultimately making lithium products way more expensive in the near future. Sodium Batteries is the way to go. Sodium itself is one of the most abundant mineral or element on earth which should make the battery way cheaper. Sodium batteries are less dense than lithium ones which should be fine if used for household energy storage. Let's leave lithium batteries to portable ones like smartphones and other gadgets.

[@pavelsulc2617]

hello, I don't think that the reserves of mineable lithium are currently starting to run out. Rather, the current demand is closer to the current maximum supply, which of course has a negative effect on the price. However, it is such alternative chemistries that can help spread market interest and offer more solutions to alleviate price pressure. The possibility of subfreezing charging may be very interesting for many places, but not for Australia.

[@MrHoolign]

I've been waiting for these Na videos =) so glad you are showing us the science of real life usage on these..... waiting to see what the 12V does, those specs sucked lol.... Love me some Andy in AU as always. You bring me smiles and brain food always, Andy. VIVA OGG!!!!

[@OffGridGarageAustralia]

Thank you very much. More salty video to come...

[@halilkacmaz6714]

Thank you so much for another amazing video, really appreciated. I just have a question I have 3 separate 11kw inverter in parallel for 3 phase and each has separate lifepo pack. Can I combine and parallel lifepo packs. Thanks

[@OffGridGarageAustralia]

Thanks for your feedback and kind words. Yes, you can definitely parallel your lifepo banks. I have done it many many times here on the channel and showed how it works. I'm running several packs in parallel myself. Here are some examples: kzbin.info/www/bejne/mJmmhJaZg9iEhdk kzbin.info/www/bejne/o5zInXiwi7Ogf6c kzbin.info/www/bejne/hqPYpGSQepl2hMk

[@avocadoarms358]

Hearing you say “shitloads” just like an Aussie… brings a tear to my eye 🥲

[@fx9750g2]

Andy, 17s for victron looks much better. Victron datasheet give us 38v-66v limit. For 17s it will be 2.23 to 3.88 volt.

[@evil17]

Thanks Andy, good job, they didn’t give you a lot to work with data wise. Doesn’t sound like too many pro’s for the price & expense of new gear to use it, 45oC sounds a bit low for our Aussie climate & the usable power wasn’t great either, maybe useful for outdoor solar lights/ street lights, etc. LifePo4 it is for the win! A few comments regarding only needing a volt meter to see state of charge, ok for a rough idea but wont indicate current usage & capacity measurements so you know if ur losing battery capacity, I like a good shunt. Cheers

[@robjohnston8632]

Great tests Andy, and so glad you started this. If their is more of a demand for these in the future, the inverters will change over time and the parameters will change too. I could not charge Lifepo batteries with my old Trace inverter and had to find the closes setting (gelcel) before I upgraded to my Victron inverter. The one thing that jumps out to me will be the cable and fuse sizing from the battery to the inverter with such a huge voltage swing for each size battery. You would probably have to use the lowest voltage rating when picking your cable size.

[@OffGridGarageAustralia]

I can't see a huge demand for them, tbh.

[@andreaspecco5399]

From personal test on 210Ah sodium-ion battery I find that this cell working fine from 2,5v to 3,75v. This is the real working range for sodium cells. For 210Ah cell: Charging from 3,75v to 3,95v and stop 1A battery absorbed 14Ah. Under discharge from 3,95v to 3,75v cell deliver 5÷6Ah. Above 3.8v the cell charges poorly, the voltage is no longer linear but rises in steps. below 2.5v approximately 38÷42Ah remain, consider that chemists recommend always leaving 20% ​​of residual charge. The end result is that a cell with a capacity of 210Ah, which has the same dimensions as 280Ah LFP, in 3.75v÷2.5v range of use delivers between 138Ah and 142Ah, and a capacity between 430Wh and 460Wh. Regarding number of cells you can connect in series it is 17S. You work between 42.5v and 63.75v, therefore falling within the range of your Victron Multiplus 2 inverter.

[@andreaspecco5399]

Test performed on 4 cells, the capacity is always lower than declared. In my case, sold as 220Ah cells, I found a real capacity between 195Ah and 210Ah. Internal resistances between 0.23mOhm and 0.40mOhm. If you charge and discharge them continuously between 1.5v and 3.95v, the prismatic cells tend to show a little belly. Tests performed with the ZKETECH AL40 like yours. Identical charging and discharging curves, obtained with 40A both charging and discharging at 20° room temperature.

[@Niurz92]

Bravo Andrea!

[@evil17]

Good info, thanks!

[@oze-bikes4life663]

Great Video 👌 I think the industry should "continue" to Research, Test & Prototype Lifepo4 Chemistries even further regarding Capacity & Wh / Kg alongside with the newer Sodium Batteries. You just can't beat that FLAT Discharge curve that only LIFEPO4 CELLS offer. 🔋🚲👍

[@Ressurrectio]

*So I have a Samlex SA1500 that I bought for my Off Grid solar to run off Alum batteries. My org intent and currently is to run off other lead acid cells because with the ALum conversion that was a flop as they did not maintain a sufficient current to make this worthwhile, however the inverter (along with 5 others remain, The SA-1500 at 12v [12v due to alternator tie in] has a range of 9.5v - 17.1v So with this version you would certainly be able to avail yourself more of the full range.* *I'm looking to upgrade to Sodium Ion's for travel since I dislike Lithium's thermal runaway inclinations. But mainly the BMS part. because I have been using 18650's in my bike light for some 9 years in a variety of temps) My interest would be at what voltage point the Sodium Ions have a cut off point by not being able to run a resistive load like even a soft start cap Frigidaire 5k btu Window AC unit because in my tests the LA's as well as the Lithium were at around 65% in my battery bank.* *Creds: over 7k mono crystalline solar panels, 3 Midnite Classic 150's, 5 Samlex Pure Sine Inverters 300, 600, 1500, 2, 2000 and 1 3000 (+ a 24yo 1500 watt Modified Sine and along a 750 and some smaller ones too) with 1 Go power 3000 I use with my Welder. Noctua cooling fans throughout. 4/0 copper welding cables with 60/40 Kester Solder and tinned copper lugs. 2 IOTA dls 90 and a 45, power supplies and also a 100 amp Powermax 220 version (for travel using unused J1772 EV chargers) Again Noctua PWM fans throughout the entire system. Makes a huge difference in alleviating noise pollution. Currently about 1500Ah battery bank)*

[@karelvandervelden8819]

Andy .........general question ; What if we use 1 big LiFpo4 cell of 3.6 volts and then step-up the voltage to 12.5v to make it practical. ? This way we do not have the multicell balancing problem. (I live on a boat and have a healthy distrust of complex BMS circuitry and except loss of efficiency)

[@swrekcfest]

It would be awesome to see the performance of LTO batteries as there are very few details about them on KZbin . Thanks for the video and great work !

[@OffGridGarageAustralia]

I have a few LTOs here. They are so expensive that not many will use them. With the high C-rate charging and discharging, they are not really useful for solar applications. I'll make some videos in the future...

[@lotsad1234]

excellent video!!! the biggest possible benefit we can get with this technology, is that when it matures it becomes substantially cheaper than lifepo and proves to be safe for transport and allows less difficulties when it comes to transportation than lifepo, (which falls into the lithium-ion category for most shipments), and so one does not mind losing 20% - 30% of capacity.

[@OffGridGarageAustralia]

I don't know, these are all marketing arguments... Will it ever get cheaper? Who will buy so many cells so they become cheaper if there is this big hardware challenge. Safer? We talked about it in the video. This is exactly what you can read out there, but without any substance or real world numbers. All this theoretical information. Like solid state batteries🤣

[@margusten1]

Wery interesting, keep testing , I have to use Sodium Batteries because at this winter I had in My garage -12C. At the moment using Li-ion batteries from Mitshubishi iMiev and they accept charging at -10C. I live in Estonia, near Finland.

[@MikeMUC]

Hi Andy, the "LumenTree feat. Trucki" 600-M/800-M/1000-M/2000-M Grid-Tie Inverters (old versions formerly known as SUN Grid-Tie) can actually use 100% of the capacity, as they have a very wide input voltage range of 22v-65v 😜

[@OffGridGarageAustralia]

Yes, correct. The SUN can do it...

[@NaamNatuurlijkniet]

Hi Andy, I remember one of your LiFePo tests in where you charge them to around 100% using a lower then max floatvoltage. Translating this to Victron, would it not be a nice test to see if you can top off this cell with a lower maximum voltage and then test what it does with the overall capacity. If this sticks, you might be able to build a 17,18,19S etc battery and not really need the maximum chargevoltage the Victron cannot supply and still fill them up. Whis would leave a larger portion of the SOC to be used in one cycle. Balancers don’t have any issues up to 20S, not sure about BMS availability. Gr. Mike

[@1kreature]

Great video! Have you had a look at the 40 or 45 Ah lithium-titanium-oxide (LTO) batteries? They can also be charged at sub zero and have a massive current capability, including fast charging due to some "magic" about the chemistry apparantly. Nominal voltage is only 2.3v and they have an upper limit of 2.8v but discharge cutoff is 1.5v. Here they are also different in that they do not take much damage from 0v discharge if done at lower currents. Otherwise they can do 10C charging and discharging! Very interesting!

[@OffGridGarageAustralia]

I have LTOs here. Fast charging is not really something we look at with solar installations, do we? I wonder why they didn't install them in EVs if they are so good...?

[@1kreature]

@@OffGridGarageAustralia Energy density is half of lithium...

[@1kreature]

@@OffGridGarageAustralia For solar lights and such the LTO's may be worth it though, as they can be charged when sub zero. This is important as it's starting to become the primary failure point of solar driven street lighting in colder countrys.

[@jkyri]

I'd like to ask you. If you are talking about e.g. 50% SOC, are you talking about half amp-hours or half kwh? Because right here, when the voltage difference is so large, the top 1% of the battery contains 2x as much energy as the bottom 1%.

[@tibib4029]

Estas son baterias de sodio ??? Ya se estan comercializando ???

[@andrewt9204]

I guess the one benefit of the wide voltage range is that it should have a big cycle life only using 50-70%. At least as good as lfp. Looking forward to more testing!

[@bubbawubba2307]

And then you have to buy a 1000 cells to have a battery big enough to last through the night lol

[@Goodkiwibloke]

That is some voltage working range!! Must be tough for EV manufacturers using sodium ion. Their HV cabling needs to be double the area of lithium battery EVs due to working voltage range. Thats a lot of extra very expensive copper in each car. NMC voltage range is 18.% (3.4V / 4.15V). LFP voltage range is 25% (2.7V / 3.6V). Sodium ion as you tested is 62% (1.5V /3.95V) = 2.5 times the current at low SoC compared to high SoC to deliver the same amount of power. Wow I guess the battery scientists will come up with some alternate chemistry to fix that issue at some stage. Because the current vo.tage range is unworkable in existing applications, as you point out Keep up the good work Andy

[@PavolFilek]

Thanks very much for this comprehensive info. But there is no info about cells DC resistance at 0 - 25 - 50 - 100 % SOC. Resistance and capacity are key features to know, if I would build NIB battery for my PV. Also it si well known fact, that we need use cells at 40 - 60, 30 - 70, 25 - 80 or 20 - 85 % SOC ENERGY, so we can use cells at VOC from 2.4 - 3.9 V, but under load it will be lower , depends on Cx rate of discharge and under charge, it will be higher, depends on temp + resistivity + Cx and also SOH, which means, if battery is old and was stored or used at HIGH SOC and HIGH TEMPS, the SOH parameter is BAD. So if we use battery from 2.25 V to 3.9 V under LOAD, we can use Studer inverter 38 - 68 V range, and 38 / 2.25 === 17 and 3.9 x 17 === 66.3 V, so we will use battery from 25 - 35 % SOC to 85 - 88 % SOC and our battery will be good for a lot of years, if we do not abuse battery at high currents, which means high temps at electrodes, separator and elyt, so proper temps range, proper Cx C/D rate, proper SOC range and we can use this battery pack for 1/2 of our life.

[@jackoneil3933]

Thanks Andy for the important Sunny and hot garage test. A fire and explosion test would get my vote if such votes were counted, but short term over-current discharge vs temperature as one might encounter in solar applications would be something I'd be interested in

[@OffGridGarageAustralia]

I don't have a climate chamber for such tests to keep them at a stable temperature.

[@jackoneil3933]

@@OffGridGarageAustralia I was thinking more of over 1C discharge tests to at normal temperature to monitor temperature increases

[@chrisr819]

That remembers me 20years ago when I scraped the first lithium battery from a military vehicle and had no idea about that chemistry and nothing was to be found anywhere....so very carefully testing is step by step how to deal with it.

[@OffGridGarageAustralia]

Yes, exactly. Very similar with LiFePO4 when I started the channel. Almost no info out there.

[@Jeronimo-eb2rf]

Hi, is it possible to set your configuration for the pace 200a bms on your website like the jk bms? Best channel for battery configuration. Following your channel from the start. Greetings from Belgium.

[@OffGridGarageAustralia]

Yes, sure... I'll do that after the sodium series.

[@Sailing-Aquaholic]

It is not just the inverter or charger. Most equipment designed for 12v will give low battery warnings at 10.8V. But on the other hand, that makes the sodium a safe battery if it can handle a wider voltage range than your equipment.

[@OffGridGarageAustralia]

And an expensive one... As you need double the number of batteries as you would with LiFePO4.

[@jonasstahl9826]

When sodium batterys are dirt cheap, just use the voltage range you need and compensate with more batterys.

[@SrDelbioLima]

I'm curious to see how sodium battery can be implemented to 48v solar inverters, because of the higher voltages and steeper drops. Will a dc boost converter do the trick?

[@philipreich7035]

Thanks Andy! I'd be interested to see a cost comparison between equivalent cells for building a 12v nominal pack and 48v nominal pack. The energy density seems pretty low compared to lithium ion and lithium iron phosphate cells too. Also they say they are rated for 3000 cycles to 80% -- that's worse than for lithium iron phosphate, which has 4000 cycles to 95%? At this point, I agree with you that it seems there is only a niche market for them for use at colder temperatures.

[@laurentsantaibambu7324]

Hi Andy, and once again thank you for this very interesting video!!👍 Really, we were able to get information about these Sodium batteries. A test that would seem very interesting to me is to do exactly the same test but with the same capacity LFP battery! And charging and discharging power exactly same! (obviously keeping the basic parameters of an LFP max discharge 2.5V max charge 3.65v) To see and compare the charge and discharge time!! I think it would be interesting, it could tell us at equivalent charging and discharging power what would be the most efficient??👍😉

[@ericglaude4953]

Fantastic information Andy, and I’m am hooked on you channel now. I love it! Can you tell me what you suspect will happen to the salties in their output performance within the manufacturers recommended cycles? Will it remain constant or will it degrade as the lithium-ion ones eventually do? Should one test the capacity periodically or simply rely on the SOC determine the battery “real” capacity ? It’s time for a beer! Cheers 🍻

[@pteepan2708]

thanks for making the test Andy, it seems like they are not a fit for my off-grid needs. your numbers shows how they are not really in competition with LiFePO4.

[@timlochner9147]

They might be good to use with a grid tie inverter with limiter. The voltage range on mine is 22 to 65 volts

[@theobauck258]

Hi Andy, as always a pretty good video about a NEW technology. Gratulations for reaching the 80k. In case of the sodium batterie at this stage of the First Generation of these cells, there is more coming up. The chemistry and the density of these batteries will raise, to make it more reliable. But i Think, the Energy density is not as good as LFP in the First stage. Liebe Grüsse

[@Full-of-Starships]

Didn't expect you to get to Sodium batteries this soon. Brilliant. Given the issues you're having, maybe sodium iron phosphate batteries may be more useful? IE more in line with LiFePO4? I believe this is the chemistry variant which is thought more suitable for stationary applications?

[@OffGridGarageAustralia]

Never heard about sodium iron phosphate. Nothing like this on the market right now which you can buy. Maybe something in the labs but will take years until they make first production cells. 99.9% of such new and groundbreaking technologies never make it to market.

[@marcoarpago]

Very interesting, thank you Andy. I will keep an eye on sodium batteries. Voltage Delta is a big problem....

[@r.b.l.5841]

I have some factory 100Ah Sodium Battery packs, 12v and 24v the inverters can use most of the voltage range generally, lower limit cut off leaves some un-used capacity but very small. The Soduim specs appear to tollerate far lower and higher temperatures than the LFP - should be good news for Australians, Texans Sudias and Scandinavians alike!

[@offgridwithpojectham]

Andy, First of all, thank you for conducting these tests in such a scientific manner. Excellent! Great information as always. I give these batteries ZERO frogs! With the Sodium batteries given the wide voltage range, it appears to me that you will loose all the advantages of a 48 volt based system. With such a high voltage range comes large swings in amperage through your system and blows all of your advantages in wire size that you get with a higher voltage system. Not to mention all the equipment issues you mentioned in running at these voltages. Maybe the technology will improve with time but at this time, as you would say, Nah... 🐸=0

[@joesmith1142]

This is very interesting. I still think the flat voltage curve of LFP would be preferable for running 12 V electronics etc (well obviously lol)... Love hearing about new battery technologies. Thanks for taking the dive into sodium. Been wondering about this tech for some time now. I wonder if these can cold crank starters for vehicle engines? Im trying to see the advantage of sodium over LFP. I guess if these can supply very large amounts of cold cranking amps, similar to lead acid they might be a good replacement for starter batteries with their very low temp capabilities and the capability to be drained to 100% with no adverse effects. This would be a huge plus and maybe a market they could take over.

[@OffGridGarageAustralia]

As per the specification, below freezing point they can only be charged and discharged with 0.2C. So, not good for starter batte.... hang on, who still uses starter batteries???

[@joesmith1142]

@@OffGridGarageAustralia Oh right haha! I forgot the Tesla is in the other garage. Yes good point on voltage. I wonder if its the same though with that large 12 volt battery on the bench.

[@JoeInBendigo]

Great Video! A good start to a new voyage of discovery! How does that charge and discharge curve compare to lead acid batteries? They may be a good replacement for Lead Acid, rather than for a LiFePo4 batteries?

[@maxtorque2277]

For any static energy storage solution, what actually matters is the total lifetime energy exchange divided by the cost. These batteries will be judged on their ability to absorb and release energy over time for a give installed cost. We moved from NMC to LIPO even though the energy density is significantly worse for the latter simply to leverage the greater cycle life that meant an overall lower cost, ie 4,000x 93% is greater than 2,500 x 100% 🙂

[@robsycko]

Looks like they would make a great replacment for automotive aplications lead acid replacment? Maybe a lack of high amp output?

[@Riddle-Me-This-Puzzles]

Wow! Good job keeping us informed/investigating this new tech that seems to be catching on

[@Riddle-Me-This-Puzzles]

PS I LOVE the discharge curve! The only thing I absolutely HATE about Lifepo4 is the discharge curve

[@ajvz77]

Hey Andy, great that you are investigating sodium ion batteries! I noticed in your video that you compare it's charging to LFP, which made me wonder what the actual chemistry is of these cells. Are they not more like NCM rather than LFP? Is it worth investigating sodium if there's still nickel and cobalt in there? Sure hope they are some kind of "NaFP"...

[@jamest.5001]

The inverter doesn't need to be built for the voltage, just a programing change. Some could possibly use a slight hardware update to work better at different voltages but a resistor change and adding a turn on the primary or secondary of the transformer is no big deal. I'm still hoping to build a inverter to power my house. Actually two inverters, wired in parallel, with one sleeping until the current rises to the point of requiring the additional current. I'm thinking a 4-6 kw then either an additional equal size inverter. Or a larger inverter that just takes over the whole load, then I can use high frequency technology on the small inverter, a true 3kw, capable of doing 3kw ,24/7 ,, ,(actually it should do about 3600-3800 before temperature in hot weather is starting to be a concern, it could easily be rated at 4kw. With a surge around 5k-5.5kw. working with a LF power house , weighing about 250 lbs minimum, capable of 15kw sustained, and easily 25kw for 20 minutes and 30kw for up to 5minutes, and motor starting current spikes lasting for 1-2 seconds, 45kw should never cause damages. A cheap company would rate it at possibly 35-40 kw. I'm searching for the electrical iron for the transformer cores, I need about 80 -100 lbs of the iron to produce a 18" O.D. core with a 10-12" center I.D. and 6"-8" thick , so the strips need to be 6-8" wide 48" to 96" long, and thin as possible, 20-24 gauge would be great. I want to wrap it with fiberglass. To create a protective layer and embed a quality nylon tube about 3mm of, wrapped with the iron, possibly using copper tubing wrapped around the toroid core, if it doesn't effect it electric all, about 6 coils inside in the id. And 6 on the of, to circulate mineral oil to cool the core using a 6:1 manifold, to bring the 3mm tubes to a single 12mm or 1/2" tube on the inside and outside, the core requires a gap between the windings and core to reduce saturation having two sides about 3.5mm from the core should be fine. With about 1.5mm in the top and bottom it should keep the core 50°f cooler than it would be otherwise, and I plan to use 3-5 thermal sensors in the core and in the windings. A in id sensor, a center mass of the iron core and one on the upper area, heat rises. So mount at the high point. And two inside the coil windings on opposite sides with possibly one near the coolant lines, and one far away as possible and also cast aluminum heat sinks for the mosfets and diodes with coolant passages set up the flow direction so it will circulate the coolant with thermal convection. Up to a large cast aluminum radiator/heat sinks resovoir to be mounted on top of the inverter, being 24" wide 8" -12" tall and 6" thick. The center 3.5-4" x 18-20" x 7" -11" tall being the resovoir, and holding the small pump that will power up if temperature rises, a long with 4x 100mm fans on the heat sinks internally, and 2-3, 120mm fans mounted on the finned and holy resovoir, the center 12-15" having 6mm holes passing through the resovoir with Finn's on the outer edge about 2" dep with 6mm gaps, with 5 mm fins. Having a 1mm groove mid fin about 20 mm deep making each fina double fin. Cast using the lost foam method. The tank lid possibly being a 6mm sheet of aluminum with 46 3mm screws, and o-ring seal. With a 35 mm fill cap with vent, the tiny pump fitting inside, using a tiny motor and a typical 30mm turbine pump, hopefully the restrictions through the pump will cause the oil to rise up the tube without the pump, with cool oil flow from the pump side, being the pump pushes oil down the 20mm line, splitting to two 10-12mm lines. The pump flowing only about 1,gph, or 4 litres per hour. The heat sinks holding about 75-100 ml, cc, the transformer holding around 500-1000ml , the resovoir holding 1-2 litres +/-. So 3-4 litres full possibly, it can run without, the mosfets having a fan on the heating with 30mm deep 1.5mm find on 3mm centers. In 50 mm thick cast heat sinks , with two 12mm coolant passages per heatsink, and the electrical conducting heat sinks possibly being copper plated, with a 4mm copper sheet soldered to it, with a nickel plate to keep it clean. I'm looking to build a 50+ year inverter the mosfets mounted to copper bars with pcbs making electrical connections, with M5-6 bolts. To easily swap out the mosfets, the 3mm copper bars easily mount to the heatsinks with m3 screws, then m3, 5/6, bolts to the main PCB. The fet boards being two piece, the actual part with fets being replaceable. Then the parts can be replaced on the modules, rebuilding them The electronics being made similar. The parts that typically break mounted on separate boards, easy to replace. The logic being shielded having a solid copper PCB as a cover, using pins and screws to mount, then copper tape around the sides. Remove all emi, EMF it should be empty proof on the logic side anyway. With ribbon cables and fly wires running through foil tape or braided tubing , connected to the ground plane of the board. And print a shop manual basically, left indie the casing, with a box made of pcb material to hold replacement boards. Protected from emi/EMF, EMP , if there is a EMP. It should be easy to replace the parts, and have it working again. The ats from small inverter to large would have a micro controller, measure the current, if the current is so high it wakes the larger inverter, and if the current goes any higher in half a sine wave, the load goes to the large inverter. The small inverter likely goes to sleep. Don't know if I would actually go this far. A couple 150 watt panels would offset the idle current losses. Having backup power from multiple sources, a 7kw ac gas generator, a 2500w gas generator. And working on a 150 amp gas DC generator, and a 300 amp diesel generator, that burns waste oils, sorry to ramble, have a awesome day!!

[@GregOnSummit]

The charge and discharge look like mirror images, especially with the little hump around 3v. Would the large voltage range be an issue with inverters? I also found out that the cell size dictates the max voltage? Strange stuff ....

[@davidconner-shover51]

could they be more like NiMH batteries? which work better when fully discharged regularly?

[@SabretoothBarnacle]

Curious to see a temperature vs charge curve to see if that bump coincides at all

[@ThanosSustainable]

that was also my thought, but then the bump also appeared on the other way around, as well as the official curves, so I guess it's a sodium thing. Wouldn't be super surprised to find out that getting to this bump actually does some damage, since it surely looks like something changes with the chemistry.

[@ernestodemario7447]

hello mr andy. I had the same thoughts as you about these sodium batteries. I expect to see a refrigerator in the garage soon...I await the next tests thanks for what you do for the community

[@OffGridGarageAustralia]

I don't have space for a fridge inside the garage😄

[@ekbuz]

I'm not sure if your tester would let you, but could you series some of the batteries and then discharge down to 1V? If you had all 4 it would get you down to ~.25V each which still isn't 0, but let you see the curve down that far at least.

[@OffGridGarageAustralia]

Yes, that's a good tip! It will work with the EBC-A20.

[@mprobison]

Hi, new viewer. Thank you for your time and effort. Suggested test: Low voltage cycle torture test. 1.5v-3.5v. Hypothesis: cell would experience maximum heat flux, and might reduce cell life due to anode breakdown. I'm a welder, not a engineer, please educate me if I'm off my rocker here.

[@OffGridGarageAustralia]

Hello, welcome to the channel. With what current though and how many cycles. 3.5V is not really low voltage any more for these cells?

[@techNick0815]

Megagut - back to the roots - endlich wieder Batterietests! :D

[@Scrogan]

Fascinating. One of the main points about sodium batteries is they’d use more readily available materials while not requiring entirely new production lines. Looks like the second point is taken care of, but do these cells contain nickel and cobalt? I think mass lithium mining is better than artisan cobalt mining, so I’ll probably be sticking to LFP. Or flow batteries, or some sort of molten metal battery.

[@pieterjanvz]

Is it possible to test with constant power , I imagine there is a increased voltage drop when current rises to compensate for the power

[@bubbawubba2307]

What is the C rating on these? Does it take more energy to charge vs lithium or comparable to lead acid? How big of cells are there?

[@OffGridGarageAustralia]

It's be in the next video.

[@TheOffGridShop]

Andy I would love have you on the podcast and talk about this. I have been doing R&D for a sodium Ion manufacturer for almost 12 months now and could share some good insights with you.

[@OffGridGarageAustralia]

Not enough time at the moment, my friend. Maybe one day we can make it work...

[@evil17]

Why dont you share some of ur good insights into this tech with comments in this forum so we can all get a better grasp of this tech & maybe you will get some more insight from viewers.

[@TheOffGridShop]

I can't share current information as I have signed Non-Disclosure agreements on the tech. The company I am working with has solved many of the problems Andy talked about. The reality is the sodium cell in this video is tech from 2011. I have the 2014 Tech in My house, and soon, I will have the 2023 Tech. @@evil17

[@PaulSpades]

Inverters with wider voltage range shouldn't be a problem to make, they just need to test and validate existing hardware. Please remember that the modern restrictions on voltage range is because lithium cells blow up. DC components with a wide range of step up and step down range work fine, although there's always an optimum range. If these cells are not a fire hazard, I can see even a .5v to 5v range for each cell, meaning that a "12v system" can range from 1.5v to 15, 3v to 30, or 5v to 50. We need to start talking about these as series configurations of maximum voltage, not nominal voltage configurations - like we used to before lithium cells (with their weird curves) became popular.

[@RobinRastle]

I love the ZKE ab20, brill for repeatable cycle testing, goes up to 17v max charge V so will do 4s lipo.

[@OffGridGarageAustralia]

Yes, best devices ever!

[@john543]

I have some cells from a different source, and they say the specs are charge to 4.1 and discharge down to 1.5. What would happen if you tested your batteries agin using 4.1 rather than 3.95? Based on past experince, a specific chemistry has specific voltages so both can't be right.

[@ando3807]

Given the limits of the Multiplus II, would it be worth using an 18S Pack charging to 3.55 V per cell and discharging to ~2.05 V per cell?

[@dirkaust5107]

What about the price? Should be the most important argument for Na cells

[@alexandershevchenko489]

> 18:56 we don't know if sodium batteries are getting stressed as well as lithium batteries one said sodium atoms are lager then lithium, thus it's harder to pack them into the graphite crystal lattice. thus we can assume that sodium atoms are more likely to destroy it and sodium life cycles (3000) compared to lifepo4 (6000-8000) seems to hint, who is more dependent on stress

[@philipng6598]

It’s a nice replacement for LiPo batteries since they don’t work in 12V systems 3s would be too low and 4s would be too high. You can charge sodium batteries in 4s configuration to 14.4V and drain it down to 10V safely. The only down side is we don’t get to get the full capacity maybe approximately 900mAh each cell according to your discharge graph. I hope in the future they are able to keep the energy denser so this can replace lithium since sodium is everywhere

[@lukas181118]

Interesting! Too much changes necessary to make this batteries work with present equipment, it will be ages before we will wildly use them or its high chance to become forgotten technology 😂 . Keep testing 🎉🎉😊

[@zeropointbalance7366]

Great video. Did the same thing with: HAKADI Sodium-ion 3V 3500mAh Nominal voltage: 3.1V Standard capacity: 3500mah Charge voltage: 4.1±0.05V

[@ThanosSustainable]

The use of a DALY BMS was a mistake. They are possibly the worst BMSs, afaik. "Self-balancing" cells do indeed "burn" extra energy. Did you notice the cells getting hot at 5mA? But then again, 5mA doesn't really give out any heat.

[@zeropointbalance7366]

Go to battle with tools you have :(... I was effectively using the Daly BMS as data logger (simultaneous voltage curve for 4 cells) that gave me some protection when I simply assembled the 4S battery- kind of a short cut to individual cell test. The balancer does not really work so it was off. In one step have a graph that shows the V-curve, the capacity and any balance issues! Charge and discharge. And even some info on the top of range (apparent squichiness? -energy burn? this may be a false interpretation?) nicely recorded on the out of box from China cells. Note I had the cells sitting on 4.15V waiting for current to go to zero in case these cells need to be absorbed for full charge. They were in a stored in safe space in case they decided to go nuclear (and my current limit at that point was 200mA to make sure it could not run away). The Daly cutoff was 4.2V I think the current went down to 20mA and the cells never got warm (to touch ). Note: it simply may have been that the cells were already balanced, Aside: As for the Daly startup issue: The only reason I needed it to wake up was to log the data... I do not really understand why they make BMSs relatively difficult to start up. These Daly's also struggle with LIFEPO4 startup. This specific one I usually have to tickle to 14.8V+ before it starts. Why? Why not just start if you hit something reasonable like 12V. I know this may be done to avoid completely draining a very low (LVD) battery but it could be implemented better. @@ThanosSustainable

[@HybridShedIraq]

Very nice video Andy as always, so big hype of sodium ion and the price is still too high, and the usable energy storage is well 70% too low... Thank you for this detailed explanation. maybe the 1500ma is to zero volt who knows.

[@mickwolf1077]

I reckon when the sodium chemistry is more widespread then charger and inverter manufactures will include the suitable parameters, maybe even fireware (firmware) updates for existing units if hardware supports. Interesting curve on them.

[@alikayacan6768]

Hi Andy, Please repeat 4x Sodium batery test with the voltage limits of PETER INVERTER. ie 10V - 15V limits.

[@rubenfernandez4070]

The Ingeteam Ingecon Sun Storage 1Play 6TL M inverter has a voltage range for batteries from 40V to 450V. Full capacity can be used in sodium batteries.

[@jws3925]

I had a solar shop that was handling sodium cells. They were very high on this chemistry and sold many of them but it went very bad for them. Eventually they offered full rebates to all the customers who purchased them. I really don't know much more than that.

[@gg-gn3re]

virtually no equipment supports the large voltage curve yet so whoever bought them is not too bright

[@MatWalter-q3h]

Yes and the first lithium car batteries burned lots of cars down too. There will be issues but the bottom line is sodium is one of the most PLENTIFUL elements in the world and lithium one of the rarest. end of discussion.

[@brooklynmurrau7926]

Nice video thanks Andy 👍cant wait for the next ones

[@OffGridGarageAustralia]

Thank you!

[@jamest.5001]

Seems there is little need to discharge below 2v -2.3v or above 3.8v roughly. I was distracted while watching this, but seems like not so great chemistry, at 1.3ah, they should be capable of 30-40 amp discharge , I just wonder if nak could make a battery (a sodium/potassium ion battery) maybe it could make the battery better. Or possibly sodium iron phosphate?

[@mikejones-vd3fg]

Interesting discharge curve, i suspect at 2.7v the sodium battery see's the lithium battery quit and gives it a boost of energy for the home stretch. We have a winner! ? Maybe this would be a good time to use a boost converter despite the losses you can keep a desired voltage while the cells discharge fullly.

[@OffGridGarageAustralia]

That's not going to work. The sodium battery will not boost the lithium battery, it will just take over the load at this point and keep quickly discharging further. To a point where the lithium BMS will disconnect and shut down you whole system due to a UVP. Really really bad to mix different battery chemistries...

[@mikejones-vd3fg]

@@OffGridGarageAustralia Oh i meant metaphorically, the soidum cell was racing the lihtium one and saw i t quit at 2.7v giving it a boost of extra energy to finish the race knowing it was going to win.

[@BobHannent]

I have some shitty solar lights which have served me well, but they probably need their batteries replacing. I am thinking about getting some of these 18650s for that use, I wonder if there are a single cell BMS for that application.

[@ElectricCarAustralia]

Very interesting stuff. Looking forward to more. 👍

[@OffGridGarageAustralia]

Thank you, Greg!

[@scottyb069]

I would like to see how they compare to the Lithium Titanate batteries. I have a little power station in the back of my ute with a 60W solar panel on the lid charging a 5s-3p LTO battery. This is used as a buffer and a removeable power pack and once charged feeds a slim 100ah litium battery through a dc-dc charger. This in theory keeps most of the cycling on the LTO battery and keeps the LiFepo4 topped up. I notice the number of cycles on the Sodium battery is only 3000, I was expecting better.

[@electro2606]

How much capacity you will have when you use a lifepo4 bms on a sodium battery? Edit so ovp 3.65 and ovp 2.5v