It turns out, this JKESS BMS is NOT made by JKBMS/JIKONG like all the other JKBMS we have tested and used for years (BD and PB series)! The JKESS BMS was sent to me by the JKBMS Official Store on AliExpress which is a distributor for JIKONG products, but this new BMS has nothing to do with JKBMS or JIKONG. It is a totally different product and not linked to, made or designed in any way by JKBMS or JIKONG. At the time of making this video, this was not clear to me. For a few months, I had in-depth discussions with the distributor about this JKESS BMS. At no point, I was told, that this is NOT a new JKBMS product. It looks like a JKBMS, it feels like a JKBMS, but it is NOT a JKBMS. It is a fake and counterfeit product. Unfortunately, the JKBMS Official Store tried to use the popularity of the genuine JKBMS (BD and PB series) to advertise, market and sell this product through their channels and I was fooled to believe it is a new genuine JKBMS. I apologise for making these false claims here in this video and platform, but it was unknowingly at the time of recording. A̶t̶ t̶h̶i̶s̶ p̶o̶i̶n̶t̶, I̶ w̶i̶l̶l̶ n̶o̶t̶ f̶u̶r̶t̶h̶e̶r̶ e̶x̶p̶l̶o̶r̶e̶ o̶r̶ t̶e̶s̶t̶ t̶h̶i̶s̶ p̶r̶o̶d̶u̶c̶t̶. I will continue exploring this new JKESS BMS as so many have requested in the comments. Thank you for your feedback I have spoken to the founder of JKBMS/JIKONG and they are still committed to further develop the firmware for the existing inverter BMS, both V14 and V15. I will keep working with JKBMS/JIKONG, providing feedback and testing new firmware versions.
Oh wow! After such deceptive behavior, I believe the first step should be to no longer send business to that distributor. Can you replace the Aliexpress link on your website with someone more trustworthy?
In any case, I guess this is good news for owners of a V14 and V15. JIKONG should have a lot of motivation to improve those two, to keep loyal customers from jumping ship.
Hi Andy, Do you know what I think about this whole situation? I believe they still haven’t finished (or maybe never even seriously started) fixing the issues with the previous BMS, but they thought it was a good idea to create a new one! I’m sorry, but in my opinion, this is not the way to operate. At this point, they needed to show more commitment, and to do so, they should have solved all the issues with the BMS inverter first. Then, they could have launched the new BMS at the same time. Instead, it seems like they’ve created a new problem to solve an old one! To me, this wasn’t a smart move on their part. If they don’t keep substantially improving the BMS inverter issues, people like me, who bought it believing in the reliability of J.K. as a company, will feel betrayed and look for alternatives. And surely, with today’s fast technological advancements, alternatives won’t be lacking! I really liked the way you expressed your concerns and how clearly and honestly you conveyed what you actually think. That’s something to be proud of and shows us that you don’t easily sell out to big companies, but prefer concrete results over empty talk. Well done, Andy, keep it up-you’re a legend! 😉😉 👋👋👋 Greetings from Italy 👋👋👋
Exactly that. Even when this New BMS would be the holy grail (which it isn’t) I would still be hesistant to buy it. The problem is that JK Has not proven trustworthy. They let users down by not communicating (not even thru resellers, I tried) and they don’t really Fix any problems. There also is a reason there is a V15 BMS.. what happened to the first 13 ? Anyway, buying into this does not seem smart. Using a cloud service without knowing if it will be there next year and so on. They lured a lot of OEM Partners in installing the V15, but JK Lets everybody down in the end. For me the company cannot be trusted to do anything of what is needed and when people are like that, they never change, you can put your money on that and people make or break companies. So, let them prove they are worthy of our trust in stead of using us as labrats. To be continued. Still looking forward to Andies top video about this BMS, but buying it ? No way 🫣
@@NaamNatuurlijknietTotally agree, they havent fixed v14 or v15, my soc % it all over the place, I was one of the first to swap over to v14 and it wasnt cheap, £185 in all to the UK. JK needs to show they care about all its customers, not just concentrating on the newest or latest model. I can safely say I will not be upgrading to this bms.
Finally a 4S BMS that has full communication and master/slave. This was missing so far as old 4S needed Peter board and also has no master/slave. Curious about your tests but if it's OK I will definitely buy it.
Great video Andy! Happy to be one of the - almost - 100.000 people that appreciate your efforts. At the point where cloud connection was mentioned, I stopped thinking this might be interesting. Still waitng for a daredevil company that makes it possible to hook up 4 strings of 18 EVE LF280K cells in parallel on a Victron (cerbo) system. After more than 18 months working without any glitch, the 123smart bms (made by the dutchies here 😀) is still the only BMS that combines 4 battery sets in the Cerbo without any issue at al. Would have liked to switch from cell level electronics to a centralised systems with cell wires to a box aside the battery cells but I am loosing confidence this will be available before I start building battery 3 and 4. Keep up the ood works und viel Spass!!
Whilst my Victron inverter uses/wastes over 70w ticking over, I'll not worry too much about the efficiency of the balancing... Ohh, not sure about the GPS function, but I do suspect the gyro sensing is more aimed at EV's where a roll over would trigger a power cut off... Not a lot of use in ESS applications, but can see why it might be there. The JK cloud is a little concerning.. Not sure I want my BMS 'online' directly, I especially don't want my BMS to demand WAN access to function (I know that's not being said, but it's a well trodden pathway for sure)... Great first look vid, thanks Andy
@@theshuff nope , don’t even think about trust to JK. I have just gone thru their AliExpress store which is advertising this new jkess product, has tons of typos. They didn’t even bothered to check spelling twice, I guarantee you first batch orders will face the same issue as the v14 inverter bms owners!!!
The analogue front-end they are using (MP2793) supports 58mA passive balancing, or can drive external transistors/mosfets for balancing. Would be nice to see a picture of the PCB to figure out if they're still doing active balancing, or just doing passive balancing and hiding that with fancy (my BS detector went wild) tech speak. Incidentally, this AFE is also the reason why they do high side switching - it's what this thing supports. I'm looking forward to the details!
Thank you very much for your analysis so far. More to come. Not sure if they ever release more tech details though as other Chinese companies are watching.
@@OffGridGarageAustralia Could you share some high quality pictures of the internals, especially the PCB? I would really like to know what MOSFETs they are using, if they have active balancing or not, other design choices they made, etc. If you can't share that publicly, you have my email :)
@@OffGridGarageAustralia By the way, the MPF4279x fuel gauge reference documentation describes the SoC algorithm they are talking about. The main issue with these is that they have a dedicated chip per battery chemistry, for example the MPF42793 is for LFP.
@@OffGridGarageAustraliaAndy can you please test it anyhow, you have it already. For offgrid it has not many new features that matters BUT for RVs and boats It offers in 4S and 8S a lot of features nowhere else available like master and slave as well as communication plus the high peak current values allow to use the BMS also on motors with high surge like engine starter, windlass and bowtruster. Additionally the jk inverter BMS stops at 200A in 24V which is too low for use with engine starters and in boating the house battery must be able to start the engine in emergency. For that we need a 300A BMS which is only available from JK without any communication and no master and slave which again we need. If you don‘t wanna test it ship it to me and I will do the test…I pay for shipping.
Glad you have plenty of LFP batteries to put at risk testing a new BMS for us. Sounds like balancing is DC-DC conversion between each cell separately and full battery stack voltage. I presume 'bidirectional' means can suck or push current to or from individual cells to total battery stack. A simpler version just bleeds moderate balancing current and pushes it to total battery stack (uni-directional balance bleed only) This is a common way done in automotive. Only issue I know of is it can sometimes result in the balancing current exchange to full battery stack screwing up the charging float/absorb trip points for charger causing charger to prematurely drop out of absorb phase when charger uses current taper off to determine absorb phase exit. Cells balance bleeding boosted up to battery stack can look like a current drop off to charger.
Excellent Overview Andy! Looking forward to a very detailed & long Review! I am looking to build a 24V 300+Ah LiFePO4 Battery Bank with this New JKESS BMS for my 3KW Deye Solar Inverter!
Interesting BMS, I have the old one and works great with your parameters (that makes a lot of sense), I wonder about the 'magic' of this one, we'll see in the next videos. Thank you Andy!
Andy that looks very similar architecture to an E&J BMS we recently tested (I believe JK and E&J are the same parent company). Switching the positive is not really an issue, however it's a bit strange to deviate from the norm so it didn't impress me. It also means more connections therefore more cost for lugs, wire and failure points. Our biggest problem with the BMS was the internal losses and subsequently the heat build-up. We found it had about double (or more) loss than the regular JK BMS we tested. We were only testing it at 12v and ran a full 3000w load to check losses. There were also a few alarming permissions noted for the App. Essentially it alluded that it can collect any data regardless of if the App is in use or not.
Increased losses is likely due to it using high side switching. P MOSFETS have a higher Rds(on) compared to equivalent N MOSFETS. The reasons they went with high side switching is because that's what that analogue front-end (AFE, the MP2793) supports.
@@upnorthandpersonal agreed, however this was a major step backwards in my opinion, at least when it comes to heat which is a high priority inside a small sealed battery like I make.
@5:24 the address on the back of the box is the same as NEEY, ANT, JKBMS, JIKONG, they are all interlinked. In the same buildings are many other businesses related to the power industry. They do this so, when something goes commercially badly wrong, the company with the debt walks away and disappears, and suddenly the premises have a new name, but same equipment and mostly the same staff. It might seem like a wild west approach, but it is how it is done in China. It is a low cost way to develop businesses, with the cost being borne by the consumers. Here in the West we do not (now) like this approach. Here in the UK it was exactly how our industry took off 100-150 years ago, with businesses going bust all over the place. What you get is rapid development, technological progress and advances at the lowest prices. Couple that with an almost non-existent commercial legal system, and anything goes. It is how the Chinese EV industry has taken off, although I doubt the state has much by way of a financial investment in BMSs (unlike the huge sums it is pouring in to EVs). America has much the same type of innovation potential, but its crippling commercial legal system negates the progress. In the EU, a massively burdensome regulatory system, coupled with a crippling commercial legal system, means any progress is very expensive and slow.
When visiting the Gobelpower booth at this year's interSolar fair in Munich, Gobelpower setup a big screen that showed Andy's test of their battery pack in endless loop -> the battery pack manufacturers are watching the videos on this channel as well. Andy, you wrote you are holding back with your opinion, but once you come out with it and also the opinion of the other people in this comment section would be interesting if any of us wants to see the JK-ESS in new battery packs as they come from Gobel, EEL, Yxiang, Jakiper and how all of them are called. I remember the battery pack manufacturers jumped pretty soon on the JK inverter BMS once it was on the market and sold it with their systems. With all the comments I already read here and also our experience with the update availability, I'm not so sure if quickly switching from the JK inverter BMS to the JK-ESS in the battery pack would really be in ours (=the customers) interest - this time (my opinion) they should at least wait half a year before doing that....
This BMS, and you did mention it, seems to store more for EV applications. Switching the positive seems more of an EV norm - I my be mistaking. TI, Analog devices, NXP, etc, all have chips that resemble and operated very much to the MPC2793. Can't wait to see the prelude of your next video on this topic.
For me that sounds that the BMS monitors the discharge and charge curve of each cell and than balances not based on the top voltage curve but countinously based on the SOC of each cell. Means it recognice slightly different capacities and „pre-charges“ each cell during the charge process so that all hit 100% SOC at the same time. Would be magic and over engineered
A friend of mine works at Sonnen (German solar battery manufacturer)...they where attempting that as well...but with passive balancers...because they have only low balance currents. I think it is not necessary with active balancing at all.
That is the basic concept of the AC storage systems from SAX Power in Germany using cascaded multilevel h-bridge inverter technology. Beside switching cells in series dynamically to creat AC voltage with an efficiency of nearly 99 %, they use a pool of cells from which on charge they select the cells with the lowest SOC, and on discharge the ones with the highest cells necessary to fit the current AC voltage in the sinus curve. This is done with 5 kHz. So they can do kind of wear leveling on all cells with overprovisioning, just like on an SSD. All cells are aging similar and the balancing is kind of automatic. This may then cause extended life-time of the complete storage, because instead of the weakest cell is defining the total battery capacity, the sum of the capacity of all single cells defines the total capacity. One can envision to do something similar for a poor DC system, but it seems to be JK left out the overprovisioning part, but it may be that depending on the calculated SOC of cells, they do switching of the cells. Normally the POWER MOSFET are just on or off and are not switched at high frequency. The mentioning of better EMC shielding may point into this direction. It may be interesting to look at the noise on the DC output with no load and under load. As they talk about a higher virtual voltage they may also discharge cell with higher voltage/SOC and convert the cell voltage up to the battery voltage and feed it back with battery voltage into the battery.
Thanks for the video always appreciated. To be fair, what is next Andy? JK_ASS as JK Affirmative Storage Solutions. I have purchased 3 version 14 and 7 version 15 inverter JKBMS. Unless there is no bring inverter BMS and get JKASS, I have zero intention to any new product.
@ Vielen dank. It is easy to get customer; but once lost our trust it is definitely hard to get the trust back. I have just go over their premier advertising in aliexpress and the very long pdf has tons of typos, which indicates that another careless product, else they didn’t even bothered to go over twice the publication.
Series pass, back-to-back MOSFET's are in positive battery line with MPS279x BMS analog frontend chips. This requires a charge pump to provide MOSFET gate drive 15v above battery voltage which is included in MPS279x chip. Only thing done with battery negative through line (P-, B-) in the BMS is series current shunt. I would prefer ability to add quality external shunt resistor, with external current shunt sense wires going to BMS inputs. Data stored in cloud is dictated by PRC so they can get access to your local network and leap-frog to other protected sites of interest to PRC.
@@JdZ-2023MPS is a Silicon Valley based company but with a lot of people with strong ties to China, wafer fabs in China, sorting facility in China, etc. With that said, I don't think MPS would bake spying in to the chip. JK might add in extra stuff to do that but I don't know. If MPS now is anything like the MPS I knew years ago, the chip should measure as good or better than the spec sheet and use lots of tricks like zero crossing switching to be efficient and get as much out of the onboard power devices as possible. An interesting thing would be to compare the JKESS circuit board to the MPS reference design.
The only thing we need is a BMS that works 100% and is error-free. What use is a BMS that is a few percent more efficient but kills my battery because it doesn't work error-free.
It is great that Sodium-ion is also included. I hope Victron comes with a MultiPlus kind for Prismatic Sodium-ion cels also. I really want a lot of those cells indoor home.
Alrighty then....... I've got v14 boards at the moment and will likely be adding v15 to the mix after the first of the year. It would be great to see those functioning as they should be before anything else happens. Mild interest in the ESS which isn't likely to communicate with the inverter model BMS's. Seems like a lot of hard coded function going on without any interaction available from the outside. Maybe not even a look into what's actually happening. Which in the end makes it all magic from the user perspective. For me, I'm very comfortable sitting it out and let the bleeding edge junkies hemorrhage money to figure out how and if the hocus pocus lives up to the hype.... Will remain tuned in Sir Andy. Good day my man!
Thanks for the video Andy, and so sorry you were hoodwinked by the CCP vendor trying to trick folks. However, I know many of us trust your ethics and will continue to be impressed with your tenacity and love for the alternative energy field. Everyone still benefits from your knowledge and I guess this trickster of a company attempted to ride off the coat tails of another firm and thought everyone would follow like the pied piper. Good for you for calling Jack Ass out. Cheers from Gary in SE Michigan.
As always, thanks for the interesting video. One question mark that's in my mind is the drain source breakdown voltage of 60V @8:30. 58.4V (which is generally the maximum charge voltage for 16s LiFePO4) is below 60V, but it's very close. I'm surprised they didn't use 80V or 100V MOSFETs. I'm sure they tested it to make sure it's sufficient, I'm just surprised by it.
If the relay contacts can be programmed to open when charging is disconnected by the BMS then this is exactly what I have been asking for to protect alternators from blowing their diodes when the BMS is used in camper and boat installations. I won't be rushing out to buy one to replace my older JKBMS's though until there is a Mk2. Never buy the Mk1 of anything! By the way Andy I got my Victron SOC calculations to be much more accurate by setting the charge efficiency to 96% instead of 100% as I could see that every night my grid charger kicked in (and out) at the same voltage and yet the SOC at this point was about 4% higher every day. Perhaps LifePo4 do not have the claimed charge efficiency of 100% or perhaps this just fudged the calculation just enough to be more accurate.
Sounds to me they did go back to passive balancing (balance during charge/ discharge). Wonder if software developing of the inverter BMS (V14/V15) is stopped and SOC measurement will not be fixed anymore. Could they make the inverter BMS software open source if they are not willing to fix/ develop the software anymore?
@@hichamelouatri566 If you have time, I would wait on how the new BMS develops. Inverter BMS was a software mass in beginning, think this will happen with the new one again. Now beside some imperfections the inverter BMS works quite well.
Cannot wait till you put that BMS through full test. Till now the only 4S BMS that has master/slave as well as full communication (assume connects to venus OS and HA). Installing mostly in boats and they still have un majority 12V. Keen to know how that BMS does.
What about the JK firmware updates we are missing ? I do not wanna buy something new when it is working. My drift has gotten even worser the last few weeks and the only way out is to fully charge within a week cause otherwise SOC is inaccurate and causes costs when running out juice like last week 2 time in the most expensive time of the day even though the soc had promised to have enough for 2 more hours.
@@OffGridGarageAustralia Danke, ich hab ein Spezialthema für Dich, der heißeste Scheiß und nur auf Deutsch. AKKU ENTLADE PAUSE = katastrophal hohe JK V15 Inverter BMS Standbyverluste ? Ich schalte mit Homeassistant ab X Uhr für Y Stunden (die günstigsten) per Automation beim JK BMS die ENTLADEOPTION aus. Uns beiden ist klar, das Haus kriegt dann günstigen Strom vom Netz und nach 5 h hab ich 3-4 kWh gespart für den Morgen, wo Tibber die 38 bis 50 Cent haben will. ABER: Ich hatte in Homeassistant für die Dauer der Pause im Chart eine horizontale SOC Linie zu sehen erwartet, wenn der Akku schon 5 Stunden frei kriegt bei dann 30% - 40% SOC, um danach auch am trüben Morgen bis 9 - 10 Uhr durchzuhalten. Rate mal, was aus den 30% geworden sind? Die waagerechte Mauer fällt zum Nachbargrundstück ab und ich hab da Verluste bis 4% in der letzten Woche erstmals gesehen. Ich hab den Balancer oberhalb 6 mV eingeschaltet und ich meine auch auf jenseits 3,35 V stehen, aber kann der Balancer 4% des Akkus in 5 Stunden verschlingen ? In 1 ganzen Nacht wären es mehr als 8 % oder nach 100 Betriebsstunden h ist der Akku leer. Kannst ja mal selber schauen, ob Du das auch beobachten kannst und was beim BMS das verursacht, denn 8% am Tage sind im 24h Betrieb auch 1 kWh Verlust, da der Akku ja mehr oder minder 247 aktiv ist und Verluste einfahren dürfte, sei es zum Laden oder Entladen. Ich bin auf Entladepause gekommen, weil der Strompreis in der Nacht augenblicklich zu hoch ist oder genauer zu nah an dem Preis, wo er später verbraucht werden soll wie dem frühen Morgen von so 7 bis 9 oder 10 Uhr. Ich brauch bei 28 Cent doch den Akku nicht laden, wenn morgen früh 36-38 Cent verlangt werden, weil ich bei 20% Verlusten auf dem Rundweg schon mit 34 Cent Gestehungskosten (ohne Kosten für Batt. Zyklen) rechnen kann. Da war es für mich smarter, bei absehbarem Engpaß am Morgen eben 3 bis 5 h Entladepause zu machen statt zu laden, aber bei 8% Verlust in 24 h sieht das ja gefühlt richtig übel aus.
Heya, 1st wanne see some testing before giving an opinion but using jk-cloud and all my info will be on the internet I don't know about that. better wait until you're done testing
This has been a well balanced channel with good and reliable info with german humor as a bonus. Now it is just Teleshopping and crap promotion monetizing reach . This channel is not only you, Andy. Is us (formerly) trusting your judgement and expertise. Go back to the point you lost that and try again.❤
If anyone could build a BMS "test" it would be you ! For example, a 4S LFE build with known good cells with 3x 120AH cells and 1x 100AH cell. With separate bi-directional (+/-) current/energy meters on each input and output as well as each balance lead it would allow you to quickly demystify all of these manufacture claims, as well as calculate BMS efficiency. I've not seen anyone do such a thing but it seems it would be pretty darn amazing !
oooh, such a contraption would also allow you to validate any efficiency claims of the different balancing boards, including external ones. Gosh that would be exciting !
Instead of measuring balancing current you can test it other way - discharge 2AH from one cell of the balanced pack and then mesure time it takes from connecting balancer to full balanced pack again. It shows us real current it use to pump discharged cell. But if I understand correctly this is less crucial if it starts balancing on any SOC level, wlit can have virtualy infinite time to do balancing if it's not top balance, and it can mess less with charger states in the process. But it's only if it works as should, not as previous ones (v14 v15)
Just checking the manufacturer datasheet of the MCU MP2793. It states that it has a passive balancer with up to 58mA per cell. Would be nice to get some more detail information how this „magical“ SOC balancing works…
Hi Andy, I am also enlightened, I buy dozens of bms from this supplier, I am really interested and it is not developed by jk, but still I say that if you have it, it is worth doing a test to see what it is all about. I will take possession of it only in November
Incredible build quality! However there are a couple of things I'm not a fan of. The data stored in the cloud? and the fact that they can push updates to directly to the BMS? These both make me question the safety of using this... That would mean they have access to potentially your network and everything else connected right? I'm trying to move as many things that I use to local control, as much as possible anyway lol. Great Review as always!
Watching a post in a channel community, I was expected to see a complete test in a next video... Will waiting it. I will have to assembly a battery before winter, so now even don't know what to choose - already old inverter BMS or new and fully unknown ESS board :/ Aliexpress claims to deliver new one for December 31 ))), by pre-ordering now :/ And yeah, price is not pleasant, and many pieces - optional for stupid price too
I understand the frustration of most of the people that the focus of JK has shifted. If I had a couple of the old JK inverter BMSes I would be frustrated too. However, I also understand JK. It seems they are looking to enter the EV market, so from their perspective this new version is with higher priority. Hopefully the old versions are not completely forgotten and with time they will fix any existing problems. The truth is, there aren't many alternatives that can do better job.
I would say that the increase of the interest to the BMS features - i.e. communication, balance, etc, created the effect of the "new cellphone" among the manufacturers. They stated to produce the completely new models, instead of "polishing" the existing ones. So soon we're gonna have big amount of the different models that might not be compatible with each other, even from the one manufacturer. But now when we have flood of the different batteries and bms on the market the compatibility of them is a huge topic. But manufacturers still play with own protocols, form factors, models, etc. Next they will put the camera and the SIM card into the bms, I guess.
I have the JK inverter BMS, and this seems to have a kind of a preview-version of the "SOC balancer", because the BMS "balances" between totally arbitrary SOC values.
I'm very very interested to explore this BMS through your videos and tests - experiments as well as previous BMSs. I just that this JKESS starts better than the JK Inverrter BMS.
I'm excited for the new form factor and quality of manufacturing. LEV's like scooters and ebikes need better BMS' so they don't catch fire while in use, and the high current option is very nice. I also like the use of the positive terminal on the battery and power side. It just makes installation a no brainer. Hopefully they don't have the positive cell connection along with the balance wires (I always thought that was unecessary)
Oh no, another BMS. Andy this new BMS seems more complicated coz too much wiring connections. To be honest I'm not looking for new 48v BMS, but now that I live in the Philippines I think that a 12v and 24v solar station are growing here coz it's affordable. There are many solar channels, but you explain everything much better and plus you're funny and that's a great balance for making videos. I like to see and hear from you about The Ecoflow systems out there and there type of batteries they use. Maybe that will double you subscribers. Love your work and details in all you do. Wish I could do the same.
I am not sure about the current limiter not being switched through the mosfet, I find that going through the mosfets for a better and safer way so you know that when the mosfets are off, it's off and no current is passing at all
Hey Andy, when doing your test could you make sure to test the heater function? the old style has a flaw to where the heater wont turn on until you hit the undertemp disconnect.
It would still be interesting to test to see how it performs. I like the size,a 50A version would fit well in some smaller 48V batteries. Shame uts not legit. Please still test !!!
Andy, why do you think LFP cells can only be top-balanced? Active JK bms can be set to start balancing at say 2,5V (this is how i set it). Only then, the 0% SOC can be truly achieved, aka bottom balance.
Tested v15.30 with LuxPower SNA 5000. The issue with stopped charging still occurs. Also it calculates 100 % SOC incorrectly. It sets 100% before high cell reaches set SOC-100% value... Hopefully they will fix these issues one day
The dry contacts. It’s labeled NC. Are you sure it’s Not Connected and not a relay output defaulting to Normally Closed? :) In electronics we usually label stuff not populated/in use as NA, not NC, as we use that for NC vs NO circuit.
first i thought they have completely messup something in the documentation, but now i think, the 4 green terminals are meant to connect to a external LCD in the driving cabin with an additional programmable dry contact IN or OUTput. But then the extension board has to be connected with another cable to the Wifi, GPS and maybe even the dry contact port of the BMS....then everything ould be extended into the drivers cabin. the additional interal GPIO Cable seems to be an INPUT for a programmable trigger But anyway...what a bummer :-D
Another new BMS... I would be happy if the inverter BMS received a firmware update to solve the existing problems. I hope JK doesn't forget that. The new one may be superior in hardware. But if the software continues to be developed just as badly, I don't see a future. First do the homework and more than 60%% would like an update for the V14 and V15.
Andy, I noticed you didn’t seem too enthusiastic about the JKESS BMS. Are there specific limitations or challenges you’re concerned about with it? I’m interested in your expectations for the BMS and would love to know if there’s anything about this model that you think might not fully meet them.
Very interesting. Thanks for giving us this overview. I am a big fan of the basic JK BMS and their balancers now that the SOC reset is better. SOC tracking still is a problem but that is inherent to LIFEPO4 (LIFEPO4 based cars need to be charged to 100% periodically due to this). They use a lot of fancy language and claims which may be lost in translation, but do they actually (all) apply to LIFEPO4? Please ask if you get a chance. "until battery pack capacity is less that 80% in the 13th year" statement makes me a bit nervous, but maybe it is just an example of one of the chemistries. In any case it will be very interesting to see your results.
Sooooo….oops? Sorry you got caught up in a marketing scheme. There have already been many posts stating how awesome this BMS is based on your review. 😢. I hope the product isn’t harmful to users who have already placed orders. I am glad you had reservations about it prior to finding out that it wasn’t really a JK product. I’d still like to know more about how it actually works in your garage-lab. I think that would be a great service to potential JKESS buyers-good performance, app, etc or bad.
My 2cnts as someone who saw a lot of pcbs with Mains and secondary Side only seperated with a small gap: To be fair, it seems powdercoated, so it is painted. Isolation gap should be absolutely fine, given the amount of air and voltage that this BMS is supposed to run at. You always can, just to be safe, put a plastic cover, capton Tape or another Isolation layer in between, nobody is going to stop you of course. But for normal Operation this is totally fine, from a design standpoint imho. - Happy to hear other thoughts with explanation tho.
So it will bottom balance as well. The litokola(?) test battery will get the most amps out of it by letting the other cells suffer from that one lesser cell and be "drain balanced" to 2.5 volts all together??? Hope we can prevent that by settings. Might be nice though in case of emergency. Grabbing a few more amps to call your kids before you loose power...or watch even more TH-cam....
Andy, could you please contact JK, and make them update the firmware on the old Inverter BMS, so that they ad a dry contact "Hi SOC" function? Curently theres only a Low SOC feature. That way i could use up exces solar in a water heater, based on SOC... For example, at 95% SOC the water heater turns on, and at 90% SOC it turns off. That would be a realy simplefied automatization for exces solar use!
I am waiting very hard for a fix for version 14 BMSes. They are nearly perfect. the only thing that is missing is the problem that they switch back from Absorption to Bulk when one cell gets to low. This should be fixed in 15.33 or 15.34 but these versions are nowhere to be seen. I still have Version 14 BMSes. Installed in paralell, connected to the cerbo, but not controling the charging because of this bug
would you give us an explantion video on how a bms works. I ask this because my head cant get around how it can measure the difference between each battery
hi, I'm talking about the PB-2A16S20P BMS, it happens to me that when it goes on charge after about a few seconds the soc reaches 100%, when it discharges the soc drops to 0% after about a few seconds; what's wrong?
Sounds interesting but the Doublespeak is somewhat dubious. (hiding behind the "Translator Curtain"). How will that Retrofic into cases with 4x RJ45 & LED's.
The SOC balance looks like the top balance I do. When testing batteries, I connect all the completed batteries in parallel, which is around 3.29v. When all batteries are tested I wait a day or two till all of them are balanced. Could the new BMS be doing something like that on the fly? All higher voltage batteries are charging all the lower voltage batteries to a calculated average voltage? Either way ... I'm happy with the inverter BMS's I have, and the EEL boxes, at the moment. They keep the car charged, so I'm good👍
Hi, there is a problem in the soc of the JK PB-2A16S20P; when it discharges it reaches 0% in a few seconds, when it charges it reaches 100% in a few seconds, does anyone know what it is?
What's the advantage of pulling through both + and - main terminals? More high current cables to worry about? Maybe because people didn't understand the wiring and kept blowing up their BMS's?
It would be great if you could remove the aluminium top plate and show the PCB, just to satisfy my curiosity. Also this balance technology just sounds like an insulated buck converter with programmable voltage. So they measure the average voltage of the cells, programm the buck converter and then charge the lowest cell with that voltage. This technology could reach up to 98% efficency, but the capacitor technology is easy 98-99% efficient in not loosing any charge but is horribly slow.
It's not the average voltage. Each cell contributes to the 'virtual voltage' which is pretty much the overall pack voltage. V1+V2+V3+V4 for a 4s battery.
@ That is the input voltage for the buck converter, but what is the output voltage? To which voltage will the most empty cell be charged? I bet they just take the average of all cells and set this as the target charge voltage of the most empty cell. So technically it is a single cell battery charger I guess.
Andy if you were to go out and buy a 16s bms right now what would be your pick, I currently have batrium and seplos 2.0 in action, but I am looking to bring online another two packs, and the bms market has evolved a lot. Also you should come down the road sometime to check out the new powerwall system, Hamish has my details.
@OffGridGarageAustralia: Andy can you please test it anyhow, you have it already. For offgrid it has not many new features that matters BUT for RVs and boats. It offers in 4S and 8S a lot of features nowhere else available like master and slave as well as communication plus the high peak current values allow to use the BMS also on motors with high surge like engine starter, windlass and bowtruster. Additionally the jk inverter BMS stops at 200A in 24V which is too low for use with engine starters and in boating the house battery must be able to start the engine in emergency. For that we need a 300A BMS which is only available from JK without any communication and no master and slave which again we need. If you don‘t wanna test it ship it to me and I will do the test…I pay for shipping.
Question: Any further info on Sodium-Ion Batteries....size, Ah, voltage, amps, etc...?> In Electronic's the Integrated Circuit Chips are usually a form of Micro-Processor which is in fact a Computer chip that controls various components and their functions automatically...... If you can find out what the Chip does it helps to understand the functional aspects of the Circuits and how it contreols certain applications....? =)
I'm stunned that they introduce a new product while they flail around not fixing the last one. Their older BMS are utterly horrible at SOC, I get a better picture looking at the voltage than even bothering to look at what the SOC says. And they can take their cloud functions and push them far up inside... I install their config app in a sandboxed profile because I don't trust it at all. I'd love to see more videos with competitors to inform my buying decisions, because I'm not likely to risk a JK purchase again any time soon.
The thing that I am very disappointed with is that they only sent you a 8S unit which means you cant really run it thru its paces in a 48V 16S battery like we use in our setups. I hope JK will correct this and get you a unit you can really use.
Apparently this is what they are working on now... They still develop the 'older' firmware for V14 and V15, but no dates have been set when we see a new release. I'm currently testing V14.23
Are they using the new lithium super capacitors (x10 capacity of tranditional capacitors) driven with the total battery voltage and then repass the "virtual" current to balance charge the lowest capacity cells? I think I get that logic as the super caps are charged by the charger but do they top up each cell during charge and discharge cycles by diverting current from the strongest cell to the lowest capacity cell? What's the bms cell wire size? For max current throughput to each cell? Interesting. Thank Andy. 100k in a blink
Eh, not really that interested, just want them to continue to improve the inverter BMS, I have two batteries now using that one and no plans to mess with a new BMS especially now all the battery boxes support the inverter bms.
@OffGridGarageAustralia probably not. Dad said your advice were getting these BMS has been brilliant frankly even the small issues with them have been not a big problem greatly appreciate all your content.
It is a shame the new unit is not backwards compatible with the screen. Previous versions used this same screen. If the data is stored on the cloud and the BMS that would be OK. However if only on the cloud then it would be dependent on the web at all times, not a preference for many. Since this is supposedly be available for EVs this seems unlikely. But then again, this is JK
Other than possibly powering the low voltage electronics what use could a JK BMS have in an EV? Even 300A at 48V isn't enough for a city car EV. The Hongguang mini is 20kW. Ah, now I see, the relay based 32S might be useful, ~100V at 500A could be made to work for a city car.
So we have a pig in a poke... We know that we don't know a lot and we have to believe that what the manufacturer says is the truth and nothing but the truth. This is where Andy's scope opens up. I believe that he will test this ESS as thoroughly as he has tested everything so far. Then we will see whether this ESS is: a sparrow or an eagle.
Never buying JK again! They totally lost my trust leaving me stuck with a faulty BMS unable to update its firmware! Despite many attempts to contact them and you, Andy!
It turns out, this JKESS BMS is NOT made by JKBMS/JIKONG like all the other JKBMS we have tested and used for years (BD and PB series)!
The JKESS BMS was sent to me by the JKBMS Official Store on AliExpress which is a distributor for JIKONG products, but this new BMS has nothing to do with JKBMS or JIKONG. It is a totally different product and not linked to, made or designed in any way by JKBMS or JIKONG.
At the time of making this video, this was not clear to me. For a few months, I had in-depth discussions with the distributor about this JKESS BMS. At no point, I was told, that this is NOT a new JKBMS product.
It looks like a JKBMS, it feels like a JKBMS, but it is NOT a JKBMS. It is a fake and counterfeit product. Unfortunately, the JKBMS Official Store tried to use the popularity of the genuine JKBMS (BD and PB series) to advertise, market and sell this product through their channels and I was fooled to believe it is a new genuine JKBMS. I apologise for making these false claims here in this video and platform, but it was unknowingly at the time of recording. A̶t̶ t̶h̶i̶s̶ p̶o̶i̶n̶t̶, I̶ w̶i̶l̶l̶ n̶o̶t̶ f̶u̶r̶t̶h̶e̶r̶ e̶x̶p̶l̶o̶r̶e̶ o̶r̶ t̶e̶s̶t̶ t̶h̶i̶s̶ p̶r̶o̶d̶u̶c̶t̶. I will continue exploring this new JKESS BMS as so many have requested in the comments. Thank you for your feedback
I have spoken to the founder of JKBMS/JIKONG and they are still committed to further develop the firmware for the existing inverter BMS, both V14 and V15. I will keep working with JKBMS/JIKONG, providing feedback and testing new firmware versions.
Oh wow! After such deceptive behavior, I believe the first step should be to no longer send business to that distributor. Can you replace the Aliexpress link on your website with someone more trustworthy?
In any case, I guess this is good news for owners of a V14 and V15. JIKONG should have a lot of motivation to improve those two, to keep loyal customers from jumping ship.
😮
Now I definitely want to see the PCB!
Not an issue. Do not be so hard on yourself. We all make mistake since we are only human.
Hi Andy,
Do you know what I think about this whole situation? I believe they still haven’t finished (or maybe never even seriously started) fixing the issues with the previous BMS, but they thought it was a good idea to create a new one!
I’m sorry, but in my opinion, this is not the way to operate. At this point, they needed to show more commitment, and to do so, they should have solved all the issues with the BMS inverter first. Then, they could have launched the new BMS at the same time. Instead, it seems like they’ve created a new problem to solve an old one!
To me, this wasn’t a smart move on their part. If they don’t keep substantially improving the BMS inverter issues, people like me, who bought it believing in the reliability of J.K. as a company, will feel betrayed and look for alternatives. And surely, with today’s fast technological advancements, alternatives won’t be lacking!
I really liked the way you expressed your concerns and how clearly and honestly you conveyed what you actually think. That’s something to be proud of and shows us that you don’t easily sell out to big companies, but prefer concrete results over empty talk. Well done, Andy, keep it up-you’re a legend! 😉😉
👋👋👋 Greetings from Italy 👋👋👋
Exactly that. Even when this New BMS would be the holy grail (which it isn’t) I would still be hesistant to buy it.
The problem is that JK Has not proven trustworthy. They let users down by not communicating (not even thru resellers, I tried) and they don’t really Fix any problems.
There also is a reason there is a V15 BMS.. what happened to the first 13 ?
Anyway, buying into this does not seem smart. Using a cloud service without knowing if it will be there next year and so on.
They lured a lot of OEM Partners in installing the V15, but JK Lets everybody down in the end.
For me the company cannot be trusted to do anything of what is needed and when people are like that, they never change, you can put your money on that and people make or break companies.
So, let them prove they are worthy of our trust in stead of using us as labrats.
To be continued.
Still looking forward to Andies top video about this BMS, but buying it ? No way 🫣
@@NaamNatuurlijknietTotally agree, they havent fixed v14 or v15, my soc % it all over the place, I was one of the first to swap over to v14 and it wasnt cheap, £185 in all to the UK. JK needs to show they care about all its customers, not just concentrating on the newest or latest model. I can safely say I will not be upgrading to this bms.
@NaamNatuurlijkniet hello thanks for your comment, which BMS would you recommend me please
Hence, I started the survey on my TH-cam channel.
Thank you for all your comments and sharing your thoughts. I'm holding back with mine...
Finally a 4S BMS that has full communication and master/slave. This was missing so far as old 4S needed Peter board and also has no master/slave.
Curious about your tests but if it's OK I will definitely buy it.
Great video Andy! Happy to be one of the - almost - 100.000 people that appreciate your efforts.
At the point where cloud connection was mentioned, I stopped thinking this might be interesting.
Still waitng for a daredevil company that makes it possible to hook up 4 strings of 18 EVE LF280K cells in parallel on a Victron (cerbo) system.
After more than 18 months working without any glitch, the 123smart bms (made by the dutchies here 😀) is still the only BMS that combines 4 battery sets in the Cerbo without any issue at al.
Would have liked to switch from cell level electronics to a centralised systems with cell wires to a box aside the battery cells but I am loosing confidence this will be available before I start building battery 3 and 4.
Keep up the ood works und viel Spass!!
Whilst my Victron inverter uses/wastes over 70w ticking over, I'll not worry too much about the efficiency of the balancing... Ohh, not sure about the GPS function, but I do suspect the gyro sensing is more aimed at EV's where a roll over would trigger a power cut off... Not a lot of use in ESS applications, but can see why it might be there.
The JK cloud is a little concerning.. Not sure I want my BMS 'online' directly, I especially don't want my BMS to demand WAN access to function (I know that's not being said, but it's a well trodden pathway for sure)... Great first look vid, thanks Andy
Thanks a lot for sharing.
"my Victron inverter uses/wastes over 70w ticking over" what system in detail please ?
98.3K I’m seeing this as it’s moving up and I’m smiling 😊 thank you for sharing your work.
Can we trust jk ? We still wait some support on their old hardware firmware !
@@theshuff nope , don’t even think about trust to JK. I have just gone thru their AliExpress store which is advertising this new jkess product, has tons of typos. They didn’t even bothered to check spelling twice, I guarantee you first batch orders will face the same issue as the v14 inverter bms owners!!!
I know, hence I did this survey before this new BMS came out.
@@OffGridGarageAustraliaand you interpreted that rather strange imo.
Survey felt like the old bms is obsolete. Buy the new bms instead.
@@OffGridGarageAustralia and you shouldn't promote new one (IMHO)
The analogue front-end they are using (MP2793) supports 58mA passive balancing, or can drive external transistors/mosfets for balancing. Would be nice to see a picture of the PCB to figure out if they're still doing active balancing, or just doing passive balancing and hiding that with fancy (my BS detector went wild) tech speak. Incidentally, this AFE is also the reason why they do high side switching - it's what this thing supports. I'm looking forward to the details!
Agree with this. There seem to be a lot more questions than answers at this time.
Thank you very much for your analysis so far. More to come. Not sure if they ever release more tech details though as other Chinese companies are watching.
@@OffGridGarageAustralia Could you share some high quality pictures of the internals, especially the PCB? I would really like to know what MOSFETs they are using, if they have active balancing or not, other design choices they made, etc. If you can't share that publicly, you have my email :)
@@OffGridGarageAustralia By the way, the MPF4279x fuel gauge reference documentation describes the SoC algorithm they are talking about. The main issue with these is that they have a dedicated chip per battery chemistry, for example the MPF42793 is for LFP.
@@OffGridGarageAustraliaAndy can you please test it anyhow, you have it already. For offgrid it has not many new features that matters BUT for RVs and boats It offers in 4S and 8S a lot of features nowhere else available like master and slave as well as communication plus the high peak current values allow to use the BMS also on motors with high surge like engine starter, windlass and bowtruster. Additionally the jk inverter BMS stops at 200A in 24V which is too low for use with engine starters and in boating the house battery must be able to start the engine in emergency. For that we need a 300A BMS which is only available from JK without any communication and no master and slave which again we need. If you don‘t wanna test it ship it to me and I will do the test…I pay for shipping.
Glad you have plenty of LFP batteries to put at risk testing a new BMS for us.
Sounds like balancing is DC-DC conversion between each cell separately and full battery stack voltage. I presume 'bidirectional' means can suck or push current to or from individual cells to total battery stack. A simpler version just bleeds moderate balancing current and pushes it to total battery stack (uni-directional balance bleed only)
This is a common way done in automotive. Only issue I know of is it can sometimes result in the balancing current exchange to full battery stack screwing up the charging float/absorb trip points for charger causing charger to prematurely drop out of absorb phase when charger uses current taper off to determine absorb phase exit. Cells balance bleeding boosted up to battery stack can look like a current drop off to charger.
Excellent Overview Andy!
Looking forward to a very detailed & long Review!
I am looking to build a 24V 300+Ah LiFePO4 Battery Bank with this New JKESS BMS for my 3KW Deye Solar Inverter!
Interesting BMS, I have the old one and works great with your parameters (that makes a lot of sense), I wonder about the 'magic' of this one, we'll see in the next videos. Thank you Andy!
Andy that looks very similar architecture to an E&J BMS we recently tested (I believe JK and E&J are the same parent company). Switching the positive is not really an issue, however it's a bit strange to deviate from the norm so it didn't impress me. It also means more connections therefore more cost for lugs, wire and failure points. Our biggest problem with the BMS was the internal losses and subsequently the heat build-up. We found it had about double (or more) loss than the regular JK BMS we tested. We were only testing it at 12v and ran a full 3000w load to check losses.
There were also a few alarming permissions noted for the App. Essentially it alluded that it can collect any data regardless of if the App is in use or not.
Increased losses is likely due to it using high side switching. P MOSFETS have a higher Rds(on) compared to equivalent N MOSFETS. The reasons they went with high side switching is because that's what that analogue front-end (AFE, the MP2793) supports.
@@upnorthandpersonal agreed, however this was a major step backwards in my opinion, at least when it comes to heat which is a high priority inside a small sealed battery like I make.
@@PowerPaulAu "this was a major step backwards in my opinion" --> absolutely.
@5:24 the address on the back of the box is the same as NEEY, ANT, JKBMS, JIKONG, they are all interlinked. In the same buildings are many other businesses related to the power industry. They do this so, when something goes commercially badly wrong, the company with the debt walks away and disappears, and suddenly the premises have a new name, but same equipment and mostly the same staff. It might seem like a wild west approach, but it is how it is done in China. It is a low cost way to develop businesses, with the cost being borne by the consumers. Here in the West we do not (now) like this approach. Here in the UK it was exactly how our industry took off 100-150 years ago, with businesses going bust all over the place. What you get is rapid development, technological progress and advances at the lowest prices. Couple that with an almost non-existent commercial legal system, and anything goes. It is how the Chinese EV industry has taken off, although I doubt the state has much by way of a financial investment in BMSs (unlike the huge sums it is pouring in to EVs). America has much the same type of innovation potential, but its crippling commercial legal system negates the progress. In the EU, a massively burdensome regulatory system, coupled with a crippling commercial legal system, means any progress is very expensive and slow.
always one step ahead ;) thanks Andy!!!
Thank you.
Can’t wait for test video and that high efficiency balance 😮
When visiting the Gobelpower booth at this year's interSolar fair in Munich, Gobelpower setup a big screen that showed Andy's test of their battery pack in endless loop -> the battery pack manufacturers are watching the videos on this channel as well.
Andy, you wrote you are holding back with your opinion, but once you come out with it and also the opinion of the other people in this comment section would be interesting if any of us wants to see the JK-ESS in new battery packs as they come from Gobel, EEL, Yxiang, Jakiper and how all of them are called.
I remember the battery pack manufacturers jumped pretty soon on the JK inverter BMS once it was on the market and sold it with their systems.
With all the comments I already read here and also our experience with the update availability, I'm not so sure if quickly switching from the JK inverter BMS to the JK-ESS in the battery pack would really be in ours (=the customers) interest - this time (my opinion) they should at least wait half a year before doing that....
This BMS, and you did mention it, seems to store more for EV applications. Switching the positive seems more of an EV norm - I my be mistaking. TI, Analog devices, NXP, etc, all have chips that resemble and operated very much to the MPC2793. Can't wait to see the prelude of your next video on this topic.
What is the status of V15? Still waiting for fixed SOC. That is over 50 percent of your responding audience.
Enjoyed the video. Thanks for sharing.
Thank you for watching!
For me that sounds that the BMS monitors the discharge and charge curve of each cell and than balances not based on the top voltage curve but countinously based on the SOC of each cell. Means it recognice slightly different capacities and „pre-charges“ each cell during the charge process so that all hit 100% SOC at the same time.
Would be magic and over engineered
A friend of mine works at Sonnen (German solar battery manufacturer)...they where attempting that as well...but with passive balancers...because they have only low balance currents.
I think it is not necessary with active balancing at all.
That is the basic concept of the AC storage systems from SAX Power in Germany using cascaded multilevel h-bridge inverter technology. Beside switching cells in series dynamically to creat AC voltage with an efficiency of nearly 99 %, they use a pool of cells from which on charge they select the cells with the lowest SOC, and on discharge the ones with the highest cells necessary to fit the current AC voltage in the sinus curve. This is done with 5 kHz. So they can do kind of wear leveling on all cells with overprovisioning, just like on an SSD. All cells are aging similar and the balancing is kind of automatic. This may then cause extended life-time of the complete storage, because instead of the weakest cell is defining the total battery capacity, the sum of the capacity of all single cells defines the total capacity.
One can envision to do something similar for a poor DC system, but it seems to be JK left out the overprovisioning part, but it may be that depending on the calculated SOC of cells, they do switching of the cells. Normally the POWER MOSFET are just on or off and are not switched at high frequency. The mentioning of better EMC shielding may point into this direction. It may be interesting to look at the noise on the DC output with no load and under load.
As they talk about a higher virtual voltage they may also discharge cell with higher voltage/SOC and convert the cell voltage up to the battery voltage and feed it back with battery voltage into the battery.
Thanks for the video always appreciated. To be fair, what is next Andy? JK_ASS as JK Affirmative Storage Solutions. I have purchased 3 version 14 and 7 version 15 inverter JKBMS. Unless there is no bring inverter BMS and get JKASS, I have zero intention to any new product.
Hence, I was holding back with my verdict about this new product😋
@ Vielen dank. It is easy to get customer; but once lost our trust it is definitely hard to get the trust back. I have just go over their premier advertising in aliexpress and the very long pdf has tons of typos, which indicates that another careless product, else they didn’t even bothered to go over twice the publication.
@@OffGridGarageAustraliaWe have noticed that. I could use another gear down. JK should move now.
Series pass, back-to-back MOSFET's are in positive battery line with MPS279x BMS analog frontend chips. This requires a charge pump to provide MOSFET gate drive 15v above battery voltage which is included in MPS279x chip.
Only thing done with battery negative through line (P-, B-) in the BMS is series current shunt. I would prefer ability to add quality external shunt resistor, with external current shunt sense wires going to BMS inputs.
Data stored in cloud is dictated by PRC so they can get access to your local network and leap-frog to other protected sites of interest to PRC.
Only assumptions and fear spreading, no facts.
@@JdZ-2023MPS is a Silicon Valley based company but with a lot of people with strong ties to China, wafer fabs in China, sorting facility in China, etc.
With that said, I don't think MPS would bake spying in to the chip.
JK might add in extra stuff to do that but I don't know.
If MPS now is anything like the MPS I knew years ago, the chip should measure as good or better than the spec sheet and use lots of tricks like zero crossing switching to be efficient and get as much out of the onboard power devices as possible.
An interesting thing would be to compare the JKESS circuit board to the MPS reference design.
You got that about the PRC.
The only thing we need is a BMS that works 100% and is error-free. What use is a BMS that is a few percent more efficient but kills my battery because it doesn't work error-free.
Oh yeah !! Thank you for your tests they were very helpful to us !
😉From France !
You are welcome!
It is great that Sodium-ion is also included. I hope Victron comes with a MultiPlus kind for Prismatic Sodium-ion cels also. I really want a lot of those cells indoor home.
Alrighty then....... I've got v14 boards at the moment and will likely be adding v15 to the mix after the first of the year. It would be great to see those functioning as they should be before anything else happens.
Mild interest in the ESS which isn't likely to communicate with the inverter model BMS's. Seems like a lot of hard coded function going on without any interaction available from the outside. Maybe not even a look into what's actually happening. Which in the end makes it all magic from the user perspective.
For me, I'm very comfortable sitting it out and let the bleeding edge junkies hemorrhage money to figure out how and if the hocus pocus lives up to the hype....
Will remain tuned in Sir Andy. Good day my man!
Thanks for the video Andy, and so sorry you were hoodwinked by the CCP vendor trying to trick folks. However, I know many of us trust your ethics and will continue to be impressed with your tenacity and love for the alternative energy field. Everyone still benefits from your knowledge and I guess this trickster of a company attempted to ride off the coat tails of another firm and thought everyone would follow like the pied piper. Good for you for calling Jack Ass out. Cheers from Gary in SE Michigan.
As always, thanks for the interesting video.
One question mark that's in my mind is the drain source breakdown voltage of 60V @8:30. 58.4V (which is generally the maximum charge voltage for 16s LiFePO4) is below 60V, but it's very close. I'm surprised they didn't use 80V or 100V MOSFETs. I'm sure they tested it to make sure it's sufficient, I'm just surprised by it.
Never mind, I wrongly assumed they were using the same voltage for 16s as it had the contacts on the BMS, but they're not.
If the relay contacts can be programmed to open when charging is disconnected by the BMS then this is exactly what I have been asking for to protect alternators from blowing their diodes when the BMS is used in camper and boat installations. I won't be rushing out to buy one to replace my older JKBMS's though until there is a Mk2. Never buy the Mk1 of anything! By the way Andy I got my Victron SOC calculations to be much more accurate by setting the charge efficiency to 96% instead of 100% as I could see that every night my grid charger kicked in (and out) at the same voltage and yet the SOC at this point was about 4% higher every day. Perhaps LifePo4 do not have the claimed charge efficiency of 100% or perhaps this just fudged the calculation just enough to be more accurate.
Will be interesting to see how well it works in practice. Presumably the software will give us some more insights once it's available.
Sounds to me they did go back to passive balancing (balance during charge/ discharge).
Wonder if software developing of the inverter BMS (V14/V15) is stopped and SOC measurement will not be fixed anymore.
Could they make the inverter BMS software open source if they are not willing to fix/ develop the software anymore?
hello thanks for your comment, which BMS would you recommend me please
It's very far away from any passive balancing.
They will continue with V14 and V15 software support.
@@OffGridGarageAustralia OK. What is the status of V15 updates to fix SOC, etc
@@hichamelouatri566 If you have time, I would wait on how the new BMS develops.
Inverter BMS was a software mass in beginning, think this will happen with the new one again.
Now beside some imperfections the inverter BMS works quite well.
@@OffGridGarageAustralia I dont see any.... Do you have a source, statement, timeline?
Cannot wait till you put that BMS through full test. Till now the only 4S BMS that has master/slave as well as full communication (assume connects to venus OS and HA). Installing mostly in boats and they still have un majority 12V.
Keen to know how that BMS does.
What about the JK firmware updates we are missing ?
I do not wanna buy something new when it is working. My drift has gotten even worser the last few weeks and the only way out is to fully charge within a week cause otherwise SOC is inaccurate and causes costs when running out juice like last week 2 time in the most expensive time of the day even though the soc had promised to have enough for 2 more hours.
Thanks for your comment.
@@OffGridGarageAustralia Danke, ich hab ein Spezialthema für Dich, der heißeste Scheiß und nur auf Deutsch.
AKKU ENTLADE PAUSE = katastrophal hohe JK V15 Inverter BMS Standbyverluste ?
Ich schalte mit Homeassistant ab X Uhr für Y Stunden (die günstigsten) per Automation beim JK BMS die ENTLADEOPTION aus.
Uns beiden ist klar, das Haus kriegt dann günstigen Strom vom Netz und nach 5 h hab ich 3-4 kWh gespart für den Morgen, wo Tibber die 38 bis 50 Cent haben will.
ABER: Ich hatte in Homeassistant für die Dauer der Pause im Chart eine horizontale SOC Linie zu sehen erwartet, wenn der Akku schon 5 Stunden frei kriegt bei dann 30% - 40% SOC, um danach auch am trüben Morgen bis 9 - 10 Uhr durchzuhalten.
Rate mal, was aus den 30% geworden sind?
Die waagerechte Mauer fällt zum Nachbargrundstück ab und ich hab da Verluste bis 4% in der letzten Woche erstmals gesehen.
Ich hab den Balancer oberhalb 6 mV eingeschaltet und ich meine auch auf jenseits 3,35 V stehen, aber kann der Balancer 4% des Akkus in 5 Stunden verschlingen ? In 1 ganzen Nacht wären es mehr als 8 % oder nach 100 Betriebsstunden h ist der Akku leer.
Kannst ja mal selber schauen, ob Du das auch beobachten kannst und was beim BMS das verursacht, denn 8% am Tage sind im 24h Betrieb auch 1 kWh Verlust, da der Akku ja mehr oder minder 247 aktiv ist und Verluste einfahren dürfte, sei es zum Laden oder Entladen.
Ich bin auf Entladepause gekommen, weil der Strompreis in der Nacht augenblicklich zu hoch ist oder genauer zu nah an dem Preis, wo er später verbraucht werden soll wie dem frühen Morgen von so 7 bis 9 oder 10 Uhr.
Ich brauch bei 28 Cent doch den Akku nicht laden, wenn morgen früh 36-38 Cent verlangt werden, weil ich bei 20% Verlusten auf dem Rundweg schon mit 34 Cent Gestehungskosten (ohne Kosten für Batt. Zyklen) rechnen kann.
Da war es für mich smarter, bei absehbarem Engpaß am Morgen eben 3 bis 5 h Entladepause zu machen statt zu laden, aber bei 8% Verlust in 24 h sieht das ja gefühlt richtig übel aus.
Can't wait your review 🎉
Me neither😄
Heya, 1st wanne see some testing before giving an opinion but using jk-cloud and all my info will be on the internet I don't know about that. better wait until you're done testing
This has been a well balanced channel with good and reliable info with german humor as a bonus.
Now it is just Teleshopping and crap promotion monetizing reach .
This channel is not only you, Andy. Is us (formerly) trusting your judgement and expertise.
Go back to the point you lost that and try again.❤
If anyone could build a BMS "test" it would be you ! For example, a 4S LFE build with known good cells with 3x 120AH cells and 1x 100AH cell. With separate bi-directional (+/-) current/energy meters on each input and output as well as each balance lead it would allow you to quickly demystify all of these manufacture claims, as well as calculate BMS efficiency. I've not seen anyone do such a thing but it seems it would be pretty darn amazing !
oooh, such a contraption would also allow you to validate any efficiency claims of the different balancing boards, including external ones. Gosh that would be exciting !
Instead of measuring balancing current you can test it other way - discharge 2AH from one cell of the balanced pack and then mesure time it takes from connecting balancer to full balanced pack again. It shows us real current it use to pump discharged cell. But if I understand correctly this is less crucial if it starts balancing on any SOC level, wlit can have virtualy infinite time to do balancing if it's not top balance, and it can mess less with charger states in the process. But it's only if it works as should, not as previous ones (v14 v15)
Just checking the manufacturer datasheet of the MCU MP2793. It states that it has a passive balancer with up to 58mA per cell.
Would be nice to get some more detail information how this „magical“ SOC balancing works…
That is basically all I got, what I already showed on the screen and what they sent in this spec sheet. It is a bit of magic🪄
Hi Andy, I am also enlightened, I buy dozens of bms from this supplier, I am really interested and it is not developed by jk, but still I say that if you have it, it is worth doing a test to see what it is all about. I will take possession of it only in November
Incredible build quality! However there are a couple of things I'm not a fan of. The data stored in the cloud? and the fact that they can push updates to directly to the BMS? These both make me question the safety of using this... That would mean they have access to potentially your network and everything else connected right? I'm trying to move as many things that I use to local control, as much as possible anyway lol.
Great Review as always!
Looking forward to first practical use/test of this BMS
Near 100k on a niche subject like this is very impressive!
Watching a post in a channel community, I was expected to see a complete test in a next video...
Will waiting it.
I will have to assembly a battery before winter, so now even don't know what to choose - already old inverter BMS or new and fully unknown ESS board :/
Aliexpress claims to deliver new one for December 31 ))), by pre-ordering now :/
And yeah, price is not pleasant, and many pieces - optional for stupid price too
Thats nice and all but how will the support be ? they realy gotta upppppppp that game.
Probably the same or worse (is that even possible??)
We will get to that later. This was just the new hardware and features.
My last JK Bms blew up after 3 months. Not happy.
Which model?
I understand the frustration of most of the people that the focus of JK has shifted. If I had a couple of the old JK inverter BMSes I would be frustrated too. However, I also understand JK. It seems they are looking to enter the EV market, so from their perspective this new version is with higher priority.
Hopefully the old versions are not completely forgotten and with time they will fix any existing problems. The truth is, there aren't many alternatives that can do better job.
I would say that the increase of the interest to the BMS features - i.e. communication, balance, etc, created the effect of the "new cellphone" among the manufacturers. They stated to produce the completely new models, instead of "polishing" the existing ones. So soon we're gonna have big amount of the different models that might not be compatible with each other, even from the one manufacturer. But now when we have flood of the different batteries and bms on the market the compatibility of them is a huge topic. But manufacturers still play with own protocols, form factors, models, etc. Next they will put the camera and the SIM card into the bms, I guess.
I have the JK inverter BMS, and this seems to have a kind of a preview-version of the "SOC balancer", because the BMS "balances" between totally arbitrary SOC values.
Nice BMS! I cannot find it on the links in the descriptions!
When is the next video scheduled? can't wait for it. Love the way you describe the tech stuff..
I'm very very interested to explore this BMS through your videos and tests - experiments as well as previous BMSs.
I just that this JKESS starts better than the JK Inverrter BMS.
Let's see what we can find when we connect it.
Thanks for the videos!
Greetings from Hungary!🤩
I'm excited for the new form factor and quality of manufacturing. LEV's like scooters and ebikes need better BMS' so they don't catch fire while in use, and the high current option is very nice. I also like the use of the positive terminal on the battery and power side. It just makes installation a no brainer. Hopefully they don't have the positive cell connection along with the balance wires (I always thought that was unecessary)
Equalization with reference to SOC-OCV curve sounds pretty cool 👍
Oh no, another BMS. Andy this new BMS seems more complicated coz too much wiring connections. To be honest I'm not looking for new 48v BMS, but now that I live in the Philippines I think that a 12v and 24v solar station are growing here coz it's affordable. There are many solar channels, but you explain everything much better and plus you're funny and that's a great balance for making videos. I like to see and hear from you about The Ecoflow systems out there and there type of batteries they use.
Maybe that will double you subscribers. Love your work and details in all you do. Wish I could do the same.
Nice new technology - but i am happy with old BMS as well as inverter BMS for all my application (sailboat & PV home)
I am not sure about the current limiter not being switched through the mosfet, I find that going through the mosfets for a better and safer way so you know that when the mosfets are off, it's off and no current is passing at all
Hey Andy, when doing your test could you make sure to test the heater function? the old style has a flaw to where the heater wont turn on until you hit the undertemp disconnect.
I'm very excited about the hocus-pocus that JK announces there 😅
It would still be interesting to test to see how it performs.
I like the size,a 50A version would fit well in some smaller 48V batteries. Shame uts not legit. Please still test !!!
10:59 The documentation you show says alarm I alarm II and alarm II? Where is alarm III? Lol
Andy, why do you think LFP cells can only be top-balanced? Active JK bms can be set to start balancing at say 2,5V (this is how i set it). Only then, the 0% SOC can be truly achieved, aka bottom balance.
The "Diverter" on the negative path is actually the current measurement shunt.
Shown at 13:10 and 16:40
Tested v15.30 with LuxPower SNA 5000. The issue with stopped charging still occurs. Also it calculates 100 % SOC incorrectly. It sets 100% before high cell reaches set SOC-100% value... Hopefully they will fix these issues one day
YamBMS with LuxPower protocol fixes this problem.
The dry contacts. It’s labeled NC. Are you sure it’s Not Connected and not a relay output defaulting to Normally Closed? :) In electronics we usually label stuff not populated/in use as NA, not NC, as we use that for NC vs NO circuit.
first i thought they have completely messup something in the documentation,
but now i think, the 4 green terminals are meant to connect to a external LCD in the driving cabin with an additional programmable dry contact IN or OUTput.
But then the extension board has to be connected with another cable to the Wifi, GPS and maybe even the dry contact port of the BMS....then everything ould be extended into the drivers cabin.
the additional interal GPIO Cable seems to be an INPUT for a programmable trigger
But anyway...what a bummer :-D
Another new BMS... I would be happy if the inverter BMS received a firmware update to solve the existing problems.
I hope JK doesn't forget that. The new one may be superior in hardware. But if the software continues to be developed just as badly, I don't see a future. First do the homework and more than 60%% would like an update for the V14 and V15.
Thank you.
Andy, I noticed you didn’t seem too enthusiastic about the JKESS BMS. Are there specific limitations or challenges you’re concerned about with it? I’m interested in your expectations for the BMS and would love to know if there’s anything about this model that you think might not fully meet them.
Your statement clarifies my concern. Thank you.
Very interesting. Thanks for giving us this overview. I am a big fan of the basic JK BMS and their balancers now that the SOC reset is better. SOC tracking still is a problem but that is inherent to LIFEPO4 (LIFEPO4 based cars need to be charged to 100% periodically due to this).
They use a lot of fancy language and claims which may be lost in translation, but do they actually (all) apply to LIFEPO4? Please ask if you get a chance.
"until battery pack capacity is less that 80% in the 13th year" statement makes me a bit nervous, but maybe it is just an example of one of the chemistries.
In any case it will be very interesting to see your results.
Sooooo….oops?
Sorry you got caught up in a marketing scheme. There have already been many posts stating how awesome this BMS is based on your review. 😢. I hope the product isn’t harmful to users who have already placed orders.
I am glad you had reservations about it prior to finding out that it wasn’t really a JK product. I’d still like to know more about how it actually works in your garage-lab. I think that would be a great service to potential JKESS buyers-good performance, app, etc or bad.
I do not really appreciate the idea to have massive positive and negative power wires so close to a thick aluminium plate, with just minimal air gaps.
My 2cnts as someone who saw a lot of pcbs with Mains and secondary Side only seperated with a small gap: To be fair, it seems powdercoated, so it is painted. Isolation gap should be absolutely fine, given the amount of air and voltage that this BMS is supposed to run at. You always can, just to be safe, put a plastic cover, capton Tape or another Isolation layer in between, nobody is going to stop you of course. But for normal Operation this is totally fine, from a design standpoint imho. - Happy to hear other thoughts with explanation tho.
@@slythee7423, 32S bms-es incoming...., well, I don't like it either
Yeah, that could get exciting 💥🔥⚡️🧨
So it will bottom balance as well. The litokola(?) test battery will get the most amps out of it by letting the other cells suffer from that one lesser cell and be "drain balanced" to 2.5 volts all together??? Hope we can prevent that by settings.
Might be nice though in case of emergency. Grabbing a few more amps to call your kids before you loose power...or watch even more TH-cam....
Andy, could you please contact JK, and make them update the firmware on the old Inverter BMS, so that they ad a dry contact "Hi SOC" function? Curently theres only a Low SOC feature. That way i could use up exces solar in a water heater, based on SOC... For example, at 95% SOC the water heater turns on, and at 90% SOC it turns off. That would be a realy simplefied automatization for exces solar use!
Сделайте это через реле
I can add it to the list, but atm they don't even fix the important bugs.
@@OffGridGarageAustraliaopen a bug tracker and put those bugs to it.
Gobel is already selling and shipping jkess 16s 150a. I'm hoping it's the pb type like I already have
I am waiting very hard for a fix for version 14 BMSes. They are nearly perfect. the only thing that is missing is the problem that they switch back from Absorption to Bulk when one cell gets to low. This should be fixed in 15.33 or 15.34 but these versions are nowhere to be seen. I still have Version 14 BMSes. Installed in paralell, connected to the cerbo, but not controling the charging because of this bug
would you give us an explantion video on how a bms works. I ask this because my head cant get around how it can measure the difference between each battery
hi, I'm talking about the PB-2A16S20P BMS, it happens to me that when it goes on charge after about a few seconds the soc reaches 100%, when it discharges the soc drops to 0% after about a few seconds; what's wrong?
Sounds interesting but the Doublespeak is somewhat dubious. (hiding behind the "Translator Curtain").
How will that Retrofic into cases with 4x RJ45 & LED's.
The SOC balance looks like the top balance I do. When testing batteries, I connect all the completed batteries in parallel, which is around 3.29v. When all batteries are tested I wait a day or two till all of them are balanced. Could the new BMS be doing something like that on the fly? All higher voltage batteries are charging all the lower voltage batteries to a calculated average voltage?
Either way ... I'm happy with the inverter BMS's I have, and the EEL boxes, at the moment. They keep the car charged, so I'm good👍
Hi, there is a problem in the soc of the JK PB-2A16S20P; when it discharges it reaches 0% in a few seconds, when it charges it reaches 100% in a few seconds, does anyone know what it is?
I am in class, and I don't have a clue what is being said, but I know I am at the right place 😂😂
What's the advantage of pulling through both + and - main terminals? More high current cables to worry about? Maybe because people didn't understand the wiring and kept blowing up their BMS's?
It's because of that AFE they're using. It only supports high side switching.
It would be great if you could remove the aluminium top plate and show the PCB, just to satisfy my curiosity. Also this balance technology just sounds like an insulated buck converter with programmable voltage. So they measure the average voltage of the cells, programm the buck converter and then charge the lowest cell with that voltage. This technology could reach up to 98% efficency, but the capacitor technology is easy 98-99% efficient in not loosing any charge but is horribly slow.
It's not the average voltage. Each cell contributes to the 'virtual voltage' which is pretty much the overall pack voltage. V1+V2+V3+V4 for a 4s battery.
@ That is the input voltage for the buck converter, but what is the output voltage? To which voltage will the most empty cell be charged? I bet they just take the average of all cells and set this as the target charge voltage of the most empty cell. So technically it is a single cell battery charger I guess.
Andy if you were to go out and buy a 16s bms right now what would be your pick, I currently have batrium and seplos 2.0 in action, but I am looking to bring online another two packs, and the bms market has evolved a lot.
Also you should come down the road sometime to check out the new powerwall system, Hamish has my details.
@OffGridGarageAustralia: Andy can you please test it anyhow, you have it already. For offgrid it has not many new features that matters BUT for RVs and boats. It offers in 4S and 8S a lot of features nowhere else available like master and slave as well as communication plus the high peak current values allow to use the BMS also on motors with high surge like engine starter, windlass and bowtruster. Additionally the jk inverter BMS stops at 200A in 24V which is too low for use with engine starters and in boating the house battery must be able to start the engine in emergency. For that we need a 300A BMS which is only available from JK without any communication and no master and slave which again we need. If you don‘t wanna test it ship it to me and I will do the test…I pay for shipping.
Question: Any further info on Sodium-Ion Batteries....size, Ah, voltage, amps, etc...?> In Electronic's the Integrated Circuit Chips are usually a form of Micro-Processor which is in fact a Computer chip that controls various components and their functions automatically...... If you can find out what the Chip does it helps to understand the functional aspects of the Circuits and how it contreols certain applications....? =)
No further information on sodium or these chips. I'm not a micro-electronics expert.
bitte, wir brauchen ein Video über die Einstellungen von Felicity BMS
I'm stunned that they introduce a new product while they flail around not fixing the last one. Their older BMS are utterly horrible at SOC, I get a better picture looking at the voltage than even bothering to look at what the SOC says. And they can take their cloud functions and push them far up inside... I install their config app in a sandboxed profile because I don't trust it at all.
I'd love to see more videos with competitors to inform my buying decisions, because I'm not likely to risk a JK purchase again any time soon.
The thing that I am very disappointed with is that they only sent you a 8S unit which means you cant really run it thru its paces in a 48V 16S battery like we use in our setups. I hope JK will correct this and get you a unit you can really use.
Hey, where is the firmware for the jk inverter bms? (((
Apparently this is what they are working on now... They still develop the 'older' firmware for V14 and V15, but no dates have been set when we see a new release. I'm currently testing V14.23
Are they using the new lithium super capacitors (x10 capacity of tranditional capacitors) driven with the total battery voltage and then repass the "virtual" current to balance charge the lowest capacity cells? I think I get that logic as the super caps are charged by the charger but do they top up each cell during charge and discharge cycles by diverting current from the strongest cell to the lowest capacity cell? What's the bms cell wire size? For max current throughput to each cell? Interesting. Thank Andy. 100k in a blink
Eh, not really that interested, just want them to continue to improve the inverter BMS, I have two batteries now using that one and no plans to mess with a new BMS especially now all the battery boxes support the inverter bms.
Thank you for sharing. You are not one of the 38% in the survey, ey?😄
@OffGridGarageAustralia probably not. Dad said your advice were getting these BMS has been brilliant frankly even the small issues with them have been not a big problem greatly appreciate all your content.
It is a shame the new unit is not backwards compatible with the screen. Previous versions used this same screen. If the data is stored on the cloud and the BMS that would be OK. However if only on the cloud then it would be dependent on the web at all times, not a preference for many. Since this is supposedly be available for EVs this seems unlikely. But then again, this is JK
Sounds like threy want to calculate SOC for each cell individually and balance them according to this calculatons.
Oooh, up to 36 cells will finally be possible! I can build my nominal 115 volt system!!!
Don't get zapped!
Other than possibly powering the low voltage electronics what use could a JK BMS have in an EV? Even 300A at 48V isn't enough for a city car EV. The Hongguang mini is 20kW. Ah, now I see, the relay based 32S might be useful, ~100V at 500A could be made to work for a city car.
There is no way a bms like this is ever used in a ev
andy , was machen die ammos? da muss doch was passiert sein :)
hello thanks for your comment, which BMS would you recommend me please
So we have a pig in a poke... We know that we don't know a lot and we have to believe that what the manufacturer says is the truth and nothing but the truth.
This is where Andy's scope opens up. I believe that he will test this ESS as thoroughly as he has tested everything so far. Then we will see whether this ESS is: a sparrow or an eagle.
Never buying JK again! They totally lost my trust leaving me stuck with a faulty BMS unable to update its firmware! Despite many attempts to contact them and you, Andy!
Relire le dernier message sur la vidéo de Andy, vous vous trompez.