The only "safe" default settings would be "unconfigured" ... So an unconfigured BMS would do *nothing* until it has been configured by the user ... Then the real bug to fix is the *volatile* settings that should stay "put" ...
Thanks Andy for the great work bringing this to the attention of users and working with JK for the fix I can't help but feel there should be an asterisk after the fixed in the video title, and a footnote saying that JK fixed the issues that customers have found so far I feel like the firmware updates is as much a blessing as a curse as it gives developers the temptation to push buggy software out and just let the customers find the issues rather than do any real testing. Anyone who plays video games has no doubt experienced this in recent years where customers are effectively paying to be beta testers I'm not going to single JK out for this, I wouldn't be at all surprised if the other BMS manufacturers had similar bad practices I wish that they would understand that if they have a major software bug that it doesn't just cause an inconvenience for the user but could lead to a serious hazard like the one JK just fixed
Here is my analysis. The issue was with the over voltage protection recovery. When you sent to LFP it sets it to around 3.54v or something. But I went to RCV to 3.45 and 3.449v for the 100% SOC like you did in your example. The NEW firmware doesn't allow you to set those if your Cell OVPR is set to 3.54. So I have to set THAT to 3.448v. But because the default settings were allowing you to set it, that's where the issue came from. I think they need to update their mobile app with error codes telling you WHY you can't sett those settings.
@@carmichaelmoritz8662 it's the overall design of the software and how it interacts with the user. It gives out the same error message regardless what settings are wrong. The FW itself has a strange logic it follows, like OVP>RCV>SOC100%>OVPR.
@@OffGridGarageAustralia maybe the people that write the code software are purposely writing in the strange logic. Logic should never be strange unless its on purpose or the person writing it is strange.
I knew you would get it figured out and work with the manufacturer to get things fixed! 🙂 Great job! It's still a bummer that it won't just leave the settings alone and not reset ANYTHING when powered off.
Agree. The BMS need to remember what it was set for. Its not like this is their first BMS. the Old and now goldie JKBMS active balance version remembers!
Thanks Original Andy!! Impressive that you receive after the weekend the new fix and then another 30 minutes your recommendation Has been Adapted as well. This is a huge Praise to your Expertise and the solar community. Well done. You deserve your andy Assi fresh beers.
@@OffGridGarageAustralia and you have the right to feel so. After all you've been doing this stuff for a long time already. It should come naturally to you to know all to do with what you do.
Well done JK ,🤞that will silence a few overreacting critics. Great effort Andy you have taken on a lot of work leasing with JK and informing the world. Are Andy's 2,3&4 on the naughty step with Jk?
Hi Andy. I must have been one of your first viewers/subscribers. I came across the channel because I had purchased the same EVE cells some 3 odd years ago at the same time you did. We went in slightly different directions - mainly due to monetary constraints my end. I am a pensioner with a very fixed income. We both chose the best battery option (hindsight is great). We both went with second hand panels. Then you went Victron, I couldn't so I went Taiwanese all in one Inverter/Charge controller. I feel like I have learnt the ropes alongside you, looking over your shoulder so to speak. You seem to love tweaking your system and making things easier to monitor and more efficient as a result - I went through that phase. Whilst I find all this very interesting and follow each of your projects, my goals are somewhat different - no UA-cam channel obviously. My goal was (and still is) to build a system that just works and requires very little monitoring. A drop in replacement for the grid if you like. Minimal maintenance, minimal monitoring. Just what is required for safety. Anyway I have an idea for a new project if you are looking for one. What about looking at the lower end full system? Whilst we are fortunate that a lot of people in Australia can afford a big, top of the line setup, some like myself are not so lucky but still want to reduce their carbon footprint and their power bill - and this is possible. Anyway, I suppose you are flat out and have lots of projects coming up - and I look forward to them - just thought I'd mention it.
Thanks a lot for your feedback and being around for so long. I really appreciate that. The problem with building such a new 'low end' system as you suggest is, what am I supposed to do with it afterwards? Where should I mount it, install it? Who should I test it? Who is paying for it? I just don't have the space for another separate system. BUT, in saying that. there will be some changes to my setup in the near future which may enable me to test other inverters as well as batteries.
After testing some cells i understood every setting needed and the fact that i don't like those cells when used to make a battery. If only they could make twelve volt cells , one could run them parallel and they would all discharge evenly. no need for a bms to read four 3.2 @3.65 volt cells to make 12 volt or three 4.0 volt cells uvp or ovp. You'd only need one cell uvp or ovp to monitor the complete 12 volt parallel system.
Thanks Andy and folks in community. Excellent job. For god sake, hopefully I can get my PC and BMS connection set up on this weekend, new RS485 module just arrived. Looks like one Andy has but it came with CD. For god sake who still have CD player, hope standard windows drivers will work.
you would have to ask the JK engineer to modify the smartphone application, about the choice of chemistry (lithium ion, lifepo4, LTO), and to add a popup confirming the change, because if you inadvertently click on another chemistry, you lose all your specific data.
This still is a risk. LTO has a lower bottom voltage so you can damage your batteries with a low voltage scenario. It fixes the top but then makes it worse. They need to be storing this properly in non-volatile memory. If they don’t have it, they need to recall all of them and add it.
The settings just needs to be written to the chips registry, why they can't set aside certain reg values for this is beyond me.... Why are the other BMS's not suffering this issue, unless they used an inappropriately sized chip for this, it should be totaly possible, this is a bandaid not a fix....
This is simply a least effort quick hack to handle the backlash of bein plain dangerous. Damaging batteries aren't that bad, and to be expected it seems.
@@GapRecordingsNamibia they are. We just haven’t found the bug that will corrupt the memory yet. What’s happening is that it encounters an invalid setting and instead of loading a safe default that turns off balancing, charge and discharge with an error but loads all other settings, it’s just defaulting to factory. They just changed the factory defaults. Bad hack to solve the real issue which is they need to understand what a settings error should result in, which is safety mode.
11:45 The RCV is meant to be lower than the OVP. The system first line of defence is to prevent the error. Even after he bypassed it, by saving the Start Balance Voltage which acidentally sets the wrong RCV to 3.65V. This makes the RCV higher than OVP. it still did not reset the BMS to default setting, even after switching it off for sometime. I feel it has been resolved
Since my new install of my big battery shelf and inverter the electric heater in my workshop has only been on low with a more comfortable temperature inside it, the extra passive warmth as a byproduct is more than welcome on these frosty mornings in 🏴. Helping me save even more money on heating too. Not too toasty warm and no equipment or cells are hot.
Good thing they responded so fast. That's a great service. I do agree however that maybe it should not work at all and show some LED or something (if it even has) or just pop a message in the app
Greatings from Ukraine! Thank You very much for Your work! I have such BMS with HW v15. I updated it to version v15.17 from v15.11 based on Your instructions, but I can not find any changelogs for this firmware. Does this bug persist in v15.17? Is there any way to read what have changed between FW versions? Thank You in advance!
Heya, good that jk so soon came with a solution hope no more surprises will be there so I can start ordering this bms. have to order a new hybrid inverter because the "old" 1 I have doesn't have canport. so the new inverter has to be modbus/can port for communication, even do it's not that important for me as I'm gried tied. w'll see what the future will bring. oh yeah I'm totally up to date now with your vlogs.
Why can't JK also release firmware updates for their old BMSs? I am sure they can be updated... at least to fix the SoC reset with an extra parameter or so... Andy, could you try your contacts? They will make the huge JKBMS owners community really happy...
The chips they used don't support that. They can be written once and for an update they may have to be taken out and re-programmed if this is possible at all.
@OffGridGarageAustralia I see... it's a shame because this hardware is not that bad at all... maybe they can add the feature to the new units... because if they compile a new firmware and share it, the community can figure out how to replace/reprogram the chip 🤔
The "issue" is still not fixed, it still rests just to LTO which has it's own set of issues towrd LFP, they can discharge to low now. This to can damage the LFP cells, so, not a fix bit a bandaid.....
Instruction how to update if an update failed and the BMS does not respond anymore: - PC Software select firmware, force update - enter PIN provided by JK support (only valid for 30mins), update is on hold at 0% - press and hold activation button - shortly press reset button and keep holding the activation button until update is finished BMS is back in the game!
@@stephanklasen2928 I managed to fix mine, its kinda simple to do You need to force update it and to do that you need to get a key from JK at the time you are working because this key lasts only for 3 minutes so make sure you contact them they can help you Btw you didnt do anything wrong it just happened to me for one out of 6 and to Andy as well
Thx again Andy for the update and all your effort! I dont understand why they reset the bms back to default profile settings after the restart - it would be much easier to debug for the user to turn off charge and discharge mosfets and trow an error (LED + CAN + RS485 + BT APP) - thats the reason why there are alarm and error functionality right?!? Thats what i would expect in case of any wrong settings - and nothing else But it is also not a real reset to factory defaults - as i can remember any of my bought JKs where delivered with Mosfets and Balancer turned OFF after the very first start. Again a quick shot, but again overall not a good job from JK here 👎👎👎 But anyway - PROST 🍻
Surely if on a cold boot for any BMS, if any saved setting parameters (or memory checksums etc) are implausible, then the system should not operate at all. It should not connect the battery outputs nor allow charge or discharge or balancing to occur. It should just sit their in this error state until the settings are corrected, resaved and the system is full rebooted??
Hi Andy I've been following your channel for a long time and I really like it, congratulations. This update solves the charging voltage problem, but what about the discharging voltage? It is necessary to program the minimum operating voltage in the inverter, and I don't know if the BMS will not change this value in the inverter itself. Well done.
I see that today they released V14.20 and V15.10 already... Hope they added the low temperature battery alarm for dry contacts... Can't wait to update the firmware this evening...
@@mateuszmazur1482 not yet, the V14. 20 log is telling me that the bluetooth is improved and something other thing is modified, but nothing about the dry intact low temperature battery alarm... So I am still waiting another version of firmware..
Something is going on out there... Today I asked official JK shop on Alli, where is my heating cable because cable is missing in a package. And they answer to me, that this is by purpose, because I don't need it, because users have some issues with heating function so they resigned from this function (sic!)... What the hack?
@@mateuszmazur1482 I know nothing about this issue. I bought my bms on November and i got the package complete, no missing cables or anything... And for me is not so important, I don't intend to use the heating function anyway...
Andy, it might be cold in the Northern Climates during certain months but we are not sweating our ashes off for months of a time either. It all depends on what you tolerate better. For me, I would rather be cold for a handful of months because I can still put on a jacket, hat and gloves and enjoy being outdoors. I cannot tolerate being outdoors in extreme heat like they have in warm weather climates. I got a taste of this living in Florida for a few years. I HATED the summers and feeling like a wet blanket if I had to be out in the heat. Give me cold any day! You indicated subscribers made comments in the community section of your site. I clicked on it and it appears all the entries are from you. Is the community section like a blog or just a place for you to add dialog?
Let’s just agree that MID temps are the best. 😁 I don’t live too far from Andy, and I “survive” our Summer. I’ve experienced extreme cold, very short days and horrible weather. I hated it. I’ve experienced stinking heat with extreme humidity. I hate it (like we’re getting at home in Brisbane right now, more like wet season Darwin than Brisbane.) However at least we can retreat to the aircon. Brisbane in Winter- usually fabulous weather- lots of sun and beautiful days. (I LOVE 💕 our Winter.) P.S. I don’t think anywhere in the world gets great weather 365 days a year.
@@FutureSystem738 San Diego, CA??? The city remains within a comfortable range between approximately 50 and 75 degrees all year long and receives just 43 days of precipitation on average.Sept 13, 2023
@@jws3925 Sounds pretty good, though I must say that the last time I visited there, circa 2000, the weather was pretty terrible. (I THINK it was late in the year … but not 100% certain.)
@@daisydonald14 it's been 5h since I made this comment. A lot can happen in the world in this time😆 Honestly, I don't know what I should think about JK right now. They have this amazing product called PBxx inverter BMS but the software seems to be the weak point right now. We (not them) discover bugs, they fix it. New bugs, new firmware updates. Since the release 3 months ago, we have seen 13 updates. Is that good or bad?
@@OffGridGarageAustralia From one side it's good, from the other it's not. At least they react to it. There are companies that don't do that. It's now normal that we're all just testers and guinea pigs, regardless of whether it's smartphones or software. And I say again, it would be bad if JK didn't do anything. Do you know what the difference is between v14 and v15 hardware version? I got my first one yesterday and it's a v15 with 15.05 software version. Thank you again for your involvement in this matter. Without you they would have had a lot of scrap now.😃 Danke ,danke
I’m looking to building a battery. Was keen on this bms. The range of problems does not inspire confidence on the brand. I need a nature and reliable bms. Suggestions?
Do you need/want coms? This JK is still the best choice out there, I would say. At least they constantly refining the product with new firmware. If no coms, go with the older style BD JK-BMS and bluetooth connection.
LTO cells have a UVP of 1,5V per cell x16 = 24V If the inverter is set to follow SOC% sent by the BMS it will destroy every battery pack at first discharge after the reset. What makes this safe? What did I miss?
LTO is 2.2V discharge right? Isn't that a bit way too low for LFP and Li-Ion? Low voltage can cause dendrite damage and short a cell, discharging it over time?
My JK BMS arrived yesterday so I power it up for checking the firmware. But my JK-PB2A16S-20P has the hardware Vers. V15.XA and the firmware Vers. 15.03. This Firmware has the same bug.. Where can I get the newest firmware?
is -60c cold? Here we just have some LN2 to warm up, it is annoying when you step in a puddle of propane though, wet boots are all ways crappy! I do indeed need heater blankets for our batteries.
Hi Andy! thanks for all your videos. got a JKBMS up and running, i have a suggestion for a new feature.. on the breakout pcb there is a push button, this can turn off the bms, but it can not turn it on.. maybe an upgrade for the next firmware could implement this button with the same features as the external single button, so we can skip that in our setup :-) and do you have a suggested document of the settings of your choice for lifepo4?., Thanks again.
The push button on the communication PCB is just a reset button to restart the BMS. This will also reset the settings (but not the password). You BMS should have a normal power button to turn it on and off?
when i press this button on the pcb it turns off the bms, i then have to use the normal power button to turn the bms on :-) i for a second thought the button on the communication board was a turn off only button, but i have since learned that it is a reset button, and now again :-) thanks @@OffGridGarageAustralia
They _do_ have non-volatile memory, what appears to be the case is that the startup routine has a validation routine that gets triggered, and if it determines the config to be invalid, it just picks some default values and continues running. The fact that a) someone wrote this and didn't consider the consequences b) someone else signed this of and thought is was OK c) nobody tested this an found the bug (_before_ it left JK) and d) wrote a "fix" that simply just changed the defaults does not inspire any confidence in these guys ability to make safety critical SW and HW. In fact, this is pretty much a smoking gun that they don't care about safety. But it's cheap, so people will continue buying it, and when they stumble across stories such as these they'll just rationalized it away. Because it's cheap.
good afternoon andy andy question, I'm a little confused about the communication via can bus between the cerbo gx and the new jk bms, you have to modify the internet cable or you can use an original one without modifying it, that's my question, I'm confused and thank you in advance, andy
You can just buy the Victron Type A cable or make your own. The manual how to make one is here: www.victronenergy.com/live/battery_compatibility:can-bus_bms-cable Just follow the Type A cable table if you make your own.
So now you might be charging your battery bank to LTO specs unconsously, which is lower than lifepo4. When in the case your bms resets unknowingly. You may suddenly run out of power because you are not fully charging your battery.
LTO voltage is 2.4V (LiFePO4 almost dead is 2.5V). You'll definitely notice something is wrong when you only have a few minutes of power left in your battery after charge.
I'm wondering if this issue could be helped by a hardware hack. Adding a bigger capacitor or replaceable battery (like CMOS batteries to store computer firmware on motherboards). It seems like there might be some volatile RAM that is getting reloaded from a default location after the RAM memory loses power. I don't know why it would be designed that way, but maybe the location where the firmware is stored only has a finite number of write operations possible so it's not wise to store user changes there?
They just need to write the values to the chips registry, unless they have used an inappropriately sized chip it should all be possible. Why do other bms's in their range not suffer from this problem, that is what the question should be, this all is just a bandaid though, and not the solution.
What has happened to Peter 2 and 3 and 4 , have they emigrated to JK wonderland ? I miss the OG late night news - at least late in the EU for sure. JK is back in the game I guess even though it had hit me hard last weekend or so cause I did not believe that. I guess I will continue with the seplos and wait till the dust has saddled on a safe and stable firmware where the issues have been fixed cause I am pretty sure that you will find them or a viewer and letting you know.
All Andy's had an RDO that day, so it was only me😎 Yeah, the Seplos has its problems as well and no charge algorithm, so, it's a very basic BMS. I guess there will be more problems with the JKs in the future... watch this space.
Well, you have to consider also one more thing - the fact it brings error message during setting the value, it might have also worked differently - sent the value, BMS set it, but reported error back. And if you did not save something other, the BMS has the value but does not show it in the app. I would test it - storing incorrect value and then quit the app, clear app cache, connect back and verify the values. But I've lost my trust into JK by this time, especially after the info from Digital Mermaid...
As for the DigitalMermaid (hello, I still like the your videos) I found it a bit exxagerated. All the version 1 and 2 bugs were known to me and I got what I knew. So, no fuss about it. As for the JK Inverter it could have gone better. BUT we were all in the market for it lusting an affordable and feature rich BMS and bought it, so this goes along with chinese mentality. And wasn't the major advantage the software updates? Here we go. Compare it to Victron (a bit of sarcasm here): well tested, flawless, not inexpensive and regular updates (plus always late with hardware releases), that I have never needed in my life. So, we all bought an early adopter product by chinese mentality and are complaining around? Well, there is always a Battrium for half a liver or a half assed Seplos if you are interested. And as I stated somewhere, my FW version resets with charging/discharging disabled, so I for sure didn't get the fire ware ;-)
Do you know about any faults for Bluetooth connection? My jk bms Bluetooth has stopped working so I can't look at settings or change anything, bms still appears to be working but unable to check individual cell voltage via the bms
Hi Andy and Folks, Offgrid, I've just built 16s 320Ah portable battery box with a JK bms and has a 10kva peak inverter inside . My problem is that i cant stop bms shutting down my power supply everytime battery reaches 1oo% multiple times per day. Is there a setting I could use to prevent BMS switching of the Inverter? Which one of the parameters would control this Inverter off function ?
Hi Andy, Any news about JK storing setup data in a flash memory instead of resetting to "safe chemistry voltages" after being resetted. I was looking at this bms but loosing all data after some days being shut off is not an option for me.
New Bug - RCV Time and RFV time reverts to the figure you have set for your RCV time if you restart the BMS, so if you set RCV time @ 0.5H and RFV time @ 6H and reset the BMS, you now have a RCV and RFV time of 0.5H
There are industry standards for risk management. They way JK reacted is suspicious. The user input is also part of risk management. ISO 14971 would almost make sense here. But at least, a structured approach to risks, hazards, failure modes, and effectiveness checks for the mitigators. Publishing the analysis would be nice.
Andy, please 🙏🏻 you can tell the manufacturer JK that they need to add, when the battery charges closer to 100%, they need to reduce the current by CAN message for Invertor such as SOLIS. It is also important to know for JK that as soon as the inverter receives information from the BMS that it is 100%, the inverter stop charging the battery I also use JBD bms version UP, and there are working perfect
@@stephanklasen2928 you don’t really understand the algorithm of how it is work As if i set 100A, the invertor will charge at current 100A until it receive from bms 100% and stop charging it. As you understand this is stupid! What does that mean, it mean that the cell will hit 100% voltage that I set in the bms, it’s around 3.6V, and if it was charging at 100A the cells voltage will drop to 3.2V, this is not real 100% charge for LFP, and this doesn’t matter which chemistry you choose, as closer you get to 100% the current must drop for best, balancing and top charging cells, this is how it work on electric vehicle, this is how manufacturers recommend to do for charging cells
@gosmart5581 the current goes down by its own when the cells get closer to the requested charging voltage. When you charge with 57.28V, which is 3.58V x16, the current will never be 100A close before the cells are full. Only if your battery is heavily unbalanced and the cells have different inner resistance they will spike above the OVP.
@@stephanklasen2928set 100A and set 57.6V, and you will see that it will charge the cells at 100A even if a voltage are 57.6 I am really sure what I’m talking about.
Then go and double check. The physics can't be cheated and you can see it on any battery that the current goes down once the battery voltage gets closer to the charging voltage of the charger. It will only charge with a higher current, if your charger is charging with a higher voltage and then stops once your battery reaches the set voltage. Maybe that is what your charger is doing.
hi Andy. I am planning to buy the new series of JK, the PB2A16S20P. The companies as EEL, Gobel and so on are obviously not happy with that BMS und use the seplos. Well, EEL came out with the version V4 of the metal boxes which can use the JK but i got the information that there is a problem with the communication of the BMS with the 48V-Inverter when you use a 2nd battery pack. question: do you know whether this problem still exists? thanks.
If I want a jk bms to communicate with the solar AIO, and I have multiple bms'es, does every bms have to be the inverter bms, or can one inverter (master) bms be used and the other non-inverter (slave)bms'es communicate with/through it?
Hi Andy. Why did you not check there reasoning for the problem and check the values you had loaded on the Bms. That failed. Before you upgraded the firmware. ? It sounds like the software interface still needs some. Work. To catch these problems. Before even trying to upload it
The JK firmware is a clusterf* and a fulltime job to work with. There is no documentation, no change logs, nothing! I cannot check every single FW they bring out for all possible failures and bugs.
On every event that the default parameters are loaded the correct fix is to stop the BMS functioning (block charge and discharge) and display a message on the screen.
Hi Andy, I just built and started my 15kwh battery pack and now got a problem with the jk inverter bms: when my phone is connected via bluetooth, even when the app is closed (app seems to run on Bluetooth LE in the background...), and I try to turn off my pack via the on off push button, then the pack beeps as supposed but will not shut down, it stays on and disables any input from the button. To make the button work again, you have to unplug the BMS power cable. Is this an old error fixed in an update or is this an ongoing problem?
Hello: are all JK BMs the same in quality (apart from power (A))? I'm thinking of getting a BD6A17S8P with LCD and bluetooth, because I don't think I need more than 3/4 kW instantaneously. Are JKBMS the best? I saw that they are equipped with an active balancer, that seems very good to me. For a lifepo4 16s 48v (51,2) 280ah eve280k battery. Thank you.
I watched your video on mixing different amp hr. Batteries. I have two 260 “Lifepo4 Lithium Iron Phosphate” now I have found 280 for $100 cheaper. I’m wanting to build up my bank. Would it be better to wire them in a separate parallel bank to my bus bar? Or just stay with the 260 batteries:-(. Or just add them to the bank I have started. Wish I had your knowledge :-).
Not sure what you mean by all of that. If they all have the same voltage, you can just parallel them together if every battery has got its own BMS and breaker.
This morning I updated my JK-PB2A16S20P, from V14.13 to V14.19 directly. Unfortunately, the only thing I needed and the reason for this update - Dry 1 or dry 2 trigger for "battery under (low) temperature" doesn't exist in this version. I need to start heating the battery even when I charge it, and using the heating function of the bms is not ok for me, as it change charging to "off" during heating... The dry contacts were the only solution for me, but I updated for nothing to V14.19, as there is no option for "battery under temperature", at least I don't see it.. Later edit: Jk confirmed that "regarding the version of 14.x, we still need further updates and improvements.". So, nothing to do than wait until V14.20 or later will add again this low temperature battery alarm trigger...
Just one question. How you deal with heating the batteries at night. If there is no power from PV so there is no sense to heat batteries and waste the energy and lower SOC. What is your approche?
@@mateuszmazur1482 I only want to keep my battery warm, I have 2 heating pads, 200 W each and I don't care if I will lower the SOC a little bit, at least my battery will not degrade because of the temperature. Couple of days before the internal temperature was 2 degrees Celsius, near frezzing, too low for me. I have a temperature controller which I want to drive through dry1 contacts from bms. Then the controller, togheter with a solid state relay open heating, until the max set temperature is reached. Then stop heating until min temperature is reached and so on... Not the best solution in the world, but it's working for me and don't need to spend any money for other things. Thank you!
Maybe it is good approche. I have to rethink my idea. I have 6 x 20W heating pads and battery is in a isolated cooler big box (as my battery is in non isolated shead). Pads are 12V rated so I put them in series for my "48V" pack. May be I will use only 5 of them to increase the voltage per pad. I assume that 56V is the maximum. Each pad have a radiator. My thoughs was to use voltage level on mppt controller. No high enought voltage, no sun. But... If pv panels are covered by snow there could be a higher voltage but still no power. So... May be your approche to keep e.g. no less than 3 no more than 5 is good...
@@mateuszmazur1482 I have a "technical" room, separated from the house, only concrete walls, steel door, nothing to burn inside, where the inverters and the battery are mounted. Battery is in a box, where the pads are also fitted, not touching the battery... It is cold inside during the winter. My pads are from a 3d printer, heating up to 180 degrees, so I need to keep them between 60 and 80 degrees Celsius with a temperature controller. The bms temperature probes - 4 pcs-are put on the battery. I will set the dry contacts to work between 12-18 degrees Celsius. In this time the pads are heated until 80, cool down to 60, heated again to 80 etc.. until the battery core reach 18 degrees C, when the bms open the dry contact. When the core cools down reach 12 C, the dry contact closes, the temperature controller start working, and the operation of heating begin again.. And so on...
Nice one! Mine are a little bit cooler. Max around 130C on 12V but I lower the voltage by putting them in a series so one problem is away. Now they have max 80C, plus radiators, plus air space between radiators and batteries. I will keep them on as you suggest if temperature is around 3C. Playing with reading voltage (for example using venusos and node-red and mppt nodes) is one more layer of potential problem/error.
I upgraded the firmware and it still resets everything! And I can't set SOC 100% at 3.45 because it doesn't follow the order of reducing values. This is absurd. I want to go back to my old firmware as I do not restart my BMS!
Yes, some values need to be in a certain order. JK argues, this is necessary to keep things safe for users and also helps the BMS to check on settings when starting up. You can always go back to an older FW.
I am still leery of JK's software development methodology, it seems to be the "seat of the pants" method. It would be interesting to see what the c/c++ code looks like.
okay so they have non write ROM chips except when loading new firmware. I don't think Digital Mermaid gives a hoot and she just hates JK cause she ordered wrong version of BMS and had to buy expensive display. Thanks Andy
So it's still a bug, it will still damage Li-Ion cells (by allowing them waaaay to low undervoltage), did you get any hint when they will actually _fix_ the bug, not just make it less serious?
Was this really a safety problem? The way I understand it, the charge controller in my hybrid Inverter would set the charge voltage for the battery system, the BMS just balances cells, does under voltage protection, over current protection and over voltage protection. If the BMS settings go to Lithium Ion, the only parameter that really creates a risk is over voltage protection but your charge controller should limit the charge voltage so that protection is redundant. Am I missing something? It's my understanding that I could run a battery bank without a BMS and it would be mostly OK until the cells become unbalanced. Then a battery balancer doesn't care what the cell voltage is, it should balance at any cell voltage. This ignores cell failure and over current protection which are big reasons to have a BMS but mostly if there are no faults and your inverter has some type of current limiting, the main thing you want your BMS to do is balance the cells. So I don't really care what the default settings are but resetting to LTO would be annoying if I have LiFePo4 chemistry. Has the resetting issue been solved? Is the JK Inverter BMS still a good choice for a 16s LiFePo4 battery bank? What else should I consider?
I have the old JK BMS 16s 2A running, it shows all the time some heating current but there is no heater connected and heating function is switched off. I'll try to talk to JK several times but I didn't get an answer. Some of you guys have any ideas?
Moin. Ja, hatte ich schon 2mal versucht, aber hatte über google ne andere Webseite gefunden. Auch was mit jikong. Aber dann versuche ich hier nochmal. Danke.
I have this same bug on my JK-B2A8S-30P. And this is not an Inverter style BMS as far as I know. I configured it all in the right logical order and did not get any errors at all. If I leave the BMS on, it seems to retain the values (at least for the short time period where I tried it). But if I switch the BMS off via the display, it resets all values to Li-Ion. So doesn't seem to be the case that this only happens if there is an error during configuration. Also: for me this happens even if the BMS is only switched off for a few seconds. So what am I supposed to do? Buy whatever hardware that is needed for updating the BMS since it is not possible over the app? What a poor product design. Definitely last time I will have used a JK BMS. So frustrating and a really bad feeling that I cannot rely on my BMS. 😳
oh thanks for the video. I have a problem: the inverter via the can port does not want to work with the inverter deye sun-6k-sg03lp1-eu . Maybe someone has how to configure it correctly so that it works?
Like# 37 - It's too bad it takes the power of Andy 1-4 before the power of an update gets initiated. Further, I hope JK adds this to their QA process to make sure this doesn't happen again...
The only "safe" default settings would be "unconfigured" ...
So an unconfigured BMS would do *nothing* until it has been configured by the user ...
Then the real bug to fix is the *volatile* settings that should stay "put" ...
If the BMS still resets for whatever reason, it has both MOSFETs turned off
It's nice to see that there is such a manufacturer
and listen with customer feedback!
You managed to fix it quickly! That's great!
I hope there is not more...
This is why I selected JK as my default choice in BMS and to recommend it to new users.
Thanks Andy for the great work bringing this to the attention of users and working with JK for the fix
I can't help but feel there should be an asterisk after the fixed in the video title, and a footnote saying that JK fixed the issues that customers have found so far
I feel like the firmware updates is as much a blessing as a curse as it gives developers the temptation to push buggy software out and just let the customers find the issues rather than do any real testing. Anyone who plays video games has no doubt experienced this in recent years where customers are effectively paying to be beta testers
I'm not going to single JK out for this, I wouldn't be at all surprised if the other BMS manufacturers had similar bad practices
I wish that they would understand that if they have a major software bug that it doesn't just cause an inconvenience for the user but could lead to a serious hazard like the one JK just fixed
Nice. I seen other UA-camrs work with battery companies to fix issues that they find. Definitely a good community.
Thank you for all you do helping inform and mobilize this community. You are an inspiration!
Here is my analysis. The issue was with the over voltage protection recovery. When you sent to LFP it sets it to around 3.54v or something. But I went to RCV to 3.45 and 3.449v for the 100% SOC like you did in your example. The NEW firmware doesn't allow you to set those if your Cell OVPR is set to 3.54. So I have to set THAT to 3.448v. But because the default settings were allowing you to set it, that's where the issue came from. I think they need to update their mobile app with error codes telling you WHY you can't sett those settings.
v17. Same here.
I can only set an OVPR
Yeah, fw still does not make a lot of sense...
@@OffGridGarageAustralia what do you mean about sense , either it works correct or it doesn't. If its working correctly, then it should make sense.
@@carmichaelmoritz8662 it's the overall design of the software and how it interacts with the user. It gives out the same error message regardless what settings are wrong. The FW itself has a strange logic it follows, like OVP>RCV>SOC100%>OVPR.
@@OffGridGarageAustralia maybe the people that write the code software are purposely writing in the strange logic. Logic should never be strange unless its on purpose or the person writing it is strange.
Thanks for all you do for this community Andy!
I knew you would get it figured out and work with the manufacturer to get things fixed! 🙂
Great job! It's still a bummer that it won't just leave the settings alone and not reset ANYTHING when powered off.
Agree. The BMS need to remember what it was set for. Its not like this is their first BMS. the Old and now goldie JKBMS active balance version remembers!
This is now fixed though and the BMS does not reset any more.
@@OffGridGarageAustralia 👍
Thanks Original Andy!! Impressive that you receive after the weekend the new fix and then another 30 minutes your recommendation Has been Adapted as well. This is a huge Praise to your Expertise and the solar community. Well done. You deserve your andy Assi fresh beers.
Thanks. I sometimes feel like I'm a JK employee already😆
@@OffGridGarageAustralia and you have the right to feel so. After all you've been doing this stuff for a long time already. It should come naturally to you to know all to do with what you do.
@@OffGridGarageAustralia 😉😉😎😎😎😎You will realize it if you get a paycheck. ✌😆😆😆😆
Well done JK ,🤞that will silence a few overreacting critics.
Great effort Andy you have taken on a lot of work leasing with JK and informing the world.
Are Andy's 2,3&4 on the naughty step with Jk?
Yeah, everyone needs to take it easy. This is all new and JK tries their best with the updates...
Hi Andy.
I must have been one of your first viewers/subscribers. I came across the channel because I had purchased the same EVE cells some 3 odd years ago at the same time you did. We went in slightly different directions - mainly due to monetary constraints my end. I am a pensioner with a very fixed income.
We both chose the best battery option (hindsight is great). We both went with second hand panels. Then you went Victron, I couldn't so I went Taiwanese all in one Inverter/Charge controller. I feel like I have learnt the ropes alongside you, looking over your shoulder so to speak.
You seem to love tweaking your system and making things easier to monitor and more efficient as a result - I went through that phase.
Whilst I find all this very interesting and follow each of your projects, my goals are somewhat different - no UA-cam channel obviously. My goal was (and still is) to build a system that just works and requires very little monitoring. A drop in replacement for the grid if you like. Minimal maintenance, minimal monitoring. Just what is required for safety.
Anyway I have an idea for a new project if you are looking for one.
What about looking at the lower end full system? Whilst we are fortunate that a lot of people in Australia can afford a big, top of the line setup, some like myself are not so lucky but still want to reduce their carbon footprint and their power bill - and this is possible.
Anyway, I suppose you are flat out and have lots of projects coming up - and I look forward to them - just thought I'd mention it.
Thanks a lot for your feedback and being around for so long. I really appreciate that.
The problem with building such a new 'low end' system as you suggest is, what am I supposed to do with it afterwards? Where should I mount it, install it? Who should I test it? Who is paying for it? I just don't have the space for another separate system.
BUT, in saying that. there will be some changes to my setup in the near future which may enable me to test other inverters as well as batteries.
After testing some cells i understood every setting needed and the fact that i don't like those cells when used to make a battery. If only they could make twelve volt cells , one could run them parallel and they would all discharge evenly. no need for a bms to read four 3.2 @3.65 volt cells to make 12 volt or three 4.0 volt cells uvp or ovp. You'd only need one cell uvp or ovp to monitor the complete 12 volt parallel system.
Thanks Andy and folks in community. Excellent job.
For god sake, hopefully I can get my PC and BMS connection set up on this weekend, new RS485 module just arrived. Looks like one Andy has but it came with CD. For god sake who still have CD player, hope standard windows drivers will work.
Hast du die beste wetter da, oder nicht?
Deutschland ist nicht so gut heute.
Gute sache hast du gemacht mit JK Bms.
👍👍👍👍👍
Thank you, yeah, pretty good weather!
Thanks Andy, 👍👍👍👍👍
Cheers, stay charged … and stay cool!
you would have to ask the JK engineer to modify the smartphone application, about the choice of chemistry (lithium ion, lifepo4, LTO), and to add a popup confirming the change, because if you inadvertently click on another chemistry, you lose all your specific data.
Yes, I suggested this a while back as well. These buttons are far too present on the settings page.
This still is a risk. LTO has a lower bottom voltage so you can damage your batteries with a low voltage scenario.
It fixes the top but then makes it worse.
They need to be storing this properly in non-volatile memory. If they don’t have it, they need to recall all of them and add it.
The settings just needs to be written to the chips registry, why they can't set aside certain reg values for this is beyond me.... Why are the other BMS's not suffering this issue, unless they used an inappropriately sized chip for this, it should be totaly possible, this is a bandaid not a fix....
This is simply a least effort quick hack to handle the backlash of bein plain dangerous.
Damaging batteries aren't that bad, and to be expected it seems.
@@GapRecordingsNamibia they are. We just haven’t found the bug that will corrupt the memory yet.
What’s happening is that it encounters an invalid setting and instead of loading a safe default that turns off balancing, charge and discharge with an error but loads all other settings, it’s just defaulting to factory.
They just changed the factory defaults. Bad hack to solve the real issue which is they need to understand what a settings error should result in, which is safety mode.
11:45 The RCV is meant to be lower than the OVP. The system first line of defence is to prevent the error. Even after he bypassed it, by saving the Start Balance Voltage which acidentally sets the wrong RCV to 3.65V. This makes the RCV higher than OVP. it still did not reset the BMS to default setting, even after switching it off for sometime. I feel it has been resolved
Yes, it is still a risk, but the BMS does not reset any more. It's in the video at 8:18
Andy, you should be on the JK payroll! Cheers!
Ah, I'm not sure. I would lose all my hair...
@OffGridGarageAustralia Always a funny guy. Of course. Very witty I get it.
I need to have this (Off-grid garage Weather 3.5amps) T-shift.❤🎉
Greetings from Nigeria 🇳🇬
It may be cold up here, but at least I don't have to think about finding an airco to keep the batteries cold, or limit my charge current ;)
Since my new install of my big battery shelf and inverter the electric heater in my workshop has only been on low with a more comfortable temperature inside it, the extra passive warmth as a byproduct is more than welcome on these frosty mornings in 🏴.
Helping me save even more money on heating too.
Not too toasty warm and no equipment or cells are hot.
In my home, the airco in the solar basement is the largest consumer I have 😂
Thanks mate, just placed an order for a JK BMS from Hankzor (they currently have a sale on ending today), some S.P.A.T. coming your way my friend...
Thanks a lot for your support.
Happy this issue has been resolved. I can't help but wonder what other critical issues have not surfaced as yet.
Yes, I'm the same.
Good thing they responded so fast. That's a great service. I do agree however that maybe it should not work at all and show some LED or something (if it even has) or just pop a message in the app
Yeah! Power to the people! Thanks, now i can go on installing
Thank you for sharing the software Upgrade. and pointing it out.
Thank you.
For JK It is a good advice to set the default settings to the strongest safety parameters concerning the battery type.
That means to "brick" it, no charging or discharging and give alarm
Please create a full descriptive video about jk-bms firmware update. What cables, how to connect the BMS to PC etc.
OK.
Yeah it gets cold here but not as cold as some places. I live next to a great lake and it keeps us warmer.
Greatings from Ukraine! Thank You very much for Your work! I have such BMS with HW v15. I updated it to version v15.17 from v15.11 based on Your instructions, but I can not find any changelogs for this firmware. Does this bug persist in v15.17? Is there any way to read what have changed between FW versions? Thank You in advance!
Heya, good that jk so soon came with a solution hope no more surprises will be there so I can start ordering this bms. have to order a new hybrid inverter because the "old" 1 I have doesn't have canport. so the new inverter has to be modbus/can port for communication, even do it's not that important for me as I'm gried tied. w'll see what the future will bring. oh yeah I'm totally up to date now with your vlogs.
Why can't JK also release firmware updates for their old BMSs? I am sure they can be updated... at least to fix the SoC reset with an extra parameter or so... Andy, could you try your contacts? They will make the huge JKBMS owners community really happy...
The chips they used don't support that. They can be written once and for an update they may have to be taken out and re-programmed if this is possible at all.
@OffGridGarageAustralia I see... it's a shame because this hardware is not that bad at all... maybe they can add the feature to the new units... because if they compile a new firmware and share it, the community can figure out how to replace/reprogram the chip 🤔
I had every faith that JK would resolve any issues & bought one regardless 💪🏆
Yep, it was clear it was FW issue rather than hardware so you made the right decision.
The "issue" is still not fixed, it still rests just to LTO which has it's own set of issues towrd LFP, they can discharge to low now. This to can damage the LFP cells, so, not a fix bit a bandaid.....
@@GapRecordingsNamibia So perhaps the 'fix' is to default to mosfets off for both charge and discharge...?
The BMS does not reset any more as they identified and fixed the bug which caused that.
Instruction how to update if an update failed and the BMS does not respond anymore:
- PC Software select firmware, force update
- enter PIN provided by JK support (only valid for 30mins), update is on hold at 0%
- press and hold activation button
- shortly press reset button and keep holding the activation button until update is finished
BMS is back in the game!
Thanks Stephan. I'll have this in the next video as well.
Hello Andy
What happened to the bricked Bms that failed when you were updating it
That would also interest me, as I toasted one of my BMS while updating to V14.17
@@stephanklasen2928 I managed to fix mine, its kinda simple to do
You need to force update it and to do that you need to get a key from JK at the time you are working because this key lasts only for 3 minutes so make sure you contact them they can help you
Btw you didnt do anything wrong it just happened to me for one out of 6 and to Andy as well
@@rockyfox7087 I fixed mine today and upgraded all to V14.19!
It will be in the next video!
Como siempre muy interesasntes tus videos, ofreciendo opciones y soluciones, gracias
Awesome again Andy!
Thanks a lot!
No worries!
Make videos Andy its too boring without you 🎉
You entertain all of us
OK, I'll make more. Just for you!
FYI
As an engineer, I would expect it to remember my settings unless I do a forced reset.
It should be shipped with the lowest battery type settings
as enginner, I would expect to start from backup-startup.txt if startup.txt is fail. When both are fail then default.
Thx again Andy for the update and all your effort!
I dont understand why they reset the bms back to default profile settings after the restart - it would be much easier to debug for the user to turn off charge and discharge mosfets and trow an error (LED + CAN + RS485 + BT APP) - thats the reason why there are alarm and error functionality right?!?
Thats what i would expect in case of any wrong settings - and nothing else
But it is also not a real reset to factory defaults - as i can remember any of my bought JKs where delivered with Mosfets and Balancer turned OFF after the very first start.
Again a quick shot, but again overall not a good job from JK here
👎👎👎
But anyway - PROST 🍻
Well, the BMS does not reset any more because they fixed the bug.
FETs are both off after a reset as far as I can tell...
Surely if on a cold boot for any BMS, if any saved setting parameters (or memory checksums etc) are implausible, then the system should not operate at all. It should not connect the battery outputs nor allow charge or discharge or balancing to occur. It should just sit their in this error state until the settings are corrected, resaved and the system is full rebooted??
The charge and discharge FETs are both off if it resets. But this is now fixed.
Andy, whch terminal is the P+ connection referred to in the manual when connecting everything to the BMS in the correct sequence? Thanks
I've seen this as well and don't know what P+ means. No further documentation🤦♂️ I just ignored it.
Hi Andy
I've been following your channel for a long time and I really like it, congratulations.
This update solves the charging voltage problem, but what about the discharging voltage?
It is necessary to program the minimum operating voltage in the inverter, and I don't know if the BMS will not change this value in the inverter itself.
Well done.
Well, the BMS does not reset any more and if it does (for another reason) it also disables all MOSFETs.
Enjoyed the video. Thanks for sharing
Good news. Thanks for information.
I see that today they released V14.20 and V15.10 already... Hope they added the low temperature battery alarm for dry contacts... Can't wait to update the firmware this evening...
Did you test it?
@@mateuszmazur1482 not yet, the V14. 20 log is telling me that the bluetooth is improved and something other thing is modified, but nothing about the dry intact low temperature battery alarm... So I am still waiting another version of firmware..
Something is going on out there... Today I asked official JK shop on Alli, where is my heating cable because cable is missing in a package. And they answer to me, that this is by purpose, because I don't need it, because users have some issues with heating function so they resigned from this function (sic!)... What the hack?
@@mateuszmazur1482 I know nothing about this issue. I bought my bms on November and i got the package complete, no missing cables or anything... And for me is not so important, I don't intend to use the heating function anyway...
The low temperature alarm to drive the relay has been in the software since v14.17, I believe.
Andy, it might be cold in the Northern Climates during certain months but we are not sweating our ashes off for months of a time either.
It all depends on what you tolerate better. For me, I would rather be cold for a handful of months because I can still put on a jacket, hat and gloves and enjoy being outdoors. I cannot tolerate being outdoors in extreme heat like they have in warm weather climates. I got a taste of this living in Florida for a few years. I HATED the summers and feeling like a wet blanket if I had to be out in the heat. Give me cold any day!
You indicated subscribers made comments in the community section of your site. I clicked on it and it appears all the entries are from you. Is the community section like a blog or just a place for you to add dialog?
I know, I used to live in these colder climates for 40 years. Both has its pros and cons.
Let’s just agree that MID temps are the best. 😁
I don’t live too far from Andy, and I “survive” our Summer.
I’ve experienced extreme cold, very short days and horrible weather. I hated it.
I’ve experienced stinking heat with extreme humidity. I hate it (like we’re getting at home in Brisbane right now, more like wet season Darwin than Brisbane.)
However at least we can retreat to the aircon.
Brisbane in Winter- usually fabulous weather- lots of sun and beautiful days. (I LOVE 💕 our Winter.)
P.S. I don’t think anywhere in the world gets great weather 365 days a year.
@@FutureSystem738
San Diego, CA???
The city remains within a comfortable range between approximately 50 and 75 degrees all year long and receives just 43 days of precipitation on average.Sept 13, 2023
@@jws3925 Sounds pretty good, though I must say that the last time I visited there, circa 2000, the weather was pretty terrible. (I THINK it was late in the year … but not 100% certain.)
For this reason I ordered two new JKs today.
I can only say: Good luck!
@@OffGridGarageAustralia Was that meant to be positive or negative?😃
@@daisydonald14 it's been 5h since I made this comment. A lot can happen in the world in this time😆
Honestly, I don't know what I should think about JK right now. They have this amazing product called PBxx inverter BMS but the software seems to be the weak point right now. We (not them) discover bugs, they fix it. New bugs, new firmware updates. Since the release 3 months ago, we have seen 13 updates. Is that good or bad?
@@OffGridGarageAustralia From one side it's good, from the other it's not. At least they react to it. There are companies that don't do that. It's now normal that we're all just testers and guinea pigs, regardless of whether it's smartphones or software. And I say again, it would be bad if JK didn't do anything. Do you know what the difference is between v14 and v15 hardware version? I got my first one yesterday and it's a v15 with 15.05 software version. Thank you again for your involvement in this matter. Without you they would have had a lot of scrap now.😃 Danke ,danke
Great information; scary information!
Yes, correct. How far can we trust this BMS?
I’m looking to building a battery. Was keen on this bms. The range of problems does not inspire confidence on the brand. I need a nature and reliable bms.
Suggestions?
Do you need/want coms? This JK is still the best choice out there, I would say. At least they constantly refining the product with new firmware.
If no coms, go with the older style BD JK-BMS and bluetooth connection.
JK was very fast but you also put the pressure on ;) Prost!🍻
They are fast in selling stuff, yeah...😂
LTO cells have a UVP of 1,5V per cell x16 = 24V
If the inverter is set to follow SOC% sent by the BMS it will destroy every battery pack at first discharge after the reset.
What makes this safe? What did I miss?
Yes, correct. Even the LTO profile can damage your battery but it does not overcharge them at least. So, I guess, this was the compromise.
LTO is 2.2V discharge right? Isn't that a bit way too low for LFP and Li-Ion? Low voltage can cause dendrite damage and short a cell, discharging it over time?
The BMS does not reset any more though.
My JK BMS arrived yesterday so I power it up for checking the firmware. But my JK-PB2A16S-20P
has the hardware Vers. V15.XA and the firmware Vers. 15.03. This Firmware has the same bug..
Where can I get the newest firmware?
On their website: www.jkbms.net/download_pc
Vielen Dank, ich hatte es nicht gefunden
is -60c cold? Here we just have some LN2 to warm up, it is annoying when you step in a puddle of propane though, wet boots are all ways crappy! I do indeed need heater blankets for our batteries.
So they have no physical way to remember settings? That's a serious engineering mistake
The BMS has a NV memory.
I am excited for the next bug 😂
lol :)
Me too!
Hi Andy! thanks for all your videos. got a JKBMS up and running, i have a suggestion for a new feature.. on the breakout pcb there is a push button, this can turn off the bms, but it can not turn it on.. maybe an upgrade for the next firmware could implement this button with the same features as the external single button, so we can skip that in our setup :-) and do you have a suggested document of the settings of your choice for lifepo4?., Thanks again.
And if you have a screenshot of your suggested settings, that would be awesome.. just installed 6 JK BMS in parallel :-D
The push button on the communication PCB is just a reset button to restart the BMS. This will also reset the settings (but not the password).
You BMS should have a normal power button to turn it on and off?
when i press this button on the pcb it turns off the bms, i then have to use the normal power button to turn the bms on :-) i for a second thought the button on the communication board was a turn off only button, but i have since learned that it is a reset button, and now again :-) thanks
@@OffGridGarageAustralia
Is just a cosmetic change in the software.
My opinion, a hardware revision, must add a non volatile memory to save seting
They _do_ have non-volatile memory, what appears to be the case is that the startup routine has a validation routine that gets triggered, and if it determines the config to be invalid, it just picks some default values and continues running.
The fact that a) someone wrote this and didn't consider the consequences b) someone else signed this of and thought is was OK c) nobody tested this an found the bug (_before_ it left JK) and d) wrote a "fix" that simply just changed the defaults does not inspire any confidence in these guys ability to make safety critical SW and HW. In fact, this is pretty much a smoking gun that they don't care about safety.
But it's cheap, so people will continue buying it, and when they stumble across stories such as these they'll just rationalized it away. Because it's cheap.
Yes, JK is using NV memory for the settings. It's all explained in the video.
good afternoon andy andy question, I'm a little confused about the communication via can bus between the cerbo gx and the new jk bms, you have to modify the internet cable or you can use an original one without modifying it, that's my question, I'm confused and thank you in advance, andy
You can just buy the Victron Type A cable or make your own. The manual how to make one is here: www.victronenergy.com/live/battery_compatibility:can-bus_bms-cable
Just follow the Type A cable table if you make your own.
good night andy thanks for the link@@OffGridGarageAustralia
Sounds like they didn't test the software changes very much with the fast response.
Probably not...
So now you might be charging your battery bank to LTO specs unconsously, which is lower than lifepo4. When in the case your bms resets unknowingly. You may suddenly run out of power because you are not fully charging your battery.
LTO voltage is 2.4V (LiFePO4 almost dead is 2.5V). You'll definitely notice something is wrong when you only have a few minutes of power left in your battery after charge.
Well, the BSM does not reset any more...
FIREWARE!? 11:05 Not sure I want fireware on mine!
Thanks Andy
Thanks, Wayne.
I'm wondering if this issue could be helped by a hardware hack. Adding a bigger capacitor or replaceable battery (like CMOS batteries to store computer firmware on motherboards). It seems like there might be some volatile RAM that is getting reloaded from a default location after the RAM memory loses power. I don't know why it would be designed that way, but maybe the location where the firmware is stored only has a finite number of write operations possible so it's not wise to store user changes there?
They just need to write the values to the chips registry, unless they have used an inappropriately sized chip it should all be possible. Why do other bms's in their range not suffer from this problem, that is what the question should be, this all is just a bandaid though, and not the solution.
The BMS uses a NV memory to store the settings. A separate button battery is not needed. It was like this from the beginning.
What has happened to Peter 2 and 3 and 4 , have they emigrated to JK wonderland ?
I miss the OG late night news - at least late in the EU for sure.
JK is back in the game I guess even though it had hit me hard last weekend or so cause I did not believe that.
I guess I will continue with the seplos and wait till the dust has saddled on a safe and stable firmware where the issues have been fixed cause I am pretty sure that you will find them or a viewer and letting you know.
All Andy's had an RDO that day, so it was only me😎
Yeah, the Seplos has its problems as well and no charge algorithm, so, it's a very basic BMS.
I guess there will be more problems with the JKs in the future... watch this space.
Well, you have to consider also one more thing - the fact it brings error message during setting the value, it might have also worked differently - sent the value, BMS set it, but reported error back. And if you did not save something other, the BMS has the value but does not show it in the app.
I would test it - storing incorrect value and then quit the app, clear app cache, connect back and verify the values.
But I've lost my trust into JK by this time, especially after the info from Digital Mermaid...
As for the DigitalMermaid (hello, I still like the your videos) I found it a bit exxagerated. All the version 1 and 2 bugs were known to me and I got what I knew. So, no fuss about it. As for the JK Inverter it could have gone better. BUT we were all in the market for it lusting an affordable and feature rich BMS and bought it, so this goes along with chinese mentality. And wasn't the major advantage the software updates? Here we go. Compare it to Victron (a bit of sarcasm here): well tested, flawless, not inexpensive and regular updates (plus always late with hardware releases), that I have never needed in my life. So, we all bought an early adopter product by chinese mentality and are complaining around? Well, there is always a Battrium for half a liver or a half assed Seplos if you are interested.
And as I stated somewhere, my FW version resets with charging/discharging disabled, so I for sure didn't get the fire ware ;-)
DM over reacted a bit. While this is just a band aid and not a solution. I do feel it should rather have been tested before public release.....
@@MichaelTheis77 That "half liver" price tag made me laugh :) Nice touch ;)
Yes, the app needs lots of improvements and more meaningful text in the pop-ups.
Do you know about any faults for Bluetooth connection? My jk bms Bluetooth has stopped working so I can't look at settings or change anything, bms still appears to be working but unable to check individual cell voltage via the bms
Which model do you have?
Hi, I have a bunch of new BD4A24S4P JK-BMS they also forget their settings. So these are also affected.
Finally some good news :)
Hi Andy and Folks, Offgrid, I've just built 16s 320Ah portable battery box with a JK bms and has a 10kva peak inverter inside . My problem is that i cant stop bms shutting down my power supply everytime battery reaches 1oo% multiple times per day. Is there a setting I could use to prevent BMS switching of the Inverter? Which one of the parameters would control this Inverter off function ?
Luv ya work Andy 🤙🇦🇺
Thank you.
Hi Andy, Any news about JK storing setup data in a flash memory instead of resetting to "safe chemistry voltages" after being resetted. I was looking at this bms but loosing all data after some days being shut off is not an option for me.
New Bug - RCV Time and RFV time reverts to the figure you have set for your RCV time if you restart the BMS, so if you set RCV time @ 0.5H and RFV time @ 6H and reset the BMS, you now have a RCV and RFV time of 0.5H
Firmware 14.19 is now available, RFV time is now stored
There are industry standards for risk management. They way JK reacted is suspicious. The user input is also part of risk management. ISO 14971 would almost make sense here. But at least, a structured approach to risks, hazards, failure modes, and effectiveness checks for the mitigators.
Publishing the analysis would be nice.
Andy, please 🙏🏻
you can tell the manufacturer JK that they need to add, when the battery charges closer to 100%, they need to reduce the current by CAN message for Invertor such as SOLIS.
It is also important to know for JK that as soon as the inverter receives information from the BMS that it is 100%, the inverter stop charging the battery
I also use JBD bms version UP, and there are working perfect
Current is caused by Voltage and resistance. Simple physics. When you lower the voltage, current drops. Easy as that.
@@stephanklasen2928 you don’t really understand the algorithm of how it is work
As if i set 100A, the invertor will charge at current 100A until it receive from bms 100% and stop charging it. As you understand this is stupid! What does that mean, it mean that the cell will hit 100% voltage that I set in the bms, it’s around 3.6V, and if it was charging at 100A the cells voltage will drop to 3.2V, this is not real 100% charge for LFP, and this doesn’t matter which chemistry you choose, as closer you get to 100% the current must drop for best, balancing and top charging cells, this is how it work on electric vehicle, this is how manufacturers recommend to do for charging cells
@gosmart5581 the current goes down by its own when the cells get closer to the requested charging voltage. When you charge with 57.28V, which is 3.58V x16, the current will never be 100A close before the cells are full. Only if your battery is heavily unbalanced and the cells have different inner resistance they will spike above the OVP.
@@stephanklasen2928set 100A and set 57.6V, and you will see that it will charge the cells at 100A even if a voltage are 57.6
I am really sure what I’m talking about.
Then go and double check. The physics can't be cheated and you can see it on any battery that the current goes down once the battery voltage gets closer to the charging voltage of the charger. It will only charge with a higher current, if your charger is charging with a higher voltage and then stops once your battery reaches the set voltage. Maybe that is what your charger is doing.
HI Andy There is Hardwate version 15.XA and shipped with V15.05 . Can you pls also put in your website firmware foe V15.XA
So, there is a new hardware version now? What is the difference?
hi Andy. I am planning to buy the new series of JK, the PB2A16S20P. The companies as EEL, Gobel and so on are obviously not happy with that BMS und use the seplos. Well, EEL came out with the version V4 of the metal boxes which can use the JK but i got the information that there is a problem with the communication of the BMS with the 48V-Inverter when you use a 2nd battery pack.
question: do you know whether this problem still exists? thanks.
Today we have -10 Celsius in Austria. Maybe you want to send us 15-20 degrees from Australia to Austria?
If I want a jk bms to communicate with the solar AIO, and I have multiple bms'es, does every bms have to be the inverter bms, or can one inverter (master) bms be used and the other non-inverter (slave)bms'es communicate with/through it?
Hi Andy. Why did you not check there reasoning for the problem and check the values you had loaded on the Bms. That failed. Before you upgraded the firmware. ? It sounds like the software interface still needs some. Work. To catch these problems. Before even trying to upload it
The JK firmware is a clusterf* and a fulltime job to work with. There is no documentation, no change logs, nothing! I cannot check every single FW they bring out for all possible failures and bugs.
Hi thanks for all you video . Maybe you have information new jk-b2a8s30p work can victron and pylontech protocole ??
On every event that the default parameters are loaded the correct fix is to stop the BMS functioning (block charge and discharge) and display a message on the screen.
What about cell UVP(V) at LTO settings? You can damage the Li-Ion or LiFePO4 also because they are discharged at lower voltages.
Yes, you can, but as I said, it does not reset any more,.
Hi Andy, I just built and started my 15kwh battery pack and now got a problem with the jk inverter bms: when my phone is connected via bluetooth, even when the app is closed (app seems to run on Bluetooth LE in the background...), and I try to turn off my pack via the on off push button, then the pack beeps as supposed but will not shut down, it stays on and disables any input from the button. To make the button work again, you have to unplug the BMS power cable. Is this an old error fixed in an update or is this an ongoing problem?
Andy, when using the monitor sotware to update the firmware, which RJ45 port is the cable to be plugged into? Thanks.
For my part I am using the second from the left, although Andy reported it doesn't work.
Thanks @@MichaelTheis77
RS485-1 and DIP set to 1000
Hello: are all JK BMs the same in quality (apart from power (A))?
I'm thinking of getting a BD6A17S8P with LCD and bluetooth, because I don't think I need more than 3/4 kW instantaneously.
Are JKBMS the best? I saw that they are equipped with an active balancer, that seems very good to me. For a lifepo4 16s 48v (51,2) 280ah eve280k battery. Thank you.
I watched your video on mixing different amp hr. Batteries. I have two 260 “Lifepo4 Lithium Iron Phosphate” now I have found 280 for $100 cheaper. I’m wanting to build up my bank. Would it be better to wire them in a separate parallel bank to my bus bar? Or just stay with the 260 batteries:-(. Or just add them to the bank I have started. Wish I had your knowledge :-).
Not sure what you mean by all of that.
If they all have the same voltage, you can just parallel them together if every battery has got its own BMS and breaker.
@@OffGridGarageAustralia That’s how I feel about your video’s most of the time he hee. Thank You! 🙏 Bless you.
This morning I updated my JK-PB2A16S20P, from V14.13 to V14.19 directly. Unfortunately, the only thing I needed and the reason for this update - Dry 1 or dry 2 trigger for "battery under (low) temperature" doesn't exist in this version. I need to start heating the battery even when I charge it, and using the heating function of the bms is not ok for me, as it change charging to "off" during heating... The dry contacts were the only solution for me, but I updated for nothing to V14.19, as there is no option for "battery under temperature", at least I don't see it..
Later edit: Jk confirmed that "regarding the version of 14.x, we still need further updates and improvements.". So, nothing to do than wait until V14.20 or later will add again this low temperature battery alarm trigger...
Just one question. How you deal with heating the batteries at night. If there is no power from PV so there is no sense to heat batteries and waste the energy and lower SOC. What is your approche?
@@mateuszmazur1482 I only want to keep my battery warm, I have 2 heating pads, 200 W each and I don't care if I will lower the SOC a little bit, at least my battery will not degrade because of the temperature. Couple of days before the internal temperature was 2 degrees Celsius, near frezzing, too low for me.
I have a temperature controller which I want to drive through dry1 contacts from bms. Then the controller, togheter with a solid state relay open heating, until the max set temperature is reached. Then stop heating until min temperature is reached and so on...
Not the best solution in the world, but it's working for me and don't need to spend any money for other things.
Thank you!
Maybe it is good approche. I have to rethink my idea. I have 6 x 20W heating pads and battery is in a isolated cooler big box (as my battery is in non isolated shead). Pads are 12V rated so I put them in series for my "48V" pack. May be I will use only 5 of them to increase the voltage per pad. I assume that 56V is the maximum. Each pad have a radiator. My thoughs was to use voltage level on mppt controller. No high enought voltage, no sun. But... If pv panels are covered by snow there could be a higher voltage but still no power. So... May be your approche to keep e.g. no less than 3 no more than 5 is good...
@@mateuszmazur1482
I have a "technical" room, separated from the house, only concrete walls, steel door, nothing to burn inside, where the inverters and the battery are mounted. Battery is in a box, where the pads are also fitted, not touching the battery...
It is cold inside during the winter. My pads are from a 3d printer, heating up to 180 degrees, so I need to keep them between 60 and 80 degrees Celsius with a temperature controller. The bms temperature probes - 4 pcs-are put on the battery. I will set the dry contacts to work between 12-18 degrees Celsius. In this time the pads are heated until 80, cool down to 60, heated again to 80 etc.. until the battery core reach 18 degrees C, when the bms open the dry contact. When the core cools down reach 12 C, the dry contact closes, the temperature controller start working, and the operation of heating begin again.. And so on...
Nice one! Mine are a little bit cooler. Max around 130C on 12V but I lower the voltage by putting them in a series so one problem is away. Now they have max 80C, plus radiators, plus air space between radiators and batteries. I will keep them on as you suggest if temperature is around 3C. Playing with reading voltage (for example using venusos and node-red and mppt nodes) is one more layer of potential problem/error.
I upgraded the firmware and it still resets everything! And I can't set SOC 100% at 3.45 because it doesn't follow the order of reducing values. This is absurd. I want to go back to my old firmware as I do not restart my BMS!
Yes, some values need to be in a certain order. JK argues, this is necessary to keep things safe for users and also helps the BMS to check on settings when starting up.
You can always go back to an older FW.
@@OffGridGarageAustralia
Yep, I tried to convince them as well. Why OVP(R) should be less than SOC (100%) is something that I do not understand.
They don't have an EEPROM to save the settings in NV memory? That seems like an engineering flaw, not a software bug.
I think they have but the bug saved the settings in the wrong order and with the next restart, the BMS reset due to finding the settings to be wrong.
As a good practice. Manufacturers should set the defaults to a safe value for any battery chemistry to avoid this. It's just common sense.
Yes, correct.
I am still leery of JK's software development methodology, it seems to be the "seat of the pants" method. It would be interesting to see what the c/c++ code looks like.
Yes, the software is very lazy programmed. Especially for a safety device!
okay so they have non write ROM chips except when loading new firmware. I don't think Digital Mermaid gives a hoot and she just hates JK cause she ordered wrong version of BMS and had to buy expensive display. Thanks Andy
They always used NV memory for the setting. That was not the problem.
So it's still a bug, it will still damage Li-Ion cells (by allowing them waaaay to low undervoltage), did you get any hint when they will actually _fix_ the bug, not just make it less serious?
Undervoltage will still damage your cells, yes. but at least it's not an overcharge.
The BMS will not reset any more as explained in the video.
Was this really a safety problem?
The way I understand it, the charge controller in my hybrid Inverter would set the charge voltage for the battery system, the BMS just balances cells, does under voltage protection, over current protection and over voltage protection.
If the BMS settings go to Lithium Ion, the only parameter that really creates a risk is over voltage protection but your charge controller should limit the charge voltage so that protection is redundant.
Am I missing something?
It's my understanding that I could run a battery bank without a BMS and it would be mostly OK until the cells become unbalanced. Then a battery balancer doesn't care what the cell voltage is, it should balance at any cell voltage. This ignores cell failure and over current protection which are big reasons to have a BMS but mostly if there are no faults and your inverter has some type of current limiting, the main thing you want your BMS to do is balance the cells.
So I don't really care what the default settings are but resetting to LTO would be annoying if I have LiFePo4 chemistry. Has the resetting issue been solved?
Is the JK Inverter BMS still a good choice for a 16s LiFePo4 battery bank? What else should I consider?
I have the old JK BMS 16s 2A running, it shows all the time some heating current but there is no heater connected and heating function is switched off.
I'll try to talk to JK several times but I didn't get an answer. Some of you guys have any ideas?
Have you contacted them here?
www.jkbms.net/contact-us
Moin.
Ja, hatte ich schon 2mal versucht, aber hatte über google ne andere Webseite gefunden. Auch was mit jikong.
Aber dann versuche ich hier nochmal. Danke.
I have this same bug on my JK-B2A8S-30P. And this is not an Inverter style BMS as far as I know.
I configured it all in the right logical order and did not get any errors at all.
If I leave the BMS on, it seems to retain the values (at least for the short time period where I tried it). But if I switch the BMS off via the display, it resets all values to Li-Ion.
So doesn't seem to be the case that this only happens if there is an error during configuration. Also: for me this happens even if the BMS is only switched off for a few seconds.
So what am I supposed to do? Buy whatever hardware that is needed for updating the BMS since it is not possible over the app? What a poor product design. Definitely last time I will have used a JK BMS.
So frustrating and a really bad feeling that I cannot rely on my BMS. 😳
oh thanks for the video. I have a problem: the inverter via the can port does not want to work with the inverter deye sun-6k-sg03lp1-eu . Maybe someone has how to configure it correctly so that it works?
Have you tried the Pylontec protocol in your inverter and BMS?
Nice we had -15°C under zero this morning near your home country - czech republic
Stay warm over there.
Like# 37 - It's too bad it takes the power of Andy 1-4 before the power of an update gets initiated. Further, I hope JK adds this to their QA process to make sure this doesn't happen again...
There will be more bugs... I'm sure...
@@OffGridGarageAustralia I manage a software shop, there sure will be more bugs 🐜
Why not have the software Auto correct certain parameters and also reset to last know correct parameters.
Auto correct? To what???
BMS is not resetting any more as the bug is fixed. It's in the video at 8:18
@@OffGridGarageAustralia that's good 👍. I must've missed the part where the bms no longer lets the user input the wrong settings.