JK Inverter BMS Float Mode and 100% SOC reset fixes!

[@StasMatvienko]

I want to say thank you from our Victron community in Ukraine to you! We met bad situation with electricity and a lot of us, I mean it is all our country are trying to install and to connect Victron and diy battery with bms in our apartments and houses. And your information about all it stuff is very helpful for us! Thank you a lot for your videos and explanations!!! We are building a big one and a small one ESS in our houses and your examples and tests are incredible helpful to live for us with lights in our apartments across all of our country!!! Keep going! Thank you a lot!!!

[@Moritz_wfa]

Good point, most of us do this as a hobby. For you guys this has a different dimension. Best of luck!

[@OffGridGarageAustralia]

Thank you very much for your feedback and sharing. I do my best to help you guys out and getting a lot of emails from viewers in the Ukraine with questions around battery and BMS.

[@kc8omg]

Thanks for sticking with this and working with them Andy, really appreciate your advocacy and persistence for everyone running these BMSs!

[@OffGridGarageAustralia]

Much appreciated

[@mikeswartz2007]

@@OffGridGarageAustralia hey. Hoping you might have a solution. I have 2 of the inverter bms’s v15 on the 15.33 firmware. They communicate just fine until you start charging. Then the slave keeps connecting and disconnecting. This makes my inverter ramp up and down constantly. Hope you can help!

[@tomberger6484]

Another severe problem with the JK Inverter BMS: far too high SOC Hello Andy, first I have to thank you for your work. I followed your recommendations and bought a 15-kWh with the Yixiang DIY box. The battery was absolutely easy to build! But I have a problem with the JK Inverter BMS that comes with the Yixiang DIY box: it reports crazy wrong SOC states. In the BMS I did set the 100% SOC value to 3.45V per cell (55.2V for the battery). My Victron ESS was set to stop discharging the battery a 40% SOC (as a precaution for winter time where for some month the battery will not reach 100% SOC). And then the ESS stopped discharging at 45% SOC with a "low battery" alarm. The voltage was only 49.5V. With such low voltage the BMS should report 0% SOC! Then I charged the battery from the grid, and when the BMS reached 100% SOC, the voltage was just 52.9V. This voltage should correspond to a SOC of 70%. I assume that when I continue charging to 55.2V the SOC will be calibrated again. But since the SOC drift happened so quickly I can't rely on SOC reports from the BMS. Now I need to check the voltage every day, and need to adjust the "mininum SOC" in the Victron ESS settings manually. My question: did anybody notice the same behaviour? And is this problem solved by the new firmware version V15.33? Greetings from Kassel, Germany Tom

[@oleggurevich2857]

Please note the voltage on the smart shunt is only 55.16V. To switch to float mode, the condition requires a steady 55.2V for 6 minutes. Since the threshold hasn’t been met, the switch to float is not triggered.

[@homie100968]

I litterally just updated my 3 pack of 280ah batteries and then saw your new video... lol i can confirm my bulk / float is NOT switching over. The requested voltage just sits in bulk.. luckily i have my mppts set to bulk only for 1 hr. My pc software the latest version also is not enabling parallel mode viewing. Thanks for your persistence Andy (Mein freund!). And keep up the good work collecting the empty bottles/cans. Well done!

[@OffGridGarageAustralia]

So, your JK is not controlling your MPPTs?

[@homie100968]

I have checked the victron mppts which report the bms is controlling them. I have set the mppts to the bulk and float voltages and it appears they do you suggest and it seems the mppts are controlling not the bms? I can spend some time this weekend and confirm. I may have my settings too tight between the 2 settings.

[@homie100968]

Ok after further testing my bms aren't controlling my mppts. I set my mppts to a higher than usual charge voltage and my system simply charged to that setting. I have now disconnected the coms cable between bms master and cerbo. Reset my system again to normal gentle charge voltages and I'm happy with that (for now)! My version 15 bms's will be fine for the time I will wait to see if the new software releases can give me confidence that the batterys can tell the mppts to slow down or even stop charging. I will make parallel switching for high voltage and temperature shutdown on the aux inputs on the cerbo to shut off the mppts. Not so hard realy. Regards Scott in sunny and 32c hot Cairns. 😊

[@TobiasZ-z7f]

thank you Andy for forwarding the problem with the float mode to JK and now there is a solution in sight for the "small" JK with 100A

[@OffGridGarageAustralia]

The solution is still not optimal but maybe better as it is right now?

[@TobiasZ-z7f]

@@OffGridGarageAustralia In any case, the current solution V15.32 is much better than V15.30. That's why JK should release V15.32 for all BMS. I don't see the problem with the SOC 100% at Victron. Greetings from rainy Franken

[@Carsten-PV-DIY]

Hi Andy, best wishes from Hamburg. Really APPRECIATING your efforts in the float Mode. We discussed in jens dc channel if we did Something wrong that this core functionality seems broken.

[@OffGridGarageAustralia]

Yeah, one thing fixed, another feature broken... will they ever get it right?

[@GregOnSummit]

Great video that fixes one of my issues ... SOC. If JKBMS has any smarts at all, they need to add a link to this channel. Either on their website or the user manual. Which means that you Andy, need to put all these JKBMS test videos into a playlist. Then you'll fly though that 100K mark. 😁 As it is, I've started a bookmark folder and will be adding all the JKBMS testing videos I can find on this channel and add some tag words to each one. This is invaluable information ... thanks for your hard work on this ... Well Done Sir 👍

[@OffGridGarageAustralia]

I have a JK-BMS playlist. Since the beginning of the channel😊 kzbin.info/aero/PLPomydD54sgA0t8KtNMNfajC3XuJzsMjN&si=drcwjqXBMx9BvdfN

[@Lobo4711]

Hi Andy, thanks for your work improving this bms. Your 100% reset proposal sounds very good! Here some thoughts how further improvement is possible. 1. Why not end RCV time if cell deviation is lower than the balance trigger voltage OR the timer ends? If cells are well balanced there is no need for the full timer running out… 2. Proposal: I noticed that the jk internal balancer always charges or discharges individual cells for 2-3 seconds. If one cell is far out of range, it takes very long to get well balanced. Why not prioritize the worst cells with longer balancing time. This would also help to reduce cells being exposed unnecessarily to high RCV for long time. Maybe jk engineers like to improve the fireware in this way 😊

[@OffGridGarageAustralia]

Well, get in contact with them and ask them for improvements😖

[@Lobo4711]

@@OffGridGarageAustralia I think I'll better wait that my programming AI is capable enough to do the job 😄

[@the_real_Stefan]

Hey Andy, great to see that there is some movement on the JK side. Did you mention the Victron “history” issue to them? This messes up data in the VRM portal. It seems they either mixed up Ah and kWh or using an odd scale factor. Most of the issues should be solvable with software changes/improvements. Hopefully we can get JK to fix these problems and make this BMS as good as possible with the existing HW.

[@OffGridGarageAustralia]

They will never fix all the issues, unfortunately. There are far too many edge cases and inverters out there. A fix for one inverter breaks something in another. Jeeez! It is impossible.

[@DannyBokma]

Thanks so much for being the spokes person for the community. And vey pleasant that you actually know all this!

[@OffGridGarageAustralia]

Thanks for your feedback.

[@tyweb9264]

Thanks Andy for keeping on this. I have 5 x HW V15 and one HW v 14. The 14 no longer communicates properly with the HW15. It comes on and off the RS485 network about every 15 seconds according to the Cerbo GX. Constantly changing the Discharge amps, and the charge amps limits. But I also noticed in Ver15:30, I cant calibrate the voltage anymore. And also the RCf time had a min of I think it was 6hrs the default. I was wanting to test same as you did but it kept saying parameter not sent. Again thanks for following up all these issues. Keep up the good work... 9 Sunny days in a row here in Perth. Tesla is 90%, the House batteries are 100% ....Loving this sun after a wet winter.

[@OffGridGarageAustralia]

Thanks for the feedback. I had this issue with one BMS disappearing for a few seconds as well. I'll chase it up with JK...

[@geozar]

Nice idea Andy about the inverter stops charging when hits 100 % SOC

[@OffGridGarageAustralia]

Yeah, not everyone is happy about it though.

[@vladislavlozitskyi671]

What I can say about my tests for this BMS and Float Mode: v15.24: for example, Your RCV is set to 55.2V. Once this voltage is reached - the RCV timer will start and the BMS will request the RFV voltage after it finishes. v15.30: for example, Your RCV is set to 55.2V. Once this voltage is reached - the RCV timer will start BUT this voltage must be >= 55.2V during all the time set for the timer. If the voltage drops even to 55.19 the RCV timer will be reset and started again when the voltage is 55.2V. Because the charging process has slight voltage fluctuations to avoid this behavior on v15.30 I have calibrated my BMS voltage to a value bigger for 0.05V - 0.07V (if the real battery voltage is 55.2V - BMS shows 55.27V). So now the voltage during the RCV timer is never drops less than 55.2V and the Float Charge Mode works as expected. Maybe this will be useful for someone=)

[@SiBex_ovh]

Nice work-a-round :)

[@OffGridGarageAustralia]

But if you turn on a load, voltage will dipp down far more than what you can compensate with that, so I rather would have the timer running regardless if the voltage fluctuates

[@vladislavlozitskyi671]

@@OffGridGarageAustralia I am not saying that this is a correct way, just a workaround. At the same time I have no any voltage drops because I am using the system as UPS for the house (always on grid, on battery when no grid)

[@hansenmoniquefilms]

Yes same problem here if you have a few cels with 0.040 lower voltage the trigger never goes, hope that Andys @OffGridGarageAustralia idea get in the JKBMS

[@geozar]

After I have installed the 15.32 I have also installed again batmon and I don't hear any beep every minute!! Nice!!

[@OffGridGarageAustralia]

That depends on how Batmon is configured. I have it set to poll every 30s only. That is, so I can still connect via BT to the BMS to change settings without having to stop Batmon first.

[@carlosgaspar74]

Hi Andy. RCV timer full cycle then 100% reset is a problem for those who charge only to 70% for example... Keep in mind that if you have big enough battery capacity and only charge to 100% once in a while to calibrate the BMS and balance the pack for example. Resetting every time to 100% could be a problem because your maintaining under 70% charge...

[@rcinfla9017]

You could have BMS stop all charging when Columb counter in BMS shows 70% but personally I would not do it. The fact you are not doing a full charge says you are going to allow SoC Columb count to degrade over time, creating variance in the amount of charge level applied.

[@OffGridGarageAustralia]

There is no benefit of charging LioFePO4 only to 70%SOC. Why would you do that, it will not prolong the life of the battery.

[@carlosgaspar74]

@@OffGridGarageAustralia I know that calendar aging is the main contributor to battery degradation in small LFP stationary energy storage units. But. But "Cycling LFP cells across a lower average SOC results in less capacity fade than cycling across a higher average SOC, regardless of depth of discharge." [1] study For example If you are in summer and you have solar production and battery capacity that you don't need cycling the battery between 75 and 100% is more detrimental than cycling in the 25, 50% range. This is more important in my opinion on big ESS because of the $$$ ROI. Maybe if... RCV timer count and cells≥3,45 or SOC≥95 then reset to 100% 🤔 Name of the study [1] The Operation Window of Lithium Iron Phosphate/Graphite Cells Affects their Lifetime Keep up the wonderful work you've been doing 💪 Thanks again

[@OffGridGarageAustralia]

@@carlosgaspar74 Yes, I have seen the videos about this study. That is totally impractical with LFP batteries though. Charge them and forget about it. They will outlast all your other equipment.

[@DominikAndreas]

During the summer I use a dynamic SoC limit set via home assistant depending on the last time the battery was fully charged. If fully charged in the last week, limit to 90%, otherwise 100%. This prevents the battery from sitting at 100% all the time while still having reasonably accurate SoC estimates. Not sure if this actually makes a difference, but easy to implement and somehow feels better 😅

[@Ale-or2gk]

Really thanks for pressing them to work on this.... SoC reset and float it's a really important thing.

[@OffGridGarageAustralia]

Yes, I hope they will get it right at some stage.

[@showme360]

Great video Andy as always, but can I make a suggestion about WARNINGS, perhaps making it clear at the beginning and end of the video would help save some people who are in a hurry to prevent any awkward situations with your valuable viewers, its just that some people may only watch the video half through. Just saying OK Cheers Ken in mid Wales.

[@OffGridGarageAustralia]

Thanks for the feedback, Ken. Most of my videos have a lot of information and it is difficult to add all these details to the thumbnail, video title or even description. Maybe I should make them shorter about one specific topic only? But then I would make like ~8 videos out of this single one. It's very hard to get it right...

[@SimonKasper]

Good work again Andy! Thanks! Do you know when we can expect v14 firmware updates with all these new versions? I have two 2x v14 BMS and getting the feeling that v14 is being sidelined by JK?

[@pavcam7031]

Hallo Simon, I´m only ver. 14.x . Do you have new info from this ver 14? Sory may anglish. :-( Thaks you

[@SimonKasper]

@@pavcam7031 No, sorry. I'm also still waiting on V14 firmware updates.. Andy: do you have any information with regard to v14 firmware releases?

[@ЁлыйПалыч]

I measured the actual cell voltage and adjusted the battery voltage in the app. This solved the problem of switching to floating mode. My cell voltages differed from the JK BMS readings by 0.014...0.02 volts (the further the cell was from the BMS, the greater the voltage deviation). As a result, I added 0.32 volts to the battery voltage readings in the app.

[@OffGridGarageAustralia]

Interesting. I need to test this out!

[@NasheedClub.]

Great video, helped me understand BMS float mode and SOC reset!

[@OffGridGarageAustralia]

Thanks. I have a video explaining all settings for this BMS. kzbin.info/www/bejne/m4u7aHaheqqklZo

[@goncalomanuel4386]

Hi Andy, Thanks very much for your work, we really appreciate it. do you have any clue as to when v15.33 will be publicly available ? Best Regards

[@ВіталійВоліченко]

Hi. Remind them that there is also version 14. Thank you.

[@OffGridGarageAustralia]

Yes, I just did. Again!

[@costipopovici2975]

any news with the new awaited 15.33 FW version ?

[@IamVlanOne]

I noticed yesterday that it was available on the google drive. Waiting for the thumbs up first.

[@costipopovici2975]

@@IamVlanOne Thanx, looking forward for Andi's eval to understand whether or not on the FW ver 15.33 SoC estimates and reset got better than before

[@joesmith1142]

Holly Molly! Thats a lot of Spat calibrating.

[@OffGridGarageAustralia]

Since Christmas, yes!

[@rulexi]

Greetings from sunny Arizona. As usual, excellent analysis. Perhaps a enable/disable timer average using mean interruption would give the user a easy option.

[@OffGridGarageAustralia]

Hello, yeah, the problem with the BMS is, that they are running out of memory for new options and features. Hence, they are using existing ones and trying to make it work. It is not ideal though.

[@reflections9409]

If you are usung victron multiplus II/Easysolar, it will now allow the BMS to be used as a voltage/current sensor but without requiring that the BMS be in control at the same time.That led me to decide that I will not go above V15.30 firmware for my JK BMS: My multiplus controls charging (with bulk, absorption, float modes), but based on the BMS reported voltage/current. That prevents voltage differences across the system and perfectly neutralizes JK issues. That is perfect for me. No more BMS control.

[@jimj3936]

So there is a solar forum that has been in communication with JK for a while making suggestions on methods that can be used to resolve a number of the JK issues, including state of charge. It included detailed methods that included suggestions on how to actually perform the calculations.

[@unionse7en]

...but, you are not naming it...why?

[@jimj3936]

​@@unionse7en Look up DIY Solar Forum

[@jimj3936]

@@unionse7en I did reply but it was taken down

[@thestoicloner]

I love this guy! Thanks again mate .....

[@OffGridGarageAustralia]

Thank you.

[@rcinfla9017]

I am assuming all the BMS requested absorb / float voltages and times only apply if you are using inverter communications. For non-inverter communications operation, the user would be required to ensure charger absorb voltage and time period are greater than values set in BMS to ensure a 100% SoC reset, Bottomline is the BMS must stay compatible with inverter comm or without inverter comm operation. Different inverter types react differently when jumping from absorb to float modes. Low freq inverters, like Victron, will usually try to immediately take any inverter loading (AC output or grid back push) and drag the battery voltage down to the float voltage level. This happens even if you have inverter in grid supported UPS mode where normal running AC output power comes from grid. Some inverters, mostly HF types, will shut down charging but will not purposely drag the battery voltage down to float voltage setting when it enters float mode. If in UPS mode they will continue to support AC output from grid, leaving battery at near absorb voltage until it decays to float voltage solely by inverter idle overhead power, and battery/BMS leakage current.

[@MrElciupacabra]

that is why I don't recommend communication with inverter.. BMS must know only protection and ballancing....charger settings are charger settings...in the charger you define battery states not in the bms

[@OffGridGarageAustralia]

Well, it is a Battery Management System. It manages the battery. What you mean is a Battery Protection System.

[@MrElciupacabra]

@OffGridGarageAustralia it manages the battery...meaning protection over cells,battery bank currents and voltage limits...but it shouldn't manage the charging by limiting the current charge ...this should be in the hands of the charger not of the bms

[@Angelorademon]

I noticed a current consumption of 0.5 A from the batteries immediately after switching to floating mode, while the load uses only the city network, not the battery. Do you know what this current is spent on? Why does this happen and what is the consumer of this current? Does it go into heat? Inverter PIP5048MKX.

[@ronhellenbrand]

Hi Andy, Will they also change the way of amps measurement for SOC calculation? The current a Multiplus II charges from battery isn't DC it is DC + a sine wave current. So the peak measurement they use now will show wrong numbers, they should implement running average current!

[@leumel900]

Isn't this a hardware problem? Like how fast the refresh rate of the internal current sensor (shunt?) is?

[@ronhellenbrand]

@@leumel900 What you see is a jump from -3,5A to 3,5A around 0A current shown by victron shunt. They sample and use peak current only elsewhere you would see everything between -3,5 and 3,5A.

[@erik79196]

Same here. If my Victron installation has a load of 2200W and the mppt produces 2400W then the battery current jumps permanent from +15 A to -15A to +15A to -15A ... so the VRM statistics data is trash. How can this be fixed?

[@ronhellenbrand]

@@erik79196 Seems if battery average current is 0,1A positive The BMS reports peak positive current, and if the current is just negative it reports minimum measured current. Because Victron Multiplus II draws a sine wave current from battery, you will see the max and min sine amps. If I disconnect AC IN this jumping is gone and I do see currents like 0,7A. JK should implement average measurement for SOC calculation!

[@urbanbeeman5402]

As always......Good job Andy

[@OffGridGarageAustralia]

Thanks!

[@urbanbeeman5402]

@@OffGridGarageAustralia By the way I got the code to work around my loss of password... thanks for your help.

[@iwanboskamp3182]

To set the SOC to 100% at RCV Timer run-out will cause issues. If the sun goes down while the timer is running and I add a high AC load to my system (Charging my EVs for example) will drop the actual SOC to way below the 100% mark. Then the auto-set after run-out will reset it to 100% which will be wrong. V15.32 works a charm :)

[@PhilippeCJR]

15.32? Where you get this one?

[@iwanboskamp3182]

@@PhilippeCJR See Andy's description above

[@MrLucize]

@@iwanboskamp3182 can't seem to find it in the drive, has it been removed ?

[@MrLucize]

see it now, after clearing the cache

[@leumel900]

Once there is a firmware version that finally works well with all those wonderful features, will I be able to directly upgrade to the latest firmware? Or should I better upgrade one by one and go through all the firmware versions inbetween when starting from a previously stored JK-PB in an older factory state?

[@OffGridGarageAustralia]

Just the latest will be fine.

[@fal6097]

Hi Andy, despite the JK BMS already having to little memory, how do you think about whether a good BMS should have "Charge UCP" (under current protection)? Especially nowadays with solar generators, we can't always make sure that the sun gives us enough charging current to exceed the cell its charge termination current. You state that LiFePo4 chemistry after charging to 3.45V and some rest will settle at 3.35 volts. In case there is very little load, BMS timers have run out over night. The cell voltage is just below 3.35 volts and weather is foggy. That could overcharge the battery all day long with 0.049C (below termination) assuming the cell will never reach 3.45 volts. (the need for something like yamBMS to prevent "over charging" is not ideal)

[@SunInLithium]

I believe the issues of the timer resetting do to a drop in voltage. In the past was in regards to the SOC 100% not being triggered. And the solution to that issue was to implant the 100% voltage reset. Instead of the timer reset. I think keeping the timer going is a much better solution. Although i don't know why this can't just be an opinion button, like the float mode is. Also, the Soc counter should be paused at 99% until the RCVT is past. So that in case the soc drifts in the wrong direction, it wouldn't read 100% soc prematurely and cause it to stop charging.

[@OffGridGarageAustralia]

Yes, I agree.

[@thesupernauta]

First of all, what a great videos you always make, with a wonderful dose of humor. and very instructive, but will there be a firmware update for the JK BMS PB2A16S15P I still have V15.30 with the not working Floate mode

[@jankristensen1513]

What is the reason for daisy chaining the battery boxes? I can see that the bms in battery box number 2 and 3 are ignored, or rather the settings.

[@angyp74]

Andy, Why can't we just the absorption timer be based on trailing current instead of a fixed timer?

[@kappa7c]

Yes this would be better, no false 100% reset if a load is switched during the timer period. The logic should be: voltage at bulk level (or at >= bulk - 2% to take accounts of tolerances) and current below a threshold = switch to floating and set the soc at 100%

[@OffGridGarageAustralia]

Tail currents are not great either. We have seen this with the PACE BMS which has the charged voltage and a charged current. Both need to be met to show 100%. If you have a cloudy day though, charging current is low and as soon as it hit the charged voltage it resets. This is wrong though as there was no absorption phase. Best would be to have a charged voltage, a tail current and a timer like in the smart shunt. But where do we stop with all this?

[@kappa7c]

@OffGridGarageAustralia tecnically, in this case there is no absortion phase, but the battery should be equally charged(same voltage with the same low end current). Maybe this can be a problem for balancing the cells if it happens frequently, so I agree that the timer should be maintained

[@afferzikes]

any news on 14.x ? - watching video now..

[@OffGridGarageAustralia]

Apparently, they are working on it. I'll chase it up.

[@adrianmar397]

Hello , i have a problem to set the com with the Voltronic inverter 8kw , it doesnt work , you have the protocols for or seting to set it up with the inverter , i have a jk Bms with 16 cels

[@cubb4726]

Thank you so much for all your effort with the JK-BMS. Hopefully you can manage to get also the newest firmware for the 100A Version. Do you know if this will be done? Best regards and everytime a little bit sun

[@OffGridGarageAustralia]

As far as I know, they are working on the versions for all of their BMS. I'll chase it up with JK...

[@MrJramirex]

I'm glad you are helping JK make the BMS better. I do have a concern, though. You are using a Victron system, so all of your findings are helping your side of things. But the JK supports all kinds of systems, and erasing or adding things to the BMS may break or introduce bugs to other systems. I'm not sure what the solution or answer to that would be, but it's something to think about when asking for logic changes.

[@OffGridGarageAustralia]

Yes, I'm fully aware of that. Hence, I'm leaning to run all my BMS without any communication. Seplos, Pace, JK old and new... I just parallel the batteries, set the protection parameters in each battery and let the inverter/MPPTs do the charge algorithm. It really works the best and removes almost all headaches.

[@SiBex_ovh]

JK is a KING :)

[@angyp74]

Andy, When do you think we will be getting V15.32 for the 150A BMS?

[@Timuragashka]

It is sad, that we have an update only for 200a version of jk bms How to be with another models?

[@OffGridGarageAustralia]

Yes, I don't know. I have to chase it up...

[@MichaelSchmidt-rn3gg]

Hi Andy, thanks for the work! Great Job! Did you hear about the Firmware for the Hardware V14? Viele Grüße

[@OffGridGarageAustralia]

Thanks, Michael. I have asked them about the V14... I'm waiting for it myself.

[@sevani777]

Hi Andy, please tell me where to get the firmware for jk bms b2a8s-30p, the regular version, not the inverter one!

[@jimpowell9205]

How do I email you about this BMS issue?

[@OffGridGarageAustralia]

Don't email me. I'm getting 20+ emails a week from people wanting more features from JK or reporting more bugs. Nothing I can do about it, sorry. Email JK directly www.jkbms.net/contact-us

[@robby60259]

Is the LDC screen on the JK Inverter BMS worth the extra cost?

[@Koernell]

No. I used it when I first time started the battery. But never again after that. But without it, I'd have an empty hole in my battery box.

[@OffGridGarageAustralia]

No screen is fine, no need for a display.

[@robby60259]

@@OffGridGarageAustralia I agree, you have showed there is already Bluetooth and RS485 communication.

[@daniele9160]

Hi, what cables should I use for the P- and B- cables of the BMS for a 16s 2p 48v system?

[@olottin]

Hi Andy, I tested the v15.30 version today: it fixes the charge/discharge direction bug on Victron VRM but all the BMS settings were reset to default during the update... I do not remember that the previous updates had the same effect, am I wrong? If I may suggest a small improvement to JK, the new way to number the cells (XX::XX) in Victron console is not useful when one have only one battery and maybe they could still use the old numbering way when only one battery is connected to the system?

[@OffGridGarageAustralia]

An update usually does not reset the settings, no. Not sure why it did for you. Totally annoying. Why is the numbering not useful? I think it gives you all the info?

[@olottin]

@@OffGridGarageAustralia I got used to the new numbering system, just needed time 🙃🙃

[@peterdiy]

Oh the suspense... I finally pulled the trigger and ordered 16 MB31 cells yesterday. ETA some time January. Being new to this (and a tad lazy) I had thought of ordering the Yixiang DIY box for them. But now with all those issues with the JK BMS that is included with the box, is that still a good choice? Any other box that is (at least almost) as easy to use, and comes with a more reliable BMS? Or is JK with the current firmware 'good enough' for a simple single battery setup?

[@AdrianMZA]

Hi Andy🙂 Do you perhaps have the latest software for the JK BMS display, which displays two additional options: Cap/Pwr? Greetings from Poland :)

[@nctarheel1980]

So I’m running the 200amp Jk with an EG4 6000xp. The requested charge volt stops as soon as 1 cell reaches the RCV. As for the float mode this inverter does not listen to this bms at all. I’m hoping this firmware will fix this. PS. Signature solar support tells me the lifepo4 cells don’t require or need to be floated. I’ve used lithium batteries since they started in the RC world. They don’t understand floating these batteries at set point voltages till amps hit zero is a thing. Andy could we get videos on other inverter systems from you?

[@OffGridGarageAustralia]

I don't have any other inverters. It would be a huge task to test other inverters. Signature Solar does not understand much about LiFePO4 batteries it seems if they are saying they don't need to be floated. If you use the Pylontech protocol in the JK to communicate with the inverter, it has some limitations and not all data may be transmitted.

[@HybridShedIraq]

I saw some new BMSES that controls the max charge/discharge current to charge the battery depending on SOC and temperature of the cells so it will be dynamic and improves the battery life.

[@OffGridGarageAustralia]

Yes, I will present the new JK BMS soon which has some of these features.

[@HybridShedIraq]

@@OffGridGarageAustralia Thank you Andy! is it possible that they add these features with firmware upgrade?

[@OffGridGarageAustralia]

@@HybridShedIraq Yes, it is possible, but unlikely. No real need to reduce the current with rising SOC.

[@CoryRobson]

I'm still awaiting a firmware release for the version 14. currently I'm still running 14.20 and cannot get my cells to full capacity. even though i have balancing starting at 3.3 if i get one cell reach 3.45 then straight away it resets to 100% soc and stops charging. balancing will still continue but it does not resume charging in the day time whilst sun is out.

[@pmacgowan]

Boy charging batteries is complex, lucky they JK has someone like you 🙂

[@OffGridGarageAustralia]

It actually is very easy to charge these batteries. Maybe we should let this communication thing go and let the MPPTs or inverter do the charging.

[@egbertegberts5384]

I've noticed that when the bms goes into float mode the Victron inverter starts discharging the battery rapidly until the voltage reaches RFV. All excess energy is dumped into the grid. To me this is a very silly action. Why not use this energy to power the AC loads in stead? It can be stopped by setting DC Coupled PV - Feed In Excess to Off (ESS setting) but this also prevents excess solar from feeding into the grid which costs me money. That's why I've disabled Float mode. My advice therefore is to enable Float mode only when you're off-grid or when feed in of a DC coupled PV is switched off.

[@OffGridGarageAustralia]

I have seen this as well in my off-grid installation. The load will consume all the energy until RFV is reached, then, the solar will kick in again and stabilise the voltage at the set point. I think, we are getting to a pint where the BMS and its software is unfixable.

[@egbertegberts5384]

@@OffGridGarageAustralia right, this should be fixed by Victron but I think it may not be easy to fix as well.

[@allstarnb1]

I have an EG4 6000XP and have the issue you are describing with the inverter stopping charge as soon as it sees 100% SOC, very frustrating.

[@OffGridGarageAustralia]

Hopefully they fix it soon.

[@malejuzasnak]

I've encountered a problem with JK and Victron system and I'm not sure where could be the problem but I'm inclined to say it is on the JK side. After the RCV timer runs out and JK sends RFV to the Victron system the externally controlled (by JK) Victron MPPT stops even though there is a connected load. So the battery starts discharging until the battery voltage drops to the RFV and at that moment the MPPT starts working again. Any idea what could the problem be?

[@OffGridGarageAustralia]

Perfectly normal. Going from RCV to RFV, the voltage needs to be lowered (55.2V -> 53.6V for example). Once RFV is reached, the MPPTs will ensure RFV will be stable and only supply enough juice to power the loads. How else would you lower the voltage in your battery than by discharging it?

[@malejuzasnak]

​@@OffGridGarageAustralia I would expect the MPPT to assist in powering the load after the RCV timer runs out even though the voltage difference between the battery and MPPT would mean the bulk of the power would be drained from the battery.

[@marcorobbe9003]

Hi Andy, also from my side thanks a lot from Germany (next to Hermann from the Teuteburger Wald). It’s so great to see the progress and I hope, the inverter BMS is usable in production mode in the near future. Or do you think, it is right now? Best regards, Marco

[@OffGridGarageAustralia]

I know Hermann. It's been a while, but I was there! Honestly, they will never get this right, I'm afraid...

[@marcorobbe9003]

@@OffGridGarageAustralia OK, next time you are there, I will spend you a fresh Beer 🍺

[@dijovi]

Does it support Pylon RS485 protocol?

[@zapatto51]

Hello. No upgrade for V14 ??? Greetings from France.

[@OffGridGarageAustralia]

I have't heard anything but asked them, so hopefully soon...

[@ser-Sound7]

Tell me, thank you, why the red error light is on, I collected the battery and I don’t see any errors

[@APatchworkCanvas]

both of my jk bms quit communicating with my pc. I updated them a few months ago. Tried to update to the latest version and the one bms wont connect at all. The other one will connect but all options in the pc software are greyed out. In the command window the pc is receiving info but not sending.

[@Gooey1000]

Hi Andy. Great video. I'll buy the same JK BMS replacing my troublesome 123 smart BMS. I have a question: how can I connect the JK BMS to my Victron Cerbo GX?

[@egbertegberts5384]

@@Gooey1000 you need a Victron VE.Can to CAN-bus BMS type A cable for this

[@Gooey1000]

@@egbertegberts5384 thanks Egbert! Is that cable just plug and play?

[@egbertegberts5384]

@@Gooey1000 yes it is!

[@Gooey1000]

@@egbertegberts5384 awesome! Thank you so much. I'm dumping my old BMS after years of trouble and want to set up an easier one. (I hope).

[@egbertegberts5384]

@@Gooey1000 one minor but important addition: in the BMS app, you should set the 'CAN Protocol No.' parameter to 4, to select the Victron CAN protocol.

[@arranpritchard9004]

I get "NA" for temperature reading from B1 B2 B3 and B4 but I do get a temperature reading for the MOS, any ideas ?

[@benoitblondeau4828]

hello i hope you are well? i need your advice. i made my solar installation with a battery that i built myself using 16 modules of 320 ah managed by a bms jk bms B2A20S20P that i configured thanks to your video. i want to make a second battery of the same design but i have to change my bms B2A20S20P which does not take parallel mode. which bms do you recommend to be able to use my homemade batteries in parallel mode? thanks for your help benoit

[@OffGridGarageAustralia]

All BMS can be paralleled. There is no actual parallel mode. Just ensure that the voltage is the same and every battery has its own breaker/fuse and BMS. I have paralleled like 10 batteries with different capacities and BMS without any issues.

[@teapot_]

Andy, how many days of holiday a year do you get from JK company ? Also you have to work at tbe weekend.....😊

[@OffGridGarageAustralia]

Ha, you wish. JK is my hobby😆

[@teapot_]

@OffGridGarageAustralia You wouldn't work these hours if it was a job! It must be a hobby, and I've got to say a very interesting and continously developing one. As new stuff comes along, we can look forward to you putting it though it's paces. You will easily hit the 100k and onward, keep them coming Andy..thanks.👍

[@maritzm]

Thanks for hard work and testing @Andy ! Quick Question: Can this BMS be used with a 15S battery bank? I was planning to build my own to add to my existing 15S bank but hear some concerns about this idea.

[@OffGridGarageAustralia]

Yes, it can handle 8s-16s LiFePO4 batteries

[@WBurg]

Question, I only have one battery bank. Which firmware version should I use? Is version 15.24 ok? What is the status of the new firmware that JK “hopefully” developed ?

[@skyeye9262]

Hi Andy, firmware V 15.33 is finally released. Please, could you check and post a short video? Thanks a lot!

[@joseo.cordero7286]

Where did you get it? It is not on their website yet.

[@thomaslinzbauer9120]

Yesterday I installed the new V 15.33 also two BMS 150A Version and use can bus with Victron mp2gx. At the moment it works.😊 I got the SW from a forum. Release date 03.11.2024.

[@skyeye9262]

@@thomaslinzbauer9120 Thanks 🙏

[@grua-vfe]

​@@thomaslinzbauer9120 did you test float mode?

[@Degrach]

Vous avez un lien pour télécharger la 15.33 ?

[@COC_CUSTOMS_MotoTonics]

Good day, have you encounter error on your jk-bms like this ? "PL Module Abnormal"

[@JeGo9]

Great video! 16:27, what if you start discharging battery before the rcv timer is up? How would it be able to reset SOC to 100% then? Some people might be constrained with needing to discharge soon after charging is done eg charging finished just before offpeak rates finish, and discharge straight away as peak rates kick in.

[@OffGridGarageAustralia]

Yes, all not ideal. What would be your solution for the problem.

[@donalain69]

I got a problem with the bms communication between jk_PB1A16S15P And jk_PB2A16S15P One of them has UART2 set to 1 and the other to 15. I updated both firmwares to 15.30, but didnt help. I cant change the settings for UART2, so no idea how i can get them to use the same protocol.. Or does it even matter?

[@robby60259]

Hello, does anyone know what the red and black wire is for on the heating interface port (P9) on the JK Inverter BMS? The Chinese translates to smart charger detection?

[@IamVlanOne]

Thank you for all the information and work on these BMSes. These are still a work in progress and I have picked up an issue on two occasions where the reported capacity and calculated cycle count changes. I believe this happened because one cell went below the SOC 0% voltage. During the charge cycle as soon as the voltage of the cells rises to above the SOC 100% value the BMS takes the current remaining Ah and reports that to be the new battery capacity and ignores the configured capacity. I do not know how to get this capacity value back to the set value. Once before I changed the capacity in the App to a new value and then back to the correct value, but this changed the cycled capacity and count to 0. Has anyone noticed this? I have some graphs with some detail. I am currently on 15.32, but this happened before on a few minor versions back.

[@bryce00723]

Greetings from QLD. Thanks for the update and test! I successfully updated my 200A V15 but am finding I cant update RCV or 100% SOC below 3.580v without "Sending Failure" am I missing something stupid? Unplugged CAN cable and have rebooted with same problem..

[@OffGridGarageAustralia]

The parameters need to be in the right order. Highest to the lowest: OVP - RCV - SOC-100 - OVPR

[@bogdanverezberger1873]

When change a high value to a lower must start from the lowes parameter value and working up to the highest.

[@saschahorstkotte3331]

Moin Andy, good job. But at the start of you video the app shows cell type "LFP". I updated 3 BMS now to 15.32 but all shows "LPF"🤔. Special version or an old screenshot?

[@OffGridGarageAustralia]

Yes, known mistake, will be fixed in 15.33

[@onapt19]

Hello from Ukraine Odesa. I have jk inverter bms v15.30, 150 amp. And Deye 6kw. I have some issues -rcv set 3,45 volt, inverter shows the same values 55,2(16s)but are charge request is 0.1 volt less only for 55.1 this is , and jk can’t change regime because voltage will not reach 3,45

[@OffGridGarageAustralia]

Yes, that's a known issue. The problem is that the inverter measures a different voltage than the BMS and there is no sync of the voltage. It's a limitation of the inverter. If connected to a BMS, it should read the voltage given by them BMS instead of measuring the voltage itself again.

[@kappa7c]

you can calibrate the voltage of the bms: when the battery is at 55.1V just put 55.25V in the calibration voltage on the jk app.

[@romanpetry1625]

Thx for all the work

[@Jmarty34]

Hi…love your videos. I am building 2 separate batteries using eve 3.2 volt 304 Ah- 4 cells for a 12 volt battery. I am confused as to what BMS would I need ? Type and AH recommendation ? Thanks

[@OffGridGarageAustralia]

Thanks for the feedback. The size of your BMS depends on the current it needs to handle. So, whatever maximum load you want to connect to it, determines what amperage the BMS needs to have. If in doubt, just get the 200A BMS.

[@66paolob]

Hallo Andy , i have inverter voltronic mks iv and now i make the new battery 16S2P whit bms jkpb2a16s20p, do you have the diagram for make the connection battery - inverter ? Thanks

[@p0l0us]

Hi Andy. I'm bit concerned you don't considering other use of JK BMS than yours, when proposing changes to JK. For instance, I use JK on the boat with two 16s batteries in parallel, charged by 250A generator (on/off only), 3000Wp solar panels (Bluesolar) or from shore 220V/380V (Multiplus 5000) all driven by CerboGX. I would appreciate if JK invertor BMS shows max 99% or 99.9% until the battery is balanced (eg. all cells 3.55V) and show 100% once complete so the generator stops immediately. On the other hand, when the boat is charged from solars or from 380V shore plug I would like to run timers and achieve the float mode (eg. over night). I'm afraid moving the 100% after the timer, like you are proposing, will break my scenario.

[@jasong7374]

I am not going to update any firmware until all the bugs are gone, good to see the progress

[@OffGridGarageAustralia]

There will always be bugs. You can only swim with the flow...

[@rcinfla9017]

Since you have communication with inverter setup, it would be nice if you can figure out what conditions trigger the 10A DC to DC converter charging reduction. I am guessing it only runs if cells are out of balance and highest voltage cell is approaching overvoltage limit. I believe charger must maintain absorb voltage level setting when charging current drops down due BMS DC to DC converter. It likely requires inverter comm to prevent inverter/charger thinking battery is fully charged and dropping to lower float voltage mode.

[@dejanlupo]

Andy, thanks for your detail analyzation. My JK BMS both series PB and B series have blocker bug. When at over charge state the JK MBS disables the charging of the battery also silently disable discharging. With silently mean not seen on screen, but no current is flowing. The same is happening when under voltage protection gets off. The JK BMS silently disallow charging. This is a blocker bug as battery will never come back to live. I have JK PB2A16S15P V15.23 and JK_B2A20S20P V11.26. Both have same issue. Do someboty noticed same issues?

[@OffGridGarageAustralia]

I have never seen or heard about this before. Are you sure the BMS causes this issue and not something else in your system? Crazy that both, PB and BD series show the same bug as their software is different...

[@ionuttimofti]

Any news about V15.33? :)

[@OffGridGarageAustralia]

Nope, the work ethic in China dies off again if I'm not constantly chasing....

[@ckvsoft]

Hi Andy It looks like firmware version 15.32 is now available for the 150A BMS. Do you happen to know where I can download it? I recently received a replacement unit for one of my inverter BMS systems due to a defect, and it already came with version 15.32 installed. However, my other devices currently in operation are still running on version 15.30. Unfortunately, I can't seem to find a download link for the newer version anywhere. Best regards from a not-so-sunny Austria, Chris

[@OffGridGarageAustralia]

Hi Chris, I haven't seen the V15.32 as an official release. I only showed the 200A version of the BMS which you can download from my website here. off-grid-garage.com/battery-management-systems-bms/ I think the next official release will be V15.33 or .34

[@hansenmoniquefilms]

Hi Andy almost perfect ! question would the float and absortion settings be the same in solar/off-grid and solar/on-grid situation be the same what do you think ?

[@OffGridGarageAustralia]

No, they should not be the same. Never. Why would you keep the cells on 3.45V when they are full until sunset?

[@WBurg]

V15. 33 is available... When will you test this Firmware-Version?

[@aposcik]

"Hi Andy, I was wondering if you could test your battery with the JkBms connected in series, 48V + 48V, to get 96V. Or perhaps someone in the community on this channel has experience connecting them in series?

[@gandalfstormcrow9605]

On the consumption calculation, I calibrated my amperage draw so charging at around 10 amp per bank, the three banks were agreeing with the smart shunt and the clamp on meter. When it goes into discharge and the Smart Shunt shows around three amps discharge, (the solar is making up the difference) the BMS's are showing a draw of around 10 amps on each of the three. One in an inverter BMS the other two are older JK BMS . This will mess with the charge state a lot. They all act the same

[@OffGridGarageAustralia]

Yeah, it's not great... you cannot win.

[@julianjiron8085]

Is the JK bms 16s 200amp A2 balancer experimental or is it ready to deploy? What do you guys think?

[@DanBurgaud]

10:36 so you saying the charger needs to be of higher voltage than the RCV. My Deye does exactly this. BUT I am not into Float to have tested this unfortunately.

[@peetsplcchannelnice6912]

Anyone have a list with te best settings for JK-BMS with lifepo4 cells?

[@OffGridGarageAustralia]

What are the best settings? It depends on your system design and needs. I have all my settings listed here. That's probably a good point to start and understand all settings off-grid-garage.com/my-settings/

[@iPeel]

The ability for the BMS to never announce 100% SoC until the absorbtion is complete would be really useful, my inverter suffers the issue that it stops charging at 100% SoC and I frequently have to play around with the reported SoC every now and then to bring batteries back to 100% charge. An option for the BMS to report never more than 99% SoC until the RCV timer elapses would prevent the inverter from shutting off charging prematurely.

[@Paul_______]

I was thinking about getting a inverter BMS for my battery. What's the difference between their inverter BMS and their regular BMS

[@leumel900]

The inverter bms (JK-PBxxxxxx) can communicate with your PV-Inverter and control it, e.g. request a specific charge voltage depending on your soc (float mode) or switch off the inverter in case some fault occurs. The "normal"/"old style" (JK-Bxxxxxx) bms, can't do that, the only communication it has is bluetooth and rs485, so you usually can only use those for setting it up but not integrate it into your pv-ecosystem or even smart home system. Still a great choice for "dumb" applications that don't require communication to the charger.

[@hanskroeger6239]

​@@leumel900and they can handle Up to 300 A continuous current.

[@leumel900]

@@hanskroeger6239 afaik only 200A max. Where did you find a 300A version?

[@CliveTrezona]

@@leumel900 If I have an old style inverter/charger that has no communication, can I still use the newer inverter bms? Or would the older JK bms' be better?

[@leumel900]

@@CliveTrezona Of course you can use it without connection, but it will miss all the functions that set it apart from the JK-B version (because those require communication). Depending on your use case, a JK-PB might just be money thrown out the window.

[@peetsplcchannelnice6912]

Question, If you have 2 or more JK-BMS connected wich one controls the signal to Victron to change the charge/absoption voltage? Is this the master or do all BMS report to Victron to change to absorption voltage.

[@OffGridGarageAustralia]

All BMS report the data to the master which communicates with the inverter or Victron.