• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

JKBMS Inverter Edition Problems/Issues | No Support / Help to fix major issues. - DO NOT BUY ! Warning (as of Oct.12.2024)

Steve_S said:
Wow, the Hardware is 15 so that's good.
The firmware 15:18 is pretty old (relatively speaking LOL) I dunno what that one's quirks/foibles & goofs are.
Your Phone App 4.16 is quite dated and if memory serves does have all the things needed for the inverter models. Likely best to update that phone app.

Just peaked into Andy's Garage's last video - comments. More issues being reported (including some fixed issues).
Other issues are being reported elsewhere as well... IS ANYONE SURPRISED ?
Click to expand...
That's strange that you think my phone app seems old. I used the APK from the link you posted in post 386 of this thread.
And it made my app work with the new inverter BMS when it wasn't working.
Maybe I need to delete the app before I load the APK.
 
JK Android App : Publish at 2024-08-22
Version 4.27.0
Build 37
www.pgyer.com

极空BMS

极空BMS 5.0.1(Build 40) Latest APK download for Android. 点击下载安装，尽享便捷电池管理！
www.pgyer.com www.pgyer.com

When I installed from Google (one of the rare times it allowed it) I had to uninstall prior to installing a new version from elsewhere. For me, GooglePlay sometimes works but not that often, insists my Motorolla G is not compatible while it is.
 
Steve_S said:
JK Android App : Publish at 2024-08-22
Version 4.27.0
Build 37
www.pgyer.com

极空BMS

极空BMS 5.0.1(Build 40) Latest APK download for Android. 点击下载安装，尽享便捷电池管理！
www.pgyer.com www.pgyer.com

When I installed from Google (one of the rare times it allowed it) I had to uninstall prior to installing a new version from elsewhere. For me, GooglePlay sometimes works but not that often, insists my Motorolla G is not compatible while it is.
Click to expand...
Thanks for the reply. I have a Motorola Edge.
 
Steve_S said:
JK Android App : Publish at 2024-08-22
Version 4.27.0
Build 37
www.pgyer.com

极空BMS

极空BMS 5.0.1(Build 40) Latest APK download for Android. 点击下载安装，尽享便捷电池管理！
www.pgyer.com www.pgyer.com

When I installed from Google (one of the rare times it allowed it) I had to uninstall prior to installing a new version from elsewhere. For me, GooglePlay sometimes works but not that often, insists my Motorolla G is not compatible while it is.
Click to expand...
Thanks for the link. I typed the URL into my Samsung Android (no prior app installed) and the install was completed.
 
AshleyL said:
For the Dry Relay contact, one cannot enter decimal value.
For example, "Cell voltage over" and "Cell voltage below". If we can enter decimal value, I can connect the Heltec 5A capacitive balancer on/off switch to the relay.
Click to expand...
Try to press the decimal point twice....I have to do this on my jk (not inverter) bms.....
 
Okay.......after waiting for few days.....Andy just posted a new video on the SOC% issue........and it is all about current calibration.....in which we already tried many times.......
 
AshleyL said:
Okay.......after waiting for few days.....Andy just posted a new video on the SOC% issue........and it is all about current calibration.....in which we already tried many times.......
Click to expand...
Is he talking about SOC drift or SOC 100% reset

Edit: watched video. Only talked about drift
 
Last edited:
ChrisG said:
Is he talking about SOC drift or SOC 100% reset

Edit: watched video. Only talked about drift
Click to expand...
Yup. I am somehow very disappointed with JK company.

Still no update or anything on the SOC 100% stop charging solution and firmware update for version 14 hardware.
 
AshleyL said:
Yup. I am somehow very disappointed with JK company.

Still no update or anything on the SOC 100% stop charging solution and firmware update for version 14 hardware.
Click to expand...
Is the SOC charging issue resolved on v15? Don’t think it was
 
Is that a BMS firmware problem or an inverter firmware problem?

Sunny Island doesn't have that problem; it has others instead (unless just operator error.)

How about making a man in the middle who jumps up and down shouting "99 ... 99 ... 99" until conditions where you actually want charging to stop?
 
Hedges said:
Is that a BMS firmware problem or an inverter firmware problem?

Sunny Island doesn't have that problem; it has others instead (unless just operator error.)

How about making a man in the middle who jumps up and down shouting "99 ... 99 ... 99" until conditions where you actually want charging to stop?
Click to expand...
Both BMS and inverter problem.

It already been discussed and confirmed that Sunny Island SMA inverter doesn't use the SOC% for charging in BMS communication mode, the inverter ignores the SOC% during charging and relies on voltage to determine the true SOC. The Sunny Island manual even mentions this. SMA only use the SOC% for "back-to-grid" functionality.

We do have someone developing the 99% solution. Ghost Battery SOC Fixer.

ChrisG said:
Is the SOC charging issue resolved on v15? Don’t think it was
Click to expand...

It is never been resolved with v15.30. I was hoping andy video would show us a new firmware resolving the issue. It turned out.....just a simple current calibration which we all already did.
 
Watched Andy's latest video and looked at the comments. Oivey.
Andy doesn't seem to think that Voltage Calibration is a "thing" even worthy of considering for the SOC drift etc.

He is right about trying to pull a Constant fixed current via Inverter. It does have to be DC to DC without the ripple.
Using a DC Supply with a fixed Amperage & Voltage is pretty much the only way to get it done.
- BTW: You have to check the Voltage at the END of the Power Supply wires as to what the actual Voltage is, relying on the Supplies "displayed" value is foolish, always check, often it will be off a bit. (wire losses etc)
Yes, the AMP Calibration does not allow for a decimal point.
- FYI: Clamps meters are NOT that Accurate, because it is a SENSE device and not an actual inline measurement. Their docs provide a +/- Accuracy value because of it.
However you do absolutely have to Calibrate the Voltage @ the Batt Pack Terminals too !

Power Supply NOTE:
"Assuming" the displayed values are accurate is foolish. Internal to the power supply they may be, but the moment it hits the wires to battery that changes. ALWAYS check the voltage provided at the end of the wires before attaching to battery pack. Even the clamps/lugs/terminals can (and often do) affect that. Crappy clamps/clips can cost you voltage & amperage !

----------------
No resolution as to WHY the BMS can "choose" to hit 100% BEFORE reaching the set 100%-SOC value in the settings. Barely brushed past that point. we know why, coulomb counting is awry, largely because it fails to measure low values properly.

Given the information Andy stated from the JK Tech who said how to Calibrate the Amperage, I am actually surprised (well not really) at that... But Andy "fell for it" (rather accepted) what this ding-a-ling said ... seriously ??? I did my electronics engineering courses in the 1980's and haven't forgotten much (sure as hell not THAT much) to know the JK Techie is a dork ! and the fellow who believed it... well.... whatever.

SADLY, I fear the Customers are on their own ! Andy as the "interface to the techies" is a FAIL - he doesn't get it / doesn't want to ?
Irony is that Several Senior EE's and other Serious "Real" Techies "were" available & willing to help JK and even did so but JK blew them off for a "YT Personality".

At least Andy is getting More Viewers, More Likes & More Subscribers therefore more $ income resulting in a happier Mrs Andy ! So ALL IS GOOD, Carry On ! NOT POSTING A LINK TO HIS VIDS - AS IT SUPPORTS NON-ACTION and zero credit/mention given by Andy to Anyone on FaceBook Groups, Forums in EU, US and even Kenya (Yeah amazed me too) - he's "The Champion" of JK Customers... LMAO... OI !
 
Steve_S said:
Watched Andy's latest video and looked at the comments. Oivey.
Andy doesn't seem to think that Voltage Calibration is a "thing" even worthy of considering for the SOC drift etc.

He is right about trying to pull a Constant fixed current via Inverter. It does have to be DC to DC without the ripple.
Using a DC Supply with a fixed Amperage & Voltage is pretty much the only way to get it done.
- BTW: You have to check the Voltage at the END of the Power Supply wires as to what the actual Voltage is, relying on the Supplies "displayed" value is foolish, always check, often it will be off a bit. (wire losses etc)
Yes, the AMP Calibration does not allow for a decimal point.
- FYI: Clamps meters are NOT that Accurate, because it is a SENSE device and not an actual inline measurement. Their docs provide a +/- Accuracy value because of it.
However you do absolutely have to Calibrate the Voltage @ the Batt Pack Terminals too !

Power Supply NOTE:
"Assuming" the displayed values are accurate is foolish. Internal to the power supply they may be, but the moment it hits the wires to battery that changes. ALWAYS check the voltage provided at the end of the wires before attaching to battery pack. Even the clamps/lugs/terminals can (and often do) affect that. Crappy clamps/clips can cost you voltage & amperage !

----------------
No resolution as to WHY the BMS can "choose" to hit 100% BEFORE reaching the set 100%-SOC value in the settings. Barely brushed past that point. we know why, coulomb counting is awry, largely because it fails to measure low values properly.

Given the information Andy stated from the JK Tech who said how to Calibrate the Amperage, I am actually surprised (well not really) at that... But Andy "fell for it" (rather accepted) what this ding-a-ling said ... seriously ??? I did my electronics engineering courses in the 1980's and haven't forgotten much (sure as hell not THAT much) to know the JK Techie is a dork ! and the fellow who believed it... well.... whatever.

SADLY, I fear the Customers are on their own ! Andy as the "interface to the techies" is a FAIL - he doesn't get it / doesn't want to ?
Irony is that Several Senior EE's and other Serious "Real" Techies "were" available & willing to help JK and even did so but JK blew them off for a "YT Personality".

At least Andy is getting More Viewers, More Likes & More Subscribers therefore more $ income resulting in a happier Mrs Andy ! So ALL IS GOOD, Carry On ! NOT POSTING A LINK TO HIS VIDS - AS IT SUPPORTS NON-ACTION and zero credit/mention given by Andy to Anyone on FaceBook Groups, Forums in EU, US and even Kenya (Yeah amazed me too) - he's "The Champion" of JK Customers... LMAO... OI !
Click to expand...
Can we solve (the biggest part) of the SOC issue with the solution"

BMS SoC Fixer -- 'Ghost Battery' ?​

I am almost ordering the 2 hardware parts. I dont even have my 3 JKBMSses yet, but something tells me I end up with this Ghost fixer solution.

Did you saw this, tried it? :)
 
Steve_S said:
No resolution as to WHY the BMS can "choose" to hit 100% BEFORE reaching the set 100%-SOC value in the settings. Barely brushed past that point. we know why, coulomb counting is awry, largely because it fails to measure low values properly.
Click to expand...

Sounds like we want BMS to rely on coulomb counter at voltages between the knees, but climbing above the knee rely more on voltage to indicate SoC. Needs adjustment for IR drop of course.

So long as some cells haven't reached 100% voltage, it could sit between 99% and 100%. Only hit 100% when balancing complete.

So Ghost Battery may fix that.


For our Sunny Island (which I haven't hooked to BMS quite yet, learning from other members at the moment),
I understand problem is JK BMS says 99.xxx% and Sunny Island suddenly shifts to something like 0.0000099xxx%
We don't know yet if there are communications glitches or firmware bugs.

A communications system ought to have error detection/correction, or check subsequent values for consistency.
 
Getting close to connecting everything in my system. Should I leave the inverter (Victron Quattro) to BMS cable disconnected and just let the Victron gear do it's thing like it normally would do?
 
HRTKD said:
Getting close to connecting everything in my system. Should I leave the inverter (Victron Quattro) to BMS cable disconnected and just let the Victron gear do it's thing like it normally would do?
Click to expand...
Personally I'd try going open loop with just voltage, to check for basic system functionality before embarking down the perhaps more complex battery comms route.

Did you mean from a Cerbo or other Venus OS device? I don't think you can plug a jk BMS into the ve.bus port of a quattro.
 
Hedges said:
I understand problem is JK BMS says 99.xxx% and Sunny Island suddenly shifts to something like 0.0000099xxx%
We don't know yet if there are communications glitches or firmware bugs.
Click to expand...
I don't get these glitches and have used several Firmware's and currently on 15.24, also my Canbus cable is only 1M long to the master SI. Sounds very much like interference on the comms, either cable related or a particular BMS. I only use the two BMS's plus a ghost, maybe once you go above a certain number of batteries the master BMS has periods its CPU is overloaded and this results in error being sent sent via canbus.
 
Brucey said:
Personally I'd try going open loop with just voltage, to check for basic system functionality before embarking down the perhaps more complex battery comms route.

Did you mean from a Cerbo or other Venus OS device? I don't think you can plug a jk BMS into the ve.bus port of a quattro.
Click to expand...

Yes, I meant the Cerbo. I haven't got far enough to actually wire anything into the Cerbo, so I have a bunch of cables still in bags and a Cerbo sitting there all alone.

With the BMS being called an inverter BMS, I get confused.
confused3.gif
 
Last edited:
After Andy's last JK videos I have the following take. At this point I don't trust Andy. JK lost its senior devs to some intern that asks Andy for how to count amps?!? Whatever Andy knows or not he hides it behind his showmanship. Not wery helpful.

Clearly the firmware isn't capable to calibrate at multiple points. At this point a calibration curve is at least needed. And the hardware might not be up to the task by the looks of it. It seems to have a treshold of ~0.4A, no temperature compensation and non-linear drift. I don't think JK is capable to fix this in firmware and who knows if they're willing to fix the hardware. Time will tell.

The one point that was interesting was the comments on how the JBD does it, as it's closer to the Victron shunt. Anyone knows?

And in one comment there was mentioned a shunt from Riedon that peaked my interest. If JK can't get this right themselves they should open up for reading an external shunt.

Riedon analog shunt PDF
Riedon digital shunt RS485 PDF
Riedon digital shunt CAN PDF

Riedon digital shunt CAN programming (youtube)
 
kommando said:
I don't get these glitches and have used several Firmware's and currently on 15.24, also my Canbus cable is only 1M long to the master SI. Sounds very much like interference on the comms, either cable related or a particular BMS. I only use the two BMS's plus a ghost, maybe once you go above a certain number of batteries the master BMS has periods its CPU is overloaded and this results in error being sent sent via canbus.
Click to expand...

OK, so try reducing battery from 16s8p to 16s2p?

"[Abnormal cancellation of coprocessor communication]" might be related to that CPU overload you suggest.

(thinking of our common friend; I haven't connected mine yet. I plan to get crimper working for battery end of custom harness first.)
 
Hedges said:
"[Abnormal cancellation of coprocessor communication]" might be related to that CPU overload you suggest.
Click to expand...
Yes it could be, in which case the solution could be to use the Sleeper85 ESP32 to take all the battery info and provide the Canbus link to the SI. This would take the computation away from the master BMS which would revert to being a slave.
 
Last edited:
HRTKD said:
Yes, I meant the Cerbo. I haven't got far enough to actually wire anything into the Cerbo, so I have a bunch of cables still in bags and a Cerbo sitting there all alone.

With the BMS being called an inverter BMS, I get confused.
confused3.gif
Click to expand...
@HRTKD ,hi there,I know several people now with victron gear and I-without any victron energy products- having decided NOT to connect the BMS to any devices but the Batteries.Reason: JK inverter BMS seems to have trouble to communicate propperly,if there are more then 1 bank connected in parallel,like master/slave...And the other reason is,not connected to each other means that you have two seperate controls:the inverter who does its charging/discharging job with its own security parameters and then you have the BMS-protection which balances(the main task) to keep your batteries from individually drifting appart-which causes over and undervolt problems on certain cells- and then the protection on the negative line,which shuts down charging and or discharging in case of wrong parameters.Once you have even protected the positive lines with adequate fuses,I formyself think that's a better solution.As for any case this should be considered by everyone themselves....For me the two factor individual protection works fine:unsure::coffee:
 
You must log in or register to reply here.

diy solar

diy solar
Back
Top