diy solar

diy solar

JK BMS owners, have you had this error yet? Heat pump electrical feedback???

brandonboosted

New Member
Joined
Jul 30, 2021
Messages
155
I noticed this morning my JK BMS is throwing a "Protection current sensor anomaly". Unsure what this meant or how to remedy I contacted the seller on Aliexpress and I am being told this is a known issue and through some translation my understanding is that the BMS is picking up feedback from the inverter and it cannot be fixed. From what I can tell it is still charging and discharging fine but the internal shunt cant read how much current is coming in and out sporadically, throwing off the battery percent which isn't a huge issue for me as I use a shunt but to some users it may be. Just sorta a heads up for new and existing JK BMS owners since I cant find any info on this and its a "known issue".

Update- Appears to happen only at lower amperage charge and discharges 0-18A, after messing around some more with some of my circuits I'm thinking it may be a feedback from my heat pump. Has anyone experienced this?
 
Last edited:
I noticed this morning my JK BMS is throwing a "Protection current sensor anomaly". Unsure what this meant or how to remedy I contacted the seller on Aliexpress and I am being told this is a known issue and through some translation my understanding is that the BMS is picking up feedback from the inverter and it cannot be fixed. From what I can tell it is still charging and discharging fine but the internal shunt cant read how much current is coming in and out sporadically, throwing off the battery percent which isn't a huge issue for me as I use a shunt but to some users it may be. Just sorta a heads up for new and existing JK BMS owners since I cant find any info on this and its a "known issue".

Update- Appears to happen only at lower amperage charge and discharges 0-18A, after messing around some more with some of my circuits I'm thinking it may be a feedback from my heat pump. Has anyone experienced this?

What type of inverter do you use?
There are inverters where the battery is not galvanically isolated from the AC side.
And so the AC component can be present on the battery side.
 
What type of inverter do you use?
There are inverters where the battery is not galvanically isolated from the AC side.
And so the AC component can be present on the battery side.
Thanks for the reply, I am using a Victron Multiplus 2 48v.
 
Thanks for the reply, I am using a Victron Multiplus 2 48v.

OK that is isolated :)
I do not have any other ideas.
I use my JK BMS with low Amps too and I did not see this error.
Yes the Amp reading is jumping up and down a bit ... but no error so far.


Try the JK BMS support directly
 
When I get my second battery connected it will be interesting to see if the both act the same or if its just the one being faulty. Thanks for the link ill give that a try and maybe I can bug them about PC to JK BMS connection while I'm at it lol.
 
OK that is isolated :)
I do not have any other ideas.
I use my JK BMS with low Amps too and I did not see this error.
Yes the Amp reading is jumping up and down a bit ... but no error so far.


Try the JK BMS support directly
It has to be my heat pump based on messing with it I cant get it to do it with it off so far, granted its 11am here so its pushing 70A at the moment. Is there anyway to test for this "feedback"?
 
When I get my second battery connected it will be interesting to see if the both act the same or if its just the one being faulty. Thanks for the link ill give that a try and maybe I can bug them about PC to JK BMS connection while I'm at it lol.

You can connect with the RS485 they give. Or you can make your own.

You can find a lot of info in this topic



It has to be my heat pump based on messing with it I cant get it to do it with it off so far, granted its 11am here so its pushing 70A at the moment. Is there anyway to test for this "feedback"?
With a multimeter you can check the AC volt on the battery terminals.
With a clamp meter you can check the DC Amps, also AC Amps and frequency.
But the best is an oscilloscope :D
 
Other than the anomaly, are you sure it's not reading the SoC correctly? It's common for BMS amperage to fluctuate wildly with an inverter, but it doesn't affect SoC accuracy. The BMS just isn't smoothing out the amperage readings sufficiently for it to LOOK constant. Even the Victron shunt will fluctuate quite a bit...
 
Other than the anomaly, are you sure it's not reading the SoC correctly? It's common for BMS amperage to fluctuate wildly with an inverter, but it doesn't affect SoC accuracy. The BMS just isn't smoothing out the amperage readings sufficiently for it to LOOK constant. Even the Victron shunt will fluctuate quite a bit...
I am positive, once the code pops up it reads 0 amps in or out and at 70-80a there is no movement in SoC.
 
You can connect with the RS485 they give. Or you can make your own.

You can find a lot of info in this topic




With a multimeter you can check the AC volt on the battery terminals.
With a clamp meter you can check the DC Amps, also AC Amps and frequency.
But the best is an oscilloscope :D
I will try this, thank you.
 
I am positive, once the code pops up it reads 0 amps in or out and at 70-80a there is no movement in SoC.

Oh, the warning actually shuts down the measurement. I wonder if that's how it detects short-circuit conditions and shuts off the BMS... so it shuts down the "protection circuit" feature altogether because it believes it can't reliably detect the current. Are there settings related to SoC (e.g. shut down BMS at X% SoC) or maximum current or things like that? I wonder if adjusting them or setting to 0 would help and/or disable the fault.

There are some simple smoothing circuits (resistor + capacitor) that people have mentioned on these forums before, mostly with respect to the Chargery BMS. That might be too advanced to implement on the BMS itself (read: soldering). I guess the external shunt setup you have is a good alternative. Does the shunt vary more erratically when using the heat pump?
 
Update- Appears to happen only at lower amperage charge and discharges 0-18A, after messing around some more with some of my circuits I'm thinking it may be a feedback from my heat pump. Has anyone experienced this?
Seems everyone is looking only at the BMS.
1) The Victron is a Quality Product and when properly installed will not cause issues. A No-Name "who knows what brand" Inverter would be far more suspicious.
2) The BMS works generally as it should with the exception of this issue and I am sure the settings have all been double-checked at this point.

--- Now the Other side of the equation. ---
?) Is the house a New Build or not and is the Heat Pump on an originally installed circuit or new circuit ?
?) Has this Circuit been checked with a "Proper Tester" ? A Ground Loop or returning feedback can mess up many things.
?) Have you double-checked the contacts/connections at the heat pump to ensure everything is tight & secure ?

ODD Question but may be telling. Do you have any GFI or CGFI Plugs or Breakers installed in the house and are they all working as they are supposed to ? IF they do not trip & pass the tester as they should, that IS a ground issue flag to chase. You would be surprised how much trouble that can cause which seems unrelated !

--- The 3rd Side. ---
DC Grounding. Panels & Racks should be grounded for protection BUT not to the electrical ground point as that is for lightning etc.
SCC's, Inverter/Chargers also have ground points and most have TWO MODES, one for Fixed Terrestrial Installation or for Mobile Installation. Virtually everyone will have a "Jumper Wire" that needs to be connected for proper operation. Failure to do so will usually not stop the equipment from operating but errors will happen and the device safety "may" be compromised, PLEASE CHECK THIS OUT AGAIN !
Note that I have seen PRO Installs & Hobbyist installs where this was not done and simply put "what a Pain to figure out" of chasing clues and should have been #1 to check. A TIP, If CGFI Breakers/Plugs trip & fail this is an excellent indicator of this problem.

Hope it helps, Good Luck.
 
I have two of the 200A JK BMSes in operation and both also produce this same "Protection (Current Sensor Anomaly)" blinking red message with 0.0A/0.0W charging/loading reported under certain conditions.

I'm 100% convinced that this issue is unrelated to DC grounding and isolation, and is a benign message that you should simply ignore. It does not indicate any sort of equipment fault/defect/ HVAC feedback and is simply a limitation of the BMS's current sensing ability.

The message is only displayed when two conditions are simultaneously present:
- The sun is out and you have power generation
- You have a heavy AC load that roughly matches the immediate DC power generation.

What happens in this case is that the AC load draws more current whenever the AC sinusoid is near its voltage peak, then drops to zero current around zero crossings of the voltage waveform. Correspondingly, your inverter briefly demands 100's of amps from your low voltage DC domain, then negligibly little and repeating at 100 or 120Hz (50/60Hz line frequency, two peaks and zero crossing per full sinusoidal period). Since the sun is out, the constant DC inflow - lumpy DC loading sums to a current profile that is constantly swapping between charging your batteries and discharging your batteries. The BMS samples the DC current asynchronously around 1 Hz (or at least this is the approximate GUI update rate), so gets confused when it sees the instantaneously sampled ADC readings unpredictably reporting + amps charging and - amps discharging in back to back samples.

The BMS software would be superior if they sampled the current well over 240 times/second and averaged the result for a "True RMS" lumpy DC current it could display and credit/debit SoC accurately with, but it instead appears to just assume zero current and signal an alarming but meaningless warning message.


P.S. rapidly oscillatory DC charging and discharging, within rated or lower C rates is 98 to 100% harmless to your batteries and does not meaningfully count against lifetime cycle ratings. This is because batteries (well, lithium ternary and LiFePO4 anyway) have substantial electrostatic capacitance that can take this load, and even if the capacitance is inadequate, the diffusion time of lithium ions is way too slow to cause compressive/expansive stresses to happen in the anode and cathode matrixes that cause cycle degredation. The only reason why any harm can come at all from these oscillations is due to there still being an AC current passed through your batteries, which resistivity creates some minor self heating.
 
I have two of the 200A JK BMSes in operation and both also produce this same "Protection (Current Sensor Anomaly)" blinking red message with 0.0A/0.0W charging/loading reported under certain conditions.

I'm 100% convinced that this issue is unrelated to DC grounding and isolation, and is a benign message that you should simply ignore. It does not indicate any sort of equipment fault/defect/ HVAC feedback and is simply a limitation of the BMS's current sensing ability.

The message is only displayed when two conditions are simultaneously present:
- The sun is out and you have power generation
- You have a heavy AC load that roughly matches the immediate DC power generation.

What happens in this case is that the AC load draws more current whenever the AC sinusoid is near its voltage peak, then drops to zero current around zero crossings of the voltage waveform. Correspondingly, your inverter briefly demands 100's of amps from your low voltage DC domain, then negligibly little and repeating at 100 or 120Hz (50/60Hz line frequency, two peaks and zero crossing per full sinusoidal period). Since the sun is out, the constant DC inflow - lumpy DC loading sums to a current profile that is constantly swapping between charging your batteries and discharging your batteries. The BMS samples the DC current asynchronously around 1 Hz (or at least this is the approximate GUI update rate), so gets confused when it sees the instantaneously sampled ADC readings unpredictably reporting + amps charging and - amps discharging in back to back samples.

The BMS software would be superior if they sampled the current well over 240 times/second and averaged the result for a "True RMS" lumpy DC current it could display and credit/debit SoC accurately with, but it instead appears to just assume zero current and signal an alarming but meaningless warning message.


P.S. rapidly oscillatory DC charging and discharging, within rated or lower C rates is 98 to 100% harmless to your batteries and does not meaningfully count against lifetime cycle ratings. This is because batteries (well, lithium ternary and LiFePO4 anyway) have substantial electrostatic capacitance that can take this load, and even if the capacitance is inadequate, the diffusion time of lithium ions is way too slow to cause compressive/expansive stresses to happen in the anode and cathode matrixes that cause cycle degredation. The only reason why any harm can come at all from these oscillations is due to there still being an AC current passed through your batteries, which resistivity creates some minor self heating.
My bms is doing this while charging my car when the sun is putting heavy current in. It will not give this error if I'm later in the day with less going g in. Today I waited till 7:30pm and it's fine.
I also found out I can't disconnect the charge at all like this. So I pull 1k I'm putting in almost 1k while charging for a couple of hours.
I think this will get better when I get 2 more and split my 3 packs to separate and all the charging and discharge numbers get split into 3. We shall see.
 

Attachments

  • Screenshot_20220509-174614.jpg
    Screenshot_20220509-174614.jpg
    240.5 KB · Views: 38
Panels & Racks should be grounded for protection BUT not to the electrical ground point as that is for lightning etc.
The electrical grounding system is for electrical protection. And by code, is where the panels and rack should be grounded to.
Lighting protection would be an entirely different system.
 
I got told by the jk people it is a silly code and effects nothing but the readout. Read through the entire post.
 
Back
Top