• 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)

1727401178510.png
Steve_S said:
Surprisingly & shockingly Andy mentioned THIS Forum... I'm positive a few had the same reaction I did when he did that...

He made an important point in his last video in regards to his 1st Bank & the NON-Connected BMS's doing exactly what they ought to when thing are properly setup. It is also one of the Rare Moments he mentions TailCurrent and it's application for transitioning to Float.

Quite Right when he covered how the Victron uses Voltage+Absorb Timer, TailCurrent (Amps being taken by batteries) to validate 100% SOC. DO I needed to go back and QUOTE a post which suggested this for the JKBMS giggles.

IRONY is the JK Inverter BMS could do that !
1) The BMS can read the Voltage @ Terminals & AVG Cell Voltage. AVG Voltage = better for using User Set 100% SOC Voltage.
2) It can read Inflow AMPS own to 0.2A. It can determine the EndAmps/Tailcurrent by doing the fast math of (xxxAH * 0.05 = TailCurrent)
----NB 0.05 is for Pack Level TailCurrent/EndAmps
3) Does not need to be "Absorb Mode Aware" as such. It only needs to know when Pack AVG Voltage = 100% SOC Setting and wait till EndAmps/TailCurrent read TailCurrent value to send the REQUEST FLOAT VOLTAGE signal.

The Coulomb Counter (hardware) in the JK is crappy ! It cannot resolve below 0.2A which represents losses in both directions as it is not registered & counted. This will most obviously affect SOC Accuracy if the batteries do not reach 100% SOC & Reset and will accumulate over time. THAT's the GOTCHA ! If you are on the equator and hit 100% SOC daily or even every 2nd day, the SOC drift wouldn't be that bad, but after 3-6 Months of not hitting 100% Daily the drift will be abysmally inaccurate. High Precision Coulumb counting can be done but at a Cost which may add $10 to the cost of the BMS (personally, I'd be OK with that IF it was truly accurate) *not that I will replace my fleet of BMS' if they issue a new HW Version.

-----------------------------------------
#SOC-100_User_Val = 3.450 (user set in BMS)
#SOC-0_User_Val = 2.800 (user set in BMS)
#Pack_AH_Val = 280AH (internally calculated 280x0.05=14A)
--
I won't bother with the pseudo code which is simple enough.
Click to expand...
The map value, OCV and value of different brands of cells are different,
We are writing BMS charging parameters according to different brands of cells,
To charge the cell to the maximum SOC value,
At the same time, multi-stage charging is added.
Sorry, I have been very busy recently,
There are many parameters to be debugged and verified. I will reply to your message sometime.
Thank you for understanding
 

Attachments

  • 1727401391145.png
    1727401391145.png
    57.8 KB · Views: 30
I have read some scientifics papers about SOC calculations base on physical model (or deep learning) and various battery parameters (Voc, current, temperature) in addition to Coulomb counting which in any way can't be accurate enough for all situations.
To name a few:
www.sciencedirect.com

State-of-charge estimation for onboard LiFePO4 batteries with adaptive state update in specific open-circuit-voltage ranges

Accurate estimation of the state-of-charge (SOC) is crucial for efficient and safe battery applications. However, existing SOC estimation methods fail…
www.sciencedirect.com www.sciencedirect.com
 
Nami said:
The map value, OCV and value of different brands of cells are different,
We are writing BMS charging parameters according to different brands of cells, WRONG ! Pointless waste of time !
To charge the cell to the maximum SOC value,
At the same time, multi-stage charging is added. LFP & LiIon General is TWO STAGE Charging ! No Fast Charging etc with these !
Sorry, I have been very busy recently,
There are many parameters to be debugged and verified. I will reply to your message sometime.
Thank you for understanding
Click to expand...

Sorry but you are chasing the impossible with this "Brand Specific Cell Differences" !
This is an Unneeded Complication and a waste of time ! unless you want to fail or using this as an excuse to fail ?

You need to focus on the "General LFP Chemistry" as the BMS supports 3 Different Chemistries, each has to be managed according to the "Chemistry Standards" and not "Brand Specific Variations".

100% SOC = When cells reach user SET 100% Voltage and the Amperage decreases to TailCurrent/EndAmps. It is JUST THAT SIMPLE ! and that applies to ALL Chemistries, even Lead Acid ! COULOMB Counting down to 0.01A is required for accuracy but also MUST to account for losses in both directions. The current Coulomb Counter registers only as low as 0.2A and that is too inaccurate.

Multi-Stage Charging ?
For WHAT ? LFP & Li-Ion in General is TWO STAGE CHARGING !
We are not using LMFP (that would be another BMS profile type), we also are not using LFP that is 3C/5C or 6C Capable which take Fast Charging (which are ONLY available for EV's)

Solar Controllers & Charger Systems handle the charging and are responsible for it, NOT the BMS. BMS should only Send a Signal when the Battery Pack is ready to transition to FLOAT mode (Constant Voltage / Variable Current)

If you are chasing this nonsense of Cell Variations having different values & Multistage charging which is truly irrelevant then you are JUSTIFYING ONGOING FAILURES and rationilizing it with these excuses with a basis of Fiction.

The REALITY IS:
- Every Cell from every production batch is different. As will every Brand & Variation !
- Every Cell with different AH ratings 100AH vs 200AH vs 300AH will also be DIFFERENT for IR, TailCurrent/EndAmps and so on.
- Every Brand & Model has uniqueness but they have the Chemistry Commonalities and THAT is what needs to be targeted.

---------------------------------------
Lame excuses coming in from Nami & Andy too (with a different twist). NONE of this bodes well for anyone. Shockingly Unbelievable and terribly worrisome.

This is quite scary if not shocking and how some get a specific mindset locked and will not debiate even when they are out of the ballpark.
The spewage of excuses starting early this summer is over the top.
Currently only a few "reasonably priced" BMS' that have the majority of these features/functions & capabilities are available. Each have their own Quirks, Issues or Foibles (Passive balancing only as 1 example) or communications compatibility but they are also working on it.

SEPLOS BMS: https://www.seplos.com/seplos-bms-3.0.html
PACE BMS: http://www.pacebms.com/en/index.php/lists/33.html
TDT BMS: https://www.tdtbms.com/products-category/smart-bms.html

Others ???
Add Link to OEM's making SmartBMS with Communications/Interactions with AIO's & Inverters, if you have them.
 
Last edited:
Nami said:
Дякуємо за розуміння
Click to expand...
Нагадаю про необхідність врахування коефіцієнта втрат при розрахунку енергії заряду і розряду.
Моя батарея ємністю 5 кВт/год, наприклад, видає реальні 5 кВт/год, але при зарядці споживає більше 6 кВт/год.
1727439782263.png
 
Quite Correct, the Coulomb Counters "require accuracy" to account for the losses during charge & discharge, which also changes relative to the Amperage flowing in/out.

* Some SmartShunts have a setting provision for Accuracy which allows a User to enter a value for corrections. For example an FLA Battery is about 88% Efficient while LFP is 98% efficient. The "efficiency" percentage is then used for calculating the losses in/out improving the SOC status readings.
 
Setting different parameters for each slightly different manufactures batteries is trying to be perfect. I don't think you need to be perfect, you just need to be good enough, and able to handle the average battery of each type and have code to reset at any voltages that are a give away that the battery is at 100% or say 10% were the voltages can be easily used to determine SOC. And don't report 100% unless the voltages clearly indicate 100% (as I believe the newer version is already supposed to do).

My JK inverter bms (on old firmware) is now appearing to be working mostly correctly because SOC was off and the battery hit a definite close to 0% voltage and the entire range reset. Those resets need to be a bit higher above 0%, and given 20% is claimed to be about 51.2 and 10% is claimed to be 48V then there is chance to reset in between those 2 since the voltages change rapidly. Mine reset with SOC of 31% and voltage 42.5, and well before it got to that point SOC could have been realized to be wrong and adjusted.
 
rhino said:
Guess they are giving up on updating the v14 hardware?
Click to expand...
Unpacking NAMI's statement:
1) The Hardware shortcomings were pointed out by Nami & Andy. No specific mention on V14 so safe to assume it's abandoned in favour of V15 which happened quick, so....

2) Nami's statement about the "Unobtainable Goals" of addressing Manufacturer Variations is the Perfect "Escape" from having to fix & finish the current boards because it is a virtual impossibility. Lamest Excuse for an escape from responsibility.

3) That "Multistage Charging" BULLONEY is yet another terrible excuse. BMS does NOT do any charging, it's a Guard Dog ! The Solar Controllers & Inverter/Chargers DO the charging, the BMS only has to signal when TailCurrent is reached at SET 100% Voltage to issue a Transit to Float signal to SCC/Inverter/AIO Systems.

THE ONLY TWO PLACES Reduced Charging is required:
When battery pack(s) are Between 0C/32F - 5.0C/41F temp where it cannot exceed 0.2C-Rate.
When battery packs are above 40C/104F where again no more than 0.2C-Rate should be applied.
{default industry norms for all Standard LFP}

-----------------=====================================================---------------
We have been LIED to Well over THREE TIMES !
We have been given a Myriad of Excuses & Justifications, majority being Hot Air !
Contradictions & Conflicting statements by JK and their "agents" to support that nonsense.
CREDIBILITY is at an ALL TIME LOW !

---> REMEMBER that Earlier on in the Thread where JK Refused to even acknowledge ANY Issues being complained about UNTIL a Vendor/Supplier sent then SCREENSHOTS of actual Emails sent to THEIR NAMES DIRECTLY outlining a long list of the issues to only then have it acknowledged. Helen used the infamous line of "I'm new here & did not know". (Psst her EMail Address has been attached to JK for over a year) so MORE LIES !

NO TRUST - FAITH or BELIEF in JiKong. Their words mean NOTHING, their actions are even LESS !
We GOT SHAFTED and more will be shafted - even with whatever new Hardware Version they come up with !
Will ANY of us buy another JiKong BMS ?
Would you replace your existing one with another version by them and LIKELY face same Bulloney ? Ready to toss a couple of hundred bux per unit X # of units ? on a "hope they do it right finally" ?

I have 8 BMS in my fleet ATM, they can ship me 8 NEW V16-Fantasy Fixed BMS' for free and I "may" consider testing ONE but they won't and neither will I.
--------------------------------------------------
I miss the days when we were working in harmony with Jason Wang @ Chargery BMS who listened AND HEARD us, developed & improved based on our collective engagements and cooperative testing. Chargery was getting serious support from the membership here, and returned it in like. Well until ONE person got pissy and kiboshed those interactions.

PS @Nami
There are over 22,000 views on this thread. A Large portion is from OUTSIDE of this forum, meaning that people globally are seeing & reading this. It is also being echoed in FaceBook, other Forums & Sites. This has an immediate EFFECT on JiKong and it is up to you how that will work out. The damage to reputation is increasing faster & faster... ask yourself, how much business are you prepared to lose ?
 
Last edited:
We have a phrase in the English language. “Smoke and mirrors”.

Basically A lot of talk with no substance from JK. October is right around the corner and we have no assurances that a fix is coming.
 
Last edited:
Damage already done... Our solar discord channel stopped recommanding jkbms for diy lfp battery. A new hardware wont help to raise reputation... you cant sell product without any support for elementary feature like soc, can communication etc... too bad i got trapped on this product :-/
 
JJJJ said:
A lot of talk. October is right around the corner and we have no assurances that a fix is coming.
Click to expand...
And the Big October Sales Bonanza is coming, so promises will made to flog the CRAP they have.
They slipped V15 in over V14 without a word, so will V16 Hardware just magically appear when the V15 junk is gone ?
- Sucker tattoos are now on Sale to ! Free if put on your forehead.
 
Steve do you still recommend the non inverter jk or are you totally soured against jk now?
 
Steve_S said:
My suppliers in China are saying the same thing, increasing amounts of people are asking for something ELSE and are saying outright they do not want any JK BMS. It was only a few 2 months ago but now it is becoming the norm.
Click to expand...
I guess they might pull another Daly stunt? Daly rebranded itself into 100 Balance BMS, still same pointless hardware.
:ROFLMAO:
 
SparkyJJO said:
Steve do you still recommend the non inverter jk or are you totally soured against jk now?
Click to expand...
The Non-Inverter BMS' which were running my fleet were OK and did what they were supposed to and pretty well. The Active Balancer in those is not as good IMO but still fine. Given that JK has been cheating out on parts and quality is variable, NO - I CANNOT & WILL NOT RECOMMEND JK AGAIN ! Their Karma is arriving and boy it's not pretty.
 
AshleyL said:
I guess they might pull another Daly stunt? Daly rebranded itself into 100 Balance BMS, still same pointless hardware.
:ROFLMAO:
Click to expand...
ROFLMAO yeah, good example ! The Glossy sure looked nice at first, then people got them and realized they got suckered.
Renogy has a huge marketing & sales glossy look but absolute junk for product and they keep selling that shyte. If they spent 1/2 their advertising budget on product refinement they may actually have reasonable stuff but Noooooo !
 
AshleyL said:
So.....I assume they gonna abuse DMCA to remove any "offending" links from search engine soon? :unsure:
Click to expand...
Google & DuckDuckGo are not likely going to accommodate that.
Would not surprise me if they try it but half of our stuff is Banned in China and they can't access it.
 
You must log in or register to reply here.

diy solar

diy solar
Back
Top