diy solar

diy solar

JK 4S 200A BMS

It was in a box inside of a shipping bag. The bag was fine, but the box was crushed. This was some of the worst packing of a valuable item that I have seen. Any bump to the box would go straight to the BMS because the single layer of very thin bubble wrap would do almost nothing to cushion the contents. The packing was not even sufficient to keep the contents from rattling around inside of the box. so that it would "clunk" as the box was rotated,
 
We will improve the packaging of the goods, but such damage is obviously the responsibility of the courier. Please test it as soon as possible. If the BMS cannot be used, we will refund it for you. We will contact the logistics provider for compensation. Thank you for your support
Improving the packaging on future shipments is good for future customers but does nothing to fix my current problem.
 
It was in a box inside of a shipping bag. The bag was fine, but the box was crushed. This was some of the worst packing of a valuable item that I have seen. Any bump to the box would go straight to the BMS because the single layer of very thin bubble wrap would do almost nothing to cushion the contents. The packing was not even sufficient to keep the contents from rattling around inside of the box. so that it would "clunk" as the box was rotated,
I think that part of the reason is that couriers, all products are transported by air, which is not treated roughly like other modes of transport. The problem in your photo is that if couriers treat goods so roughly, even using multi-layer packaging can't be avoided. Thank you for your advice anyway. Even if this is the first time we've met this situation. We will also improve the packaging.
 
I am afraid that it is DOA. I have it wired up and when I push the button, it beeps and the red light starts flashing. I now have the current app v4.7.3 which opens just fine but that is it. The device shows up on my Bluetooth page but the scan button on the app does not do anything and no devices show up on the list.
 
Can anyone tell me how you connect to the JK with the app.
  • The BMS is turned on and the red LED is flashing.
  • I have v4.7.3 loaded.
  • On my phones bluetooth page it shows "JK_B2A8S20P".
  • If I click on that device, I get a message "An app is needed to use this device" then says "pairing" and the red LED on the BMS quits flashing for a bit.
  • In the JK BMS app, there is a big green button that "scan" but it does not appear to do anything and there are no devices shown on the Device List.
What am I doing wrong and what can I try. Or is the BMS just broken?
 
Can anyone help with this.
I sent this to info@jkbms.com, no reply.
I then posted on Hankzor BMS store and was linked to the git hub documentation.
I made the mistake of changing the settings password from 123456.
As a result I can't even turn on the BMS in the app.

"I am trying to set up this BMS.
I set the cell count to 16 from 20 and changed the password.
I can not now go further than the password verify screen.
It does not accept the password i set, or any other password.
Is there a password reset procedure or a factory default reset I can use?"

Many thanks
 
In the JK BMS app, there is a big green button that "scan" but it does not appear to do anything and there are no devices shown on the Device List.

Make sure you have location permissions etc. allowed for the app. Bluetooth needs that. Don't pair or anything with your phone.
 
My original intent was to mount the BMS with the sense wire plugs on the bottom, but the sense wires are not long enough to reach the far end of the 2p4s string of batteries. I am considering putting the plug-in terminals on the top instead of the bottom but then it creates a problem of how to route the two #2-AWG wires from the battery - to the B- terminals in the other side of the end cap.

Any suggestions on the best route?

One thing that I was wondering about is can the BMS be mounted with the mounting ears on the outside and the face against the plywood. Can the leads be connected to either the front or back of the terminals?
 

Attachments

  • B- wiring.png
    B- wiring.png
    539.3 KB · Views: 51
My original intent was to mount the BMS with the sense wire plugs on the bottom, but the sense wires are not long enough to reach the far end of the 2p4s string of batteries. I am considering putting the plug-in terminals on the top instead of the bottom but then it creates a problem of how to route the two #2-AWG wires from the battery - to the B- terminals in the other side of the end cap.

Any suggestions on the best route?

One thing that I was wondering about is can the BMS be mounted with the mounting ears on the outside and the face against the plywood. Can the leads be connected to either the front or back of the terminals?
The back sides of the terminals have solder "bumps", and the 4 mounting plates are on only one side. (In my case, it is the side opposite the paper label.) Your photo is out-of-focus in this area, but seems to be upside down (showing the threaded screw sleeve prominently, rather than the flat mounting plates). My compressed box 'end-plate' is a lot wider than yours: I've also got the BMS sitting adjacent to the end plate but it's mounted horizontally.

In my case, both the "B-" ports and the balancing wire port are near the top of the pack, allowing for shorter "reach" lengths among the balancing wires. I built new shorter "B-" cables from AWG-2 welding cable, (and actually should have built them shorter than I did.). The P-" ports along the bottom edge of the BMS, about 1-3/4" above the bottom of the pack. These "P-" cables are not yet upgraded, and they connect into a "coulomb counter" monitor.
- - -
(In a couple of weeks following my "main testing" and report, I observe that the BMS reported "capacity left" dwindles over time -even though the pack is not being discharged, and the voltage is still quite high. (The monitor and App show "3% Remain Battery", paired with just "6.9Ah Remaining Capacity" - but the pack Voltage is 13.29V, 3.324V per cell balanced perfectly and not being discharged. That should be around 80% SOC (room temperature), 80% remaining battery and 180+ Ah Remaining Capacity. I'm waiting to see if it does something crazy at "0% Remaining Capacity" - hoping that it's just a bad report, calculated wrong.
- - -
Anyway, the battery pack - and everything else on the testing bench - moves into the Travel Trailer in a couple of hours. Here's a photo of BMS sitting on the side of the pack, it's only held in place by the "B-" cables.
 

Attachments

  • BMS-in-position-not yet-taped.jpg
    BMS-in-position-not yet-taped.jpg
    299.1 KB · Views: 36
The back sides of the terminals have solder "bumps", and the 4 mounting plates are on only one side. (In my case, it is the side opposite the paper label.) Your photo is out-of-focus in this area, but seems to be upside down (showing the threaded screw sleeve prominently, rather than the flat mounting plates). My compressed box 'end-plate' is a lot wider than yours: I've also got the BMS sitting adjacent to the end plate but it's mounted horizontally.
...
Thanks. the "humped side is on the label side so that same as shown in your photo. Am I correct that the connectors must be attached to the hump side? I am thinking that many of my challenges might be solved by rotating the BMS 90º like you have done or maybe -90º so that the plugs are on the left.
I am still working on what I want to use for my settings. This is what I plan to start with.
 

Attachments

  • BMS terminals.jpg
    BMS terminals.jpg
    110.6 KB · Views: 30
  • hwse settings.png
    hwse settings.png
    176.4 KB · Views: 32
Thanks for the sharper photo! You've got the wire attached on the correct side. Your settings look pretty good to me, except for the Discharge Overcurrent Protection Delay: You don't want that to be any longer than needed to account for reactive loads and motor starts creating temporary high current within an Inverter. I've got mine set down to only 2 seconds, although 3 seconds might be a better value for that. After 3 seconds, you've probably got some kind of "load" which you didn't intend to support at all.
 
The only spike load that I might have is the inrush from a 500W anchor windlass. I have always been in the habit of running the motor while pulling the anchor Running the engine while using the windlass served two purposes. 1. to support the battery for the relatively high load from the alternator. 2. to have a means to direct the boat (even when leaving under sail). I will continue that even though I have a much bigger bank of batteries. The engine running with the alternator should mitigate the inrush concern or at least keep it below 350A. ;)
 
I am somewhat pleased to report that the erroneous status values of "0% remaining battery" and "0% remaining battery Ah" do not cause the BMS to shutdown. (My '230Ah' battery pack was in fact at about 65%-75 SOC when these figures first dwindled down nearly zero, over a period of about 11 days not-in-use.) My placement of the couiomb counter does not count energy consumed by the BMS itself, but @ up to 3w per hour that should not have been more than 800 watt-hours. The "12v" battery pack has a capacity of about 3000 watt-hours, and the end voltage while otherwise disconnected (and before reconnecting to a charger) lines up with the 65-75% SOC estimate.

I am of course less pleased by the increasingly incorrect SOC reported in the cellphone App "status". Maybe, at some time in the future, I will move the coulomb counter to INCLUDE power consumption by the BMS itself. (More wires to cut, with more lugs to add in ...).
 
I am somewhat pleased to report that the erroneous status values of "0% remaining battery" and "0% remaining battery Ah" do not cause the BMS to shutdown. (My '230Ah' battery pack was in fact at about 65%-75 SOC when these figures first dwindled down nearly zero, over a period of about 11 days not-in-use.) My placement of the couiomb counter does not count energy consumed by the BMS itself, but @ up to 3w per hour that should not have been more than 800 watt-hours. The "12v" battery pack has a capacity of about 3000 watt-hours, and the end voltage while otherwise disconnected (and before reconnecting to a charger) lines up with the 65-75% SOC estimate.

I am of course less pleased by the increasingly incorrect SOC reported in the cellphone App "status". Maybe, at some time in the future, I will move the coulomb counter to INCLUDE power consumption by the BMS itself. (More wires to cut, with more lugs to add in ...).
This small current draw will not be picked up by shunt. And by looking at voltage that is rough estimate. Your SOC is guessing. Try to find better shunt that will keep memory of SOC even if not used for many weeks. But extend periods of not charging or discharging will not substitute lifepo4 chemistry losses.
 
My BMS showed up today.
Unfortunately, the cover plate / heat sink was pretty mangled in shipping. Two of the mounting ears were bent with one almost 180º. Both the top and bottom plate are also bent outward between the case screws which seems to indicate that the FETs have probably been crushed somewhat.

Compared to the packaging that we get with the cells this was very weak. It was in a thin bubble wrap bag inside of a single layer cardboard box with the loose bits packed down around it. I have received $20 electronics that were packaged in solid boxes with form fitting foam over an inch thick inside. I expected better packaging than this.
damn it man... they can't spend the $0.2 on a piece of foam?
 
damn it man... they can't spend the $0.2 on a piece of foam?
That is what I thought. The same day I got three other pieces of electronics from AliExpress. They ranged in price from $1 - $6 and all better packaging. (Antistatic bag and double layers of bubble wrap.)
 
This small current draw will not be picked up by shunt. And by looking at voltage that is rough estimate. Your SOC is guessing. Try to find better shunt that will keep memory of SOC even if not used for many weeks. But extend periods of not charging or discharging will not substitute lifepo4 chemistry losses.
I'll SWAG idle JK power consumption to lie somewhere between 0.3 watts and 2.0 watts in idle, because the Bluetooth radio is active at all times. On my "12v" battery pack, current supporting those power figures would range between .02 and .16 Amps. It is a bad assumption, on your part, to assume that my current shunt/coulomb counter modules are unable to track such small changes in VA over a period of days or weeks. With effort, I could measure the JK consumption with a pair of "before" and "after" versus after shunt/coulomb counter modules, but it's 99% irrelevant to my original complaint.

What you accuse to be a mere "guess" at SOC was confirmed upon recharging the battery pack, with the coulomb counter showing a bit under 80A taken in before "full" charging voltage was reached (on all cells in good balance, due to JK's high balancing current capability).

Although a lead-acid battery stored at STC temperature will lose capacity quickly, LFP batteries discharge at a much lower rate. The self-discharge rate of a particular cell varies according to the original storage SOC and temperature, along with the characteristics of the particular cell in question. But unless stored HOT, a properly constructed LFP cell charged to at least 50% SOC will last for months before going low from self-discharge.
- - -
Over a period of less than 2 weeks, your misstated "issues", even if present to a tiny degree, would be irrelevant to my original complaint: The JK App shows quickly rapidly dwindling capacity numbers, clearly becoming significantly "wrong" in just a couple of days, and it degraded all the way to "nothing left" when the battery pack was actively used and then left at more than 50% full.
 
I'll SWAG idle JK power consumption to lie somewhere between 0.3 watts and 2.0 watts in idle, because the Bluetooth radio is active at all times. On my "12v" battery pack, current supporting those power figures would range between .02 and .16 Amps. It is a bad assumption, on your part, to assume that my current shunt/coulomb counter modules are unable to track such small changes in VA over a period of days or weeks. With effort, I could measure the JK consumption with a pair of "before" and "after" versus after shunt/coulomb counter modules, but it's 99% irrelevant to my original complaint.

What you accuse to be a mere "guess" at SOC was confirmed upon recharging the battery pack, with the coulomb counter showing a bit under 80A taken in before "full" charging voltage was reached (on all cells in good balance, due to JK's high balancing current capability).

Although a lead-acid battery stored at STC temperature will lose capacity quickly, LFP batteries discharge at a much lower rate. The self-discharge rate of a particular cell varies according to the original storage SOC and temperature, along with the characteristics of the particular cell in question. But unless stored HOT, a properly constructed LFP cell charged to at least 50% SOC will last for months before going low from self-discharge.
- - -
Over a period of less than 2 weeks, your misstated "issues", even if present to a tiny degree, would be irrelevant to my original complaint: The JK App shows quickly rapidly dwindling capacity numbers, clearly becoming significantly "wrong" in just a couple of days, and it degraded all the way to "nothing left" when the battery pack was actively used and then left at more than 50% full.
I will be more than happy if you can provide shunt that has ability to measure 200+ A and it is capable to measure very small current draw. You have probably misunderstood my post. I was not referring to the JK bms issue but your overall expectations that you should expect from JK BMS . And especially if you looking on the shunt that can have tolerance to less than 1% +/- SOC. None of Chinese BMS have sophisticated algorithms specifically designed for one type of chemistry that take in account all variables to present true SOC. Cells drifting when not used and unknown batch of cells in pack are biggest unknown. So my apologies if I did make you feel bad.
 
Back
Top