LG A9 CordZero Vacuum Batteries

• 

  [Deactivated] on 02/02/2023 - 14:10

  How often did you swap batteries? Did you keep the spare one in the charger?

• 

  impoze on 02/02/2023 - 14:29

  didn't swap too often as it was usually only used a few minutes at a time.

  I also found this old comment from @ifreddo

  ifreddo on 24/06/2020 - 22:01
  I own one of these and they're great. The ease of use over a corded vacuum more than makes up for the small dust compartment.

  Just a word of warning: I've had 2 batteries go dead on me with this model. Both replaced under warranty. I've realised that keeping the second battery in the 2nd charge slot while only using the first battery may have led to it dying. So I now keep the 2nd battery in a storage cupboard and swap them over every few months to ensure both batteries get a workout.

• 

  mshanann on 02/02/2023 - 15:21

  I think you mean 25.55 volt rather than 2.5V

  If the battery is 2000mAH - I'd expect 7 x 18650 cells in 7s1p configuration

  If it is 3000mAH or higher, then you could have 14 x 18650 cells in 7s2p configuration

  Just because the voltage is reading at ~25 volt doesn't mean the cells are not shot.

  You could try and charge the batter to capacity, then do some type of independent load test on it, using something like https://www.aliexpress.com/item/3256804909652648.html

  If the capacity is really low, then you could disassemble the battery pack and try and determine if individual cell(s) are bad or too discharged. They are all likely spot welded in though - and the battery housing is typically very hard to get in to. You could check facebook marketplace for anyone advertising reclaimed 18650 cells or reconditioned batteries and tool repair that may have the equipment to test the battery for you. Cells themselves are cheap at about $3.

  • 

    impoze on 02/02/2023 - 17:20

    It was def 2.5v, not 25v
    Used two different multimeters as well

    • 

      mshanann on 03/02/2023 - 11:20

      Possibly packs were drained - normally though any half decent BMS will have over discharge protection.

      Given you have nothing to lose, and assuming you cannot get them to repair/replace for free, I'd split one open and then try and individually charge the cells inside.

    • 

      BrettBullock on 28/05/2023 - 09:29

      Mine as well 2.5v at the terminals however cracking the casing open and measuring the actual ends of the batteries it reads 26.5v .
      There's plenty of juice in the batteries but somewhere in the circuitry it is not allowing it through.
      I could understand dying batteries from constantly discharging them too fast then leaving them flat but this is something different ,
      This is a BMS issue.
      All these people buying new batteries from the manufacturer and their batteries are possibly fine just a circuit board issue. Anyways I'm going to change out all the cells anyways and I will get back to you all

      • 

        impoze on 28/05/2023 - 11:07

        Yeah I tried that but couldnt get it to run, theres a korean video on youtube that did it successfully though

        I've ended up buying a dyson

      • 

        Embaloo on 16/06/2023 - 14:19

        Any progress with this, Brett? Frustrating!

        As with many others, I've got a sigle light coming on when charging, which turns off after maybe 3-5 mins, never charges properly. Both supplied batteries are the same.

      • 

        edrift on 02/04/2024 - 16:32

        Did you end up getting it working?

• 

  Embaloo on 25/04/2023 - 13:24

  I am trying to troubleshoot this for my neighbour too
  Their batteries have 2018 on the label but the unit has been close to unused for ages, and very little use initially. Pretty frustrating.

  • 

    impoze on 25/04/2023 - 15:55

    I opened up the packs and tried to refresh some cells,

    Put it back together but still couldn't get it to run

    • 

      Embaloo on 17/06/2023 - 11:31

      Nice attempt, yeah saw a youtube video for the dyson rechargables which worked for that one. Frustrating its not working for these.

• 

  gr43m3 on 07/07/2023 - 17:47

  @impose, I have the same problem, both batteries are dead, at the same time

• 

  Tomoro on 15/02/2024 - 13:49

  Hi, mshanann, I found your post on the LG cord zero battery pricing conspiracy very helpful& I watched the Korean YouTube refit- I’m keen to try. Just wondering if you can help me find which battery in particular I can purchase to get the required/ safe parameters- amps, voltage.
  Thanks

  mshanann on 02/02/2023 - 15:21
  I think you mean 25.55 volt rather than 2.5V

  If the battery is 2000mAH - I'd expect 7 x 18650 cells in 7s1p configuration

  If it is 3000mAH or higher, then you could have 14 x 18650 cells in 7s2p configuration

  Just because the voltage is reading at ~25 volt doesn't mean the cells are not shot.

  You could try and charge the batter to capacity, then do some type of independent load test on it, using something like https://www.aliexpress.com/item/3256804909652648.html

  If the capacity is really low, then you could disassemble the battery pack and try and determine if individual cell(s) are bad or too discharged. They are all likely spot welded in though - and the battery housing is typically very hard to get in to. You could check facebook marketplace for anyone advertising reclaimed 18650 cells or reconditioned batteries and tool repair that may have the equipment to test the battery for you. Cells themselves are cheap at about $3.

• 

  Tomoro on 15/02/2024 - 13:55

  I had to post comment as I’m new and cannot message you ‘mshanann’ directly

• 

  Tomoro on 18/02/2024 - 17:32

  Hi, mshanann, I found your post on the LG cord zero battery pricing conspiracy very helpful& I watched the Korean YouTube refit- I’m keen to try. Just wondering if you can help me find which battery in particular I can purchase to get the required/ safe parameters- amps, voltage.
  Thanks

  mshanann on 02/02/2023 - 15:21
  I think you mean 25.55 volt rather than 2.5V

  If the battery is 2000mAH - I'd expect 7 x 18650 cells in 7s1p configuration

  If it is 3000mAH or higher, then you could have 14 x 18650 cells in 7s2p configuration

  Just because the voltage is reading at ~25 volt doesn't mean the cells are not shot.

  You could try and charge the batter to capacity, then do some type of independent load test on it, using something like https://www.aliexpress.com/item/3256804909652648.html

  If the capacity is really low, then you could disassemble the battery pack and try and determine if individual cell(s) are bad or too discharged. They are all likely spot welded in though - and the battery housing is typically very hard to get in to. You could check facebook marketplace for anyone advertising reclaimed 18650 cells or reconditioned batteries and tool repair that may have the equipment to test the battery for you. Cells themselves are cheap at about $3.

• 

  edrift on 01/04/2024 - 20:04

  Just found out my battery also died.
  I noticed it was flashing the led in the charging stand. The voltage is measuring 2.5v and it's impossible that I've drained it dead.

  I've torn down and replaced plenty of 18650 cells in other packs before and even invested in a spot welder.

  Problem is that if the bms is dead then I'm stuck.

  This LG battery looks like a pita to open up too.

• 

  edrift on 02/04/2024 - 16:29

  Ok so I checked YouTube and found the Korean video of the guy cutting open the battery.

  I used a Dremel with cutoff wheel to cut open the pack.

  At the terminals it was measuring 2.5vdc

  At the battery positive and negative it was about 27.5vdc.

  5 cells were about 4v each
  1 cell was lower at about 3.65v

  I decided to see if I could manually force a charge into the terminals by sending about 29vdc in. No luck.

  I decided to then disconnect and reconnect the negative terminal. Bad idea. This dropped the terminal voltage down to 0.6vdc

  Then I decided to manually balance the low cell but the terminals are still at 0.6v instead of 2.5v.
  I think the bms is dead or shut down.

  If you look at the pack there are 3 tiny holes in it. I suspected they were probing points and I was right. After opening the pack there are 3 empty pads there marked SDA GND SCL. Other than ground, no idea what the other two are for.

  [Img]https://files.ozbargain.com.au/upload/119657/111577/17120352992232610112732873712918.jpg[/img]

  • 

    Stonemull on 04/07/2024 - 19:25

    +1

    My parents just gifted me one with these battery issues. They bought the same one again for some reason…

    One battery totally dead when in the vacuum, the other only lasts a few minutes. I cut open the bad one with a hacksaw and measured 5 cells at 4.14v, 1 at 3.86 and 1 at 3.95v. Charged up the 2 lower ones with a power supply set to 500mA and 4.2v till they hit about 4.1v, roughly reassembled pack and tested it in the vacuum for 5 minutes or so, then placed it back on its charger. Next morning the 5 good batterys only measured 4.04v and the 2 lower ones were back under 4v again. Will test it further yet.

    Good comment regared SDA and SCL, that is an I2C connection so I went down the rabbit hole. Scraping the conformal coating off the main chip and it is an interil chip marked 94202, so I reckon the ISL94202 is a perfect match, https://au.mouser.com/datasheet/2/698/REN_isl94202_DST_20190… a Standalone 3 to 8 Cell Li-Ion Battery Pack Monitor

    The 4 big mosfets gave me some strife till i recognised the symbol as Alpha and Omega semiconductors, so I think 4 off AON6260 at 85A each is a good match https://www.aosmd.com/res/datasheets/AON6260.pdf only scraped off the top of one but I expect all 4 are discharge cutoff making it a 340A max pack.

    managed to read the battery voltage registers, bat 4 is the unused slot. made up a little header with 3 pogo pins, had to add 4.7k pullup resistors to 3.3v and used an esp32 CYD.

    this is my cut in half "bad" battery after a reasonable length vacuum session.

    Bat 0 0d25 3.944322
    Bat 1 0d11 3.920879
    Bat 2 0d34 3.961905
    Bat 3 0d2f 3.956044
    Bat 4 0000 0.000000
    Bat 5 0d2e 3.954872
    Bat 6 0d1e 3.936117
    Bat 7 0d35 3.963077

    this is my untouched "good" battery

    Bat 0 0d76 4.039267
    Bat 1 0d8c 4.065055
    Bat 2 0d8a 4.062710
    Bat 3 0d8d 4.066227
    Bat 4 0000 0.000000
    Bat 5 0d8b 4.063882
    Bat 6 0d19 3.930256
    Bat 7 0d8c 4.065055

    include "Wire.h"

    define I2C_DEV_ADDR 0x28

    define CYD_SDA 22

    define CYD_SCL 27

    uint32_t i = 0;

    void setup() {
    Serial.begin(115200);
    Serial.setDebugOutput(true);
    Wire.begin(CYD_SDA, CYD_SCL);
    }

    void loop() {
    delay(5000);

    Wire.beginTransmission(I2C_DEV_ADDR);
    Wire.write(0x90); // register to read address
    Wire.endTransmission(false);
    uint8_t bytesReceived = Wire.requestFrom(I2C_DEV_ADDR, 16);
    if ((bool)bytesReceived) { //If received more than zero bytes

    union {
      uint8_t temp[16];
      uint16_t vdata[8];
    };
    
    float volts[8];
    Wire.readBytes(temp, bytesReceived);
    
    for (uint8_t batnum = 0; batnum < 8; batnum++) {
      volts[batnum] = vdata[batnum] * 1.1721611/1000.0;
    
      Serial.printf("\nBat %u %04x %f", batnum, vdata[batnum], volts[batnum]);
    }
    
    Serial.printf("\n\n");
    

    }
    }

    • 

      edrift on 04/07/2024 - 20:24

      Wow you really put some effort into this.

      I gave up.

      My next plan is to convert it to use 18v power tool batteries. I'm planning to buy a Dyson to ozito adaptor then convert that to be used in the LG

      I don't think there's an LG to ozito adaptor

      • 

        Stonemull on 05/07/2024 - 11:56

        Few hours work and I have a functional vac :) Wrote the code while I was balancing the 2 cells and it didnt take long with the datasheet. Updated my code this morning to get more data decoded https://codefile.io/f/DLD7kr2hpM

        Just be aware there maybe a problem with the ozito packs as they are only 5 cells (18v), it looks like the original Dyson V6 is 6 cells and rated 20V and this LG battery is 7 cells and 28v plus. So it may well be underpowered or cause issues with the motor driver going that route. 7 cells is an awkward size to find.
        Something like this would be equivalent to 2 LG packs https://www.aliexpress.com/item/1005006443750696.html

        Program now dumps the following..

        Register dump 0x00 to 0x4b
        2a 1e aa 0d ff 19 55 0b ff 0e 00 05 ff 0d 00 05
        02 08 f4 05 14 02 02 48 01 38 c8 50 55 08 aa 0d
        2b 00 ab 01 01 08 64 04 f2 0b 93 0a b6 04 3e 05
        3b 04 88 04 17 0b ca 0a 0a 03 1c 03 f2 0b b2 0b
        66 06 10 06 ff 09 02 fc 41 ef 04 3f

        Vcell COV 4.250256v
        Vcell OVR 4.100220v
        Vcell UVL 2.999560v
        Vcell UVR 3.400439v
        Vcell OVLO 4.499926v
        Vcell UVLO 1.500366v
        Vcell EOC 4.199853v
        Vcell LVCL 1.500366v
        Balance CBMIN 2.500220v
        Balance CBMAX 4.100220v
        Balance CBMIND 0.050403v
        Balance CBMAXD 0.500513v

        Status and data dump 0x80 to 0xab
        00 00 40 40 00 00 00 00 04 00 ff 0f 00 00 0f 00
        25 0d 11 0d 34 0d 2f 0d 00 00 2e 0d 1e 0d 35 0d
        d0 04 7f 08 67 08 a9 07 28 0b a1 09

        BAT1 SOC 3.944322v
        BAT2 SOC 3.920879v
        BAT3 SOC 3.961905v
        BAT4 SOC 3.956044v
        BAT5 SOC 3.954872v
        BAT6 SOC 3.936117v
        BAT7 SOC 3.963077v
        CELMIN 4.800000v
        CELMAX 0.000000v
        IPACK 0.017582
        VBATT 27.583297v

        • 

          impoze on 05/07/2024 - 12:31

          @Stonemull: amazing, have no idea what this is but great to see you were able to get it working

        • 

          edrift on 05/07/2024 - 17:12

          @Stonemull: Thanks for the pointers about the ozito and dyson battery.

          I couldnt remember how many cells the Dyson had.
          I assumed incorrectly it's the same as the LG and that the adaptor must have a booster chip of some sort to step up the voltage.

          How did you actually fix your battery in the end?
          Did you just rebalance the cells?

          I got a bit lost when you explained you were reading the battery registers through the data port.

          Couldn't you have just manually probed each battery?

          • 

            Stonemull on 06/07/2024 - 01:27

            @edrift: It was fixed in the first paragraph of the first post where I checked each cell, found 2 low and manually charged them up. Once I did that that battery worked fine for the 5 or so minutes I tested it at while vacuuming.

            But that did not explain to me why my parents had to go and buy another $600? vacuum cleaner as it is not that old and the batterys failed, well one did. So I want to get to the bottom of that. Once I looked into what the chip was and could see that I can monitor the cells over time pretty easily in both of my packs and my parents new one and see whats going on.

            The interesting thing is that the batttery is not dead, just unbalanced. In this state it stops charging when the fullest cell hits 4.2v (Vcell EOC 4.199853v parameter, "end of charge") and stops discharging when the flattest cell hits 3v, (Vcell UVL 2.999560v), but this was not the case with the battery I had. You could charge it and put it in the vac and it did nothing. I still have not gotten to why it did this.

            The interesting parts from the data above is the Vcell EOC 4.199853v parameter. There is no full state for the entire battery, it simply stops charging all of the cells when any one cell reaches 4.2v.

            From the datasheet…

            If any cell voltage rises above the VCELL EOC threshold, the device enters an End-of-Charge condition. This sets
            bit “0x81.7 VEOC” on page 73 and the “EOC Pin (35)” on page 99 is pulled low. The EOC output can signal an
            external MCU or enable an LED circuit to signal a fully charged pack.
            An VEOC condition also effects cell balancing operation, see “0x4B.0 CB_EOC” on page 70 for more information

            The last line is talking about whether the battery keeps trying to balance the cells after it has completed charging. WHich is bit 0 of register 0x4b, from above (last register in forst block) that is set to 0x3f so LG has programmed the thing to keep balancing once charged.

            So I figure there is an issue with the cell balancing section , CBMIN, CBMAX etc and reading through that you find ..

            If the difference between CELL N and the lowest voltage cell is larger than “0x20-21 CBMINDV” on page 50, and
            the voltage on CELL N is between CBMINDV and CBMAXDV, cell balancing of CELL N is enabled.
            Note: Cell balancing for CELL N is not enabled if its voltage is less than CBMIN or greater than CBMAX

            Last line says that the cell balancing is not enabled over CBMAX which is 4.1v, which has me a bit confused, it will charge till any one cell hits 4.2v but stops balancing on the highest voltage cells once they hit 4.1v, so the last part of the charging has no balancing on the highest cells, though I think I am missing something here, I just pulled the pack off thats been on charge all day and the cells all measure between 4.04 and 4.09v, CBMIND is 0.05v so according to that all cell balancing is below the delta voltage and disabled. I cant see why the cells are all under 4.1v and not closer to 4.2v.

            I don't know exactly why the pack was disabling itself and I don't know why the cells only hit 4.1v, oh, doh! I bet thats the max voltage that the charger can supply. 7 x 4.1v would be 28.7v. On the side of the pack it says voltage 25.55v (3.65v per cell) and limited charge voltage 29.4v (4.2v per cell)

            after wading through the data one more time the only thing i can think of is that the low cells were hitting that 3.0v threshold and that is why it shut off, I probably will have to read the data with the cells under load to get to the bottom of it. that would mean at least one cell needs replacing.

            • 

              edrift on 06/07/2024 - 07:41

              @Stonemull: Thanks this is useful.

              From memory i think I saw similar in that the cells were not balancing.
              I thought the whole point of Abby BMS was to keep cells balanced.

              However having said that I've recently had issues with an 18 month old byd battery pack for Kogan robot vac.

              It runs 4s2p config using LG 18650 cells.

              Basically after only 18 months the pack failed because 4 cells of 8 were dying.i think there was an internal short in each of those as the pack as a whole was getting really hot on the charger.

              After disassembling them I charged each cell separately and recorded the voltage at max charge.
              After some time, I measured the voltage again, and the 4 faulty cells were self discharging and staying warm despite the ambient temperature being quite cold. Those cells discharged down to maybe 3.7v doing nothing, from 4.1v

              So it's possible in the LG cordzero pack that the cells are just gone and need to be replaced.

              • 

                Stonemull on 06/07/2024 - 10:34

                @edrift: After sleeping on it I think its voltage sag on the lower voltage cells under load causing the issues, with no current there was a difference of 0.25v between highest and lowest cell, I can totally see a cell at 3.86v dropping to 0.5v more than a cell at 4.14v when you draw something like 100A at start up. Thats a difference of 25milliohms between the best cell and the worst. So when that CBMAXD = 0.5v constant is hit, ie instantly, the pack shuts off.

                The other interesting thing I can see, when a couple of cells are low but the max charge voltage for the entire pack remains constant, then that means the highest voltage cells have more than the 4.1v the charger seems to aim for. As my original measurements of 4.14V show (used a fluke 12 for this) at this point we have exceeded the 4.1v the balancing is allowed to operate at (CBMAX = 4.1v), so the only way to get the cells below 4.1v is to discharge the pack slightly, however the vacuum refuses to operate.

                I suspect a fix without opening the case maybe to apply an external load, maybe a couple of 12v globes in series, drag the pack down a bit then recharge on the charger where the balancing will be allowed to operate again. Do this a few times if needed and I reckon a dead pack could be revived.

                Now I think that leaving the spare pack on charge is the issue, the cells are close to that 4.1v balance limit, the second one exceeds 4.1v it is no longer discharged and balanced so it stays higher voltage, at the same time the total pack voltage remains the same, so all the other 6 cells can drop slightly before the charger trickles in more current. When it does, the highes voltage cell still charges as its above 4.1v so the cycle continues. Eventually a second cell reaches 4.1v .. same happens again. Now only 4 cells can balance the lowest one.

                If the packs are swapped over, or removed from charge, then they all drop in voltage below that limit and can balance.

                I did see someone mention that they take the pack off charge and store it elsewhere when its done and theirs was many years old, I think I may do the same thing.

              • 

                Stonemull on 07/07/2024 - 16:07

                @edrift: battery is still bad. If I use the higher speeds or turbo mode it shuts off almost instantly.

                • 

                  edrift on 07/07/2024 - 19:00

                  @Stonemull: Yeah need to replace the cells.

                  When I can be bothered I'll try that. But I think my BMS might be bad too.
                  I still have one half working battery so hopefully that lasts a bit more.

                  It wont work on the highest turbo mode at all, but will work on the medium turbo mode for a short time.

                  I'll probably go back to a Dyson eventually. They just seem more reliable and the existence of aftermarket batteries is great

• 

  Tarashak19 on 15/10/2024 - 23:37

  Good afternoon. I flashed the ESP32 with the code provided here. When reading information from the battery, I get this:
  Register dump 0x00 to 0x4b
  00 00 cd cd cd cd cd cd cd cd cd cd cd cd cd cd
  cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd
  cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd
  cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd
  cd cd cd cd cd cd cd cd cd cd cd 00

  Vcell COV 0.000000v
  Vcell OVR 4.141245v
  Vcell UVL 4.141245v
  Vcell UVR 4.141245v
  Vcell OVLO 4.141245v
  Vcell UVLO 4.141245v
  Vcell EOC 4.141245v
  Vcell LVCL 4.141245v
  Balance CBMIN 4.141245v
  Balance CBMAX 4.141245v
  Balance CBMIND 4.141245v
  Balance CBMAXD 4.141245v

  To be honest, I'm not good at coding and probably won't be able to figure it out myself. Therefore, I ask for your help.