วีดีโอ

You need just the balancing wires from the cells. If you use same capacity cells should be good. Adding bigger cells might work as well, but needs to be tested

Yes, it's a beautiful town. :)

Hi, from my experience, the cells must be around 3.2V for the BMS to wake up. If the bms was closed and you charged the battery directly, the bms will fail to open. Try discharging it with a dummy load until 32V, check that voltage on each cell is right and try removing and reinserting the bms. If that does not work, order a bms from aliexpress, is easier than trying to reprogram or fix it.

Hi, thank you for your support! At the bottom of the github, there are 2 links, one for stripe and one for paypal. github.com/alex3dbros/megacnc

Disconnect the battery from the DC DC and use a 48V charger to get cells above 3V. While charging the bank, measure each cell one by one to make sure they don't get overcharged. If you have any dead cells in the pack, they must be revived or replaced.

Honestly if that’s during sunshine hours, I couldn’t care less. But if that’s at night while it’s discharging, that’s a lot relative to the 5 kWh capacity.

You are welcome! Thanks for your feedback!

The software is free, open source actually. But this image must be maintained, uploaded, pay for traffic. For those that mind paying 5 Euro, can watch the video on setting this up on their own on VM or Raspberry Pi

Some of the cells were discharged and some not. Applying voltage on the discharge lines might wake up the battery, but it will not balance the cells. Charging individual cells from the series will ensure that all cells were fully charged and the pack is balanced.

@@thecelldoctor8890 Yeah those cells that discharge like that are weak ones and proably bring down the rest of them by not being able to charge fully to 4.2, replacing one could bring back lots of capacity. I have an e45 i found in the dump with an error, turns out the internal one just wanst charged, because the external one wasnt making connection to it because it got loose. Taking off the external battery and charging the internal one fixed it but getting the external battery was the trickky part, i see I wasnt the only one with issues heh, i had to use a screw extractor to get it out, probably another reason it was thrown out, it looks like they also tried to get in there, eventually stripping it. Anyway i only get 12km range with this battery pack, sure its not going to be the full 22km like it was new but i think replacing a weak cell or 2 may bring it back closer to 20. So this information is very valuable thank you, even to build a new one from scratch, lots of good info here thanks.

You are right

Salut, da, pe deepcyclepower.com/

Hi, make it print using the phomemo software first using bluetooth, then it will work on the pack calculator as well.

Well, the main issue is the bidirectional DC to DC, is not the protocol. Protocol can be sniffed in a few days most probably, but you need to provide 400 V DC from the battery and charge it from the 400V buss as well (bidirectional DC to DC), redo the BMS part as well to communicate with the other DC to DC. (It uses 2 DC to DC's until it gets to inverter). Seeing the complexity of the system, I decided to stop the investigation as it's not worth the effort. There are other inverters out there which work directly with low voltage 48V banks and BMS'es like JK BMS deepcyclepower.com/product/jkbms-smart-active-balance-bms/ communicate directly with them. I used Growatt 5000 offgrid inverter in one of my installs with a 15 Kw battery that we built and it works like a charm: deepcyclepower.com/product/48v-280ah-metal-box-kit/

This version has no GUI. You can run a VM with ubuntu if you need gui version

@@thecelldoctor8890 I ended up using my Ubuntu desktop version and installing using your instructions. Thank you!

The Long BMS board has markings for each cell. Positive to negative and which cell number goes where. It's a 10S pack. I recommend understanding how series and parallel works first before redoing the pack. It's 10S 2P. Once you understand this, you will figure out how to assembly them back.

It's a PCB we developed. Can order one here: www.deepcyclepower.com/old/product/10s-battery-balancer-megabalancer/

here is the link, he used the liitokala 4 then an electronic charge to respect the cell builder test protocol : th-cam.com/video/j4mdaokeM9M/w-d-xo.htmlsi=RLUoHvM_EWWJeXwD&t=266 - I hope YT will allow the link

Hi, it depends on the testing current. If discharged with less current, the cell will get discharged slower and have higher capacity due to low volt drop. It depends how the heat is dissipated when discharging, a lot of factors including contacts of cells if rusted or have nickel strips, environment temp can contribute to different results compared with datasheet. It also depends if the device have been calibrated

@@thecelldoctor8890 Yes , I get it , the guy wanted very logically to manage the testing respecting the datasheet environment. Clearly the result showed as good a weak battery if measured with the Liitako 4. BOw you gave us a lot of work between the update and the new software ! All the best

@@thecelldoctor8890 " If discharged with less current, the cell will get discharged slower and have higher capacity due to low volt drop" is it the same when you regulate the "chemistry" to 0.5 1 , or 2 A ? the results differ (a little ?) Up to today I get similar results after similar cycles, tell me, if I have not enough power, do the MCC refuse the 1amp choice (my impression) ? BTW it is for discharge and for charge ?

Yes, that's possible, but you need to have a BMS installed to safely charge the cells as well. The DC DC has role of a BMS as well

You are welcome! Thanks for your feedback!

If you connect 1 slot at a time, yes, can do that. If you connect multiple slots to different cells in that pack, megacell charger will get damaged permanently

​@@thecelldoctor8890 Thank you ! Too bad ! I was vaguely expecting to use the balancing wire to send them a low current so the cells can begin to charge . But I do not understand if constant voltage or constant current is needed in that case. THe power unit you use in direct mode, do you regulate voltage or is it intended for fixed foltage and some not too high current max ? , (and those batteries have only 10 cells each ?)

@@thecelldoctor8890 hello, I was looking at your video on facebook where you state that you try to use non conductive pry bar. Last year I bought a zipped bag with many tools to repair cars external skins and other where plastic pry bars are useful. The kit was about 15€ and very useful to dismount modernTV (many of them have no screws). I will send you a linkwith Gmail if I find it now. Best regards

Thank you!

Hi, Thanks for your feedback! I did not see any dehumidifier inside, there's something on the lid that looks like a fire protection cell which should pop if it senses fire to extinguish it. Or maybe I am wrong and it's a dehumidifier. I did tare down the DC DC Part off-camera for learning purposes, but It's too much work to be worth the effort of building that part to connect other batteries. I can look for the photos I got and send on telegram or whatsapp, visit deepcyclepower.com and can find my number there.

@@thecelldoctor8890 There is a fire extinguishing bag in each Huawei battery module, not a de-humidifier. That's why there are those 8x temperature sensors. There's also a battery optimiser inside each 5kWh battery. I saw it at a trade show where they had replaced the metal casing with Perspex.

@@kalbarriman The temp sensors are for the BMS and cell monitoring. The fire extinguisher cell will pop when there's fire present or goes over a certain temperature, it's not linked to the BMS in any way, it's standalone

:)) Yes it does. Thanks for making it!

Were are you from Michael ? In EU I can ship these kits with no problems

I am in the US... so probably not possible?

@@michaelcharge4783 I can talk with my supplier and ask for shipping price, I think we find a solution to ship to US as well. If you are interested to get the kit, email me at deepcyclepower@gmail.com with shipping details and I will provide you with shipping options and cost.

Hi, If it's connected to same network with megacell charger, than no need for wifi. About the instalation, it looks complicated but is not that hard once you get through the virtual machine step. In the future there will be images for VM's and raspberry pi. But database is stored on that VM so I still need to teach people to use the console for future updates.

@@thecelldoctor8890 Thank you, yes I need to look in that Virtual machine use (I thought we where using at last one under windows, without knowing it !)

It will work withouth any problems. I chose gui version as is easier for me to use

¿Para cargar este tipo de batería o para probar las celdas? Producimos un dispositivo llamado cargador megacell. Puedes ver más información en deepcyclepower.com

@@thecelldoctor8890 pero como llama la maquina que carga las bateria . No la fuente hablo de la que conecto ala fuente

It's a board that we built www.deepcyclepower.com/old/product/10s-battery-balancer-megabalancer/

Ese es se puede perder a Colombia Y me podría explicar como la coloco a carga .

Som rád, že video bolo užitočné!

Hi, it did not die. Working now on a new software, the hardware got upgraded to PRO with charge control up to 4.5A for each slot, discharge up to 3A controlable on each slot CC, CV, CR. For more details, check deepcyclepower.com

Yes, it worked without problems

No, removing the BMS did not make any issues.

@@thecelldoctor8890 I see, You made pictures. Do You have picture from fans voltage, flowrate etc and also can give me dimensions? I wish order the new silent fans before opening.

That is also doing DC DC up to 700 Volts. It will raise or lower the voltage depending on the needs of the inverter. This allows connecting 3 phase or single phase inverters. The DC DC's are bidirectional, used for charging from the high voltage buss as well. The main advantage of high voltage bus is lower current, thus using thinner wires for the entire system.

Thank you!

I did not tried those as I don't have any recycled. To buy new, it's better to buy the battery directly assembled

That is something to try I suppose. But I assume they have some values in the brain unit to check state of charge. That's basic for every smart BMS. In theory it might work if they did not add safety values for stopping at a certain capacity threshold or if there will be enough interest from some programmers to change those values and allow for higher capacity.

@@thecelldoctor8890 Hi, thank's alot for your quick answer. Actually I don't own this battery but I have several thousands of 18650 LiIon cells from recycled ebike batteries. Furthermore I got alot of old USV lead batteries. If I assemble a big block of those LiIon cell wit 13 cells in series and put a independent BMS on them so I get around 50 volt wich will fit perfectly to the original LPF pack. The idea behind is to buffer them. But if there is a supervision of the capacity of the original battery pack I doubt that it will work. Do you try this idea once out for yourself? Kind regards

No I did not try that. But usually connecting BMS'ses in chain and mixing chemistry is not a very good idea as they have different voltages for charging. Adding a LifePo4 pack in parallel and set both packs at same voltage before connecting might work if the capacity metering don't stop at 2 Kw. Maybe in the future when these batteries start dying or become unbalanced and people starts selling them as Used, will experiment more, but for now I did not afford to break a 2000 Euro battery

@@thecelldoctor8890 Yes I know, you are allright. But at the levels of "48" Volts there are closely similar voltages. 16S LIFEPO have max. 52,8 Volt, 24S lead acid have max. 2,2 Volt for each cell and 13SLiIon have 53,3 Volt with 4,06 V each cell. So they are close together. If you take the LIFEPO as master the LiIon will be slighty underchared and the lead. If you put independend BMS circuits on them so only the max. Voltage will be controlled for each cluster. I thing there might be no problem from this site. Okay I have to check it out first but I'm pretty shure it will work well. What do you think about?