After 15 return trips to work and back (approx 33km each) two of the cells in my Heter 48v10A battery pack died. Without hesitation, and at no cost to me, Heter sent me 2 new cells to replace them. They arrived in a flat envelope within a week - waiting for me on my return from a week of back country skiing in Kosciuszko National Park.
I proceeded to replace the dead ones with the new ones in the battery pack. To ensure that I recabled the battery pack correctly, I first photographed the pcb which connected the battery cells and then carefully labelled the three black charging leads and the three BMS sensor leads which I removed.
The battery pack outside its case.
The PCB connecting the battery cells, charging wires (black) and BMS sensor wires (coloured)
I desoldered the broken cells, absorbing all the solder with braided solder wick and them reassembled the pack with the new cells. With a 200Watt soldering iron, this exercise went very smoothly.
After reassembling the pack, I checked the voltage across each pair of pins in the BMS sensor header and each read between 3.24v and 3.33v. Readings for the two new cells corresponded with the measurements I had taken prior to assembling them into the pack.
Issue 1. BMS failed
I connected my 5 globe load to the external connectors of the battery pack - and nothing. I checked the unloaded terminals with a voltmeter and got a reading of only 26v - dropping to less than 1v when I attached my load. I double checked all the connections to the sensor wires against my labels and the reference photographs, plus again checking continuity from each individual battery cell terminal. All connections proved to be correct. Looks like somehow I had killed the BMS. I decided to proceed to condition the battery pack keeping the BMS in place because the sensor wire header provided a convenient way to individually measure the voltage of each cell. After conditioning the battery pack I intended removing the BMS and sending it to Heter for assessment and repair.
I then prceeded to cycle the battery pack several times using my string of 5 12v 23Watt globes. These drew 1.8A when the battery was 54v and 1.55A at 48v. After three short cycles (approx 1 hr on the load and then recharging until the charger LED went green) all cells were showing the same voltage. After completing the charge, once the charger light went green, the surface voltage of the cells was between 3.59 and 3.72v with most at 3.64v. After 5min of load to burn off the surface charge, all cells read 3.34v.
Then I dedided to try to determine the pack capacity. To ensure a full charge, I left the charger on for 1 hour after the charger LED turned green. Then connected my load and started timing the discharge. Minimum rated capacity for each cell is 10000mAH so with my 1.7A load the lights should burn for at least 5.9 hours.
After the first three hours of discharge(at approx 1.7A) I checked the voltage across each cell every half hour. At that stage, the cells were all between 3.28v and 3.34v unloaded. After 5 hours the cell voltages ranged from 3.15v to 2.88v unloaded. With the load on, the battery pack measured 48.1v and load current was 1.55A. I decided to leave it drain for another 30 min - bad decision.
Issue 2. Buldging Battery Pack
The voltage across the battery pack had dropped to 18v. Individual cells measured between 0.8v and 1.58v with one cell at 3.02v. Integrity of the cell cases was OK with none burst or broken.
I attached the battery charger and turned it on. After 5 min the cells appeared to be charging normally and after 30 min all cell voltages were over 2v. It continued charging normally, but, in this shape the cell pack is no good to me so I will now send it all back to Heter for analysis.
More piccies and info available at http://det-hem.ourpix.biz/main.php?g2_itemId=34711
I proceeded to replace the dead ones with the new ones in the battery pack. To ensure that I recabled the battery pack correctly, I first photographed the pcb which connected the battery cells and then carefully labelled the three black charging leads and the three BMS sensor leads which I removed.
The battery pack outside its case.
The PCB connecting the battery cells, charging wires (black) and BMS sensor wires (coloured)
I desoldered the broken cells, absorbing all the solder with braided solder wick and them reassembled the pack with the new cells. With a 200Watt soldering iron, this exercise went very smoothly.
After reassembling the pack, I checked the voltage across each pair of pins in the BMS sensor header and each read between 3.24v and 3.33v. Readings for the two new cells corresponded with the measurements I had taken prior to assembling them into the pack.
Issue 1. BMS failed
I connected my 5 globe load to the external connectors of the battery pack - and nothing. I checked the unloaded terminals with a voltmeter and got a reading of only 26v - dropping to less than 1v when I attached my load. I double checked all the connections to the sensor wires against my labels and the reference photographs, plus again checking continuity from each individual battery cell terminal. All connections proved to be correct. Looks like somehow I had killed the BMS. I decided to proceed to condition the battery pack keeping the BMS in place because the sensor wire header provided a convenient way to individually measure the voltage of each cell. After conditioning the battery pack I intended removing the BMS and sending it to Heter for assessment and repair.
I then prceeded to cycle the battery pack several times using my string of 5 12v 23Watt globes. These drew 1.8A when the battery was 54v and 1.55A at 48v. After three short cycles (approx 1 hr on the load and then recharging until the charger LED went green) all cells were showing the same voltage. After completing the charge, once the charger light went green, the surface voltage of the cells was between 3.59 and 3.72v with most at 3.64v. After 5min of load to burn off the surface charge, all cells read 3.34v.
Then I dedided to try to determine the pack capacity. To ensure a full charge, I left the charger on for 1 hour after the charger LED turned green. Then connected my load and started timing the discharge. Minimum rated capacity for each cell is 10000mAH so with my 1.7A load the lights should burn for at least 5.9 hours.
After the first three hours of discharge(at approx 1.7A) I checked the voltage across each cell every half hour. At that stage, the cells were all between 3.28v and 3.34v unloaded. After 5 hours the cell voltages ranged from 3.15v to 2.88v unloaded. With the load on, the battery pack measured 48.1v and load current was 1.55A. I decided to leave it drain for another 30 min - bad decision.
Issue 2. Buldging Battery Pack
The voltage across the battery pack had dropped to 18v. Individual cells measured between 0.8v and 1.58v with one cell at 3.02v. Integrity of the cell cases was OK with none burst or broken.
I attached the battery charger and turned it on. After 5 min the cells appeared to be charging normally and after 30 min all cell voltages were over 2v. It continued charging normally, but, in this shape the cell pack is no good to me so I will now send it all back to Heter for analysis.
More piccies and info available at http://det-hem.ourpix.biz/main.php?g2_itemId=34711
