xenonhost wrote: ↑
Wed, 04 Apr 2018, 15:00
I have bought 11 months ago 48 pcs Winston LifePO4 for a friend. They worked well, but 2 weeks ago 18 pcs got swollen and they are all shortcircuited.
How did you have these 48 cells configured into a battery? Did you first parallel them in groups of 3 (making so-called "buddy triples"), then connect 16 of these triples in series via diagonally opposite corners? That is the recommended way to do it. Alternatively, you might have first connected them into series strings of 16, then connected those strings in parallel. It is more difficult to get equal current-sharing that way, and you need 3 times as many cell-monitoring-units in the BMS.
Where, in relation to each other, were the 18 that died? Did they consist of 6 buddy-triples?
I need a BMS, something for remaining 16 pcs connected to PIP.
Yes. You need a BMS. Every lithium-ion battery needs to be protected by a BMS that monitors the voltage and temperature of every cell and prevents charging when any cell goes overvoltage, overtemperature or undertemperature, and prevents discharging when any cell goes undervoltage or overtemperature. Given that you are in Europe, perhaps this supplier can help:
On a system with 5 PIP4048 in parallel, for the last week it doesn't change from grid to solar. I have setup from Watchpower the charging only from grid 2A, (i think it's 10A for the whole 5 pcs) because from solar I don't have such low value and I've lost 2 groups of Winston 16s 40Ah by charging with 20A/PIP.
This sounds like you had them connected in strings of 16 first, then paralleled those 3 strings. But charging them at 100 A alone would not destroy them. Even one
string of 40 Ah Winstons can take 100 A (2.5C) with no problem, provided all the cells are balanced.
Given that you didn't have a BMS, did you at least manually balance them when you first installed them?
Coulomb and I do this by setting the absorb voltage to 55.2 V (3.45 V per cell) and monitoring the individual cell voltages with a multimeter while charging, until the first cell goes over 3.45 V. Then we use headlight bulbs with alligator-clip-leads to bring down the voltage of the highest cells. If any cell reaches 3.6 V, we temporarily reduce the absorb voltage setting, just enough to prevent that. We keep moving the headlight bulbs around to the highest cells until all cell voltages remain between 3.44 and 3.46 volts, even after all headlight bulbs have been removed for half an hour. You may need to temporarily raise the float voltage to the same as the absorb voltage, to keep them on charge for long enough to complete this initial manual balancing.
We then leave the absorb voltage set to 55.2 V (3.45 V per cell), the float voltage set to 53.7 V (3.355 V per cell) and the low voltage cutoff set to 50.8 V (3.175 V per cell).
What absorb("bulk"), float and cutoff voltages did you use?
Anyone having a clue why it doesn't switch to solar anymore?
Setting number 16 is charge source priority. A value of "Cut" or "uti" here would prevent solar charging whenever the grid is available.
If you are not running our patched firmware
(LFP version), then there are a number of other problems that could be occurring.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).