I will use a lab power supply to measure current and voltage for charging and battery tester for discharge.
Batteries Charging Routine
- Thread starter Az.
- Start date
Seen as the original poster want to open a Pandoras box, it seems that 'memory effect' is still possibly a factor to be considered. It has been suggested below that some cells charged on lower voltages to extent their lifetime benefit from a full 4.2v charge every now and again as it recovers the capacity lost due to the 'memory effect'.
This is illustrated in the graph below. One particular LG cell (red line) loses an additional 10% of its capacity over 50 cycles but this capacity is recovered when it gets a full 4.2 volt charge which he does every 50 cycles. Other cells also benefit but to a lesser extent. The recovery from the 'memory effect' ( rise in capacity after full 4.2 volt charge) can be seen on most of the cells but it is very low on some cells (e.g. Samsung 50E top blue line)
endless-sphere.com
This is illustrated in the graph below. One particular LG cell (red line) loses an additional 10% of its capacity over 50 cycles but this capacity is recovered when it gets a full 4.2 volt charge which he does every 50 cycles. Other cells also benefit but to a lesser extent. The recovery from the 'memory effect' ( rise in capacity after full 4.2 volt charge) can be seen on most of the cells but it is very low on some cells (e.g. Samsung 50E top blue line)
Cycle life tests of High Energy density cylindrical cells
"Standard Power Rate" test (1C continuous discharge) note: the 1C discharge test is performed at ca 90% DoD (4.15V-3.00V) Test eqiupment: - 2pcs ZKETECH EBC-X0510 - BF-2A four-wire cell fixture (10A version with Nickel plated contacts) - tested in room with relatively stable temperature 23 ±2...
endless-sphere.com
Last edited:
I couldn't get any sensible charger figures because things kept switching off. As far as I could figure out, the consumption was about 6w , once the battery is fully charged, but I can't be certain. After both the BMS and charger switched off, and after the BMS reset, I connected my lab charger bwith a much higher voltage and current limited to 0.5 amps, which should be about 30w, but it could only push 6w inton the battery for whatever reason
I have completed the test on my 20Ah 48v battery. The discharge from 54.6v to 53.3v at 0.5A (0.025C) was 29.8wh, which is 3.1% of the total.
Interesting result. On the surface, there appears to be a difference between your figures and Padja. However, when you dig down into Padja figures, he uses the spec of 2.5v to 4.2v @ .2c as a reference point. This figure seems to be compared to his test usage which is 3v to 4.15v @ 1c to get his DoD and whats aparrent on the graph.( I think)
But his overall intention is optimistic and good in trying to demonstrate that long life can be achieved by very little voltage and capacity reduction if good quality high energy cells are used in the first place.(imo)
Last edited:
Saneagle's test is much easier to understand compared to Padja's though. Saneagle also explains the point in plain English. You may lose up to 3% of capacity if you charge to 4.1V but you gain in longer battery life.
Last edited:
And any capacity you lose at the bottom end isn't normally important because most of us never run our batteries down that far anyway, but when you give up charge at the top end, you're more likely to be in the position of having to run your battery right down, which is where the real danger to the life of your battery is, especially if you're not going to charge it back up to where it does the cell-balancing.
The whole idea of extending battery life by short charging is just complete BS. People are just concentrating on one factor with blinkers on. If you consider everything, you're more likely to lose than win.
I disagree with that conclusion. How often your battery is completely flat in the last 5 years? Statistically, people put their battery on charge far too soon for fear of running out on the next outing rather than when their battery is flat. Charging when the battery is 80% or more full does not do your battery much good. If your battery needs rebalancing, chances are that it's been made with poor quality cells.
When was the last time you saw a pack out of balance?
If you consider everything, would you agree that lower voltage equals also to lower fire risk beside longer life?
I dont see that knowing the apparent amount of charge lost when a battery is charged to 42V and then discharged to 41V is of much value, its not the same as charging to 41V and then discharging from there and thus measuring the batteries real actual capacity.
Here is the discharge plots with a load of 16.7ohm, so circa 2Amps, of my 21700 10S1P; first from a full charged battery (to 42V). Shown is volts, Ahr drawn, minutes;
Rest volts 41.7.
41V, 0.01Ahr,0min
40V, 0.384A,10min
39V,1.16Ahr,31min
38V,1.63Ahr,44min
37V,2.25Ahr, 62min
36V,2.72Ahr,75min
35V,3.23Ahr,91min
34V,3.7Ahr,105min
33V,3.97Ahr,114min
32V,4.2Ahr,122min
31V,4.41Ahr,129min
30V,4.59Ahr,135min
And now the same discharge when the battery charge had been stopped at 41V;
Rest volts 40.4V
40V,0Ahr,0min
39V, 0.295Ahr,8min
38V,0.748Ahr,20min
37V,1.26Ahr,35min
36V,1.75Ahr,50min
35V,2.25Ahr,64min
34V,2.72Ahr,79min
33V,2.97Ahr,87min
32V,3.18Ahr,94min
31V,3.37Ahr,100min
30V,3.53Ahr,106min
So the 'capacity' used from 42V full charge till the battery dropped to 41V was only 0.01Ahr, or circa 0.2% of the overall capacity, so not much information there.
But the practical test, simulating what a user might do, as in charging to 41V and then stopping charge, shows a capacity loss of 23%.
Clearly the battery is aquiring a far bit of capacity in the last 1V of charge to 42V.
For eBike batteries, has anyone charged them to 41V, stopped the charge, and then measured the battery capacity ?
Last edited:
My Sans chargers output is 42V when the LED on it turns green but the battery voltage is between 41.3V and 41.5V. Saneagle uses
How do you charge your battery to 41V? Do you use an adjustable PSU? Saneagle discharged his pack a lot more slowly to avoid voltage sag interfering with capacity measuring.
Last edited:
I use a power meter and a bit of patience, when the voltage indicated on the battery goes to 41V, I stop the charge.
Strange question really.
Saneagle test is very well done.
He found that between 42V and 41V (for a 36V battery) you'd lose 3.1%. It's believable because the slope Voltage/Capacity is highest between 41V and 42V and also below 35V so the loss is between 0% (if the slope is infinite) and 10% (if the slope is the same for the 41V-35V segment).
The 24% reduction you found can't be explained.
Last edited:
I am asking about your own batteries or those that you have fixed. This is because BMSes have active balancers that work continually in the background.
This is the problem. All sorts of strange things are going on between voltages you measure, the battery sinking the voltage, the way the charger works and the way the BMS deals with things. The voltage you measure on the charger depends on how your measuring device interprets the pwm and how much the charger people spent on smoothing the pwm.
Yes, nearly all batteries go out of balance when you run them down. I've checked that many times when I've done cycling tests. It always looks worse than it is because it only takes a small amount of charge to make a big difference to the voltage when you're down there. Typically, you see one or more at 3.0v and some around 3.3v or more. You must have some scrap batteries. Check it out.
I don't understand what you mean about background balancing. All normal BMSs only balance at the top end. The only thing that happens at the bottom is that the battery is shut down when the first cell reaches LVC.
Can you please explain what do you mean by flat battery? I thought BMS should protect battery from that eventuality (unless battery is left unused for a long time).