Cycle Testing Observations…

I have had several discussions about how batteries age during their use. Winny was so kind as to translate an article on battery cycle testing for me (many thanks Andreas), and I decided to see if I could answer some questions that I have had.

Cycle testing takes a large commitment of time and equipment. There is not a fast way to do this, and commercial equipment is designed around “standard” parameters that do not necessarily reflect the way we use our batteries.

I have been told that fast charging reduces cycle life, however I have been unable to get any specific numbers. This became the objective of this test.

I took two Sanyo 2500 mAh cells, performed a forming charge of 200 mA for 16 hours, cycled them 5 times, and proceeded with the testing.

Two chargers were compared. The Energizer 15 minute charger was the fast charger, and the Sanyo NC-MQH01U was used as the slow charger by utilizing one of the slow charge slots. The Sanyo charger charges at roughly a 1.0 amp rate in that slot.

The first test was done after charging both cells on the slow slots of the Sanyo charger. This became my base line capacity. The cells came out reasonably close in capacity and the testing continued.

I saved the data collected every 25 cycles, and it is presented below.

The procedure involved charging the cell on its respective charger, letting it rest for 20 minutes, then discharging it at 2.5 amps down to 0.8 volts.

High capacity cells are not the best choice for high charge or high discharge rates. A cell is usually considered fully discharged at 0.9 volts, and going to 0.8 volts is a bit of an over discharge. These values were chosen to accelerate the aging process and reduce the number of cycles needed to see a trend. I also only let the cells rest when I was asleep. I believe you can get improved life by less rigorous usage.

A general consensus from a variety of battery manufacturers suggest that you can expect around 500 cycles from high capacity batteries, and around 1000 cycles from lower capacity batteries. Lower capacity batteries are those with less than 2200 mAh of capacity, however Energizer was the only one that suggest a 2200 mAh cell as lower capacity. Most consider 2000 mAh cells as the high end of the lower capacity cells.

Keep in mind that when the battery manufacturer comes up with these numbers, they are doing a 5 hour discharge and a 14-16 hour slow timed 0.1C charge.

Also, keep in mind that all of the information I have been able to find indicates the number of cycles to reach around 60-80% of the cells original capacity. It is understood that the mid point voltage will be reduced, but the performance is measured in reduction of capacity.

In non regulated flashlights, the voltage retention under load is very important to the brightness of the light. It has been interesting to watch the voltage drop during cycling.

Here is the data from the slow charge through 150 cycles.

Here is the data from the fast charge through 150 cycles.

Here is a comparison of the slow charged cell at 125 cycles with the fast charged cell at 100 cycles. Based on this limited testing, it looks like you lose around 25 cycles by fast charging.

In my notes I observed that the fast charge cell was about done after 125 cycles. I then switched to “topping off” the fast charge at a slower rate on the Schulze charger. This did not seem to have much effect.

At 148 cycles, the fast charge cell was no longer able to be charged on the Energizer 15 minute charger. I would put the cell in, and would get a blinking red error light. The final charge cycles were done on the Schulze.

Out of curiosity, I did another cycle (number 151) on both cells. This time I charged both cells on the Sanyo charger and did the discharge at 1.0 Amps.

Conclusions:

Fast charging high capacity cells, followed by fast discharging to an over discharged state is harder on the cells than slow charging followed by the same conditions.

The mid point voltage drops with use and it seems to drop faster with fast charging.

Cycle performance based on a percentage of remaining capacity does not tell the whole story.

Cells stressed and damaged by heavy usage may still be suitable for lower drain applications.

The performance lost by cycling is permanent.

There does not appear to be an increase in the self-discharge rate of high cycled cells.

Here is round 2 of this effort. I was wondering how smaller capacity cells would handle this. I took some Titanium 2000 mAh cells and decided to test them under the same conditions. The same chargers were used, and the discharge remained at 1C, which, in this case is 2.0 amps.

I did change one thing. After 50 cycles, I decided to do a 0.1C charge for 16 hours after every 25 cycles. These cells seemed to be able to handle the cycles a lot better than the higher capacity cells did. I am not sure if the forming charge made any difference or not.

There was no problem charging on the Energizer 15 minute charger, and while the cells have lost some capacity, they still seem to be in reasonable condition.

Here is the data from the slow charge through 150 cycles.

Here is the data from the fast charge through 150 cycles.

Here is a comparison of the slow charged cell at 125 cycles with the fast charged cell at 100 cycles. Based on this limited testing, it looks like you lose around 25 cycles by fast charging. This is the same result that we saw with the Sanyo 2500 mAh cells.

Once again, I did another cycle (number 151) on both cells. This time I charged both cells on the Sanyo charger and did the discharge at 1.0 Amps.

Conclusions:

Lower capacity cells handle cycle testing better than higher capacity cells.

Lower capacity cells handle fast charging better than higher capacity cells.

It seems that fast charging performance drops off faster than slow charging performance, buy around 25 cycles.

Based on this limited amount of testing, I can believe that these cells may be able to go 300-500 cycles. Looking at the data from the 1.0 amp discharge at cycle 151, the slow charged cell is at 95% of the original Amp Hours and 91% of the original Watt Hours. The Fast charge cell is at 94% of the original Amp Hours and 88% of the original Watt Hours. Cycle testing usually goes to 80% of the original Amp Hour capacity, so these cells still have a ways to go.

It would appear that the price you pay for higher capacity is a loss of power and cycle life. Of course, this test is designed to accelerate the process. Charging and discharging at lower rates may give you some additional cycle life.

Round 3

This time the testing involved Sanyo Eneloop cells. These cells have a low self discharge rate and are supposed to maintain about 85% of their initial capacity after a year of room temperature storage. These cells may be sensitive to extended trickle charging as well as high charge rates.

These 2000 mAh cells were charged on the Energizer 15 minute charger which uses a 7.5 amp charge rate. That works out to charging at around 3.75C. Discharging was done at 1 amp this time, and every 25 charge/discharge cycles a 16 hour charge at 200 mA was done.

It was interesting to observe the performance of these cells. They started out having an internal resistance of 0.026 ohms. They held, or slightly increased in capacity over the first 100 cycles, then they started to fade. Looking at the graph you can see this fade in cycles 125 and 150.

I then charged them back up and let them sit for 30 days. It appears that this ultra rapid charging has effected the cells ability concerning low self discharge. Also, the cells internal resistance rose to 0.340 ohms. Discharge 151 on the graph has a very interesting shape to it. The Energizer 15 minute charger now rejects these cells. When I charged them up and put them into my camera, I got a low battery warning after 15 shots. I was able to coax another 8 shots before the camera powered down. I usually get several hundred shots from a set of Eneloop cells.

It looks like these cells can handle 100 – 125 charge cycles on the Energizer 15 minute charger, but after that they are about done. The Titanium 2000 cells came in better than these cells did, but they are not low self discharge. Overall, I was impressed that they held up as well as they did, but for longer cycle life it would be better to keep the charge rate in the 0.5 – 1.0C range.

Here is the graph.

Tom

## Bookmarks