SilverFox
Flashaholic
Once upon a time, long ago…
Engineers and scientists huddled together trying to figure out what was going wrong with the power supply on a satellite in orbit around the earth. The battery capacity had been selected so normal use would only require 25% of the total capacity, however, when they tried to use some of the unused capacity, it was not available.
Testing in the laboratory initially failed to reveal similar problems, and it wasn't until they followed the exact regime that the satellite was using that they were able to duplicate the problem.
It appeared that when a sintered cell NiCd battery is conditioned by repeated shallow discharges, it seems to remember that it only has been using a portion of its capacity. Trying to use more capacity results in problems and this became known as the "memory effect."
News of the "memory effect" swept the world and it became the answer to all rechargeable battery problems. I did an informal poll asking people of all ages and walks of life what they thought the main problem was with under performing or dead rechargeable cells. One of the early, if not the first, thing mentioned was the "memory effect." People remember the "memory effect" and still believe that it is a major contributor to rechargeable battery problems.
How do you reproduce the "memory effect?"
The laboratory testing revealed that in order to produce a strong "memory effect" you had to specifically follow this "recipe."
Start with sintered NiCd cells. Other constructions were not as successful in reproducing the "memory effect."
Next, you need to define a strict discharge cycle, which does not fully discharge the cell. Each discharge cycle needs to be within 1 – 2% of each other.
The charge cycle must only recharge the battery to 100%. No overcharging is allowed.
The charge rate must be slow. 10 – 12 hours is a good time frame for charging when you are trying to reproduce the "memory effect."
That's it…
Violating any of these steps reduces the "memory effect" and sometimes can completely eliminate it.
Where did the rest of the capacity go?
The laboratory testing continued and scientists discovered that the remaining capacity was still available, but at a reduced voltage. This was not a "memory effect," but more accurately a "voltage depression." Unfortunately, by the time they came out with this, the term "memory effect" had permeated everything and is still with us today.
Let's take a closer look at voltage depression.
The reason the voltage drops is because of an increase in internal resistance within the cell. Looking for a cause of this increase in internal resistance, scientists dissected a voltage depressed cell to see if they could observe any changes that had taken place.
When cells are new, the crystals on the plates of the cell are around 1 micron in size. The size of the crystals on the plates of the cells that were showing voltage depression were 50 – 100 microns, or larger. These larger crystals resist breaking down to give up their energy. In addition, the longer the larger crystals remain intact, the harder they get, and it becomes even more difficult to break them down.
This form of voltage depression is reversible, as is the "memory effect." Usually a single complete discharge at a low current draw will break down the large crystals and you are back to normal. Sometimes it takes a couple of cycles. It is dependant on the discharge current. Higher currents take more cycles.
Now, how do you know if your battery problems are being caused by the "memory effect?"
Here is a list of questions to help you determine this…
1. Are you using sintered NiCd cells?
2. Have you done several similar (within 1 – 2% of each other) discharge cycles to less than a full discharge?
3. Is your charge limited to 100%? {i.e. no overcharge}
4. Does your charger take 10 – 16 hours to complete a charge?
If your answer to any one of these questions is "NO," your battery problem is probably not caused by the "memory effect."
Most of us are using NiMh cells, but we do have power tools that utilize NiCd cells, and some of those may utilize sintered construction. There is a possibility that you are using these cells in your power tools. If you have a turbine engine aircraft, you are most likely using a vented sinter plate NiCd battery.
The next point is difficult to do. A variation of more than 1 – 2% in the duty cycle drastically reduces the "memory effect." In normal day to day use, it is very difficult to adhere to this tight of a schedule. It is possible, but extremely difficult to do.
Next comes charging. If you are using a charger that terminates using negative delta voltage, you are applying a slight overcharge every time you charge. If your charger utilizes a top off charge to bring the cells up to full capacity, you experience a small overcharge every time you charge. The scientists studying the "memory effect" initially were unable to reproduce it because they fully charged the cells after each shallow discharge and the slight overcharge that is involved in normal charging erased the "memory effect." If you are using a modern charger, this point will be the stumbling block to developing the "memory effect" in your cells.
Finally, how long does it take to recharge your cells? If you are following the battery manufacturers recommendations and are charging in 1 – 2 hours, you minimize the production of the memory effect in NiCd cells. However, a similar effect can be observed if you leave your cells on the charger to trickle charge for an extended period of time.
What about NiMh cells?
The Nickel plate also shows a voltage depression when the cell is conditioned by repeated shallow discharge, but it is not as pronounced as the sintered Cadmium plate. Replacing Cadmium with Metal Hydride eliminates a good portion of the "memory effect," but there is still some left due to the Nickel plate.
Yes, you can show voltage depression with NiMh cells. With clarity and uniformity in mind, let's also call this the "memory effect."
Sanyo shows a way to create the "memory effect" in NiMh cells. It is directly related to discharging to exactly the same voltage which is higher than the 1.0 volts per cell that is considered a full discharge. They show 20 cycles, discharging to 1.2 volts. On the 21st cycle there was a voltage depression as the voltage dropped below 1.2 volts. Things were pretty much back to normal on the 22nd cycle after the discharge to 1.0 volt per cell.
The charge and discharge rate used for the Sanyo tests were a one hour charge, rest for one hour, and discharged in one hour. The charge was terminated using –10 mV as the value for negative delta voltage termination.
This indicates that if you are using your NiMh batteries exactly the same each cycle, and are terminating the discharge at 1.2 volts per cell, you should plan to do a full discharge every 20 cycles.
There you have it… The "memory effect" is actually a form of voltage depression that is very difficult to reproduce under normal use. If you are experiencing difficulties with your batteries, it is probably not due to the "memory effect." If you do happen to produce the "memory effect," it is completely reversible with a discharge/charge cycle.
How can we take better care of our NiMh and NiCd batteries?
New batteries and batteries that have been stored for extended periods of time need to be conditioned, or formed. The standard charge is a good way to do this. The standard charge involves a 16 hour charge at a charge rate equal to one tenth the capacity of the battery.
An alternative to this is to do around 5 cycles using a 2 hour charge and a 2 hour discharge.
Heat is the enemy of Nickel based batteries. In cold weather, their performance drops, but once they are back up to room temperature, the performance comes back. Heat, on the other hand, can cause permanent damage.
Ideally, if your cells never got above 86 F, you would be taking very good care of them. Unfortunately, it is nearly impossible to keep temperatures below 86 F, so the idea is to minimize the heat damage as much as possible. Storing cells in a flashlight on the back window of a car in the middle of summer in the sun, is not the best idea.
Charging is another time when cells can be over heated. At the end of the charge cycle, cells heat up. Battery manufacturers recommend charging at a rate that completes the charge in 1 – 2 hours. If your cells heat up a lot during a 1 hour charge, there may be something wrong with your cells, and you may have to reduce the charge rate to keep them cool.
Ok, cells warm up at the end of the charge, but how hot is too hot? There isn't a specific answer. The RC people want "hot off the charger" performance and tolerate cell temperatures up to 140 F, or maybe a little higher. The charger manufacturers generally have a high temperature cut off at around 130 F. The battery manufacturers show graphs indicating healthy cells should end up at around 104 F or less when performing a one hour charge at room temperature. They go on to indicate a maximum charging cell temperature of 140 F.
If you are looking for an absolute number, it looks like cells should be kept below 140 F. Cooler is better, but it looks like hot is 140 F, or above.
The same goes for discharging. If you discharge at a rate that causes your batteries to heat up to high temperatures, you can expect reduced life from them.
As cells wear out, their internal resistance increases. Cells with higher internal resistance will get warmer during charging. Keep in mind that there are other things going on within a cell that can cause higher temperatures during charging (and use), but usually the cause is a higher internal resistance. If your cells heat up during a one hour charge, you may want to back off to a two hour charge. If they still get hot during a two hour charge, you may want to consider recycling them and getting some new cells.
Nickel based cells are happy when they are being used. Storage is hard on them. When fully charged we have a large number of small crystals whose natural tendency is to join together to form larger crystals. When they do this, some of their ability to give up energy is lost and this contributes to the self discharge of these cells. The large crystals are harder to break down than the small crystals, and this causes a rise in the cells internal resistance. This contributes to voltage depression and sluggish performance. If you reduce the state of charge, this activity is greatly reduced.
Storing cells is a balancing act. The best way to store cells is to have them in a discharged condition and stored at around 40 F. At cooler temperatures, the chemical activity within the cell is greatly reduced, but does not completely stop. You can't just leave them, or they will continue to run completely down resulting in an over discharged state that damages the cell. When you store cells in a discharged state, you need to do a charge/discharge cycle every 30 days to keep the chemistry active. If you want to store for longer periods of time, you can discharge the cells, then put a partial charge back on them.
You can also store cells fully charged. If you do this, keep in mind that it can take up to 3 charge/discharge cycles to get them back up to their initial performance.
To enjoy the best performance and life from your cells, it is not recommended to leave them on a charger trickle charging. Some chargers have a very low maintenance charge that may be OK, but a lot of them have too high a trickle charge rate. Not only will you aid the growth of large crystals, but keeping the cell warm may accelerate electrolyte dry out. If you want to keep your cells ready for use, put your charger on a timer and only charge for a short time each day. This is much better than constant trickle charging.
What about the new Low Self Discharge cells?
These cells don't seem to show any ill effects from being stored fully charged. Early tests indicate that their performance after six months of storage is comparable to their initial performance. It is too early to make any generalizations, but it looks like slowing down the chemical reaction internally within the cell is just as effective as storing a normal NiMh cell in a discharge state.
So, there you have it…
If you are having problems with your batteries, it probably isn't due to the "memory effect." If it is, a simple full discharge will eliminate the problem.
Use your batteries, and if you must store them, store them in a discharge state. Remember to do a charge/discharge cycle periodically on your stored cells, and they will last a long time.
Finally, to the best of your ability, try to keep your batteries cool.
Here are some excellent references on the "memory effect."
References:
Some Ramblings about NiCd Batteries, by Ken Nichimura
http://www.electronics-lab.com/articles/files/ABOUT_NICD_BATTERIES.PDF
Memory Effect in Stationary NiCd Batteries? Forget about it! by Jim McDowall
http://www.battcon.com/PapersFinal2003/McDowallPaperFINAL2003.pdf
Tom
Engineers and scientists huddled together trying to figure out what was going wrong with the power supply on a satellite in orbit around the earth. The battery capacity had been selected so normal use would only require 25% of the total capacity, however, when they tried to use some of the unused capacity, it was not available.
Testing in the laboratory initially failed to reveal similar problems, and it wasn't until they followed the exact regime that the satellite was using that they were able to duplicate the problem.
It appeared that when a sintered cell NiCd battery is conditioned by repeated shallow discharges, it seems to remember that it only has been using a portion of its capacity. Trying to use more capacity results in problems and this became known as the "memory effect."
News of the "memory effect" swept the world and it became the answer to all rechargeable battery problems. I did an informal poll asking people of all ages and walks of life what they thought the main problem was with under performing or dead rechargeable cells. One of the early, if not the first, thing mentioned was the "memory effect." People remember the "memory effect" and still believe that it is a major contributor to rechargeable battery problems.
How do you reproduce the "memory effect?"
The laboratory testing revealed that in order to produce a strong "memory effect" you had to specifically follow this "recipe."
Start with sintered NiCd cells. Other constructions were not as successful in reproducing the "memory effect."
Next, you need to define a strict discharge cycle, which does not fully discharge the cell. Each discharge cycle needs to be within 1 – 2% of each other.
The charge cycle must only recharge the battery to 100%. No overcharging is allowed.
The charge rate must be slow. 10 – 12 hours is a good time frame for charging when you are trying to reproduce the "memory effect."
That's it…
Violating any of these steps reduces the "memory effect" and sometimes can completely eliminate it.
Where did the rest of the capacity go?
The laboratory testing continued and scientists discovered that the remaining capacity was still available, but at a reduced voltage. This was not a "memory effect," but more accurately a "voltage depression." Unfortunately, by the time they came out with this, the term "memory effect" had permeated everything and is still with us today.
Let's take a closer look at voltage depression.
The reason the voltage drops is because of an increase in internal resistance within the cell. Looking for a cause of this increase in internal resistance, scientists dissected a voltage depressed cell to see if they could observe any changes that had taken place.
When cells are new, the crystals on the plates of the cell are around 1 micron in size. The size of the crystals on the plates of the cells that were showing voltage depression were 50 – 100 microns, or larger. These larger crystals resist breaking down to give up their energy. In addition, the longer the larger crystals remain intact, the harder they get, and it becomes even more difficult to break them down.
This form of voltage depression is reversible, as is the "memory effect." Usually a single complete discharge at a low current draw will break down the large crystals and you are back to normal. Sometimes it takes a couple of cycles. It is dependant on the discharge current. Higher currents take more cycles.
Now, how do you know if your battery problems are being caused by the "memory effect?"
Here is a list of questions to help you determine this…
1. Are you using sintered NiCd cells?
2. Have you done several similar (within 1 – 2% of each other) discharge cycles to less than a full discharge?
3. Is your charge limited to 100%? {i.e. no overcharge}
4. Does your charger take 10 – 16 hours to complete a charge?
If your answer to any one of these questions is "NO," your battery problem is probably not caused by the "memory effect."
Most of us are using NiMh cells, but we do have power tools that utilize NiCd cells, and some of those may utilize sintered construction. There is a possibility that you are using these cells in your power tools. If you have a turbine engine aircraft, you are most likely using a vented sinter plate NiCd battery.
The next point is difficult to do. A variation of more than 1 – 2% in the duty cycle drastically reduces the "memory effect." In normal day to day use, it is very difficult to adhere to this tight of a schedule. It is possible, but extremely difficult to do.
Next comes charging. If you are using a charger that terminates using negative delta voltage, you are applying a slight overcharge every time you charge. If your charger utilizes a top off charge to bring the cells up to full capacity, you experience a small overcharge every time you charge. The scientists studying the "memory effect" initially were unable to reproduce it because they fully charged the cells after each shallow discharge and the slight overcharge that is involved in normal charging erased the "memory effect." If you are using a modern charger, this point will be the stumbling block to developing the "memory effect" in your cells.
Finally, how long does it take to recharge your cells? If you are following the battery manufacturers recommendations and are charging in 1 – 2 hours, you minimize the production of the memory effect in NiCd cells. However, a similar effect can be observed if you leave your cells on the charger to trickle charge for an extended period of time.
What about NiMh cells?
The Nickel plate also shows a voltage depression when the cell is conditioned by repeated shallow discharge, but it is not as pronounced as the sintered Cadmium plate. Replacing Cadmium with Metal Hydride eliminates a good portion of the "memory effect," but there is still some left due to the Nickel plate.
Yes, you can show voltage depression with NiMh cells. With clarity and uniformity in mind, let's also call this the "memory effect."
Sanyo shows a way to create the "memory effect" in NiMh cells. It is directly related to discharging to exactly the same voltage which is higher than the 1.0 volts per cell that is considered a full discharge. They show 20 cycles, discharging to 1.2 volts. On the 21st cycle there was a voltage depression as the voltage dropped below 1.2 volts. Things were pretty much back to normal on the 22nd cycle after the discharge to 1.0 volt per cell.
The charge and discharge rate used for the Sanyo tests were a one hour charge, rest for one hour, and discharged in one hour. The charge was terminated using –10 mV as the value for negative delta voltage termination.
This indicates that if you are using your NiMh batteries exactly the same each cycle, and are terminating the discharge at 1.2 volts per cell, you should plan to do a full discharge every 20 cycles.
There you have it… The "memory effect" is actually a form of voltage depression that is very difficult to reproduce under normal use. If you are experiencing difficulties with your batteries, it is probably not due to the "memory effect." If you do happen to produce the "memory effect," it is completely reversible with a discharge/charge cycle.
How can we take better care of our NiMh and NiCd batteries?
New batteries and batteries that have been stored for extended periods of time need to be conditioned, or formed. The standard charge is a good way to do this. The standard charge involves a 16 hour charge at a charge rate equal to one tenth the capacity of the battery.
An alternative to this is to do around 5 cycles using a 2 hour charge and a 2 hour discharge.
Heat is the enemy of Nickel based batteries. In cold weather, their performance drops, but once they are back up to room temperature, the performance comes back. Heat, on the other hand, can cause permanent damage.
Ideally, if your cells never got above 86 F, you would be taking very good care of them. Unfortunately, it is nearly impossible to keep temperatures below 86 F, so the idea is to minimize the heat damage as much as possible. Storing cells in a flashlight on the back window of a car in the middle of summer in the sun, is not the best idea.
Charging is another time when cells can be over heated. At the end of the charge cycle, cells heat up. Battery manufacturers recommend charging at a rate that completes the charge in 1 – 2 hours. If your cells heat up a lot during a 1 hour charge, there may be something wrong with your cells, and you may have to reduce the charge rate to keep them cool.
Ok, cells warm up at the end of the charge, but how hot is too hot? There isn't a specific answer. The RC people want "hot off the charger" performance and tolerate cell temperatures up to 140 F, or maybe a little higher. The charger manufacturers generally have a high temperature cut off at around 130 F. The battery manufacturers show graphs indicating healthy cells should end up at around 104 F or less when performing a one hour charge at room temperature. They go on to indicate a maximum charging cell temperature of 140 F.
If you are looking for an absolute number, it looks like cells should be kept below 140 F. Cooler is better, but it looks like hot is 140 F, or above.
The same goes for discharging. If you discharge at a rate that causes your batteries to heat up to high temperatures, you can expect reduced life from them.
As cells wear out, their internal resistance increases. Cells with higher internal resistance will get warmer during charging. Keep in mind that there are other things going on within a cell that can cause higher temperatures during charging (and use), but usually the cause is a higher internal resistance. If your cells heat up during a one hour charge, you may want to back off to a two hour charge. If they still get hot during a two hour charge, you may want to consider recycling them and getting some new cells.
Nickel based cells are happy when they are being used. Storage is hard on them. When fully charged we have a large number of small crystals whose natural tendency is to join together to form larger crystals. When they do this, some of their ability to give up energy is lost and this contributes to the self discharge of these cells. The large crystals are harder to break down than the small crystals, and this causes a rise in the cells internal resistance. This contributes to voltage depression and sluggish performance. If you reduce the state of charge, this activity is greatly reduced.
Storing cells is a balancing act. The best way to store cells is to have them in a discharged condition and stored at around 40 F. At cooler temperatures, the chemical activity within the cell is greatly reduced, but does not completely stop. You can't just leave them, or they will continue to run completely down resulting in an over discharged state that damages the cell. When you store cells in a discharged state, you need to do a charge/discharge cycle every 30 days to keep the chemistry active. If you want to store for longer periods of time, you can discharge the cells, then put a partial charge back on them.
You can also store cells fully charged. If you do this, keep in mind that it can take up to 3 charge/discharge cycles to get them back up to their initial performance.
To enjoy the best performance and life from your cells, it is not recommended to leave them on a charger trickle charging. Some chargers have a very low maintenance charge that may be OK, but a lot of them have too high a trickle charge rate. Not only will you aid the growth of large crystals, but keeping the cell warm may accelerate electrolyte dry out. If you want to keep your cells ready for use, put your charger on a timer and only charge for a short time each day. This is much better than constant trickle charging.
What about the new Low Self Discharge cells?
These cells don't seem to show any ill effects from being stored fully charged. Early tests indicate that their performance after six months of storage is comparable to their initial performance. It is too early to make any generalizations, but it looks like slowing down the chemical reaction internally within the cell is just as effective as storing a normal NiMh cell in a discharge state.
So, there you have it…
If you are having problems with your batteries, it probably isn't due to the "memory effect." If it is, a simple full discharge will eliminate the problem.
Use your batteries, and if you must store them, store them in a discharge state. Remember to do a charge/discharge cycle periodically on your stored cells, and they will last a long time.
Finally, to the best of your ability, try to keep your batteries cool.
Here are some excellent references on the "memory effect."
References:
Some Ramblings about NiCd Batteries, by Ken Nichimura
http://www.electronics-lab.com/articles/files/ABOUT_NICD_BATTERIES.PDF
Memory Effect in Stationary NiCd Batteries? Forget about it! by Jim McDowall
http://www.battcon.com/PapersFinal2003/McDowallPaperFINAL2003.pdf
Tom
