Nickel Cadmium Batteries - The Truth About Nicads

This article is primarily directed at those who use nicad batteries in their portable PCs and mobile phones. However, the information which follows is equally relevant to nicads used in camcorders and other domestic appliances.

NOTE: None of the the following information applies to Lithium-Ion batteries.

Nickel Cadmium batteries are here to stay with us for a while yet so most of you will use them sooner or later. The nickel-cadmium cell was invented by the Germans during the last war. They were first produced by Deutsche Akkumulatoren Gesellschaft and were originally known as "Deacs" (from DeAk) in the same way as vacuum cleaners are still referred to as "Hoovers". (Note: One unit is a cell, two or more cells make up a battery.)

At the end of the war, the patent for the nickel-cadmium cell was offered as a part of German war reparations but the patent was rejected by the United States Patent Office because it was presumed to be much the same as the NiFe (nickel-iron) cell invented by Thomas Edison 50 years earlier. The NiFe cell was a poor second to the lead-acid cell as an accumulator because it would not hold a charge very well and would only generate 1.2 volts in comparison to the 2.0 volts of a lead-acid cell. However, a NiFe cell is virtually indestructible. It has steel plates and a steel case. If a cars were fitted with a NiFe batteries instead of lead-acid batteries, the battery manufacturers would soon be out of business. A NiFe battery could outlast several cars. Lead-acid cells have plates which are less robust than papier maché‚.

The original nickel cadmium cell had a big advantage over the conventional lead-acid and NiFe cells - it could be sealed because it didn't produce gas when being charged. (The sealed lead-acid cell was not invented then.) It also had the advantage that it could withstand a very heavy discharge rate without being damaged. This second factor has made model electric car racing possible; fully charged batteries are usually discharged within four minutes. Lead-acid batteries would be ruined by such treatment.

The sealed nickel-cadmium cell has three disadvantages when compared with a 2 volt lead-acid cell. A lead-acid cell voltage drops from 2.1 volts to 1.9 volts during 95% of its discharge cycle. A nicad generates only 1.25 volts on average, but its output voltage falls from 1.35 volts to 1.1 volts during discharge and it weighs more for the same ampere-hour capacity. On the other hand a nicad cell will last two or three times as long as a lead acid cell if it is treated correctly. (Some cells I use in my torch are over 20 years old.)

A modern NiCad cell is made like a double swiss roll inside. (Imagine chocolate and lemon sponge layers rolled together with cream in between.) Both 'plates' are made of steel. The cadmium plated cathode is an open net and the nickel plated anode is a fine woven mesh. The two plates are separated by a thin layer of porous plastic which acts like blotting paper. The separator is soaked in a potassium hydroxide solution. All nicads are fitted with a seal which acts as a safety valve to release gas generated within the cell. The amount of cadmium used is very small (it is just a coating) so the cell is not as unfriendly to the environment as some people believe. (Cadmium is widely used as an orange/yellow colouring agent for plastics and as a coating for screws.) The construction of a nicad cell should be borne in mind because it affects the response of the cell when it is misused.

Cadmium from the cathode is converted into cadmium hydroxide which goes into solution as the cell is discharged. The cadmium is plated back on the cathode when the cell is charged up again. It sounds simple but the charging/discharging process does not restore the cathode to its original state unless certain conditions are met. A slow discharge will remove cadmium from the cathode in a fairly even manner. A slow charge will cause cadmium to be more heavily plated on the parts of the cathode which are closest to the anode. This effect creates whiskers of cadmium which can grow during repeated charge cycles through the separator until they touch the anode. When whiskers have crossed the separator, the cell is shorted so that it cannot be charged any more. When charging stops, the whiskers are eroded a little by self discharging so that the short is removed but the cell ends up with a very much reduced capacity.

Both hydrogen and oxyen gases are generated through charging when a cell is already charged but they can recombine again to form water if the charge rate is low enough. This trick is achieved by having an anode which is much bigger than the cathode. If the cell is overcharged on a fast charger, the two gases will be generated more quickly than they can recombine. The safety valve releases the gas pressure but water is lost and the cell's capacity is reduced as a result. If the cell is overheated by very fast charging and discharging, steam can be generated and lost through the safety valve, and the plastic separator may melt. Very slow charging and discharging can cause the electrolyte to stratify so that its chemical activity is reduced. This gives the effect of reduced cell capacity. This problem also occurs if the cell is left unused for a long time. Stratification can be reduced and the cell's capacity restored by several cycles of fairly fast charging and discharging.

I have deliberately referred to cells so far because problems arise with batteries of several cells. No two cells in a battery are identical so some will have higher capacities than others. In a battery of cells, the battery will be effectively flat as soon as the weakest cell is discharged. This will leave most cells with some charge left in them.  If the battery is further discharged, the weakest cell will be partially charged in reverse. Cadmium will be deposited on the anode and removed from the cathode. When the battery is charged again, the reverse charging of the weakest cell has to be overcome before proper charging can commence again. As a result the weakest cell is frequently undercharged and the process snowballs.

IT IS THEREFORE VERY STUPID TO FLATTEN NICAD BATTERIES COMPLETELY!

It shortens battery life and reduces total capacity. I don't know where the idea originated that nicads should be totally flattened. This notion has caused many otherwise salvageable batteries to be ruined. Whoever first published the idea that flattening NiCad batteries as a "good" thing should have all the ruined nicads heaped upon his head. I trust that this explanation of nicad battery operation will convince users that nicad batteries should never be discharged completely. Individual cells can be flattened but I fail to see any advantage in doing this anyway.

Nicads thrive on heavy use. They are ideal for electric screwdrivers and model cars. Careful fast charging and rapid discharging prevents stratification and reduces the rate of growth of whiskers. The ideal nicad charger is one which charges for 90% of the time and discharges for 10% of the charging time. The easiest way to do this is to employ a 'leaky' half wave rectifier in the charger unit. (Few commercial chargers employ this technique. If they did, the electricity suppliers would soon object.) On one half cycle the rectifier passes current normally and the battery is charged. On the other half cycle, the rectifier prevents current flow but a bypass resistor allows a small reverse current to flow and discharge the battery slightly. This charging method prevents the growth of whiskers altogether. Stratification can be avoided by charging the battery fast for the first 2/3 of the charge and charging normally for the final 1/3 of the charge. Nicads are made so that they can be charged indefinitely at the normal rate so it is possible to restore the charge in the weaker cells in a battery without damaging the cells that are already charged. The charging instructions on the side of every nicad indicate that the cell should be given a charge which is 40% more than the cell's capacity. Part of this is because of the inefficiency of the charge/discharge process and part is to help to prevent the weaker cells in a battery from being undercharged.

Some more advanced chargers include heat sensors. These detect when cells in a battery start to warm up. The conversion of electrical energy into a chemical change generates very little heat so charging cells stay cool. Charged cells have no more chemical to change so all the electrical energy is turned into heat and the cell gets warmer. A charged cell can accommodate the heat generated during normal charging but it rapidly deteriorates if it is overheated by fast charging for the reasons mentioned earlier.

Those of you who have the requisite electrical knowledge who would like to make up a whisker- preventing charger for their nicads can base the design on the following rules: (The discharge current should be 1/10 of the charge current.)

1. The transformer AC output voltage should be around 1.5 to 2 times the charged battery voltage.
2. Cells should be normally charged at 1/10 of their capacity rating. e.g. A 2Ah cell should be charged at 200mA or 0.2A and discharged (during charging) at 20mA or 0.02A
3. The rectifier diode should be rated at 5 to 10 times the normal charge rate.
4. The series charging resistor value Rc is calculated as follows: (V = volts, I = amps, W = watts).
Rc = (V transformer - V battery)/(I charge X 2)   Wattage W => I X R
(The current I charge is doubled because charging only takes place during forward 1/2 cycles.)
5. The rectifier shunt discharge resistor Rd is calculated as follows:
Rd = (V transformer + V battery)/(I discharge X 2) Wattage W <= 1W
6. These rules also apply to fast charging at 10 times the normal charge rate. Fast charging should be carefully monitored and stopped as soon as cell warming is detected.

As an example I will show how the values of the components are calculated.

Let us assume that you want to charge a set of four AA size nicads at fast and standard rates.
Each cell will have up to 1.5 volts across it during charging making 6 volts for the charging battery. The transformer output AC voltage could therefore be 12 volts.
The normal charge rate for 800mAh AA cells is 80 mA. The fast charge rate can be up to 800mA.
The charging voltage is effectively 12 volts - 6 volts = 6 volts.
The normal charge resistor would be 6/0.08 = 75ohms for full wave charging but 37.5 ohms for the half wave charging required.
The fast charge resistor is therefore 1/10 of the normal charge resistor and is 7.5 ohms for full wave and 3.75 ohms for halfwave..
The normal charge discharge current is 8mA and the fast charge discharge current is 80mA.
The discharge voltage is 12 volts + 6 volts = 18 volts.
The normal discharge resistor is therefore 18/0.008 divided by 2 = 1125 (1.1k) ohms. The fast charge discharge resistor is 110 ohms.
The rectifier diode should be rated at up to 10 times the fastest charge rate. A 1N5401 diode will suit. (It is quite common.)

The above example has a major shortcoming. A 1N5401 diode has around 0.5 volts across it when it is conducting. This reduces the charging voltage to 5.5 volts. In practice, a nominal 12 volt transformer will have an output that is slightly higher than 12 volts. The very small transformers that are used in many plug-in adapters usually have a much higher voltage output at low currents. Note that plug-in adapters normally use full wave rectification and are therefore unsuitable for use for a "Dirty DC" charger.

A charger can be built using car bulbs as charging resistors. These have the advantage of showing how the charge is progressing and they can easily handle the heat dissipation problem. To calculate the resistance of a 12 volt car bulb use the formula: 144/wattage (V^2/W=R) A 6 watt tail light bulb = 144/6 = 24 ohms. A 21 watt stop light bulb = 144/21 = 6.8 ohms. Do not worry if the bulb resistances do not quite match the charging resistance required. It is best to have bulbs with voltage ratings as high as the transformer output voltage to prevent them burning out if the output is shorted. Two similar bulbs in parallel have half the resistance of one bulb. Two bulbs in series have twice the resistance of a single bulb and will withstand twice the voltage.


A Basic "Dirty DC" Nicad Charger


Basic "Dirty DC" Nicad Charger Circuit
(This is for European power supplies. US & Canada use 110VAC supplies)

Using Lamps as Charging Resistors
Lamps have the advantage that they can dissipate the heat generated more easily than conventional resistors.
Resistances for 12 volt Car Lamps
2.2W 65ohms - 2.5W 57ohms - 3W 48ohms - 6W 24ohms - 18W 8ohms - 21W 6.8ohms - 36W 4ohms - 48W 3ohms
Two or more lamps in parallel or series can be used to obtain other useful values. It is advisable to make sure that the lamp voltage is at least as high as the transformer voltage. Use two identical lamps in series if the transformer voltage is greater than 12 volts. This will prevent lamp burnout if the output is accidentally short circuited.

If a laptop computer or video batttery pack seems to have lost capacity, charge it normally then discharge it with a car trafficator 21W bulb. (Use a tail lamp 6W bulb for mobile phone batteries.) Stop discharging as soon as the light output starts to dim. A good discharge rate is 10 times the normal charge rate. Repeat the charge/discharge process several times but let the battery cool off if it shows any signs of getting warm. Most battery packs are fitted with a safety fuse. If you short circuit a charged battery pack, the battery pack has had it unless you can repair a plastic container which has been broken open to replace the fuse.

A fully charged nicad cell will show just over 1.45 volts across its terminals while it is being charged at the normal rate. This voltage will fall to around 1.38 volts after being disconnected from a charger for half an hour. It will fall further to around 1.30 volts within a week without being used. A cell is effectively flat when the voltage has fallen to 1.15 volts. These voltages should be used to check if all the cells in a battery are working OK. A five cell battery should show around 6.9 volts when it is disconnected from the charger. If it shows 5.5 volts, one cell is dead. It is seldom possible to restore dead cells to normal working. A battery which contains one dead cell is effectively useless. If the cell is replaced, the new cell will still be working when another of the older cells has died.

If you need a new nicad battery, check with third party suppliers to get a better deal than the appliance manufacturer offers. The most popular size of cell used in bigger battery packs is the R size which is rated at 1.8 ampere hours. If you need a new battery pack, check to see if there is another make of battery which has the same voltage and ampere hour rating at a much lower price. If the shape of the pack is different, use the cells from the new pack to replace the cells in the original pack. Solder can be used to attach connections to the welded-on strips but never try to solder directly to the body of a cell. If the solder takes, the cell will have been ruined by overheating.

If the nicad pack in your portable computer will not last long enough for your needs, obtain a larger one of the same voltage and obtain a connector plug which is identical to the one on the original battery. Attach the connector plug to a suitable piece of twin cable and connect the cable to the additional battery. Make absolutely sure that the polarity is correct before plugging the connector into your computer. The external nicad will normally operate the computer in the same way as the internal battery or the mains power unit. The computer will now be a bit less portable but it will run it for hours without the battery going flat.

If you can use a soldering iron for electrical work, you should be competent enough to organise an additional battery supply. If you do not know the relevance of 60/40 solder, do not on any account play with your portable computer's power supply system. Get someone who is competent to do it for you.

ADDITIONAL NOTES IN RESPONSE TO EMAILS
"Memory" and Stratification

The so-called "memory" of NiCd cells is in fact no such thing. (This not the "whiskering problem that needs a different solution.) You have to remember that the electrolyte in a nicad cell is in a state that is somewhere between a jelly and a paste. In other words it does not move around much. As a result a form of stratification takes place where the more active chemicals in the electrolyte remain close to the anode and the less active chemicals remain close to the cathode. This is the opposite of what is needed.

This phenomenon has been known since NiFe cells came into wider use. They have free flowing potassium hydroxide solution as an electrolyte. I remember from my dim and distant days in the RAF more than 40 years ago when I was told by an electrician how NiFe batteries were treated in the battery charging room. The term he used was "lazy" for cells that did not hold the full charge. The cure for this included physical shaking of the battery to stir up the electrolyte followed by several fast charging and fast discharging cycles.

A NiCd cell is the direct descendant of the NiFe cell and suffers from many of the faults of the NiFe cell. Its main advantage is that it can be sealed. This is possible because the gases liberated during charging recombine to form water if the charge rate is not too high. As a NiCd cell is sealed, it is impossible to do anything to stir up the electrolyte physically. All that can be done is to give it a series of vigorous charge and discharge cycles - preferably using a "dirty DC" charger. However, all charging and discharging should be stopped as soon as the cell becomes noticeably warm.
I hope that this makes things a bit clearer with regard to the alleged "memory" problem.

NiMH cells must be treated differently. They do not like being repeatedly charged when they have not been discharged. The cell will rapidly lose capacity if it is repeatedly charged without being NORMALLY discharged. NEVER short NiMH cells.

Whiskering
The whiskers that grow from the cathode through the insulation material to the anode can limit the charge that a cell will accept. A very high current of very limited duration can vapourise these whiskers and stop them from shorting the cell when it is only partly charged. A whiskered cell will never pass the charge test described below. Whiskering can sometimes be cured by using a large (2000uF) capacitor that is charged to around 3 times the normal cell voltage. The charged capacitor is connected across the terminals of the charged whiskered cell. The cell is then charged normally to see if its capacity has improved. This process can be repeated after charging again if the capacity has not been improved. This will not work with cells that are flat. It only works if the whiskers are limiting the amount of charge.

How long will Solar Lights Shine?
The amount of charge that you can put in the solar light's NiCd cell is limited by the brightness of the sunshine, the solar cell capacity and the day length.

Solar lights use an electronic circuit to step up the 1.25 volts (average) of a NiCd or NiMH cell to around 2.5 volts or more to run a LED that takes at least 1.8 volts to work. White LEDs take over 3 volts to work.

You could find out what power is available by monitoring the power available from the solar cell on a bright day. Use a 3.5V torch bulb in place of the NiCd cell to measure the voltage available and the current taken. Then repeat the tests again on a dull day. You will be surprised at the difference. Then you could estimate the
average amount of power available during an average day. Multiply that by 0.7 and you will get some idea of how much charge the cells are getting. The amount will be the current available when the voltage is greater than around 1.6 volts across a bulb times the number of hours of sunshine times 0.7.

NiCd and NiMH cells have the advantage that they can be charged and discharged rapidly. They are poor at slow rate discharges. NiCd cells became popular because they can be charged gently indefinitely without gassing. A typical charging arrangement is to charge the cells for 1.4 times their capacity. A 1 Ah cell would normally be charged at 100mA for 14 hours but could be charged for 20 hours or more at 100mA without damaging the cell.

Changing from NiCd to NiMH will only be of significant use if the solar charging capacity during an average day greatly exceeds the capacity of the NiCd cell. A typical NiCd cell has a capacity of between 600 and 800mAh. NiMH cells have capacities from 1,300mAh to 2,600mAh. I doubt if the solar cells in your units could provide more than 1,000mAh during the average day. This will provide you with around 700mAh of power. If the LEDs use around 200mA with the step up circuit, just over three hours running time is about all that one could expect.

AAA Cells

AAA cells vary in capacity. A typical NiCd AAA cell has a capacity of around 220mAh. That means it should be charged at the basic 1/10C rate of 22mA for 14 hours. NiCd cells will normally accept indefinite charging at the 1/10C rate. A faster charge (from flat) could be at the 1C rate for 50 minutes followed by 5 hours at the 1/10C rate. Any attempt to try to charge a NiCd cell completely at a high rate is doomed to failure. It is always best to do a 2/3 charge at a high rate followed by a 1/3 charge at the 1/10C rate.

Checking Nicad Cells and Batteries With a Digital Meter
Beware of checking the charge on NiCd cells with a modern digital multimeter. The voltage indicated by the meter alone at the Volts DC setting will give a poor indication of the state of charge. Always use a shunt resistor across the meter terminals to check the voltage of a cell on load. A torch bulb or a car lamp bulb is good for this purpose. A fully charged cell should show at least 1.3 volts for more than 10 seconds when being discharged at the 1C rate one hour after charging has finished. A meter with a "Battery Test" position is better than one without unless you use a shunt resistor. Alternatively you could construct you own cell testing meter from a milliammeter with appropriate series and shunt resistors for cells of varying capacities.

Loss of Electrolyte
Many Nicads lose capacity because of loss of water. This happens when a Nicad get hot and steam is released through the safety vent or a cell is charged too quickly and the gases generated escape through the vent. Electrolyte loss can be checked by weighing a suspect cell and comparing the weight with a known good cell of the same make and capacity. The water in a cell represents around one fifth of the total weight of an AA cell. A cell that weighs only 80% of the weight of a good cell has probably lost most of its water.

Salt Deposits
Older Nicads often show a white deposit around the vent. (Normally this is near the positive terminal.) This deposit is potassium carbonate. This occurs usually because the safety vent has opened at some time to release gas pressure because of overheating or overcharging. The vent then does not seal properly and a small amount of potassium hydroxide escapes. The potassium hydroxide quickly changes into potassium carbonate by taking carbon dioxide out of the air. The potassium carbonate can be removed with a damp cloth and drying the cell afterwards. The cell should be weighed to check for electrolyte loss before returning it to normal service.

"Good" Makes of Nicad
I have found that Uniross and Yuasa Nicads (and NiMH cells) aregenerally very good. Some other makes are not so good. The quality of manufacture can be estimated by the voltage shown on a digital multimeter (without a shunt) when a new cell has just been fully charged. Yuasa cells often reach 1.44 volts immediately after charging. Uniross cells usually manage 1.42 volts. Lesser makes struggle to reach 1.4 volts.

How to Check the "Goodness" of Your NiCads
A modern digital multimeter on its own is very poor at indicating the health of NiCd cells. The best way to check is to use the digital multimeter to measure the voltage while a significant current is being drawn from the cell. This should be done by connecting something like a 2.5V torch bulb from positive to negative at the same time as the meter.
If you know the resistance of the torch bulb you can work out the current drawn by the bulb from the voltage shown on the meter. If you then measure the voltage of the cell without the torch bulb connected you will get a higher voltage reading. Then you must do some arithmetic to work out how good the cell is. I will give you an example.

Let us say that the cell voltage reads 1.27 volts with the bulb connected and 1.31 volts without the bulb connected. The bulb resistance is 8.3 ohms (2.5V 0.3A) . This results in a current of 153mA at 1.27 volts. The voltage difference with and without the bulb is 0.04 volts. This gives a cell internal resistance of 0.26 ohms. (0.04/0.153 = 0.26) The higher the cell resistance, the worse the cell. The lower the resistance, the better.

If you check each cell in this way you will find the internal resistances of all the cells. These resistances will give an indication of capacities of the cells. The higher the resistance, the lower the capacity. You must also remember that the capacity of a NiCd cell varies with the way that you use it. A fully charged cell discharged at 0.1C (C/10) will be flat after 10 hours. At 0.2C (C/5) the cell will be flat after 4 hours 10 minutes. At 1C it will be flat in 33 minutes.

Let us assume that we have a pack of four cells with a capacity of 1Ah. Let us assume that the internal resistance of each cell is 0.25 ohms. This gives a pack resistance of 1 ohm. The nominal voltage of the pack is 5 volts. If the pack is discharged at 0.1C, the resistance of the load is 50 ohms. The internal resistance of the pack at 1 ohm is only 2% of the load resistance. At 0.2C the resistance of the pack is 4% of the load of 25 ohms. At 1C the resistance of the pack is 20% of the load of 5 ohms. As a very rough and ready guide, the resistance of a battery pack in relation to the load resistance is an indication of the capacity loss
to be expected with various loads. The charts below have been nicked from a well known NiCad manufacturer's website.

Now you know how to test the cells for their internal resistances, and you know what current is drawn when you test your battery pack's capacity, you can see if the loss of capacity that you measured is related to the internal resistance of the battery pack.

NiCad Characteristics
           
           

Miscellany Page
Buying a Printer

Cleaning Printer Printheads
The Essential Bits and the Less Essential Bits
How It works
Windows 3.11 for Workgroups, Windows 95, Windows 98, Windows 2000 and Linux
Nickel Cadmium Batteries
Winemaking


-- Thank you for visiting Wilf's Nicad Page. --

Please send your comments to: Wilf James - wilf dot james at ntlworld dot com