Raymond UK":3869j2g1 said:
It's sometimes called memory effect (sometimes battery effect) and is derived from cyclic memory. The battery seems to remember the smaller capacity.
It's an urban myth actually. It's true that there
is a memory effect, but it's only been observed in communication satellite batteries (which used to be NiCd and may still be). Comms satellites are usually geostationary, meaning they are on an equatorial orbit, with a period of approx 24 hours, so they appear stationary in the sky when viewed from the Earth's surface.
The cell memory issue was caused because they went through extremely regular charge-discharge cycles as they orbited through the day and night sides of the planet. Here on the ground it's almost impossible to replicate that usage model (outside a lab).
There are two other issues, however, that do affect NiCds:
The first is poor cell matching in batteries and it applies to all rechargeable chemistries really. Each cell in a battery needs to have identical electrochemical performance to the others, so that they can be charged and discharged as a set*. In use, if there is a weaker cell, it will flatten first and then become a load on the rest of the battery (and the cell itself will suffer reverse charge to an extent, which will damage it further.
Once it's weakened, a normal charging routine will cause it to be overcharged before the rest of the cells are completely charged. Again this damages it further (water is electrolysed to hydrogen and oxygen, which are vented and don't recombine, so the cell slowly dries up). Once one cell dies, the whole battery usually becomes useless.
The second issue is dendrite formation. These are spiky crystals that grow between the plates in the cell. They are conductive, and will short out the cell once they get long enough.
It's often recommended that NiCd cells are stored in a fully discharged state, but I've used good quality cells (Saft) where the manufacturer specifically recommends not fully discharging them (I suspect this is because of the small risk of reverse polarization).
In both cases, the battery almost always suffers single-cell failure.
I'd expect that flattening them through an electric motor (the on-site behaviour you mention) would actually be counter-productive, as you'd wear out the battery more than you'd protect it.
The three reasons why Lithium ion batteries have taken over are:
1. better energy density overall, so more energy stored in a given space/weight,
2. a higher cell voltage (3.6V instead of 1.2V for NiCd), so fewer cells are needed,
3. Cadmium isn't used, so the cells are (theoretically) less polluting. In practice it's not quite that simple...
NiCds are generally more robust than Lithium ion, and don't usually catch fire or explode when damaged. "Wet" versions are still used in industrial applications, as they can be maintained (you add distilled water typically, much like a car battery). I'd happily re-cell my older NiCd batteries (I've done it a lot in the past), but I'd think twice about trying it with a lithium ion one.
HTH, E.
*I remember one very, very expensive
battery system in the 1980s, which used a microcontroller in the battery to charge and monitor the individual cells separately. It worked extremely well, but IIRC, each battery pack cost over 1,000 quid and was for a very specialist, very high current application.
) on charge for the last 25 years. I don't recall reading that you shouldn't.
