Dan's Quick Guide to Memory Effect, You Idiots
(with apologies to Stephen Notley)
Originally published 2002 in Atomic: Maximum Power Computing Last modified 03-Dec-2011.
I have, upon occasion, heard people explain that the nickel metal hydride (NiMH) battery of cordless drill X is superior to the nickel cadmium (NiCd) battery of cordless drill Y, because NiMH has no "memory effect", but NiCd does.
I have heard other people talk about how the lithium ion battery in their laptop's better than NiMH because lithium ion has no memory effect, but NiMH does.
I have not yet heard anybody extolling the virtues of running your cordless drill from a truck battery, because lead acid batteries don't have memory effect but everything else does, but I'm sure someone has. Probably someone with an impressive collection of trusses.
People have all kinds of strange religious beliefs about the rituals that must be performed to exorcise the Memory Monster. If they were just dancing around their laptop, MP3 player or cordless drill waving incense sticks then I wouldn't care. But they're usually doing unnecessary charge/discharge cycles. Which is bad.
If you flatten a battery before you recharge it - some people flatten their batteries manually, and some chargers do it automatically - you greatly reduce the life of the pack. A given battery pack may last for 500 full cycles, or 2000 partial ones. If you're fully flattening the battery by actually using it, then fair enough. But flattening it as part of the recharging process is goofy.
"Memory effect" is now used as a general term for anything that makes a battery not deliver its full capacity. What the term originally referred to, though, is a phenomenon that's probably never actually been observed in consumer hardware.
True memory only happens in sintered plate NiCd cells (which aren't necessarily the sort of NiCd you're using, and are of course completely different from any kind of NiMH cell), and it only happens when you precisely discharge a cell to exactly the same level over and over again, and recharge it without any overcharge. True memory effect happens in satellite power systems, electronics test labs, and practically nowhere else.
Cheap trickle chargers always overcharge if you leave them long enough, and quality consumer NiCd chargers also slightly overcharge, because the slight voltage drop that happens when you do that is what they use to pick the end of the charge cycle. So even if you're using sintered plate NiCds, which you're probably not, your charger will cure memory effect anyway.
What people nowadays call memory effect is a combination of two things.
One - cell aging. Batteries don't last forever. The older they get, the less capacity they have. Live with it.
Two - "voltage depression". Voltage depression is a problem with NiCd batteries and, according to some sources, also with NiMH, and it doesn't affect the battery capacity much at all. Rather, the battery voltage drops unusually quickly as it discharges. Gadgets that monitor their battery voltage therefore think the battery's flat earlier than they should. There may be lots of capacity left at the slightly depressed voltage, but the gadget doesn't know that and flashes its "low battery" warning.
Fully discharging cells cures voltage depression, but if you fully discharge a whole battery then the stronger cells in the battery will "reverse" the weaker ones. The weaker ones go flat first, and then get charged backwards by the others. This is very bad for the weaker cells, and will kill a battery pack quick smart. So don't do it.
Fancier discharging hardware lets you set a voltage to discharge to - say, 0.9 volts per cell. A NiCd or NiMH cell that's down to 0.9 volts under moderate load has practically no charge left; it's very nearly dead flat. But stopping the discharge at that point, rather than letting the pack slump down to zero volts, should save weak cells in the pack from any significant reversal. Well, unless they're so weak that the pack's toast anyway.
There's still not a whole lot of point to doing this, though, unless you've got a scientific battery care regimen and want to start every charge from a precisely known state.
Radio control enthusiasts often cell-by-cell discharge their battery packs to nothing, in order to be able to achieve the absolute maximum super-punchy charge. But they often then go on to use that whole charge up in five minutes or less. Sometimes much less. If you've never welded a battery connector through overcurrent, then you're not in that class.
When people cycle their batteries for no reason and kill them early, they have a problem. Rechargeable batteries are hazardous waste. In most places, including here in Australia, you're not allowed to just chuck them in the bin. Instead, you're meant to take your dead batteries to a recycling centre.
Which, helpfully, doesn't exist.
Various places that sell lead acid batteries will accept them for recycling, but if you want to get rid of ordinary loose cells or consumer battery packs that use the other rechargeable chemistries, there's nowhere in Australia to go. The government requests that you just put your dead rechargeables on the mantlepiece, for the time being.
If you don't want to do that, and also don't want to break the law and make Australia's landfills even nastier than they already are, then the best idea I've been able to come up with is to subvert the Mobile Phone Industry Recycling Program. They send batteries overseas for recycling, and they're only meant to take old phones and their batteries, but I don't believe they have armed guards on their recycling bins. Since one NiCd or NiMH or LiI is much like another, you could just drop the lot in the phone battery bin when nobody's looking.
If you get busted, though, I never met you.
References
The sci.electronics FAQ on NiCd memory
The full repairfaq.org NiCd Battery FAQ
