Newsgroups: sci.electronics.components
From: Henry Spencer <henry@zoo.toronto.edu>
Subject: Re: NiCd Memory
Date: Fri, 10 Oct 1997 12:53:09 GMT

In article <876445141.30318@dejanews.com>,
 <c9304617@alinga.newcastle.edu.au> wrote:
>      I've heard aboout two different sorts of memory in a NiCd battery.
>One is by cycling WITHOUT overcharging and only occurs after many hundred
>cycles, and can be wiped out simply by discharging to 1V per cell or
>overcharging...

Basically correct -- although as far as I know, a discharge is the only
way to get rid of it -- but this effect, the true memory effect, occurs
*only* when the battery is repeatedly discharged to *exactly* the same
point.  Satellite batteries are about the only real batteries whose
charge/discharge cycle is controlled precisely enough to exhibit genuine
memory effect.

>      The other memory I have heard of is caused by cumulative overcharge
>and is called voltage depression and can occur in only about 50 cycles.

This is not memory effect at all.  The battery still retains essentially
all of its capacity; it merely delivers it at a slightly lower voltage.
It's often mistaken for memory effect because a lot of stupid circuits try
to sense the charge state of NiCds by measuring their voltage -- a poor
method at best, since NiCd voltage curves are pretty flat.  Much of the
folklore about NiCd memory comes from stupid circuits incorrectly deciding
that a battery is empty when in fact it's nearly full but suffering from
voltage depression.

>...It too can be wiped out by cycling to 1V per cell.

Note:  NiCd discharging should be done on the individual cells, not on the
battery as a whole.  If you try to discharge the battery as a whole, it's
quite likely that some cells will hit full discharge a bit early, and the
remaining ones will then reverse-charge them... which kills NiCd cells
permanently.
--
If NT is the answer, you didn't                 |     Henry Spencer
understand the question.  -- Peter Blake        | henry@zoo.toronto.edu