On Mon, 2009-02-23 at 10:09 +0000, Chris G wrote:
Well go and search on Google, the *first* hit I got for a "NiCd memory effect" was this one:-
http://www.repairfaq.org/ELE/F_NiCd_Memory.html... and the next hit says exactly the same, as do lots of others. They're from reputable places too.
Ok so "memory effect" is bad (but common) terminology for an effect I have personally witnessed and measured where a NiCd cell's apparent capacity is reduced by a re-profiled discharge curve caused by repeated part charge/discharges. I'll agree with you for the sake of argument that it isn't a true "memory" effect but regardless you can add more curve to the discharge slope by part charges or part discharges which in turn limits the effective usable capacity of the cell in many applications.
Up until a short while ago during a big clearout I even had graph stats from computerised charging equipment that would quantify this. I used to many years ago produce and race high performance chassis for the (highly competitive) Model racing clubs...we are talking £1k for a 10th scale rolling chassis here so naturally we took every step to get as much performance as we could from the 1800mAH NiCd cells we were limited to. Even the cell manufacturer data sheets backed this up and I can assure you we weren't overcharging.
Ok so that doesn't quantify with "normal" use as the charge/discharge times we were working with were in the order of 20/5 minutes, but when I have measured such an effect..plotted the changes in the discharge curve and had supporting documentation from the cell manufactures. I am afraid I am going to take that over a couple of hits on Google, even if it does put me in the "There will always be people who spark off another flame-war by posting that they have seen "memory" and they will not hear otherwise." category, but I don't care :)