|
|
|
|
|
|
|
|
|
|
Valen
Experienced Roboteer
Joined: 07 Jul 2004
Posts: 4436
Location: Sydney
|
Ni-Cd and Ni-Mh basically just undergo electrolysis when they are overcharged, they produce H2 and O2 out of water, this increases internal cell pressure. There is a Platinum recombination catalist in the cell however which acts to recombine the H2/O2 into water.
NiCd is more robust in terms of surviving overcharge in general.
NiMh is more sensitive.
I believe the cells can get up around 40PSI at full charge.
Temperature rise begins at about 80% charged. Typically when doing a quick fast charge from room temp bats the cutout is about 50c as I recall.
(IE that temp is indicative of 100% charge.)
Angus is right to try to cool the cells however, if you have a good DP charger you will get more capacity into a cell if you force cool it to room temp. But i'm thinking that effect will be in the 5-10% range. Still its best to put cold cells into a bot than warm ones anyway and your better off not temp cycling them twice. (charge, cool , use , cool = bad Vs Charge, use, cool)
The new NiMh's that have lower internal resistance than Ni-Cd are looking good for bots, but thats a recent development, Before nothing could supply the current of the NiCd _________________ Mechanical engineers build weapons, civil engineers build targets
|
Wed Feb 08, 2006 6:23 pm |
|
|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
There is a *lot* of conflicting Tech Info on battery cells out there..
Everything from the so called "Memory Effect" (that doesnt actually occur in practice) through to people who
still
market "Burp Chargers" to the gullible as "The latest Tech Secret" years after they've been tested by the battery manufacturers and proven to be bunk.
There is "some" evidence to the effect that NiCd (and possibly NiMh) cells used in High discharge applications (like ours), actually work *better* if you thump power into them at a reasonably fast rate, rather than gradually trickle them up.
Supposedly, its to do with the crystalline structure formed by the contents as they take up charge. Slow gentle charging encourages the formation of larger crystals that have a smaller reactive surface area (meaning less peak current) than a finer more broken-up larger-surface-area crystalline structure that a faster charge is reported to encourage.
I think the notion of "the slower the charge the better" comes from the very-low-current applications (like your RC Transmitter and Receiver batteries) where they are being charged by dumb plug-packs with no delta-peak end-of-charge sensing..
In low-current applications ONLY
, a slow charge (.1c or less)
1. prevents serious cell heating (and damage) when someone leaves their pack on charge for 48 hours to "squeeze that little bit extra in" (not knowing that overcharging causes symptoms that are usually known as the dreaded memory effect - correctly called "voltage depression")
2. Allows unevenly discharged cells in a multi-cell-pack to equalise as they all gradually come up to the same voltage/charge state. If you charged them faster, then the cells that reached full charge before the others would start to trend downwards (the negative voltage peak), possibly causing your peak detection charger to false-peak from some cells going down while others are still going up.
This is why when you first construct a pack from cells, it *is* a good idea to give them a slow charge (even better a *individual cell* discharge to 0.5v each prior to pack assembly first, then followed by the slow charge after you assemble them into a pack). This should bring them all close to the same state of charge, and you can then charge them as a pack in future and be reasonably confident they are all at roughly the same discharge/charge level.
*If* you have time to fiddle, inidividually discharging the pack cells every now and then (once every 6 months or so with average bot use) by means of a set of crocodile clips is a good way of bringing them all back into line, or 2nd-best, occasionally do a *slow* (.1c) charge to equalise them, followed by a quick (1C+) discharge and charge to "flex their muscles" a bit and stir up the chemistry.
Liquid Lead-Acid batteries are well known to exhibit this effect as well, since in their case, the fast-charge stirs up the chemicals in the electrolyte, prevent "acid stratification" from the heavier substances settling to the bottom of the cell. Not likely to be a problem in our application where the batteries get the hell shaken out of them.. This effect probably doesnt apply to Gel or Absorbed-Glass-Mat SLA batteries that we use anyway, but it illustrates the point that there are often unexpected effects going on in battery chemistry that arent immediately apparent.
Also, keep in mind that a lot of the advice out there is intended to maximise battery *cycle life*, which isnt really a big factor in combat robot applications. Your batteries are much more likely to get the stuffing pounded out of them, shorted out, overheated from overload and hurried-between-round charging, or sawed by Scoopy-Doo long before they die of old-age. Telling a combat soldier in a battlezone that they really shouldnt smoke because it takes ten years off your life is unlikely to worry them too much.. Power-to-weight Performance is more important to us than a comfy retirement.
Correct Care and Feeding of batteries is a controversial subject.. Its like asking Pet Breeders what food is best, each will have their own preference and reasons for believing that their choice is best. The above is just *my* opinion. Read up and make your own choice (and if you discover something new, let us know..). A couple of good starting points are here
http://www.camlight.com/techinfo/techtips.html
http://en.wikipedia.org/wiki/Nickel-cadmium_battery
*some good info for the Wiki in this thread, I'll go through and compile some of it - thanks for the input everyone* _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Wed Feb 08, 2006 10:39 pm |
|
|
|
|
|
|