> Does anyone have a good reference about end of charge voltage for
> NiMh batteries depending on the temperature ?

> Reading one of Russel's posts I found his idea of using a shunt
> regulator to control the "termination" of the battery charge great but was
> not able to find any references anywhere I searched for what would be the
> best voltage to clamp. My application does not need 100% of the batteries
> capacity but temperature variation may be big and I need the batteries to
> last as long as possible. 1.4 volts "fells about right" but I would prefer
> to find some references to begin testing.

I'm researching this at present - also for niCd.
It's substantially more complex than one would like it to be.
It varies somewhat with the specific call, drops with increasing temperature 
and rises with charge rate. if you can stop charging and let it settle or 
even load it slightly then you may get a better idea but I don't have that 
luxury.

I'm notionally setting about 1.35V plus a delta V for charge current so that 
a heavily charging cell will not trip out before a slower charging one.

For a typical cell there is a fairly hard knee point which they will not 
rise above and setting the trip point just above or just below there will 
lead to very different results.
If you have the luxury of significant spare capacity (and this helps cell 
longevity greatly, then setting the voltage to about 1.35v is liable to be 
good.

Note that the useful voltage also depends on what you do when it is reached.

I am trialing both a series regulator that backs off current when set point 
is reached or a clamp regulator that shunts charge current at this stage. 
The implications for the battery vary. I'm also looking at hysteresis so the 
setpoint is dropped somewhat when reached.

All this sounds rather complex - and it is- but the complexity is largely in 
the design process - the end result will cost almost nothing and be very 
simple. It should end up with current varied reference voltage plus 
hysteresis. No tem[perature compenation yet. For a single cell design I'm 
using the TLV431 1.25 V shunt regulator as reference and for an N cell 
design the TL431.

Total cuoff may not be necessary - if you can muchly reduce charge curremt 
you may get down to "safe infinite period trickle current"  assuming such 
exosts for your cell type and size.



   Russell



-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist