[NTLK] This 'n that on batteries...

From: Frank Gruendel (Frank_Gruendel_at_t-online.de)
Date: Sat Dec 08 2001 - 01:26:51 EST


> I thought it might be that the rechargeable battery (Newton accessory)
> was loose or losing contact as it bumped about in my pouch.

The way you describe it, this may well be the problem. At least this is
a typical problem with the 2x00 family, the battery contacts seem to
be quite fragile and sometimes get bent.

> I put in another rechargeable battery
<snip>
> I can use it for a short while before it says the
> battery needs recharging

You might want to try a couple of charge/discharge cycles. Charge for like
12 hours (ignore whatever the Newton says battery-state wise), discharge
until the Newton turns off.
If that doesn't help, you can refurbish the pack, see the hardware section
of www.pda-soft.de for info on how to do this. If you don't dare to do it
yourself, I can do it for you.

> You should NEVER discharge a battery pack (with
> NiCDs,NiMH) below about .6V/cell,

This is correct, but the Newton won't allow you to do this anyway. It will
automatically shut off way before this voltage is reached.

> .8V/cell is perfectly acceptable.

Well, this depends on which datasheet you read or which site you browse to.
Most datasheets I know set that line at about 1.0 volts. Some even at
1.1 volts. I agree, though, insofar that 0.8 volts probably wouldn't do
much harm.

> When this happens, this empty cell can go into cell "reversal" and is
> being charged "the wrong way". Further discharge of the PACK will
> certainly destroy the battery(that is now empty) AND the pack will never
> behave the same.

Although true, this doesn't happen very often in real life. Usually cells
put into a pack come from the same production batch and thus behave and
age more or less the same. It rarely happens that one out of the four cells
decides to behave significantly worse than the other three, but if that
happens, your best option is to refurbish the pack, not trying to use
it in a way that makes life easier on the bad cell.
Whenever I refurbish a larger amount of packs, I spend about 4 hours
per cell on evaluating the exact discharge capacity and match those
that are most similar, this makes cell polarity reversal even less
likely.
Using rechargeables that are not part of a pack is another cup of tea of
course. It can easily happen here that cells with a different charge
state are used.

> (Incidentally, you'll notice on any pack of you buy Alkalines, you are
> told never to replace just one cell in a device, you should replace all
> the cells at the same time. The cell reversal issue is again why you are
> told to do this).

Assuming that you mean non-rechargeable alkalines this wouldn't be much
of a problem, you'd just throw the
bad cell away and replace it with another. The problem here is that the
internal resistance increases with discharge and you won't get as much
current through the empty batt as you would maybe like to.

> Both NiCds and NiMH have a steep discharge curve, so if you actually watch
> the voltage as you discharge, the pack/cell voltage will at first slowly
> fall then as you near .6V/cell the voltage(cell/pack) will drop off very
> steeply.

All cells I know have the fast drop way before 0.6 volts. More like around
1.0 volts. This is why it is recommended not to discharge them below that
voltage in normal use.

> A 1kiliohm resistor across the pack/cell is typically a sufficient
> load(don't violate the power rating on the resistor either! Use a 1/4 or
> 1/2 Watt to be safe).

Oh no, it's not. 1 kiloOhm across a fully charged pack (around 5 volts)
results in a current of about 5 milliAmperes. I guess there's no
mode (except for sleep mode) in which any Newton will draw as little
current. This current would by the way result in 0.025 Watt, so
there'd be no need for a 1/4 or even 1/2 watt resistor.
For real-life measurement the current should probably be in
the range of 20...100 milliAmperes, this would mean 250...50 Ohms.
When using 50 Ohms 1/2 watt would be a good idea, unless you are
interested in first-hand experience of how fast a fully charged pack
can burn a resistor.

> Also, for those curious about charging, to make sure you never damage
> NiMHs OR NiCDs they should be charged in an "intelligent" or "peak
> detecting" charger. These chargers carefully watch the cells while they
> are being charged and know when to stop to prevent over charging.

Couldn't agree more...

> if a charger says it can charge NiCds but doesn't say NiMH,
> then only use it for NiCds, DO NOT use it for NiMH.

Couldn't agree more...

> If a charger says it can charge NiMH but doesn't say NiCd, then it is
> however OK to charge both NiCds & NiMHs.

Well, I politely disagree. The charge requirements for NiCd and NiMH
are different. Your best be is to charge NiMH with NiMH chargers and
NiCd with NiCd chargers. Or both with a charger that is intelligent
enough to find out automatically which type it is charging.

> Overcharging either cell type results in excess heat building up and the
> cells will "vent" -- literally leaking.

Yep. This is what the temperature sensors in all Newton models are
supposed to prevent ;-)
For MP2k and eMate this sensor is in the pack, for the rest it is in
the battery compartment.

> However, since TTL circuitry (which I'm pretty certain the power supply of
> the Newton is) requires .7 volts or more to activate, you should never be
> able to drain any particular cell in a newton's battery pack below that
> level.

Ermm... well...

If none of the few brain cells in which my remaining knowledge of TTL
circuitry
is stored is on vacation, TTL needs a voltage of less than 0.7 volts to
recognize a digital "0" and a voltage of 4.5 ... 5.5 volts to recognize
a digital "1".
So assuming for a second the electronics is TTL powered directly from the
batteries, this would mean you shouldn't discharge the total of
the four (serially connected) batts to less that 4.5 volts or 1.125 volts
per cell. That voltage being way above the allowed discharge voltage for
NiCd and
NiMH, this wouldn't exactly make sense.
This is why the Newtons have a switched voltage regulator. Its job is to
ensure that the voltage the Newton electronics sees stays at 5 volts as long
as the battery voltage is higher than about 3.7 volts (measured on a
120 and on a 130).

> If you wish to drain a battery outside the newton, determine, using
> ohms law, what resistor will provide a current draw from the pack that the
> newton usually pulls under a full charge, combine that with a diode (LED)
> and connect that combination to the terminals of the pack and let it go
> until you can no longer see light from the led.

You will have a hard time doing this with a MP2k pack. There is no way to
attach a clip or something to the contacts unless you have veeeeery
tiny clips. You'd probably be only marginally more successful with an eMate
or OMP/MP100 pack.

> The pack should be discharged to .7 volts.

Nope, it should not. It would be less hassle to throw it out of the
window or flush it down the drain. It should be discharged to
0.7 volts PER CELL or 2.8 volts per pack, provided 0.7 volts is what
you find in the cell's datasheet as the minimum discharge voltage.
As I described above, I wouldn't discharge the pack below 4.0 volts.

Frank

Newton software and hardware stuff at www.pda-soft.de

--
This is the Newtontalk mailinglist - http://www.newtontalk.net
To unsubscribe or manage: visit the above link or
	mailto:newtontalk-request_at_newtontalk.net?Subject=unsubscribe



This archive was generated by hypermail 2.1.2 : Wed Jan 02 2002 - 12:01:24 EST