Can I model a 100% charged lead acid battery as a store of X coulombs
and use a microcontroller to integrate the current out through a shunt
or hall sensor to keep track of depth of discharge? Or is there some
other tricky factors to consider?

On 8 Aug, 18:03, BobG <bobgard...@aol.com> wrote:
> Can I model a 100% charged lead acid battery as a store of X coulombs
> and use a microcontroller to integrate the current out through a shunt
> or hall sensor to keep track of depth of discharge? Or is there some
> other tricky factors to consider?

Yes. Technically straightforward but the gods are with you if you can
estimate within say 100% of the true charge state. The buggering
factor is the current not running pro rata with the remaining
charge.
Yuasa have a good PDF handbook on their SLA batteries. The graphs
show various discharge currents versus resulting time to a flat
battery. There's about a 10:1 variation in apparent Ampere hours
capacity, dependant on how fast you flatten the battery.

Having said that and if you have the graphs for the particular battery
in use and the insane patience to translate them into stored lookup
tables + ambient temp'+ battery temp' monitoring. You may see a +/-
25% estimate accuracy.
It's much easier and probably as accurate, to just WAG from the
battery voltage

I knew there had to be a catch. Nothing's quite as easy as it seems is
it?

On Wed, 08 Aug 2007 10:03:42 -0700, BobG <bobgardner@aol.com> wrote:

>Can I model a 100% charged lead acid battery as a store of X coulombs
>and use a microcontroller to integrate the current out through a shunt
>or hall sensor to keep track of depth of discharge? Or is there some
>other tricky factors to consider?

Yes, there are tricky factors to consider.

google "Peukert's Equation (Peukert's Law)".

Peukert essentially says that the faster you discharge a battery, the
less total energy you can extract from it.

A lead-acid deep-cycle battery is rated in ampere-hours, assuming the
battery is discharged over a period of 20 hours. If you discharge at
twice the "20 hour rate", you will only get about 89% of the rated
energy.


--
Peter Bennett, VE7CEI
peterbb4 (at) interchange.ubc.ca
new newsgroup users info : http://vancouver-webpages.com/nnq
GPS and NMEA info: http://vancouver-webpages.com/peter
Vancouver Power Squadron: http://vancouver.powersquadron.ca

On 2007-08-08, BobG <bobgardner@aol.com> wrote:
> Can I model a 100% charged lead acid battery as a store of X coulombs
> and use a microcontroller to integrate the current out through a shunt
> or hall sensor to keep track of depth of discharge? Or is there some
> other tricky factors to consider?

if it's a wet battery (liquid acid and not gel) you may do better to
measure the density of the electrolyte... yuasa make a filler cap that
changes colour dependant on this...

Bye.
Jasen

Circa Wed, 08 Aug 2007 19:10:12 -0700 recorded as
<shtkb39q7lgp449oeslq6icatspapvul7u@news.supernews.com> looks like Peter
Bennett <peterbb@somewhere.invalid> sounds like:

>
>A lead-acid deep-cycle battery is rated in ampere-hours, assuming the
>battery is discharged over a period of 20 hours. If you discharge at
>twice the "20 hour rate", you will only get about 89% of the rated
>energy.

And it gets even more complicated than that! Different batteries will have
different time frames set for their final amp-hour rating. The lead-acid
storage batteries I work with are rated to an 8-hour discharge rate, as
opposed to your stated 20-hour rate. (The major difference is probably due
to the difference of deep-cylce vs. backup.)

An added complication is that the battery's capacity will change depending
upon how it is discharged. I don't have the data at hand, so can't quote
the equation, but basically, if the battery discharge is not uniform, and
is heavier during one part of the discharge than another, total capacity is
changed from the uniform 8-hour rate, and the change differs dependent upon
whether a light draw is followed by a heavy draw, or vice-versa.

The data for each particular battery should be obtained from the
manufacturer, and as a previous poster noted, the simplest method of
determining discharge state is to measure cell voltage and compare to data
in hand, and the accuracy of this method is usually good enough.

On Aug 9, 5:29 am, Jasen Betts <ja...@free.net.nz> wrote:
> On 2007-08-08, BobG <bobgard...@aol.com> wrote:
> if it's a wet battery (liquid acid and not gel) you may do better to
> measure the density of the electrolyte... yuasa make a filler cap that
> changes colour dependant on this...
=============================================
Any idea how to read electrolyte density with a microcontroller? PH
sensor?
Hydrometer rigged up to a gearmotor to squeeze the bulb and a camera
and frame grabber and dsp to see where the ball is floating? Or
something like that? How to you get this rig over the cells? An x-y
table?

BobG wrote:
> On Aug 9, 5:29 am, Jasen Betts <ja...@free.net.nz> wrote:
>> On 2007-08-08, BobG <bobgard...@aol.com> wrote:
>> if it's a wet battery (liquid acid and not gel) you may do better to
>> measure the density of the electrolyte... yuasa make a filler cap that
>> changes colour dependant on this...
> =============================================
> Any idea how to read electrolyte density with a microcontroller? PH
> sensor?
> Hydrometer rigged up to a gearmotor to squeeze the bulb and a camera
> and frame grabber and dsp to see where the ball is floating? Or
> something like that? How to you get this rig over the cells? An x-y
> table?

A digital refractometer would probably work best (if you
could afford it):
http://www.kpatents.com/pdf/downloads/p ... ochure.pdf

On 2007-08-09, BobG <bobgardner@aol.com> wrote:
> On Aug 9, 5:29 am, Jasen Betts <ja...@free.net.nz> wrote:
>> On 2007-08-08, BobG <bobgard...@aol.com> wrote:
>> if it's a wet battery (liquid acid and not gel) you may do better to
>> measure the density of the electrolyte... yuasa make a filler cap that
>> changes colour dependant on this...
>=============================================
> Any idea how to read electrolyte density with a microcontroller? PH
> sensor?
> Hydrometer rigged up to a gearmotor to squeeze the bulb and a camera
> and frame grabber and dsp to see where the ball is floating? Or
> something like that? How to you get this rig over the cells? An x-y
> table?

I'd start with one of the yuasa magic eye filler caps and look for a
way to measure the colour, (RGB LED and a phototransistor maybe)

The cap incorporates a light-pipe and a 2 part floating hygrometer that
exploits two parts of differing shape that change alignment in reponse
to the density of the electrolyte.

Bye.
Jasen