Re: My Battery Monitor

["Darrell N." <darrelln@datalog.ab.ca>]

Lee wrote:

> That's a good solution. For a hobby project, I'd use something that can be
> removed if you need to fix or change it.

If it's cheap enough, it's a throwaway, no need to repair it.

> > you can build in calibration constants in software (ain't it wunnerful) to
> > ensure they all read the same regardless of component tolerances.
> This works to eliminate initial error, but not drift.

Hmmm, temperature drift can be handled in software, (although not 
easily) but not time related drift.  This can only be minimized by a 
decent burn in and regular calibration.

> Yes you can. No connector is more reliable -- so use *NO* connector! :-)

Ok, picky, picky.   8^[
 
> I would have wire leads coming out of a potted module. No connectors any
> where near the batteries.

Ya gotta make a connection to the battery itself somehow.  Ya gonna 
twist the wires together and black tape em? 

> For data, I like the idea of fibre optics. An LED or phototransistor can be
> put in the module. Use pieces of inexpensive acrylic fibre optic as "wire"
> between the modules and a central controller, several feet away. Such a
> system would work even submerged in battery acid.

I like it too, but it has a big drawback - cost.  Just thumbing through the 
Digikey catalogue, they have some nice cheap plastic fibre transmitters 
and receivers, but it's one way communications.  You'd have to have 2 
sets of transmitters and receivers and two fibres for 2 way 
communication.  A 1 meter link with fibres is about $6.  2 gazintas and 
2 gazoutas per module = $24.  Bidirectional tranceivers are off the scale 
price wise...

> You try making stable measurements of battery voltage when it is being hit
> with 100+amp charging and discharging transients!

But why would you have to have an accurate battery voltage reading 
under those conditions?  It's like trying to measure how much liquid is in 
a tank being drained and filled and shaken up all at the same time.  Not 
practical...  You yourself recommended not to try to measure a battery's 
OC voltage until after it had sat for a while...

> Do you really need all that complexity? All we're trying to do is
> measure a battery's voltage :-)

I guess we're back to the rotary switch and a Radio Crap multimeter.  
And I thought guys who talked on the phone while driving were 
dangerous!   8-o


darrell n
Calgary, Alberta, Canada