Al salaam a'alaykum John

01 Dec 03 09:18, John Tserkezis wrote to All:

 JT>   (for the benefit of Bob as well)

 >>  BL>  Remember to use thick wire...

 >> Indeed, but how thick is thick?

 JT>   Thicker than what they recommend.

Bob has only recommended "thick" :-)

[...]
 >> They're gel cells.

 JT> Still lead-acid.  They use a gel electrolyte that recomposes the 
 JT> gas on production.  They should never release gas unless they're abused, 
 JT> or way past their use-by date.  That's why even though they are 
 JT> technically called 'sealed' they should never be actually sealed.

The cells do have an "emergency" vent.

[...]
 >>  BL>  My guess is that it disconnects if the voltage gets to *high* (to
 >>  BL> prevent over-charging), with a limited current that drops to zero 
 >>  BL> as the battery gets up to 13.8 volts (2.3V per cell).

 >> It uses PWM to adjust the effective charge rate.

 JT> They are float charged.  That is, the output voltage is fixed (most 
 JT> of the time anyway. The normal "fast" charge current is
limited, but it 
 JT> doesn't stay at this point unless the batteries are flat, and it's 
 JT> limited to a very low value when the batteries are VERY flat).

I thought it charged harder when the cells were flatter.

 JT> It keeps track of the amp/hours going in and out to monitor the 
 JT> overall capacity more accurately.

[...]
 >>  BL> and this can be a problem with NiCad batteries and the fancy 
 >>  BL> gel-cells, 

 >> yes?

 JT>  Less of a problem with NiCads, but they don't like to use nicads 
 JT> anymore, cadmium and landfill and all that.  If you overcharge a 
 JT> lead-acid, it dries out.  You WILL kill lead acids by far quicker if you 
 JT> discharge one or more, but not all of them.  Half the cells will be 
 JT> grossly discharged, the other half grossly overcharged.

I'd rather not do that, these cost me a few bob..

 >>  BL> The three non-flat cells *will* be overcharged at the beginning of 
 >>  BL> the charge cycle,

 JT>   As above, not a good idea to do on a repeated basis though.

 >>  BL> they will fizz more gas, 

 >> These are not supposed to fizz - they're sealed.

 JT>   Oh yeah they do.  It's classed as abuse- they do release gas.  
 JT> Actually, it's part of the normal charge cycle.  They are charged to 
 JT> higher than the recommended cycle charge point, (higher than long term 
 JT> voltage) and kept there for a short period of time.  Then the charge 
 JT> voltage is brought down to the normal float point.

OK.

[...]
 >>  BL> A flat battery is 2.0V but it varies with temperature. 

 >> There is a temperature sensor attached to one cell and wired to the 
 >> regulator.

 JT>   That affects the charge voltage.  It varies depending on 
 JT> temperature.  You don't normally see it on chargers, because they don't 
 JT> care.  On a solar system, you're trying to get every bit of energy into 
 JT> and out of the batteries.

Indeed.

 >>  BL> That's why the *best*way to test a charge battery is with a 
 >>  BL> hygrometer (the float in a bulb).

 >> Nowhere to stick the hygometer....

 JT>   Ahem.  This is AVTech.  Need I remind you where it needs to be 
 JT> stuck?

I thought of that, but I don't think it will indicate the state of charge
of the cells under discussion.....

 >>  BL>  Over a few years, the life of your top three cells will be 
 >>  BL> reduced, but you can only expect 5 years out of a battery anyway 
 >>  BL> (whether you put water in or not). 

 >> These are guaranteed for ten, life expectance of 15....

 JT> Tell Bob how much they cost.  Let him do the solar economics math 
 JT> with a life reduction of more than half.  I would not recommend in 
 JT> anything that would reduce their life more than normal operation.  It 
 JT> will cost you dearly.

I bought the cells at mate's rates for $1260.40 for the six.

 >>  BL> If it worries you, swap the top three and bottom three cells 
 >>  BL> around every six months.

 >> The buggers weigh more than 20kg each. I'm not moving them any more 
 >> than I have to.

 JT>   Besides, you'll be abusing the top and bottom halves in a different 
 JT> way every six months now. :-)

Hmmm - I wonder if Wacko Jacko is interested in this method of abuse....

Regards,
David

--- Msged/LNX TE 06 (pre)