RE: (ET) batterys

["David Roden (Akron OH USA)" <roden ald net>]

On 11 Apr 2003 at 20:32, Humphrey, Timothy wrote:

> ... it seems that Barry is concerned with not "severely" overcharging
> his batteries. I understand that with Floodeds this only causes them to 
> use
> more water and does not necessarily damage them. 

Overcharging is not as rough on flooded batteries as it is on "sealed" 
(VRR) 
batteries for only one reason: because you can replace any water lost from 
flooded batteries.  Other than that, overcharging is equally damaging for 
the reasons I cited.

No question, the GE charger overcharges.  How bad is it?  A well-cared for 
golf car battery, carefully charged, should last 750 cycles or more.  If 
you 
use your ET once a week, that would be 15 years!  ET batteries usually 
last 
about 5-6 years, though, sometimes as much as 7-8 if reduced capacity is 
acceptable.

But that's still pretty decent service, no?  ET battery amortization costs 
 
maybe $40 a year.  Not bad.  So one could argue that the GE charger and 
its 
crude, dumb-lummox timer charge control is perfectly adequate for a 
machine 
that's going to be used no more than perhaps 50 times a year.


> However,  wouldn't Harry's
> controller provide a sense of peace. Knowing that your batteries will get
> charged, will not get boiled and will be kept topped off.   

I'm not so sure.

I didn't realize that it was supposed to ^replace^ the timer.  Such an 
installation would be unwise, in my opinion.  Unless I misunderstand how 
Harry's gadget works, it's a battery maintainer, not a true charge 
controller.  

Charge control is not an exact science, believe it or not; but some useful 
 
strategies have been worked out over the many years that people have been 
using lead acid batteries.  They vary from crude to pretty sophisticated.

        1. Dumb timers: The user estimates how much energy has been used 
from the 
battery, and sets the timer for roughly the amount of time needed to 
replace 
it.  Normally the charger also has some crude form of current limiting 
(typically just the resistance of the transformer, diodes, and wiring).  
This method is used in the simplest golf car chargers from Lester, Hobart, 
and the like.

        2.  Simple constant voltage (regulated or taper charging): 
Sometimes used 
in conjunction with the above scheme.  Output voltage is limited to 
prevent 
gross overcharging.  This can be: a ferroresonant transformer and 
capacitor 
circuit, as used in the GE charger; some kind of electronic regulator, as 
in 
the wall wart chargers for hedge trimmers and such; or a "cycle-dropping" 
controller, as in those blinking-LED automotive battery chargers.

        3.  IU:  Constant current until a threshold voltage is reached, 
indicating 
80% charged; then constant voltage until the current falls below a 
threshold; then shutoff.  Generally the threshold voltage and current 
require adjustment for battery temperature, which is often done 
automatically with temperature sensors.  They should also be adjusted as 
the 
battery ages, which can be (but seldom is) done under microprocessor 
control.  The Zivan chargers usually use this scheme or the following one.

        4.  IUI: Same as IU above, with an additional low-current constant 
current 
phase added before shutoff.  This extra phase provides cell equalization, 
and is limited by some combination of on-charge voltage and/or time.

        5.  dV/dT: Constant current until a threshold voltage is reached, 
followed 
by a lower constant current charge maintained until the rate of on-charge 
voltage rise (dV/dT) falls below some value.  This is the scheme used by 
Lestronic chargers for a good 25 years.  It has the advantage of requiring 
little or no adjustment for battery temperature and age. 

Harry's controller doesn't exactly implement any of these algorithms, 
although it appears to be somewhat similar to the cycle-dropping (C-D) 
method.  Let's look a little more closely.

A C-D controller charges at whatever current the charger can manage until 
the on-charge voltage hits its limit. Then it shuts off and waits for the 
voltage to fall.  When the voltage falls to some minimum value, it turns 
the 
charger on again.  The effect is to provide gross voltage regulation at 
the 
high voltage limit.  

Instead of limiting voltage as the C-D controller does, Harry turns on the 
charger and runs it for a fixed length of time -- 5 minutes.  Then he 
shuts 
it off, waits for the voltage to fall, and repeats.  The low voltage limit 
is more important to his scheme than the high limit (in fact there is no 
high voltage limit except that imposed by the charger's output).

It's described this way: "If the battery voltage drops below 38 volts, the 
controller turns on the charger for approximately 5 minutes. Then it turns 
off the charger until the voltage again drops below 38 volts."

Harry's device looks like a pretty effective alternative to a float 
charger 
for maintaining older batteries that are to be unused for a period of time 
--
 for example, when you store the tractor for the winter.  But I am not so 
sure that it would be a good choice as the only charge controller.  

In theory it doesn't sound too far off.  It should cycle so that the 
charger 
is on most of the time at first, with the off cycles gradually becoming 
longer.  

My concern would be -- and understand that I haven't tried it -- that it 
would have at least part of the disadvantage of a float charger.  

A float charger is great for maintaining batteries at full charge (mostly 
compensating for self-discharge).  It's a good choice for standby 
applications such as computer UPSes.  

But in a cyclic application (like ours), a float charger's voltage is too 
low to fully charge the batteries in a reasonable period of time.  At 
float 
voltage it can literally take ^weeks to months^ to charge a battery from 
about 80% to 100%.  This is obviously not good.  

It's not clear to me that Harry's controller fully overcomes this float 
charging limitation.   It appears to me that, ^if used as the sole charge 
controller^, Harry's controller would undercharge the pack unless you 
allowed a long time between uses of the machine.

How about battery life?  Harry says:  

"I have had 2 in trouble free service now for 11 years each on an E8 and 
an 
E12. My battery sets have both been changed once, at about 6 years. I only 
need to add water once per season. They have also kept the batteries 
charged 
properly over the winter."

I might point out that 6 year battery life is about the same as one can 
expect with the GE charger.  But I do think the device has some utility as 
a 
maintainer, as he points out.  However, I say again:  You don't need a 
maintainer with batteries unless they are to be stored for several months 
and/or they are fairly old and subject to high self-discharge.  If that 
matches your situation, then Harry's controller could be a good choice to 
keep the pack in good shape while it sits for a month or more.  

But if I got one, I wouldn't use it to ^replace^ the timer.  Instead, I'd 
connect it ^in parallel^ with the timer's switch contacts.  I'd use the 
timer to charge the pack after using the tractor, then let Harry's 
controller maintain the pack.  But -- I'd probably still pull the charger 
plug if I were going to use the tractor again in a week or two.

Just to be clear her, I'm not "dissing" Harry's controller.  I think it's 
an 
interesting and well thought out design which has a useful application.  I 
just am not so sure I'd want to use it in place of the GE's timer as the 
only charge control.

Now, maybe others here have experience that shows how well it works as the 
sole charge controller.  I would urge anyone who has one, has it connected 
as recommended, and has monitored its workings to post the results.  How 
long does it keep cycling before the off time gets to be, say, several 
hours?  Hours?  Days?  Weeks?  How long does it take to bring the pack up 
to 
a consistent, fully equalized, 1.280 to 1.300 SG?  


David Roden - Akron, Ohio, USA
1991 Solectria Force 144vac
1991 Ford Escort Green/EV 128vdc
1970 GE Elec-trak E15 36vdc
1974 Avco New Idea 36vdc
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Thou shalt not send me any thing which says unto thee, "send this to all   
  
   
thou knowest."  Neither shalt thou send me any spam, lest I smite thee.    
  
       
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =      
  
 
Est. yearly US cost to safeguard Persian Gulf oil supply: $50 billion      
  
       
Est. 2001 value of US crude oil imports from Persian Gulf: $19 billion  
                                -- Harper's Index, April 2002              
      
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =