# VRLA versus flooded lead acid, absorption stage

Discussion in 'Electronic Basics' started by kell, Sep 11, 2005.

1. ### kellGuest

I'm designing a battery charger to charge motorcycle batteries at 14.5
or 15 volts during the absorption stage, when the battery draws one or
two amps. Then, as a drop in the current drawn by the battery
indicates to the charger that the battery has taken full charge, the
charger will lower its output voltage to about 13.5 volts to float the
battery.
I wanted to build the charger for batteries ranging from 5 AH to 12AH,
about the size of batteries I can use in my bike. (I can use a battery
as small as 5 AH because the bike kick starts.)
I tested a couple of lead acid batteries on constant voltage charging
to see if the charging current drops off during the absorption stage,
as I have read it does. I used a 7815 voltage regulator. I bulk
charged the batteries, then put them on the regulator. I tested a VRLA
first.
With 15 volts at the battery terminals, a 7 AH valve regulated lead
acid battery had a charging current that declined from more than an amp
initially all the way down to 70mA before I disconnected it.
Next I tried a 12 AH flooded battery. I equalized this battery about a
week ago, so it had a pretty full charge when I started the test. At
15 volts, this battery settled down and drew just over half an amp all
afternoon, and didn't decline any farther than that.
So at full SOC and after the current stabilizes, the flooded battery
draws more than 7 times the current the VRLA does at the same voltage,
although the flooded battery has less than twice the capacity the VRLA
has.
Either I need to build a complicated charger with a switch to select
between two charging protocols, or I have to build a charger that
functions much less than optimally on VRLA's in order that the float
function kick in when I have a flooded battery connected to it.
I just wanted to ask if flooded batteries always draw so much current
even after they level off. I don't have any more batteries to test.

2. ### BanGuest

You can charge both battery types with the same method, but you have to
strictly observe the maximum charging voltages.

max. charging voltage(T=25°) should not be more than 14.4V in the absorption
stage, even if the valve battery can use a higher voltage. The current
should also be limited to <c/5, which means 1A for the 5Ah battery. A good
adjustable regulator like LM317 can be used, which then can be switched to a
lower voltage (13.8V) for trickle charge.
You need a current monitor to do the switching, any high-side current
monitor can be used (ZXCT1009, AD8210). The switchover should be at <C/10,
meaning 500mA for the 5Ah battery, no need to change this for the other
sizes.

What you did(overvoltage 15V) will destroy the batteries fast and will
deteriorate the acid level through gassing. You need to study the data
sheets a bit deeper and get a better grip of electronic circuits or else
just look for suitable circuits on the net.

3. ### ehsjrGuest

You need to re-think the design. Use voltage level to
change the charge mode, not current sensing. An op-amp
comparator is a very easy, cheap and accurate circuit
to detect a specific voltage and change output state.

There are many different charger configurations available.
The intent of this post is to focus on the op-amp
comparator, but I also included one possible charge regulator
for completeness. It is the simplest charge regulator
configuration you can build, consisting of just two parts:
an LM350 and a resistor. Since you want a step in the
charging algorithm (rightly), one extra resistor is included,
and it is selected by the comparator circuit. Finally, one
diode was added across the LM350 to protect it in the event
of a power failure while charging.

Here's a comparator using 1/4 of an LM324 that drives
a relay. You adjust the pot (100K) to transfer the relay
at the exact voltage you select. The comparator circuit
is connected to the battery terminals. The relay contacts
are connected to the regulator, shown in the second diagram
below. You don't need to use relay output - there are
many variations. I chose the relay output for simplicity.

+ -----+---+---------+-------------+----+
| | | | | POT = 100K
| | +----- | ----+ [D1] [RY] D1 = 1N400x
| | | | | | | RY = 12v relay
| | | |\ | [R3]10M +----+
| P | | \4 | |
10K [R1] O--+--3|+ \ | e/
| T | >1---+----[R2]---| PNP
| | | / 1K c\
+-- | ----2|-/11 |
| | |/ | |
6v [Zd] | LM324 | |
| | | |
Gnd ---+---+---------+------------------+

Here's a circuit to regulate the charge *current*
in the steps you select. I'll assume 2 amps for the
first step - you can select whatever you want for
the float current. This circuit connects to the
charging source and to the battery.

1N400x
+--------|<----------+
| |
| ------- |
+ ---+---in| LM350 |out---+
------- |
Adj [R4] .6 5W
| |
| +---o o
| | \ RY-1
| [R5] o
| | |
+----------+----+
|
[BATTERY]
|
Gnd -----------+

The LM350 is configured as a constant current regulator.
R4 - .6 ohms - programs it to produce 2.08 amp when the
relay is de-energized and R5 is shorted out. When the
voltage rises to the point set by the 100K pot, the relay
energizes and R5 is placed in series with R4. You determine
the float current (I) you want, then figure the value of R5
by R5 = (1.25/I) - R4 and pick the closest standard value
resistor. For example, a 5 ohm resistor for R5 would limit
current to ~223 mA; 10 ohms would limit it to ~118 mA.
Wattage is computed by (I^2*R) - add a safety margin of
at least 50%

Ed

4. ### mikeGuest

I like a good technical discussion as much as anybody. But consider this.
The first time you press the starter button, that tiny motorcycle
battery is gonna
have it's guts wrenched out. Then it's gonna have the crap charged out
of it when you drive off. All the care and precision you put into
charging it won't make much difference ten miles down the road.
Even if the only bike you have doesn't electric start, the problem
can be much the same. Motorcycles are NOT easy on batteries.

I put hundreds of hours and hundreds of dollars into building computer
controlled charger/discharger/graphing hardware and software. I've
watched nicad and NIMH and LiIon batteries charge and discharge for
hours on end. I know a LOT about how they (mis)behave. But at the end
of the day, about all I accomplished was to busy myself for hundreds of
hours. Turns out that the outcome is much the same over a wide range
of charge parameters.
For the investment, I could have replaced all my batteries several times
over.

Sometimes, all you need is any old charger and an egg timer to tell you
when to shut it off.

mike

5. ### BanGuest

The question is if the bike has a proper regulator or not. Of course the
cheapest have just a diode in the alternator and overcharge the battery and
thus reduce its lifetime considerably, but it doesn't need to be like this.
If your battery is *not* overcharged, because the bike has an electronic
regulator, the battery will last even a couple of years and thus you might
save a significant sum in replacement parts.
Another question is how much will a simple 2 phase charger cost more than
just a transformer with rectifier? LM317 costs 0.6\$, the current monitor
with comparator will be less than 2.50, and the additional bigger
transformer and caps will be maybe 3 bucks. So the whole "fancy" technology
costs less than 10\$ additionally with a piece of veroboard to solder the
parts. How much comes a new battery? 15-25 bucks?
So is it worth the effort? The decision would be clear to me. When the bike
stands in the garage most of the time you will have to add the comfort of
always having a full battery without leaking sulphuric acid etc.

6. ### john jardineGuest

Nice one Mike!.

In a simialr vein, have just been tasked to *de*-design a sealed lead acid
charger. The existing '4 stage' TI (UC????)chip based switcher design is
just too complicated/expensive/unreliable for use in the outback. Customer
needs reliability/cost rather than squeezing in last few % of possible
charge. Sometimes fancy technology is its own worst enemy.
regards
john

7. ### ehsjrGuest

Exactly. Use a basic, simple, current requlator. Two
stinking parts. Shut the charger off when done. If the egg
timer won't work ("But I might not be there"), "done" is
*easily* sensed with a comparator. Case closed. But you
can't sell that idea to those who want to get the last ounce
of perfection out of their chargers.

Ed

8. ### kellGuest

cool off, brother.
Same goes for the rest of you aggressive yoodles.