Flooded Lead Acid Battery Desulfation...
2nd Gen Unit (Based
on the Classic - Alastair Couper HP#77)...
Before going very
far into this installment, I'd like to introduce "Alastair Couper". AC
designed a simple Battery Desulfator that was published into the public domain
in issue #77 of "Home Power" magazine. Either in the article or from
his web-site AC notes that the purpose of the circuit is to extend the life of
Flooded Lead Acid batteries, by rectifying one of the most common failures due
to sulfated plates.
Not wanting to get
too far over my own understanding of the subject, FLA batteries rely on an
electrolyte solution between a series of lead plates. When a battery is being
used the sulfuric acid solution is delivering sulfur on to the plates as the
acidic solution progressively becomes weaker.
If a battery is
consistently being discharged very deeply, it successively has less & less
storage capacity, due to the inability of the sulfur to return to solution in the
electrolyte. This is like the trolling motor battery that would allow you to
spend the entire day on the lake, and only a year later lasts for less than 2
hrs.
I did build a unit
from the plans using the exact components that AC listed... It has worked when
ever needed over the years until erroneously connected with reverse
polarity...
After having
rebuilt it, I opted to take some liberties with the parts listing, AC outlined
the basic premise of the circuits functional blocks and which were critical.
Since the original
unit was soldered-up on a pref-brd, I wanted to make a nicer PCB version,
pictured to the right.
I don't want to
get into the actual schematic or even too much detail on the modifications to
the original design. The full article is very easy to find via your favorite
search engine... But I wanted to provide a link to a forum that I spent several
hours reviewing...
There are numerous
variations on the AC Classic, as well as more explanations of how the circuit
operates than I would ever want to recount.
I'm assuming that
you've looked at the AC Classic and/or the FAQ linked above...
This is the
front-end or PWM part that fires the "Tank Circuit" that does the
actual work of returning the sulfur back into the electrolyte solution.
This PWM is way
different from the "Classic" but does offer a very broad range of
Trigger pulse widths.
The reason I opted
to deviate so far from the AC Classic was to avoid purchasing any parts
whatsoever. Lets assume that we are in the depths of the Post Apocalyptic Era, I
would still have the vast store-house of junk from which to build circuits as I
needed to.
The AC Classic,
called for 2 coils or inductors rated at 220uh and 1000uh respectively, AC
himself encouraged a degree of flexibility in the substitution of parts, but
there is an implied ratio of values between the parts, and the width of the
triggering pulse would also be varied to accommodate any substitutions.
These were the
first set of coils that I hand wound on a pair of torroid cores.
There was once a
time many years ago, when if called upon I could have calculated the number of
turns required to achieve a specific inductance given the properties of the
core, the dia of the wire etc...
But today, I
congratulate myself on being able to remember how to use the oscilloscope, and
comprehend the basics of the circuit.
For these purposes
let it be said that Torrodial cores offer high flux densities, but rather than
counting the turns, I approximated the length of the wire (12ft &
3ft)...
The Switching is
accomplished by an IRF P-Channel MosFET, the Heavy diode was from an
intermittent Wiper Control brd, and the Blue Capacitor (680uF 63VDC or there
abouts) was also scavenged from another board in the pile.
Also incorporated
into this version is the use of Coaxial Cable leads to the battery, there was
discussion on the FAQ, of improved results due either to the shielded nature of
the cables or the reduced loses from the solid copper center conductor.
Three (3) days
later, I've finally wound a reliable coil pair, and have tuned the circuit for
optimal performance.
The original 12ft
length on the larger core is still intact but the primary or tank coil was
increased to 12ft (from 3ft of wire) and wound tightly onto a much smaller
core...
I don't think that
I could be any more vague, or offer less aid to anyone trying to duplicate the
setup. Should I think of it, I'll measure the wire diameter (certainly a
critical variable).
Although the
system is intended primarily for Flooded Cells, it seems to work fine on small
Sealed cells as well.
And this is the
critical output waveform that I'd been in search of.
This is exactly
what is hitting the plates of the battery, a 35V spike that carries a whollup...
When the leads are connected a spark is quite visible.
If the Triggering
pulse is too wide, the coils, diode and/or MosFET begin to get hot, and there is
an audible whine from the circuit.
In this display
the circuit had been running cool, for over an hour.
This particular
project was a nightmare from the word go, especially as it was not in the
rotation of planned events. But the addition of 13 Golf cart batteries to
theworkshop.ca accelerated it's priority. I felt compelled to nab the batteries
as they were an excellent deal, and would be yet another small step in my plans
to reduce our (family) load on the grid.
With a functional
proto-type, I hope to build 2 more units and work on recovering several
batteries at once.
The blue tube is
an empty potato chip container that now houses the Desulfator.
The battery that
it is connected to is from a Cub Cadet garden tractor. The battery (like all my
batteries) is numbered and has a chart that I dutifully fill out every time it
is either run on a "Desulf" or Charge cycle.
This particular
battery has remained discharged for 2 years prior to coming to me. It read 8V,
ultimately it will be called upon to power a single CNC Mill and PC for at least
a 2 hr run.