Do not buy the ones with the little floating balls or with an
arrow pointer made of plastic. They're worthless. A hydrometer looks
like a glass turkey baster with a floating glass vial with lead shot in
the bottom of it and a little paper scale on top. Make SURE you get a
hydrometer with a THERMOMETER in it. The electrolyte specific gravity is
very dependent on temperature, which renders the little plastic testers
useless. [ . . . ]

Hmm. There's a lot of voo-doo (emphasis on doo) floating around about
wet cell batteries, so I'll just parrot what the Trojan master
distributer in this region had to say last time I chatted with them.

Don't use the glass vial hydrometers, a broken glass (full of sulphuric
acid!) hazard on boats. Get a *good* plastic hydrometer, the disc
pointer type (the only type Trojan master distributers sell).

The S.G. of the electrolyte is *slightly* dependent on temperature, in
the typical boat environment (60 to 90 degree F. range), not
compensating for temperature would result in an error of +-0.008 or so.
You'd have tough time reading that on your hydrometer. Much bigger
error results using the above quoted measurement procedure, instead,
fill and drain the hydrometer 2 to 4 times before pulling out a sample,
or you'll get a false reading due to stratification of the electrolyte
in the battery. Tap the hydrometer to knock off any bubbles on the
pointer. Orient the tester vertically.

This nonsense of charging the batteries at 100A for 20 minutes is just nonsense. [ . . . ]

That would be a 30ah charge; about right for a motorcycle battery,
indeed nonsense. Good batteries can accept 20% of their rated capacity
when in a deeply discharged state(50%); a 500ah bank can be charged at
a 100 amp rate with minimal loss of service life. They can be charged
at much higher rates (40%) if a shorter service life is acceptable.
Charge rates must be dropped rapidly as the battery nears full charge,
of course. Temperatures should not exceed 120 degrees F., and probably
will just rise slightly at a 20% charge rate. Indeed, slow charge
rates are beneficial concerning service life, but time is money; it'll
take 3 hours or so to charge a 50% discharged 500ah bank starting with
a 20% charge rate, conversely, at a 20 amp rate it'll take 15 hours.

Modern wetcells don't use water unless something is perking them with overcharging.

If your batteries aren't using some water, you are undercharging them.
More batteries are damaged due to sulphation from undercharging than
from overcharging. Mild gassing (boiling is a misnomer) is a healthy
part of the charging process; it aids in mixing the electrolyte, and is
an indicator that the batteries are becoming fully charged at the end
of the charge cycle, when charge rates should be 1-5% of the capacity
of the bank. Periodically, or when the S.G. of the cells differ
significantly, an equalization charge should be performed, with
significant gassing.

If the charger brings the cells up above 1.280, check the charger voltage at full charge because it's overcharging the cells.

??? The absolute value of a fully charged battery is determined by
the battery chemistry when manufactured. After a dozen or so cycles on
a new bank, fully charge the bank, top off the batteries and write down
the S.G. value. That's the number for *your* batteries, and cannot
be exceeded by overcharging, unless you add or spill acid from the
battery (acid, not water). Might be 1.300 for batteries manufactured
for the tropics, might be 1.260 for batteries sold in Fairbanks,
Alaska.

Use only DISTILLED water to top them up to just above the plates.

Top your batteries up to the measurement rim in the vent holes,
significantly above the plates. A portion of the plates which uncover
for a significant period of time are permanently damaged and will never
again contribute to the battery's capacity. The extra electrolyte is
there to guarantee the plates are immersed when water loss from normal
gassing occurs, and the batteries are not level on a moving platform, a
sailboat on one tack for a few days, for example.

I suspect you may have a "dead cell", one whos acid has been converted to lead sulphate crystals that have precipitated out of solution into the space for them at the bottom of the cell.

There are many reasons for a "dead cell", one of which is lead sulfate
crystals falling off the plates to the cell bottom, ultimately shorting
the bottom of the plates. Precipitation is not involved. Other
processes can result in an open cell, a short between plates, etc. If
you have a dead cell, you're screwed, time to buy a new battery.

Please recall I'm only the messenger; the message is from the Trojan
guy, who deals with thousands of commercial customers who work the hell
out of their batteries, and have sharp pencils concerning charge rates
and service life. In essence, he says maintain water levels, charge
them vigorously at least periodically, and try to hold the discharge
depth to 50% in normal service.

More straight info available at:
http://www.trojanbattery.com/Tech-Su...intenance.aspx