There are basically three types of lead-acid batteries:
Car, Marine, and deep cycle. (Internal construction is a different
A car battery is designed the thinest and more numerous plates. This
provides more surface area and a lower internal resistence. This is why the
plates tend to get destroyed when too discharded.
A marine battery is designed with thicker and less numerous plates. This
gives it a little higher internal resistence, but better discharge
A "deep cycle" battery has even thicker plates and fewer of them, providing
even more internal resistence and much better discharge behavior.
Now, what makes you say that the batteries I have are not deep-cycle? They
are certainly not marine grade and certainly not for starting an auto
mobile, and yes, they are also good for wheel chairs and scooters.
Because UPS and fire-alarm batteries are made to be continiously
charged. Deep cycle batteries are more expensive and therefore
are typically not used for these applications.
This is just from what I have gathered over the years.
Maybe I'm wrong.
Well, I'm all about "why." I've done a bit of research on battery technology
over the years, and if you know something I have not come across I'd like
to read it. I just have a quark about hear-say science, like "you are safe
in a car, during a lightning storm, because of the rubber tires*." I'm sure
the batteries I have are deep cycle according to what I know, and I reject
the notion they they are not without any supporting information that can
*And yes, I know it is *NOT* because of the tires. I almost shot my TV when
they said this widely believed and obviously wrong nonesense on CSI.
OK. From my experience with UPSs, the batteries die after only a few
times of deep useaga (power failure with nobody present to turn
things off so the cells were drained).
Heck, I've seen a UPS killed because their usage was just a
bit above the charging rate (one of our artists got a bigger
monitor than was planned for).
Replacing the batteries fixed the first case, and replacing the
UPS with a bigger one fixed the second.
I haven't had experience with motorcycles in years, but if
they're like car batteries, they aren't the same if they've
been drained even once.
From what I know of business, extra features cost extra money.
If the normal useage doesn't need it, then they'll drop
Deep cycle useage makes a battery that isn't as good at
being a motorcycle or UPS battery (where constant charge
and complete draining means other badness). And it costs
more to make a battery that is both deep cycle and has
On the other hand, wheelchair batteries are meant to be
driven into the ground.
This exactly mirrors my experience. I have no reason to believe that all
SLA batteries for a UPS are deep cycle, or that all SLAs are deep
cycle.. UPS's are a competitive product. I'm not sure they'd always use
the more expensive deep cycle construction for a device that only rarely
This isn't to say some of the better UPS's may not use deep cycle
batteries. I'm sure the commercial and industrial ones do. The consumer
APCs that I've hacked certainly didn't. They failed to keep a charge
after about a dozen full cycles. It seems reasonable that to keep costs
down they used thinner plates. Thinner plates=less lead=less $$$.
SLAs for solar backup, however, are made for deep cycle. They're more
expensive, but I've used a 12volt AGM battery for about two years and
deep charged it dozens upon dozens of times, and it still holds a charge
like it did when new. Far better than the motorcycle batteries I used to
use. For one thing, they periodically dried out.
As I have posted to Jay, there are many depths to "deep cycle." The internal
construction of the battery defines the depth of deep cycle.
The are differences in the electrolite, AGM, gel, and liquid sulfuric acid,
there are differences in plate construction, plate thicknesses and whether
or not they are solid, etc. "Deep cycle" is applied to batteries that are
neither "starting" nor "marine." Starting batteries should not be
discharged more than about 10%, marine no more than about 50%, and deep
cycle no more than about 80% of capacity.
Some deep cycle batteries may be able to survive getting cycled lower than
20%, but they cost a lot more.
Also, you can extend battery life a lot by not deeply discharging the
battery. If you get a larger battery and only discharge it to 50% of
capacity, it may last (if I remember correctly) twice as long.
UPS and fire alarm applications are a "very" bad situation from a battery's
perspective. Being on a charger 24x7 will kill any battery.
Like I said, UPS systems constantly charge the battery, even the systems
that periodically test the battery and apply a charge when needed will kill
a battery in a couple years because of over charging.
I don't recall if bikes have deep cycle or not.
I'm not sure that "Deep Cycle" batteries are more expensive to make. They
generally have fewer and thicker plates. I can see a "cheap" battery with
thinner plates, but it would not take long for the various testing houses
(UL, CA, etc.) in the fire alarm business to rate these as poor
My cousin partially owns a fire alarm company and of the things that real
estate holding companies skimp on, fire alarms aren't one of them. No cost
is too high compared to insurance premiums.
Few lead-acid batteries will survive complete draining, a good deep cycle
battery will typically survive discharge to 20% of capacity.
A deep cycle battery that has low resistance is expensive, true. To be deep
cycle, it need to have thicker plates so they don't warp or disintegrate
There are a lot of different types of lead acid batteries and a lot of
different depths to "deep cycle." For instance, an open non-maintenence
free battery with caps that use lead-antimony for very strong plates and
can thus be discharged deeper than a sealed and/or gel battery with
lead-calcium plates which are chemically more stable and will have longer
What, I think we are talking about is "battery life," i.e. how much
discharge can it have and how many, and how long just sitting on a shelf.
Lead acid batteries don't have "memory" which is good, but suffer a great
deal of damage on deep discharge.
Gel type deep cycle batteries will fail sooner than AGM batteries (AGM
absorbed glass material, sulfuric acid soaked fiberglass), pure lead/liquid
batteries typically out last both.
AGM and liquid batteries are a problem in a hobby robot that may tip, so
sealed gel type is what is called for. Since I am using a gel battery, I
have to expect about 2-4 years life and 40%-50% of charge cycle.
For such a pedant (e.g., "There are a lot of different types of lead acid
batteries and a lot of different depths to 'deep cycle.' "), this is sort of
a ridiculous thing to say, isn't it? At the limit, you could have a human
being in the loop as part of the "charger". He would decide when the
battery should be charged using whatever optimum algorithm there might be.
If being on that "charger" would kill any battery, what wouldn't? Now just
replace the person with a smart charger.
And why is trickle charging such a bad situation from the perspective of a
lead acid battery?
Why? There are a lot of different types of lead-acid batteries and they
provide different resistences to deep cycle damage.
When ever there is current flowing through a battery there is a chemical
reaction happening, there is *always* some amount of damage that happens to
the battery. Some less damging than others. Within normal charge range, the
battery suffers very little damage. During overcharge or deep discharge the
battery suffers greater damage.
It is hard to make a battery charger that does not over charge the battery
or cycle the battery a lot. It just is. Battery charge decays with time. A
UPS tries to to keep it at 100% charge. The battery is constantly
discharging by some small percent (with decay and voltage monitor current
drain) and being recharged to full when the tricle charger detects the
voltage drops below a certain point. This is a small discharge/charge
cycle, and over time they add up, more so than charging and disconnecting
it and keeping it on a shelf to be charged next week.
Still, a UPS battery will last a year or two, maybe even three and be
There are dozens of battery charging chips available which claim to use
near-optimal charging algorithms. Do you suspect that the offerings of, for
example, Maxim or TI are inaccurately described?
Assuming that there is some problem with trickle charging, why is it hard to
make a charger that charges a battery once a week? Prohibitive with Linux
maybe, but it just might be doable with a microcontroller.
Just a nit, but this depends on the charger. Being on a _crappy_ charger
24/7 will kill a battery. A battery can be left pretty much indefinitely
on a charger properly designed for the battery chemistry. Invariably, a
good charger is usually worth the money to buy or the time to build.
(Replies: cleanse my address of the Mark of the Beast!)
As I posted in another thread, this isn't really true. A battery will
naturally discharge over time. Constantly charging/discharging +- 1% will
add up over time, and a battery left on a shelf and tricle charged one
night a month will probably outlast a battery left on the charger 24x7
A "reference" that supports exactly "this," no, it is my own conclusions
based on the science of chemical energy storage and the data I have seen.
Fact: All charge/discharge cycles of a battery damage the battery to some
Fact: The number of charge/discharge cycles a battery takes has an additive
effect on battery life. The deeper the discharge the greater the damage,
but number of cycles have an affect as well.
Fact: Lead-calcium acid batteries are a more chemically stable, providing
longer shelf life.
Fact: Plates in gel or ACM batteries are typically lead-cacium plates.
Assumption: the battery is not over charged or discharged beyond some lowest
Conjecture: reduce the number of charge/discharge cycles a battery has,
reduce the amount of damage.
Batteries last longer in storge than when in service in a UPS or fire alarm.
This indicates that charging, no matter how well designed has a negative
affect on the battery.
There are tons of books that describe lead acid batteries, and batteries in
general. You can even google for white papers.
So you are saying a Gel, SLA (AGM) or simple lead acid will last longer on a
a good trickle charger 24x7 that it will being trickle charged once a
All lead acid batteries degrade over time and have a finite life no matter
how they are handled, however, there are methods of handling that will make
them last longer. I think we agree on this.
So, I think we have the basis of a bet.
We need to acquire 2 identical SLA (gel or otherwise) made in the same
batch. Two identical tricle chargers. One connected to one battery 24x7,
and the other connected to the other battery one 24 hour period every
calendar month. At the end of a year, which of the two batteries will have
the greater capacity?
Are you up for the bet? I am, to me its a no brainer, I'll take the battery
charged once a month.
I said "a good charger". To give you an idea of exactly how well a good
charger works, REGIS has been using the same 5ah SLA (electronics) since
November 2000. For most of that time, the battery is in use around 12-14
hours a day, 7 days a week, and is on the charger nightly. The battery
has been frequently deep-discharged on more occasions than I care to
count. Only recently is this battery showing possible signs of fatigue,
and this is no special battery (Panasonic LCR1-125P1) . The same can be
said of the 7ah sla that powers the motor, although it is charged much
less frequently (and often deeply discharged).
So if I did nothing with the battery other than left it on a shelf and
charged it monthly would it last longer? Possibly -- I certainly
wouldn't take your bet. Even if you didn't win, I would have tied up a
$60.00 charger for a year. (Not to mention that, honestly, I don't know
you well enough to actually wager anything.)
Basically, if the battery is allowed to discharge, you'll end up with
some amount of sulfation on the negative plate (not good). But charging
the battery always carries some risk of erosion of the positive plate
(also not good). A good, temperature-compensated float charge algorithm
_should_ strike a reasonable compromise between the two, however, making
the real difference academic. In fact, some float charge algorithms
monitor the battery voltage and only apply a charge when the voltage
drops below a preset value, essentially mimicking your "charge it once a
month" approach -- the real difference between the two being that the
user isn't required to remember to charge the battery once a month.
(Replies: cleanse my address of the Mark of the Beast!)
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.