[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: (ET) Duracell Batteries
So, should I buy the Trojans or the Duracells?
Mike in KY
-----Original Message-----
From: David Roden [mailto:etpost drmm net]
Sent: Thursday, June 18, 2015 2:17 AM
To: elec-trak cosmos phy tufts edu
Subject: Re: (ET) Duracell Batteries
On 17 Jun 2015 at 12:17, bushman165s aol com via Elec-trak wrote:
> most GE chargers that I repair put out about 42VDC.
Is that measured open circuit, or when it's connected to the battery? It
makes a difference, because of the way voltmeters average pulsating DC.
You want to read the voltage it fires into a fully charged battery.
That's
what matters, and most likely it'll read higher than the open-circuit
voltage. For gels, this voltage should be between 42.3 and 42.9 volts.
If it's too high, you might be able to adjust it somewhat by changing the
value of the capacitor. I've never tried this, so I don't know if you
could
get enough wiggle room.
Even if the GE charger does produce the right output voltage, you also
need
to eliminate the battery taps that imbalance the batteries, so you don't
have to beat them over the head to equalize them on every charging cycle.
Converting the lights to run on 36v would be pretty easy; just use
forklift
bulbs. The lift is more of a challenge. A big honkin' regulator might
work,
burning off the excess voltage as heat. I'd have to see how much current
the lift draws at peak. A more efficient but more expensive solution
would
be a DC:DC converter with a 14-18 volt output.
Still another idea : give the lift its own smaller onboard battery. If
you
replace the GE charger with something more modern and smaller, you might
have enough room for an extra 12v deep cycle battery up front.
> I think I could use one of the Landis controllers with the setpoint at
> 42.5VDC or so to keep them from overcharging.
Ehh, not so fast! I'm not so sure that's going to help much with gel
batteries. The Landis controller doesn't quite work that way; it's a bit
oddball.
To explain what I mean, first a little about basic battery charging
control,
that is, ways that chargers fill up a battery right to the brim, not
either
undercharging or overcharging. If you already know this stuff, just skip
ahead.
One of the simplest methods is to just regulate the charger's output
voltage. If you limit it to the battery's theoretical fully-charged
voltage
(ideally compensated for temperature, because a battery's fully charged
voltage depends on its temperature), it *can't* overcharge the battery.
As long as the battery voltage is less than the charger's voltage, current
flows into the battery. As the battery charges, its on-charge voltage
rises, and the charging current tapers off. Thus this is called a taper
charger.
When the current falls to about 0.05C (2% of the battery's 20 hour amp
hour
capacity expressed as amperes), we consider the battery fully charged.
If properly set up, taper chargers are pretty safe for the battery. Their
main fault is that they can be maddeningly slow. It can take days or even
weeks to completely fill up a big battery.
So the engineers have developed ways to increase the current in the bulk
phase (up to 80% SOC, gassing voltage), without overcharging later.
Generally they use some kind of electronics to monitor and regulate
voltage
and current. From there they diverge on best strategies to do the job;
this
is where you get such alphabet soup as IU, IUI, and DV/DT. (Look them up
if
you're interested.) What they all have in common is the electronics,
which
of course adds to the charger's cost and complexity.
At the bottom end of these smarter chargers, carrying a C+ average, we
find
a relatively simple design that I've seen called a "cycle dropping"
charger.
A cycle dropping charger starts with an output voltage high enough to
charge
the battery fairly rapidly during its bulk phase (up to gassing voltage).
To prevent overcharging with this higher charging voltage, the controller
monitors the battery's on-charge voltage. When it reaches the fully
charged
voltage (for our purposes, let's say 42.6 volts), the charger shuts off
its
output. The battery's surface charge dissipates, and its voltage falls.
When it falls below some lower threshold voltage (there's some hysteresis
in
these voltages; let's say it's 42.1 volts), the charger restarts.
This cycle repeats many times. As the battery approaches full, the on
time
shortens and the off time lengthens, eventually reducing the average
charging current to the value I mentioned above (0.05C).
At this point the charger *should* turn off, but I've never seen a cycle
dropping charger that actually does. Instead, the manufacturers will
usually tell you that it "maintains the battery at full charge." You
should
read this as "overcharges the battery if you leave it for very long." Ask
me how I know this.
As I said before, cycle dropping is kind of crude as charging algorithms
go.
But if it's used carefully (I always put these chargers on timers), it
actually can work tolerably well for many applications.
So now we come to Harry Landis's controller. I may have missed some, but
I've never seen any other charge controller like it in the marketplace.
The
best way I can think of to describe it is that it's a cycle dropping
charger
in reverse.
It uses a threshold voltage, but it defines a *minimum* voltage to turn
*on*
the charger, instead of a *maximum* voltage to turn it *off*. Any time
the
battery falls *below* some voltage value, the charger turns on and runs
for
a fixed time (5 minutes, IIRC).
You can see that this isn't what you really need for your gel batteries,
because there's no actual top limit on the voltage. In fact the Landis
controller is more of a battery maintainer than a cyclic charge controller.
I know some folks here use it for a charge controller, but that's not what
it's really designed for. If it works for you, great, but I personally
wouldn't (and don't) use it that way.
That gives me an idea. Are you up for building your own charge controller?
Here's a schematic diagram of the charger used in Comuta-Cars in the early
1980s.
http://www.evdl.org/docs/c_car_charger.jpg
This circuit is for a 48v battery, but it should be easy to adapt to a 36v
battery. I'm not an EE, but I think all you'd have to do is change the
value of R4, and maybe of R6. In fact I just might try that for my New
Idea
rider.
A nice feature of this circuit is that L1 (a small 12v indicator lamp)
lights up when the charger is off, so you can see how the charge is
progressing. You could replace this lamp with a resistor and LED.
Anyway, to finally cut to the chase, if you want to go with the gels, I'd
recommend you either addd some smarts to the GE charger, or replace it
with
a modern smart charger. If you do neither of these, at least make sure
the
GE stays below 43 volts when charging the battery.
Either way, I also recommend getting rid of the battery taps to fix the
battery imbalancing problem.
David Roden - Akron, Ohio, USA
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Note: mail sent to the "etpost" address will not reach me. To send
me a private message, please use the address shown at the bottom
of this page : http://www.evdl.org/help/
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
_______________________________________________
Elec-trak mailing list
Elec-trak cosmos phy tufts edu
https://cosmos.phy.tufts.edu/mailman/listinfo/elec-trak