Re: (ET) Found the problem with the charger

["David Roden (Akron OH USA)" <etpost drmm net>]

On 15 Jul 2012 at 12:53, The CZ Unit wrote:

> By the way under load the voltage (AC) across the capacitor drops, my
> guess is that is the current control. Neat. 

Sort of.  Here's Lee Hart, explaining ferroresonant transformer function 
on 
the EVDL.  He provides some additional info that you may find useful if 
you're going to be tinkering with a lithium battery in your ET.

It's interesting how a constant voltage transformer works. Roughly
speaking, it's a standard transformer but wound with a slightly lower
primary voltage. Say, a 100vac primary for use on 120vac. The
transformer will saturate on the AC line peaks. When it does, the output
voltage falls to zero.

By itself, this would regulate the output voltage. However, the primary
current goes way up when the transformer saturates! So, there is a large
extra inductance in series with the primary. When the primary saturates,
its voltage falls to to zero, and all the AC input voltage appears
across the inductor. This limits the primary current. When the line
voltage moves past the peak, the inductor gives up its stored energy.
Since it's an inductor, it doesn't lower efficiency like a resistor would.

But this inductor would give [the charger] a poor power factor. So, there 
is 
a large AC capacitor across the secondary. It adds capacitive reactance to 
cancel out the series inductor's inductive reactance, moving the power 
fact 
back toward 1.0.  

The inductor isn't a separate part; they design the transformer to
include it. It also has special low-loss core materials because it will
operate in saturation. Saturation produces the characteristic "buzz" of
a constant-voltage transformer.

The simplest way to alter the voltage of a constant voltage transformer
is to change the number of secondary turns. This can be done by adding a
few turns, which you switch in series (aiding or opposing) to change the
voltage. There is almost always room for more turns.

The other way is to change the value of the AC capacitor. Less
capacitance lowers the output voltage, and more capacitance increases
it. This method is only good for perhaps +/-10% because it also affects
the power factor and voltage regulation characteristics.

The easiest way to adjust the capacitor value is to put a phase
controlled light dimmer in series with it. The phase angle adjusts the
apparent capacitance, and thus the output voltage. Note that the light
dimmer needs to be rated for at least 240vac, because the secondary and
capacitor voltage is often 240vac (or more)!

Most constant voltage transformer chargers are deliberately designed to
have a "taper" charging algorithm. They behave like a fixed voltage
supply with a series resistance. It delivers maximum current at the
lowest pack voltage, and the current falls linearly as the battery
voltage rises. But it has two extra useful features; The output is
intrinsically current limited (won't deliver more than about 150% into
even a shorted battery), and zero-current voltage is only about 120% 
higher.

But when it's desired, [a ferroresonant transformer charger can be 
designed 
to] have a nearly flat voltage vs. current curve. Output voltage can be 
within +/- 2% for output currents from 1/4 to full load, and for a +/- 20% 
change in AC line voltage. This makes a good power supply, but a poor lead-
acid battery charger. It could be useful for charging lithium batteries, 
however.