Re: Opinions [Isolated Charger]

["Joe Smalley" <joes@worldfront.com>]

Comments inserted:

> > Essentially no charger is 95% efficient, whether isolated or
> > non-isolated. 70-80% is probably typical for isolated chargers, though
> > 90% is attainable. A non-isolated charger can be perhaps 5% better. Of
> > course, there is no limit to how LOW the efficiency can be; the "wall
> > wart" chargers supplied with many consumer products are like 10-20%
> > efficient.
>
> Based on the heat given off by my PFCW-20, I'd tend to put it up there at
> 95+%.
>
> It warms a liter (sp?) of water about 25 deg C in operation, while moving
15
> kwh from wall to pack. I unfortunately didn't measure how much my water
> heated up when I did a full 20 kwh charge - I've only gotten the pack that
> low once, it's kind of hard to do.

I would like to see the results of your tests. Doing it with only 15 kWh out
of the pack is fine.

> Anyway, Joe Smalley probebly has some 'hard' efficiency numbers. How about
> it, Joe? This thing 95% efficient, given my pack voltage [268] and line
> voltage [220] ?

There are so many combinations of input voltage, output voltage and throttle
settings that I have not tested them all. This is one combination I have not
tested.

The rough calculations for the losses are:
Two diode drops times the current = ~40 watts.
Two IGBT drops times the current = ~60 watts.
IGBT switching losses = ~152 watts
Inductor losses = ~18 watts
Hotel power = ~15 watts (including ~5 watts of fan power that Sheer does not
have)

Total losses is ~283 watts. If the charger is drawing 5000 watts, then the
efficiency is 94.3%. This is just short of Lee's predicted maximum of 95%.
If we dropped the switching frequency in half and doubled the size of the
inductor core, then the charger could exceed 95% efficiency. This would make
the charger height grow a half inch and the charger would cost $100 more. We
thought the lower price and lower housing height were more important than
another percent efficiency. Engineering is the art of compromise. As you get
closer to 100% efficiency, each percent gets more costly.

> > The simple fact is that people don't care about efficiency; they only
> > care about what they pay. In most cases, they buy low efficiency
> > products because they cost less, and never add up the long-term energy
> > cost due to that low efficiency.

If you are saving 1% of your 10 cent a kWh energy, it would take 100000 kWh
of energy to pay off the difference in efficiency. If you use 10 kWh per day
commuting (40 miles at 250 Wh per mile) then it would take 10000 days (27.4
years) to pay off the increased cost.

Joe Smalley
Rural Kitsap County WA
Fiesta 48 volts
NEDRA 48 volt street conversion record holder
joes@worldfront.com