[Amps] Power factor correction

[Manfred Mornhinweg]

A quick note to all those who are still interested in that power factor 
theme:

Yesterday I repaired a 22kW motor drive, like there are millions being 
used in industry. This is a quite modern one, more modern than the ones 
I dealt with at my former job. So I was expecting to find it having 
active power factor correction - but it doesn't! Instead it has plain 
simple passive PFC using nothing else than a swinging choke between the 
rectifier and filter capacitor, and according to its specifications it 
achieves a power factor of 0.98 with that simple method!

This is a surprise for me. I would have expected something less good 
from passive PFC, with 0.98 and higher being reserved for active PFC.

For those of you who don't know these "drives", as they are commonly 
called: They are basically a line-connected rectifier, filter, followed 
by a three-phase IGBT bridge, along with a microprocessor circuit that 
implements quite sophisticated control, and drives the IGBTs. The output 
is three phase, variable frequency, voltage and current, created by 
pulse width modulation at a few kilohertz, that powers a motor, whose 
speed, torque and even angular position can be accurately controlled.

These things exist in power levels from a few hundred watts to several 
hundred kilowatts, and probably even higher. The one I repaired 
yesterday is for 22kW, accepting an input of 380 to 480V, three phase. 
For the doubters among you, all its power electronics are of course 
directly connected to the mains. It is intended to be installed by 
competent electricians, and used by anyone. It's made in Finland, and 
meets all European directives applicable at the time of its manufacture, 
in 2007 - including leadless solder, much to my pain, because with its 
higher melting point it was a chore to unsolder the IGBT module with its 
24 thick rigid pins rated for 50 amperes each.

There are some more good hints we can take from units like this. For 
example, the input rectifier is rated at 1600V peak inverse voltage, and 
the beast is rated for up to 480V AC input. That means, they are using 
less than twice the diode voltage, relative to the peak AC voltage. And 
this is for a highly reliable industrial application. That shows how 
exaggerated it is to use 4 times higher diode voltage in a ham amplifier.
Also the current rating of the diodes: The bridge is rated for 53A 
continuous DC output. The actual DC current used is up to 45A. So the 
headroom they use is really small.

Despite that, this unit failed from poor installation, not from anything 
attributable to its design or manufacture. The installer failed to use 
the cable restraints provided by the equipment. He left the heavy cable 
hang free from the contacts! Eventually a wire worked loose, made a 
short to  the chassis, creating a formidable arc that caused secondary 
arcs throughout the power circuitry, killing essentially all 
semiconductors in the area.

Even the best design and construction of equipment cannot prevent users 
from killing it through stupid or careless actions. And that's as true 
for industrial equipment as it is for ham gear!


Manfred


========================
Visit my hobby homepage!
http://ludens.cl
========================