Hi Martin,
Thanks for the info on the circuitry of the switching supply. I've never
seen the actual schematic for one, so I was going by discussions I've heard
on the air. Your block diagram description was excellent and I've saved it
for future reference.
As I've said before, I've had very good luck with switching supplies, though
I have had a couple of really noisy switching wallwarts that I just had to
throw away!
73
Bob Tinney, K8LR, [log in to unmask]
Skype, bobtinn
Life is full of challenges, that's what makes it interesting!
----- Original Message -----
From: "Martin McCormick" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Friday, November 14, 2008 3:17 PM
Subject: Re: switching supplies cause interference
Bob Tinney writes:
> I may be wrong on this, but I believe that the 110 volt 60 hertz line
> voltage is converted to a frequency of from 20 to 100 kilohertz which is
> then sent through a very small transformer to deliver the output voltage
> that is then rectified, filtered and regulated. The higher switching
> frequency allows for a very small transformer with very few turns of wire.
> The old TV flyback transformers used to work in the same way except they
> stepped up the voltage to over 20000 volts which was then applied to the
> annode of the picture tube.
In a switching supply, the AC line voltage first goes
through a bridge rectifier and in to one or more filter
capacitors so it comes out as about 170 to 180 volts DC. The
capacitors charge to the peak voltage of the wave form which is
why it is that high.
The next stage is the transistorized chopper which is
just like a TV flyback transformer setup except that the flyback
runs at the horizontal sweep frequency while switching power
supplies do run much higher frequencies so you aren't likely to
hear them but they sure can make radio noise. They are basically
a several-hundred-watt transmitter. The transformer is a
step-down transformer, all right but the voltage may be too high
when rectified. There is a very good regulator that produces the
desired voltage such as 12 volts or 5 volts and the output of
this is compared with the DC output of the supply. That produces
an error voltage, let's say, positive for too high and negative
for too low.
That voltage is fed through a low-pass filter back to
the oscillator circuit and hopefully will adjust either the
frequency or the duty cycle to reduce or increase the output
from the transformer to try to keep it as close as possible to
the same as the high-quality regulator.
And that's a general description of how switching
supplies work. They are called that because they switch the DC
on and off and then use chokes and RC filter networks to turn
that high-frequency pulsating DC in to nice pure DC that
hopefully is free of any of the RF components that made it.
Sometimes, switching supplies will even make sounds when
they are in a bad situation. Several years ago, here, we lost
one phase of power going in to the building that housed all our
network gear at the time. The voltage probably dropped to about
half what it was supposed to be and I remember standing in a
large room full of very unhappy switching supplies. They were
trying to regulate but just couldn't and their oscillators were
very much in the audible range. Imagine about 50 muffled smoke
alarms going off, each one a slightly different pitch. It was
weird to say the least.
It was also about 2 to 3 weeks before we were able to
get all our network gear properly working again. All that low
voltage caused the devices we used at the time to loose parts of
their configurations but not all so that one would think
something was okay until people started complaining about things
that were wrong.
Anyway, switching supplies are really nice when they work
but can make RF noise and fail in odd ways at times.
Martin McCormick
Systems Engineer
405 744-7572 Stillwater, OK
Information Technology Department
Network Operations Group
|
