Maker Pro
Maker Pro

How to detect transformer saturation?

  • Thread starter larry moe 'n curly
  • Start date
L

larry moe 'n curly

Jan 1, 1970
0
Is there a way to detect when a switching power supply's isolation
transformer is starting to saturate? I've read, in a Linear Technology
application note, that it can be done with a current probe and looking
at the shape of the ramp -- straight line is OK, curved like a parabola
is bad. But is there an easier way that doesn't require a current
probe?
 
C

Charles Schuler

Jan 1, 1970
0
larry moe 'n curly said:
Is there a way to detect when a switching power supply's isolation
transformer is starting to saturate? I've read, in a Linear Technology
application note, that it can be done with a current probe and looking
at the shape of the ramp -- straight line is OK, curved like a parabola
is bad. But is there an easier way that doesn't require a current
probe?

Put a series resistor (0.1 ohms or so) in and connect an o-scope across it.
 
N

n cook

Jan 1, 1970
0
larry moe 'n curly said:
Is there a way to detect when a switching power supply's isolation
transformer is starting to saturate? I've read, in a Linear Technology
application note, that it can be done with a current probe and looking
at the shape of the ramp -- straight line is OK, curved like a parabola
is bad. But is there an easier way that doesn't require a current
probe?

heat?
 
M

Mark

Jan 1, 1970
0
if you are working on power supplies and you don't have a current
probe, get one, you'll be glad you did...

Mark
 
Most SMPSes have a current sensing resistor at the emitter or source of
the main output device. Just scope across that.

If it doesn't have one, Charles is right, you can add one. You must
however think about two things, one is that it must be a small
resistance, and two if the chopper is bipolar, you need to make sure
that degeneration of the drive current does not occur. What's more if
the chopper is bipolar, you need to figure that some of what you're
seeing is base current. It is not advisable to do this at the collector
or drain, it's a shock hazard and radiates too much. You can't go to
the AC rectifiers because of the nature of the signal you need.

I cannot completely agree that the current waveform of the chopper
being parabolic indicates transformer satuation in every case. Certain
time constants, snubber schemes and other things chosen by the designer
could cause an innacurate conclusion on your part. I think you can get
really meaningful data just from the efficiency. But then you might
have your own reasons for wanting what you want.

I'd be interested in the application, or at least the capacity of this
PS, is it like huge or something or are you trying to miniturize ? Just
curious.

JURB
 
Top