# SMPS transformer saturation

Discussion in 'General Electronics Chat' started by Disaster plus, Feb 19, 2013.

1. ### Disaster plus Thread Starter New Member

Jan 29, 2013
3
0
Hi,

I am trying to build a 200W 100V switched mode power supply. I have found a transformer that I think is suitable, but I'm worried about it saturating. Is there a way I can relate the specified magnetising inductance to the maximum operating conditions that cause saturation?

I came across an equation in a TI application note (bottom of page 3) that seems to do this, but it doesn't seem right to me:

Lmag>______Vin(1-D)______
[ΔILx0.5xFPWM/(Ns/Np)]​
As far as I can tell this equation seems to describe the inductance required to store enough energy for the output during the Toff period, but I thought that no energy is stored in the transformer for bridge SMPSs.

As you can see, I am very confused, and some pointers would be very appreciated. Thanks,

DP

2. ### crutschow Expert

Mar 14, 2008
20,283
5,739
According to the data sheet, that equation is to determine the minimum magnetizing inductance of the transformer so that the circuit operates properly and the magnetizing current is not unacceptably high. The magnetizing current is a necessary (but parasitic) part of transformer operation. Ideally you want it to be near zero but that requires a large transformer, so in practice you build the transformer as small as possible while still having an acceptable value of magnetizing current.

3. ### JMac3108 Active Member

Aug 16, 2010
349
67
Make sure you're using the right kind of transformer for the SMPS topology that you're using.

If its a forward converter or bridge topology then it uses a regular transformer. If its a flyback, the transformer is completely different.

Since you didn't mention the topology I'm just saying to check. At 200W its probably not a flyback, but you never know.

4. ### Disaster plus Thread Starter New Member

Jan 29, 2013
3
0
Thank you. I was planning on using a half-bridge topology with a diode bridge on the output.
How can I determine an acceptable magnetising current?

5. ### crutschow Expert

Mar 14, 2008
20,283
5,739
The equation you referenced has a magnetizing current value built into it. You don't have to concern yourself with what it actually is.

Typically you design so that the core is near saturation with the maximum input voltage and the minimum frequency. That will usually give an acceptable value of magnetizing current.

6. ### nigelwright7557 AAC Fanatic!

May 10, 2008
527
74
If the transformer hits saturation there will be large increase in current and will probably make your SMPS IC go into reset.

For flyback SMPS you need a gapped transformer anyway which will help stop tendency to saturate.

For half bridge you don't need a gapped transformer just the right amount of primary turns to give maximum output current.

7. ### kubeek Expert

Sep 20, 2005
5,297
996
Where does the datasheet state the saturation magnetization, or core type and dimensions?

8. ### crutschow Expert

Mar 14, 2008
20,283
5,739
That is given by the transformer manufacturer.

The IC manufacturer may recommend some particular designs for an SMPS design but otherwise you're on you own. It requires a knowledge of core types and transformer design to roll you own.