Ferrite transformer saturation point?

[Llamarama]

Hello everyone, I've been stripping down a couple of dead boards and salvaged about a dozen ferrite core transformers from them to use in a SMPS circuit to drive an old mini ITX board. I have a load of different sizes and have the dimensions of all of them, but is there an easy way to calculate the saturation current? I don't know what the ferrite material constants are, and i've only worked it out from inductance a couple of times with simple rods rather than these E and I structures.

Anyone got any pointers? Thanks, Mike.

 

[crutschow]

Drive them with an adjustable amplitude sine-wave and monitor the input current. When the current starts to distort and increase rapidly at the peak, that's the saturation point.

Alternately, apply a small step voltage to the input and monitor the current increase with time. Where the slope of current ramp starts to rapidly increase, that's the saturation current.

 

[Llamarama]

Thanks! Just measure with an oscilloscope and a multimeter?

 

[crutschow]

Thanks! Just measure with an oscilloscope and a multimeter?

You can use an oscilloscope to measure the voltage across a small resistor in series with the transformer to determine the current. A multimeter is too slow for such measurements.

 

[dr pepper]

You can use the universal transformer equation to work out the flux density in the core, ferrite is safe around 100mT, some cores can take 250mT.

By winding 10 turns and measuring the inductance you can caluculate the al factor, you can also use this as your test winding, you'll need a high freq, 30 - 40 khz.

If you scroll half way down this page you'll see an inductor test bench, I have made one of these, most usefull, I put the sense resisitor in the ground leg o fthe fet, so my 'scope ground is at ground level.

https://www.dos4ever.com/flyback/flyback.html

 

[MrAl]

Hi,

A nice way to test inductors is to build a buck converter and use the device under test as the inductor.
As you increase load, you'll see the inductor current start to spike up quickly after a certain point, and that's the point where it starts to saturate. The buck converter doesnt even need feedback, just some protection for the pass transistor so that it does not burn up if the inductor saturates too much.

It is also good to know the DC resistance, and inspecting the wire tells you a little about the DC current it can handle.

 

[EinarA]

The above methods all require a scope, something many home builders do not have. I have a circuit that only requires a voltmeter, but I have sent it in to EDN and am not sure if giving it here would violate their rules about prior publishing.

 

[dr pepper]

ok, can you describe hot to use it then?

 

[chemelec]

You can use the universal transformer equation to work out the flux density in the core, ferrite is safe around 100mT, some cores can take 250mT.

By winding 10 turns and measuring the inductance you can caluculate the al factor, you can also use this as your test winding, you'll need a high freq, 30 - 40 khz.

If you scroll half way down this page you'll see an inductor test bench, I have made one of these, most usefull, I put the sense resisitor in the ground leg o fthe fet, so my 'scope ground is at ground level.

**broken link removed**

Just 10 Turns does Not give a very accurate AL Value.

Check this out:
http://chemelec.com/Projects/Inductor/Inductor.htm

With E & I Cores, Gapping them is usually Better. With a .001 to .004 Inch Gap.

 

[dr pepper]

Nice info, like you say on your site 10 or 20 turns gives you a ballpark idea, this has always been ok for me.

 

[Llamarama]

Thanks for the extra ideas. I'm very limited on equipment, but I do have an analogue scope from the 70s. I think I'll mess around with them tomorrow. One things for sure, I'll get better at winding bobbins either way!

 

[Llamarama]

Thanks for the extra ideas. I'm very limited on equipment, but I do have an analogue scope from the 70s. I think I'll mess around with them tomorrow. One things for sure, I'll get better at winding bobbins either way!

 

[throbscottle]

That's really interesting. I understand flyback converters a bit more now

 

[dr pepper]

I used to be scared of flybacks too, but they are not that bad to build, at least in simple form.

Your 70's 'scope should be fine, anything with 1mc's or more bandwidth out to do the job.

Still not sure how you can measure saturation with a meter, you'd must detect the change of di/dt, maybe with a filter.