Have coil, need core data

[walker1]

I have found a ready made, all glued up 170uH coil of unknown origin in a box and would like to use it in a buck switch mode converter.

The output current need to be 7-10A, so 170 uH should be more than enough. (50uH has been found acceptable elsewhere)

However, I am not sure the coil can cope with the DC current involved. The wire is plenty thick enough, but the magnetic core .. !

The core is what Ferroxcube refer to as an RM and has the markings 3B8 and A250.

http://www.ferroxcube.com/prod/assets/rm14i.pdf

is the best fit, I have found so far mechanically.

3B8 is most likely the core material, but I can not find info on it anywhere through Google or elsewhere.

Is there a way to test the coil to see if the core can handle the DC current without the field collapsing before I blow up my switch transistor ?

 

[walker1]

Found an old Philips catalog (1991). 3B8 seems to be one of their materials.
If the coil can be used is still unknown.

Philips does not seem to make ferrit cores anymore, or .. ?

 

[cbarn24050]

Hi, the dc current rating is soley dependent on the wire size, cooling and allowable temp rise. The saturation point of the core depends mostly on the air gap in the core (if it has 1). You will just need to experiment.

 

[ScottyUK]

Hi Walker1,

I have a bit of data on this core:

3B8 is indeed the core material and is specifically intended for DC-polarized applications.

Based on an RM14 core, as listed in Philips Databook MA01 'Soft Ferrites', the core with an Al value of 250 has an airgap of 950um.

Core Factor C1 = 0.353mm^-1
Effective Volume Ve = 13900mm^3
Effective Length le = 70.0mm
Effective Area Ae = 198mm^2
Minimum Area Amin = 168mm^2

Typical conditions:
Bmax >=315mT @ H=250A/m, f= 25kHz

Consider the allowable ripple, the switching frequency, and the energy storage capability of the core. The three things are inter-related: decide two of them and the third is calculable.

cbarn,

The DC rating is also governed by the maximum permissible flux density in the core. The air gap changes the permeability, and thus the required MMF, but not the maximum flux density in the core.



----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[cbarn24050]

Hi Scotty, your not right there. DC effects are thermal, saturation does not damage the core although it does change the inductance. Increasing the air gap means that saturation occurs at a higher current level assuming that the turns remain the same. This means that you cannot predict the inductors performance from knowing the core size and material unless you also know the air gap.

 

[Warpspeed]

This will probably horrify the purists, but....

After calculating the ac flux density and current density in the wire, I would just fit it to the circuit and monitor the current waveform in the inductor. Gradually increasing the load while watching the rate of current rise and fall.

It should be a nice clean strait edged triangle wave if the inductance remains fairly constant. If the current starts to curve upwards into a spike at increasing load, the core is saturating and more airgap will be required.

Provided temperature rise of core and winding is acceptable, it is an as rough as guts practical method of proving the suitability of a seemingly suitable inductor.

 

[walker1]

I just had a go at

http://schmidt-walter.fbe.fh-darmstadt.de/smps_e/smps_e.html

It seems I will be able to use the coil at 50 kHz, if I stay below 7.5 Amps or so, which is acceptable.
RM14, Al=250 from Siemens, listed as number 55 under "Coil Data", seems to have the same spec. as the Philips thing. And it turns up green!

A great site by the way!
(and in English too!

 

[Warpspeed]

Excellent !

Another little rule of thumb trick you can try.

Set it up running at some reasonable combination of input voltage, output voltage and load. Try adjusting the operating frequency with a variable tuning capacitor while monitoring the input current. Sometimes a definite dip can be found where combined circuit losses are at a minimum.

This takes into account all the "funny" factors of core loss, skin effect, switching losses, rectifier reverse recovery, and probably the phases of the moon.

Sometimes it makes little or no difference, but on odd occasions it can lead you to a definite unsuspected problem. At the very worst you might be able to squeeze a bit more efficiency out of it by optimising the operating frequency.

 

[ScottyUK]

Hi cbarn,

I think we are saying similar things: the change in inductance when the core enters saturation does matter in most applications, which is why we normally try to avoid it by introducing an air gap. The core still saturates at whatever flux density - 350mT or so - but as you rightly say it takes more turns i.e. more MMF to reach saturation when an air gap is present. So yes, the air gap is crucial to the performance of the core with a DC current in the winding, which is why I noted the airgap specified by Philips for this particular is 950um.

The RM14 core is available with quite a range of air gaps and corresponding range of inductance factors Al from 250nH with a 950um airgap to to 7100nH with a zero airgap. Al is a useful parameter for initial design assuming that the core is not saturated:

Inductance L = N^2.Al where N = number of turns.

The other parameters I quoted make it a whole lot easier to calculate the performance of the core rather than trying to manually integrate them along the flux path, although this can certainly be done. There is a really good treatise of this in the book 'Soft Ferrites' by Edgar Snelling, now sadly long out out print. It is a really useful reference for anyone designing ferrite cored wound components.




----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[cbarn24050]

Hi Scotty, Snellings book is very impressive on the book shelf, given how large it is, the only snag is that it is now so old that most of the materials in it are long out of production. The 3C8 handbook from Phillips, probably also out of print, has a completely new design format for SMPS output chokes as well as all the usuall stuff on transformers and topology.

 

[ScottyUK]

Hi cbarn -

The '3C8 handbook' - is that the same as the old MA01 book from 1991, or something later? Snelling's book is still a worthwhile read - quite a lot of the materials are still available, although newer materials like 3C85 aren't listed. The book's real value is when you are designing big transformers using multiple U or I-section cores. The maths hasn't changed much either - it is just as hard now as it was then.



----------------------------------

If we learn from our mistakes,
I'm getting a great education!