diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Power Supplies (http://www.diyaudio.com/forums/power-supplies/)
-   -   Gapping an SMPS transformer core. (http://www.diyaudio.com/forums/power-supplies/220252-gapping-smps-transformer-core.html)

treez 23rd September 2012 07:47 PM

Gapping an SMPS transformer core.
 
I need to put a 1.2mm gap into an SMPS transformer ferrite...(otherwise it would saturate), but this gap size is so big that 3D magnetic theory is needed to calculate the new Reluctance value......(is this true?)

Therefore, how do i calculate the new total core reluctance value with the 1.2mm gap in it.

The core is PQ5050-3C96 and 33 turns and a gap of 1.2mm "should" give it a inductance of 341uH and I(sat) of 9A7.....but will it?

Its and inductor of a boost PFC stage.

PQ5050 datasheet
http://www.ferroxcube.com/prod/assets/pq5050.pdf

Elvee 23rd September 2012 08:07 PM

Quote:

Originally Posted by treez (Post 3176124)
I need to put a 1.2mm gap into an SMPS transformer ferrite...(otherwise it would saturate), but this gap size is so big that 3D magnetic theory is needed to calculate the new Reluctance value

In principle, 3D theory is always needed when you add a significant air gap.

But I am not sure theory is your best option: magnetic reality is difficult to model and theorize, and when you have an approximate value, the best solution is to try calibrated spacers on real pair of cores.
You can start with approximations, like this one: Designing high-current chokes is easy | EDN and refine them with actual tests.

One thing has to be remembered: magnetic components tend to be forgiving, provided you give them enough turns.
Suppose you implement a too small gap: the inductor will saturate at about the same instant (inductance higher, but Isat smaller), but more sharply.
Since the magnetic path doesn't have an homogeneous section, some parts will saturate earlier, creating "emergency air gaps" as needed by the induction created.
This will of course slightly increase the magnetic losses, but in limited parts of the circuit, and the overall effect will be small.

It is better not to abuse of this self compensating mechanism, but a +/-30% tolerance on the inductance value will be transparently absorbed, and anyway Al tolerances are not tight either.

treez 23rd September 2012 08:26 PM

Thanks Elvee,

ill bear that in mind..but also, does anyone know of gaspped ferrite cores bigger than the ETD54/28/19 core by Epcos?

http://www.epcos.com/inf/80/db/fer_07/etd_54_28_19.pdf

...this seems to be the biggest off-the-shelf gapped core in the world.

Surely there is bigger than this?

The advantage of these sold-gapped cores os that they come with an AL value and you can work out your saturation current more exactly than having to use the 3D magnetic theory

powerbob 24th September 2012 02:16 AM

Ferroxcube has the option of ordering gapped cores also. Their full line catalog shows E cores up to E80/38/20 can be ordered with gaps. They do not offer ETD cores pre gapped.

Ferroxcube - ferrite cores, bobbins & accessories has no direct link to catalogs.


All times are GMT. The time now is 07:22 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2