STMicroelectronics A 200 W ripple-free input current PFC pre-regulator with the L6563S EVL6563S-200ZRC EVL6563S-200ZRC Fiche De Données

   

   

   

16/39

 Doc ID 17273 Rev 1

where, in this case, L is the measured inductance of the AC winding. Add 5 % to the 
result, to account for the slight coupling improvement that there is as the DC winding is 
entirely in place, and round up to the next integer. Let this number be N

2

         

7. 

After removing the layer used for the preliminary measurement and the core, wind the 
N

2

 turns of the DC winding. Do not permanently fix the end of the wire because N

2

 

should be greater than the value that meets the zero-ripple condition and some turns 
need to be taken out in the following step

8. 

Reassemble the core and adjust the air gap so as to get the required value L1. 
Measure the leakage and the magnetizing inductances (see 

) and check if 

(7) is met. If not, remove one turn and repeat the step until the condition in 

 

is close to being met

9. 

Connect the coupled inductor to the converter, power it on and measure the ripple on 
the DC winding. It should be quite small. If it looks like that in 

 the turn number 

of the DC winding has been reduced too much. If it is not possible to add more turns to 
the DC winding, either the air gap must be reduced, if possible (paying attention to 
saturation!), or the turns number of the AC winding must be reduced. If the ripple looks 
like that in 

, where the ripple is 180° out-of-phase, there are still too many 

turns on the DC winding and they should be further reduced. Also in this case it is 
possible to fine tune by adjusting the gap, if possible. Only now can the DC winding 
wire end be permanently fixed to the bobbin pin. Record the final values of N

1

, N

2

 and 

l

gap

.

Note that the first four steps of this procedure are the same as those for a conventional 
inductor. Starting from

 

step 5, the procedure becomes empirical and quite tricky and might 

also be time-consuming. Fortunately, this work needs to be carried out only for the first 
prototype.

There is still an important practical issue to consider. When the two half cores are 
assembled, even if they are kept together well, they are typically quite loose inside the 

LK

1

2

L

L

L

N

N

−

=

         

         

!-V

8QGHUFRPSHQVDWHGNQ



GFZLQGLQJZLWKWRRIHZWXUQV

GFZ LQGLQJFXUUHQW

DFZ LQGLQJFXUUHQW

JDWHGULYH

!-V

2YHUFRPSHQVDWHGNQ

!

GFZLQGLQJZLWKWRRPDQ\WXUQV

GFZ LQGLQJFXUUHQW

DFZ LQGLQJFXUUHQW

JDWHGULYH