cupoftea
Advanced Member level 5
Hi,
Would you agree, that for an offline , 120kHz, two transistor forward for 100-265VAC, 130W,
There is basically only one realistic way to wind the gate drive transformer.
Bifilar pri/sec is out of the question as too much interwinding capacitance.
So you must Sandwich wind.
That means using 0.25mm enamelled copper wire and an EFD15 core.
The two series halves of the primary that do the sandwich will be 25 turns each, 50 turns total,
giving 1.95mH with 3F3 core of AL=780.
So then, its 1:1, and so sec will be also 50 turns...so that's 2 layers wound inbetween the primary half coils.
There is no other realistic way of doing it...I mean, you can use a smaller core, (if your fingers are nimble enough)
but the principle would be the same.
Would you agree? There is no other sensible way. Doing it on a torroid is just too much manufacturing awkwardness.
Would you agree, that for an offline , 120kHz, two transistor forward for 100-265VAC, 130W,
There is basically only one realistic way to wind the gate drive transformer.
Bifilar pri/sec is out of the question as too much interwinding capacitance.
So you must Sandwich wind.
That means using 0.25mm enamelled copper wire and an EFD15 core.
The two series halves of the primary that do the sandwich will be 25 turns each, 50 turns total,
giving 1.95mH with 3F3 core of AL=780.
So then, its 1:1, and so sec will be also 50 turns...so that's 2 layers wound inbetween the primary half coils.
There is no other realistic way of doing it...I mean, you can use a smaller core, (if your fingers are nimble enough)
but the principle would be the same.
Would you agree? There is no other sensible way. Doing it on a torroid is just too much manufacturing awkwardness.