Forums

Combined resonant inductor and transformer in LLC halfbridge

Started by powerampfreak June 28, 2014
I am designing a halfbridge LLC converter.
Due to different reasons I want to use transformer leakage as resonant indu=
ctor.
As production are concerned, what are the best way to achieve the tightest =
results?
I have made some experiments, with a 2-sectioned bobbin, but as soon as the=
 transformer are wound, it's no way of "trimming" or adjusting the leakage.
I then made a new transformer on one section, where I simply adjusted the l=
eakage to my specification by winding the primary on the left side of the b=
obbin, and the secondary on the right side. Then I moved a few turns of the=
 secondary on top of the primary, until the leakage was spot on.
Will this add any drawbacks to the design?

Best regards
PAF
On Sat, 28 Jun 2014 09:11:41 -0700, powerampfreak wrote:

> I am designing a halfbridge LLC converter. > Due to different reasons I want to use transformer leakage as resonant > inductor. > As production are concerned, what are the best way to achieve the > tightest results? > I have made some experiments, with a 2-sectioned bobbin, but as soon as > the transformer are wound, it's no way of "trimming" or adjusting the > leakage. > I then made a new transformer on one section, where I simply adjusted > the leakage to my specification by winding the primary on the left side > of the bobbin, and the secondary on the right side. Then I moved a few > turns of the secondary on top of the primary, until the leakage was spot > on. > Will this add any drawbacks to the design? > > Best regards PAF
My gut feel on this is that unless you have very high production quantities, the pain of getting the "one, cheap" transformer dialed in is going to far out weigh (and out-cost) the pain of just using a good, tightly coupled transformer, and an external, swappable, coil for the "leakage". If you must have just one transformer, then your two-sectioned bobbin, with the turns swapped around as necessary to achieve what you want, is probably best. I would beware of changes due to winding style (this is a good part of the "pain" I refer to in the above paragraph). You probably want to get a test article from the transformer vendor (or from your production shop if your company is winding it), then test. Then KEEP TESTING as you ramp up production. Then impress DEEPLY upon your transformer winders that any changes may mess you up. Then periodically check. And keep a big garbage can handy for the reject boards when they don't listen. -- Tim Wescott Wescott Design Services http://www.wescottdesign.com
Thanks Tim, for your input.
The design does not have space for a separate inductor, the transformers will be wound "in-house" and production volumes quite low.
I believe it will be possible to have close follow up of each wound transformer, where each one will be put to a separate inductance test, in prior mounting line.
This will minimize the container of rejected boards!  :)