High Voltage Forum

High voltage => Transformer (Ferrite Core) => Topic started by: TDAF on April 16, 2018, 02:08:46 PM

Title: Winding E cored transformers
Post by: TDAF on April 16, 2018, 02:08:46 PM
My question is what's the optimal method to wind transformers on E cores with minimum leakage inductance??
I'm talking about the double-e cores.
Title: Re: Winding E cored transformers
Post by: profdc9 on April 16, 2018, 05:02:55 PM
Winding is a trade-off between leakage inductance and capacitance between the turns of a winding and between the primary/secondary.  Which effect dominates depends on the frequency, with low frequencies being leakage inductance, and higher frequencies being winding capacitance (especially > 1 MHz and RF).  Unfortunately, modeling leakage inductance is hard, the only way I really way know to model it is to use finite element techniques.

1.  Bifilar/multifilar turns have the least leakage inductance, but have the most winding capacitance.  This is often controlled in RF transformers by making the turns part of a transmission line, as to make Guanella/Ruthroff transmission-line transformers for ununs/baluns.  These can be difficult to wind, as these types of transformers typically use toroids and are not very amenable to using bobbins.  With high voltages, strong insulation must be used on the wires, for example, Teflon-coated wire.

2.  Layered turns.  For example, you can alternate between layers of the primary and secondary.  This also has low leakage inductance but high capacitance between the primary and secondary.  Less difficult to wind than RF transformers, but it can be challenging.  With high voltages, strong insulators such as phenolic paper or oil-impregnated paper are placed between the layers.

3.  Concentric primary and secondary.  This is probably the easiest to do of the methods that achieve a lower leakage inductance.  The primary and secondary are wound on separate bobbins and then one is placed inside the other on the central leg of the E.  For the self-capacitance of each winding, one can either traverse both directions when winding, or always return the wire back to one side before winding the next layer.  Returning the wire produces less self-capacitance of the winding, but the return wire must be well insulated from all the turns it crosses, adding bulk to the layered turns, so it is harder to keep the diameter of the winding small enough to fit inside another winding or inside the core window.  Traversing both directions adds capacitance, which may be ok if the frequency is low enough, or may make the resonance frequency of the winding too low.

4.  Primary and secondary side-by-side on center leg.  This has somewhat more leakage inductance and required strong insulation between the primary and secondary.  This is typically used on microwave oven transformers, however, these actually deliberately include magnetic shunts to increase leakage inductance because this limits current. 

5. Primary and secondary on separate legs.  For example, primary on center leg of E, secondary or secondaries on outer legs.  This is easiest but has the highest leakage inductance.

Keep in mind too that MnZn ferrites have a low bulk resistivity, but are used because of their high permeability at the typical frequencies used by switching converters, so you may need to insulate the windings well from the ferrite itself or you get arcing through the core (this happened to me frequently until I started building my own bobbins from high-density polyethylene, but mine were probably 30 kV+).    This also applies for ferrosilicon laminated transformers.  For many designs, like flybacks or MOTs or LLC converters, a certain amount of leakage inductance is a feature rather than a bug, because it can be used for current limiting or power regulation as long as the stored energy in the inductance does not harm the switches.

Dan
Title: Re: Winding E cored transformers
Post by: TDAF on April 16, 2018, 06:40:45 PM
That was of great help
Thank you
I'll be using some 80*80mm transformer cores with window of 20*20
I'm making these transformers for the qcw inverter
Pardon the stupid question but how is one supposed to clamp the 2 e cores together??
Title: Re: Winding E cored transformers
Post by: AndreiRS on April 16, 2018, 07:26:54 PM
I use polyester tape. Any tape that won't melt at higher temperatures is ok. Even that black insulation tape may be good. Then after that, you should dip it into insulation varnish. The polyester tape is good because it stays there when you use the varnish. Insulation tape sometimes falls off.
SimplePortal 2.3.6 © 2008-2014, SimplePortal