Can copper foil be used on the primary of high-voltage fly-back transformers?

Thread Starter

SiCEngineer

Joined May 22, 2019
442
I have often seen copper foil used in high power transformers for the primary and secondary windings, especially at higher frequencies where the AC effects and losses are dominant. In general, how plausible is the use of such copper foils in higher voltage designs such as a fly-back step-up transformer? The transformer will operate at least 500kHz, with a 250VDC input and 4 x 750VDC output windings that are connected in series and to voltage doublers.

Is there an argument that could be made that the foil is suitable for the primary but not the secondary, or could these foils be used for both, or none at all? Why? I am unsure on how the capacitance of the transformer is effected by such winding geometries -- still searching for some more data on this. I was thinking of using foil on the primary where the higher current is and I cannot source Litz wire, and then using standard low gauge wire that is triple insulated on the secondaries since it is low current to achieve the isolation between the primary and secondary windings. Teledyne for example has 13kV triple insulated wire but this is also proving difficult to source in lower quantities for a prototype.
 

Thread Starter

SiCEngineer

Joined May 22, 2019
442
Have you calculated how many turns you need? That might be the limiting factor.
I designed one with Litz originally that had 18 primary turns, and 4 x 44 secondary turns. I can always redesign slightly, but I would like to minimize the number of layers so keep turn count as low as possible.

So that gives around 14V per turn of the copper foil, not too bad?
 

Thread Starter

SiCEngineer

Joined May 22, 2019
442
Maybe - I don’t know what core you are using.
Let's assume that a suitable core has been chosen to ensure the core doesn't saturate at that turns count and with the inductance the transformer has, etc. My question is just about the suitability of copper foil as a replacement for Litz wire in high frequency converters when the input voltage and output voltage are not very low - I've only really seen them used in low voltage applications, which gives me the inkling that is where they are most suited and I am heading in the wrong direction.

SIC
 

Ian0

Joined Aug 7, 2020
9,671
Let's assume that a suitable core has been chosen to ensure the core doesn't saturate at that turns count and with the inductance the transformer has, etc. My question is just about the suitability of copper foil as a replacement for Litz wire in high frequency converters when the input voltage and output voltage are not very low - I've only really seen them used in low voltage applications, which gives me the inkling that is where they are most suited and I am heading in the wrong direction.

SIC
Proximity effect may be the biggest problem.
 

Thread Starter

SiCEngineer

Joined May 22, 2019
442
Proximity effect may be the biggest problem.
That's true - I am not concerned about Skin effect but proximity is certainly the biggest concern from me also. That's why I wanted to use Litz in the first place since the currents are usually much better distributed across the wire. But I can't get hold of any Litz that is useable at this frequency, and single strand wire just isn't going to cut it so I think I either drop the frequency quite a bit for the prototype, or I instead use copper foil - again assuming it is OK with this turn count and voltage on the primary.
 

ronsimpson

Joined Oct 7, 2019
2,989
When I made a transformer using a strap of copper, the resonant frequency of the primary was very low. There is a very large capacitor between layers in the winging. I think it resonated at 30khz.

Running at 500khz you need to minimize inter winding capacitance.
 

Thread Starter

SiCEngineer

Joined May 22, 2019
442
When I made a transformer using a strap of copper, the resonant frequency of the primary was very low. There is a very large capacitor between layers in the winging. I think it resonated at 30khz.

Running at 500khz you need to minimize inter winding capacitance.

Yes exactly. I am trying to do that. I have been trying to see how the use of foil effects this parasitic parameter of the transformer. I am using a clamped resonant fly-back so the increased leakage inductance that comes with a decrease in capacitance is not an issue so much...
 

ronsimpson

Joined Oct 7, 2019
2,989
Show us the structure of the transformer. There are some little known things that might help. Schematic of what is around the transformer will also help.
 

Thread Starter

SiCEngineer

Joined May 22, 2019
442
I think Litz wire or multiple strands of fine insulated magnet wire would be easier to get working the way you want than using copper foil.
agreed, but can’t seem to source any Litz wire that would do the trick in the U.K.

Triple insulated wire tends to be quite large diameter over the insulation, but I am looking. Again the trick is finding something that is fine enough to not have severe losses at AC. Teledyne MicroFlex is very large MOQ or out of stock but still looking for other manufacturers of HV Wire. I am considering going a larger core than is required but even with a 50% copper fill it seems difficult to get enough wire in there without winding loads of additional layers.
 
Top