Its not about skin depth. It is about eddy currents induced in the cross-section of the wire, particularly bad where the wire is right on the core because of the high flux gradient. There was a talk given at an Society For Information Display annual Symposium in San Francisco by a fellow from Tektronix in the early 1980's (Tek made CRT displays at the time) that explained the problem.

Shortly that after I made some CRT deflection (flyback) circuits running at 64 kHz and about 12 amps P-P through the primary. The cores became painfully hot. I tried switching to a multistrand primary. It cooled things down to where I could actually touch the core without feeling pain. (That paid for the seminar instantly.) I have since used this technique on other designs. Transformer suppliers are aware of this and won't look at you like you have a third eye if you ask them to do it this way. A reference to Litz wire might help.

 

Is that different from the eddy currents that cause the proximity effect?

Its not about skin depth. It is about eddy currents induced in the cross-section of the wire, particularly bad where the wire is right on the core because of the high flux gradient. There was a talk given at an Society For Information Display annual Symposium in San Francisco by a fellow from Tektronix in the early 1980's (Tek made CRT displays at the time) that explained the problem.

Shortly that after I made some CRT deflection (flyback) circuits running at 64 kHz and about 12 amps P-P through the primary. The cores became painfully hot. I tried switching to a multistrand primary. It cooled things down to where I could actually touch the core without feeling pain. (That paid for the seminar instantly.) I have since used this technique on other designs. Transformer suppliers are aware of this and won't look at you like you have a third eye if you ask them to do it this way. A reference to Litz wire might help.

Is this the same eddy currents that are responsible for the proximity effect?

 

I don't know. What is the proximity effect?
From the name of the effect I suspect it is the same thing but that is only a guess.

 

I don't know. What is the proximity effect?
From the name of the effect I suspect it is the same thing but that is only a guess.

Wikipedia has quite a nice explanation
https://en.wikipedia.org/wiki/Proximity_effect_(electromagnetism)
It's a very good reason to use single-layer windings.
Does the effect to which you are referring apply to single-layer windings?

 

Perhaps the same or similar.

 

Perhaps the same or similar.

It was a bit of a revelation when I found out about it.
My solution was to design single-layer windings, after I saw the price of Litz wire!

 

0.4 mm is much too large to be next to the core at 65 kHz. Many strands a finer magnet wire would create lower eddy current losses across the wire. You can make multistrand wire from regular enameled magnet wire with a twist per cm or so to help keep the wire tidy. A good starting point would be to use many strands of #30 or #32 to get the copper area you came up with earlier in the design.

thanks i'm working on a new transformer in which the strands are more and thiner

 

Skin depth at 65kHz is 0.255mm, so 0.4mm wire seems OK to me.
I think you need more inductance (a smaller gap), because at the moment your transformer is only working Two-thirds of the time (all the time it isn’t ringing), meaning that current during that time is 50% greater than it need be. Or, alternatively, you could run the same transformer at 100kHz, but you’d need to change the IC.
The current waveform is messy. You need the earth probe attached directly to one end of the sense resistor to get an accurate waveform, not connected to some random circuit ground, and a shorter earth lead.
I’m a big fan of quasi resonant designs at that power level as it eliminates that dead time when the transformer isn’t doing anything. L6565 for example.

i think there is something wrong with my current waveform. i have chosen the nearest points to the sense resistor for monitoing. but still there is a lot of noise measured. i don't know if that would be a reason for transformer heat or not. if i choose a larger inductor i may loose DCM operation since the 350uH is the critical conduction boundary value for this power and frequency.
i have bought UCC28742 which is Resonant valley switching controller. would it be a better idea to use that?

 

There‘s something making a mess of the current waveform - a lot of ringing when the MOSFET switches on. What value is your gate resistor?
How good is the coupling between windings?

I‘d approach the transformer by choosing a wire diameter (whether it be single strand or Litz) so that the winding takes up EXACTLY a single layer, same for the secondary - by winding several strands side by side.
I’d be really interested to know how much difference Litz wire makes for single layer windings.

Always worth trying a quasi-resonant! (it improves efficiency at high power by eliminating dead time, and also prevents it drifting into continuous current mode)

 

the gate drive resistor is 27ohm at turn on and 2 at turn off. should i make it more to slow down the turn on?
i have used 3 strands od 0.25mm wire for primary and in order to reduce leakage inductance i have tried to have all windings wound in single layer. (single layer for each half of primary and single layer for secondary)

 

There‘s something making a mess of the current waveform - a lot of ringing when the MOSFET switches on. What value is your gate resistor?
How good is the coupling between windings?

I‘d approach the transformer by choosing a wire diameter (whether it be single strand or Litz) so that the winding takes up EXACTLY a single layer, same for the secondary - by winding several strands side by side.
I’d be really interested to know how much difference Litz wire makes for single layer windings.

Always worth trying a quasi-resonant! (it improves efficiency at high power by eliminating dead time, and also prevents it drifting into continuous current mode)

i want to thank you for your time and effort. i really appreciate it

 

You're welcome - your transformer design looks a lot like mine (except for the rubber band holding the two halves of the core together) Occasionally, I get one that doesn't work like it should, and sometimes I never work out why.
I have two huge spreadsheets, one for push-pull converters and one for flybacks. The push-pull spreadsheet gets the temperature rise spot on, the flyback one generally underestimates.
Maybe, I should take Dick's advice and spend my hard-earned money on some Litz wire.

 

Please draw how the transformer is made.
Example:
------------------------two layers of tape
ct--oooooooooo-----B+ 350V
----------------------3 layers of tape
oV--ooooooooooo*---|>|--- secondary + 24V
-----------------------3 layers of tape
ct--oooooooooo*----- to MOSFET
core
* winding phase
ct is the center tap of the primary.

example: two layers for secondary.
-------------------------3 tape
ct000000000000000000*----|>|----24v cd secondary
-------------------------1 tape
ct0000000000000000000--------0v de secondary
--------------------------3 tape

Sorry my paint is not working well today.

 

I did not buy Litz wire in those days. The place in which I worked had metal poles supporting the roof (a factory in a warehouse). I tied the end of the magnet wire to one pole, then walked to and around the other pole, then back to the first one and continued until I had the desired number of parallel strands then cut one end lose and chucked it up in a drill and put some twists on it, stretching the cable a little after the twisting so it would not untwist.

Doing this meant that I had plenty of flexibility in selecting the number of strands and the wire diameter. And it is a heck of a lot cheaper than genuine Litz.

When doing this when consulting from home I would place my son out in the street in front of the house holding a screwdriver to loop the wire over...

 

I always thought Litz was something rather cleverer than simply a few strands of thin wire twisted together. Or have I just been believing too much marketing hype?

 

No, you've got that right, but some versions of Litz wire use a "serving" which is an additional layered insulating coating.

 

You're welcome - your transformer design looks a lot like mine (except for the rubber band holding the two halves of the core together) Occasionally, I get one that doesn't work like it should, and sometimes I never work out why.
I have two huge spreadsheets, one for push-pull converters and one for flybacks. The push-pull spreadsheet gets the temperature rise spot on, the flyback one generally underestimates.
Maybe, I should take Dick's advice and spend my hard-earned money on some Litz wire.

i would be so thankfull if you share the spreadsheets you have. about litz wire there is design note by TI that does not recommend use of Litz wire for frequencies bellow 150KHz. i have attched the file for you. read the wire type section

 

Please draw how the transformer is made.
Example:
------------------------two layers of tape
ct--oooooooooo-----B+ 350V
----------------------3 layers of tape
oV--ooooooooooo*---|>|--- secondary + 24V
-----------------------3 layers of tape
ct--oooooooooo*----- to MOSFET
core
* winding phase
ct is the center tap of the primary.

example: two layers for secondary.
-------------------------3 tape
ct000000000000000000*----|>|----24v cd secondary
-------------------------1 tape
ct0000000000000000000--------0v de secondary
--------------------------3 tape

Sorry my paint is not working well today.

Hi
------------------->Two Layers of Tape
Auxiliary Winding (9 turns)
------------------->Two Layers of Tape
CT----------------> 350V (25Turns made of multiple strands as second half of primary )
------------------->Single layer of grounded Copper Shield
------------------->Two Layers of Tape
OV---------------> Secondary Ground (24V - 14turns of multiple strands)
------------------->Two Layers of Tape
------------------->Single layer of grounded Copper Shield
------------------->Two Layers of Tape
Drain of MOSFET----------------> CT (25Turns made of multiple strands as first half of primary)
Core

 

Here's the spreadsheets. They all assume a sandwich of two single layer sections of primary in series and a single layer secondary.
There is no checking for nonsense values, and no calculation of parasitics.
However, having quickly glanced at the TI design note, I might have to make some adjustments to the core loss calculations. I had always thought that the equations that linked core-loss to peak-to-peak flux excursion were a bit dubious, and now there a better explanation.
As an aside, has anyone ever tried to make a flyback converter on a ferrite toroid by cutting a gap in it?
I've tried LLC resonant designs on a gapped toroid, but with very little success.

 

Here's the spreadsheets. They all assume a sandwich of two single layer sections of primary in series and a single layer secondary.
There is no checking for nonsense values, and no calculation of parasitics.
However, having quickly glanced at the TI design note, I might have to make some adjustments to the core loss calculations. I had always thought that the equations that linked core-loss to peak-to-peak flux excursion were a bit dubious, and now there a better explanation.
As an aside, has anyone ever tried to make a flyback converter on a ferrite toroid by cutting a gap in it?
I've tried LLC resonant designs on a gapped toroid, but with very little success.

Thanks a lot