Author Topic: Offline flyback transformer with nylon "wire" to pack out bobbin? (gap issue)  (Read 498 times)

0 Members and 1 Guest are viewing this topic.

Offline ocset

  • Super Contributor
  • ***
  • Posts: 1516
  • Country: 00
Hi,
The attached is a former cross section of a 140W Flyback transformer.
Vin = 228VDC
Vout = 24V
Iout = 5A
Fsw = 43kHz
(BCM)

Core = Ferroxcube E42/21/15 3C90 gapped to 0.85mm (centre leg) giving AL = 315.
It is a single primary coil, which is sandwiched inside two secondary coils, SEC1 and SEC2.
SEC1 and SEC2 are in parallel with each other.
NP/NS = 24/10
PRI = 24 turns of 7/0.24mm TEX-ELZ
SEC1 = 10 turns of 2 strands of 7/0.3mm TEX-ELZ
SEC2 = 10 turns of 2 strands of 7/0.3mm TEX-ELZ
LP = 181uH, LS = 31.5uH
I(pri) peak = 6A.
I(sec) peak = 14.4A
The core is packed out with 0.6mm diameter fishing line to a depth of 1.2mm (to keep turns away from centre leg gap)

I am wondering how awkward practically its going to be to wind and terminate the thick secondaries? As you know, TEX-ELX isnt purchaseable in small quantities so we cant even "have a go". Usually send off to winder to ask their advice but this proj doesnt have go -ahead yet.
Do you know?

TEX-ELZ
Triple Insulated Wire - Litz Type TEX-ELZ|Insulated Winding Wires|Furukawa Electric Co., Ltd.

E42/21/15
http://ferroxcube.home.pl/prod/assets/e422115.pdf
 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 1958
  • Country: gb
  • Test instrument designer/maker G1YWC.
    • Electron Plus
Hi,
The attached is a former cross section of a 140W Flyback transformer.
Vin = 228VDC
Vout = 24V
Iout = 5A
Fsw = 43kHz
(BCM)

Core = Ferroxcube E42/21/15 3C90 gapped to 0.85mm (centre leg) giving AL = 315.
It is a single primary coil, which is sandwiched inside two secondary coils, SEC1 and SEC2.
SEC1 and SEC2 are in parallel with each other.
NP/NS = 24/10
PRI = 24 turns of 7/0.24mm TEX-ELZ
SEC1 = 10 turns of 2 strands of 7/0.3mm TEX-ELZ
SEC2 = 10 turns of 2 strands of 7/0.3mm TEX-ELZ
LP = 181uH, LS = 31.5uH
I(pri) peak = 6A.
I(sec) peak = 14.4A
The core is packed out with 0.6mm diameter fishing line to a depth of 1.2mm (to keep turns away from centre leg gap)

I am wondering how awkward practically its going to be to wind and terminate the thick secondaries? As you know, TEX-ELX isnt purchaseable in small quantities so we cant even "have a go". Usually send off to winder to ask their advice but this proj doesnt have go -ahead yet.
Do you know?

TEX-ELZ
Triple Insulated Wire - Litz Type TEX-ELZ|Insulated Winding Wires|Furukawa Electric Co., Ltd.

E42/21/15
http://ferroxcube.home.pl/prod/assets/e422115.pdf
Yes it is.
https://powermagneticsstore.co.uk/shop/furukawa-triple-insulated-wire.html
*New Product - 5V TC reference* available now.
https://electron.plus Power Analysers, VI Signature Testers, Voltage References.
 

Offline ocset

  • Super Contributor
  • ***
  • Posts: 1516
  • Country: 00
Thankyou very much.
Also, any thoughts on the use of fishing line to pack put the bobbin, as in top post? (reduce eddy currents in the wire due to fringing field from the core gap)
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 17181
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Traditional is just using margin tape.  It doesn't melt in operation or when curing varnish, and is porous to soak up varnish.

I would expect fishing line per se has a lot of shrinkage when heated, or maybe melts easily or something.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: Ian.M, ocset

Online TimNJ

  • Super Contributor
  • ***
  • Posts: 1304
  • Country: us
Try 3M 44T-A if you want thick margin tape. Highly unlikely that the fishing line is allowable under your magnetics' vendors UL insulation system. Check with them for what insulation systems they have access to, and then you'll know what you can and can't use.

Also, unless I'm missing something from your layer stackup, you don't have to use TIW for both primary and secondary. Often times, use normal enamel wire on the primary and use TIW on the secondary. For a step-down converter, you'll need more turns on the primary, so the extra TIW thickness adds up more. You can definitely do it the other way around too.

You, of course, can use TIW for everything, but your bobbin copper utilization will be much worse, and the cost will be higher.
« Last Edit: May 04, 2021, 12:20:01 am by TimNJ »
 
The following users thanked this post: ocset

Offline fcb

  • Super Contributor
  • ***
  • Posts: 1958
  • Country: gb
  • Test instrument designer/maker G1YWC.
    • Electron Plus
Thankyou very much.
Also, any thoughts on the use of fishing line to pack put the bobbin, as in top post? (reduce eddy currents in the wire due to fringing field from the core gap)
I didn't think you were serious about the fishing line.  Don't use it for all the reasons others have pointed out.
*New Product - 5V TC reference* available now.
https://electron.plus Power Analysers, VI Signature Testers, Voltage References.
 
The following users thanked this post: ocset

Offline jonpaul

  • Frequent Contributor
  • **
  • Posts: 503
  • Country: fr
Use only UL/CE ETS rated insulating materials for safety and compliance.

In 50 years PSU design we never had any issues with gap flux requiring such a spacer

j
Jon Paul
 
The following users thanked this post: ocset

Offline ocset

  • Super Contributor
  • ***
  • Posts: 1516
  • Country: 00
Quote
In 50 years PSU design we never had any issues with gap flux requiring such a spacer
Thanks, yes, but was that because you always used say 38 strand of #38 AWG Litz?....in that case, as you know, no need for the spacer.
As you know, The probelm is that that kind of Litz  is difficult to source, especially with  offline insulation ratings.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf