+ Post New Thread
Results 1 to 7 of 7
  1. #1
    Advanced Member level 5
    Points: 20,661, Level: 34
    Achievements:
    7 years registered

    Join Date
    Jan 2011
    Posts
    3,381
    Helped
    1207 / 1207
    Points
    20,661
    Level
    34

    Integrated planar transformer

    Anyone here done an integrated planar transformer before? By integrated, I mean that all the windings are on your main PCB, instead of having a separate PCB for windings. My PCB is already going to have 8-10 layers, so this seems like an effective way to minimize cost and size. Also the core materials I'm interested in using tend to be in planar shapes, as opposed to RM sizes.
    Click image for larger version. 

Name:	planartypes.png 
Views:	6 
Size:	31.4 KB 
ID:	155388

    The issue is mass assembly. I can find practically no resources on how this is done for integrated transformers. The type of cores I want to use (EQ25) don't even offer spring clamps/clips from the manufacturers. Most suggestions I find are either "just glue the things together," or suggest clamping it with custom mechanics. Both sound labor intensive (and introduce possibility for poor yield), and the latter introduces additional costs. I'm concerned about vibration in my application as well, and can't find any real discussion of this.

    Anyone have any experience to share on this?

    •   AltAdvertisement

        
       

  2. #2
    Advanced Member level 5
    Points: 32,919, Level: 44
    Achievements:
    7 years registered

    Join Date
    Sep 2008
    Location
    cambridge
    Posts
    6,848
    Helped
    496 / 496
    Points
    32,919
    Level
    44

    Re: Integrated planar transformer

    I did one like that but it was very low power......it was because my PCB was connected to a 50khz power bus and had to slowly charge a battery from it...the mech eng guys fitted up the core halves for me.......they used a special sponge with the correct youngs modulus to pressure the core halves together without kind of slamming them....the core was kind of glued into the enclosure on the sponge, and the mech guys made it kind of like a lid, so that i closed the core halves together, and the sponge made the core halves pressured together without damaging the ferrite.......i shoved the windings in the pcb, and sufficient slotting to enable the core halve protrusions to go through the pcb, and that was it...sorry if not too helpful.

    I think for high power it would have needed the ferrite surfaces to be fettered, or whatever you call it when ferrite surfaces need to be polished extremely smooth.


    The guy that got the special spange (i couldnt find it anywhere on the web) was a guy from a consultancy in Hoddesdon. Ill ask him if you want.
    Otherwise, write to these people...
    http://www.himag.co.uk/contact_us
    ...and pretend you will pay them for the work, and ask for details of thow they woudl do it.
    Last edited by treez; 8th September 2019 at 20:00.


    1 members found this post helpful.

  3. #3
    Advanced Member level 5
    Points: 30,406, Level: 42
    BigBoss's Avatar
    Join Date
    Nov 2001
    Location
    Turkey
    Posts
    4,458
    Helped
    1339 / 1339
    Points
    30,406
    Level
    42

    Re: Integrated planar transformer

    I believe that the frequency is too low to couple two layers that are layed down on two different layers.So efficiency of a such transformer will be so weak..



    •   AltAdvertisement

        
       

  4. #4
    Advanced Member level 5
    Points: 20,661, Level: 34
    Achievements:
    7 years registered

    Join Date
    Jan 2011
    Posts
    3,381
    Helped
    1207 / 1207
    Points
    20,661
    Level
    34

    Re: Integrated planar transformer

    Quote Originally Posted by treez View Post
    I did one like that but it was very low power......it was because my PCB was connected to a 50khz power bus and had to slowly charge a battery from it...the mech eng guys fitted up the core halves for me.......they used a special sponge with the correct youngs modulus to pressure the core halves together without kind of slamming them....the core was kind of glued into the enclosure on the sponge, and the mech guys made it kind of like a lid, so that i closed the core halves together, and the sponge made the core halves pressured together without damaging the ferrite.......i shoved the windings in the pcb, and sufficient slotting to enable the core halve protrusions to go through the pcb, and that was it...sorry if not too helpful.
    Yeah I know if I made my own clamp, I'd have to put some sort of cushion, maybe like a soft thermal pad. But I was hoping to minimize the number of manually assembled custom parts. That's why simply gluing it together sounds attractive. But I can't find any references to work off of.

    The guy that got the special spange (i couldnt find it anywhere on the web) was a guy from a consultancy in Hoddesdon. Ill ask him if you want.
    Otherwise, write to these people...
    http://www.himag.co.uk/contact_us
    ...and pretend you will pay them for the work, and ask for details of thow they woudl do it.
    Thanks, I'll keep this in mind.

    - - - Updated - - -

    Quote Originally Posted by BigBoss View Post
    I believe that the frequency is too low to couple two layers that are layed down on two different layers.So efficiency of a such transformer will be so weak..
    Operating frequency is >1MHz, and I've already tested handmade prototypes which work great.



  5. #5
    Super Moderator
    Points: 260,190, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    45,437
    Helped
    13824 / 13824
    Points
    260,190
    Level
    100

    Re: Integrated planar transformer

    Expoxy resin is fine, you need a temporary clamp while curing.

    Some assembly hints in this old Philips (now Ferroxcube) application note. There may be more detailed recent application notes.


    1 members found this post helpful.

  6. #6
    Advanced Member level 5
    Points: 20,661, Level: 34
    Achievements:
    7 years registered

    Join Date
    Jan 2011
    Posts
    3,381
    Helped
    1207 / 1207
    Points
    20,661
    Level
    34

    Re: Integrated planar transformer

    I did come across that same app note, lots of good rules of thumb, but nothing really specific :/

    I would definitely need some sort of jig to apply pressure between the two halves while setting/curing. Could you recommend an epoxy that sets in <1 hour?

    The major risk with gluing is affecting the core reluctance. I don't intend to have any gap, so even a thin layer on the pole faces might be an issue. But I'm not sure putting blobs of glue on the outside of the cores will be secure enough...



    •   AltAdvertisement

        
       

  7. #7
    Super Moderator
    Points: 260,190, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    45,437
    Helped
    13824 / 13824
    Points
    260,190
    Level
    100

    Re: Integrated planar transformer

    Although the app note suggests not to apply epoxy resin to the core gap, I usually do to achieve highest strength. As far as I'm aware of, most transformer manufacturers also go this way. I prefer a slow curing standard epoxy (like Araldite), curing at elevated temperatures. Warmed epoxy becomes highly fluid resulting in a minimal glue film.

    For smallest parasitic air gap, a dry surface and clamps may be better anyway. With EQ25, this means custom made clamps.


    1 members found this post helpful.

--[[ ]]--