• Welcome to engineeringclicks.com
  • Flex PCB design

    Discussion in 'The main mechanical design forum' started by adih, Mar 20, 2013.

    1. Bob Burns

      Bob Burns Member

      Joined:
      Mar 2013
      Posts:
      5
      Likes Received:
      0
      Hi Adih,

      Usually the PWB fabricator will calculate the different lengths for the flex arms, to create the "hump". The designer sends in the gerber and print, with the gerber files all the same length, but indicates on the print that the part is to be manufactured as a bookbinder rigid flex. Once, I did see a customer do all of the calculations themselves. They hired an intern for the summer, and that was his summer project! But I have only seen it that one time; it was quite the project and the intern really threw himelf/herself into it!
       
    2.  
    3. Mark Stapleton

      Mark Stapleton Active Member

      Joined:
      Jan 2012
      Posts:
      38
      Likes Received:
      0
      @Bob Burns, if that is so and the PWB fabricator actually creates Gerbers with different lengths for the flex portions on each layer (i.e., double layer) then how do they laminate these different lengths together? In two separate processes? If so then even the slightest misalignment of the first hard section laminated would cause a huge misalignment in layers of the 2nd hard section laminated. This is why I presume you are correct in mentioning custom tooling in the "bookbinder" rigid flex fabrication process. What Adih is apparently intending to do will seem to result in causing his employer considerable expense for minimal gain.
       
    4. Bob Burns

      Bob Burns Member

      Joined:
      Mar 2013
      Posts:
      5
      Likes Received:
      0
      Hey Mark,

      The thing about rigid flex, is that you already have to laminate coverlayer onto each flex arm anyway. We add tooling holes to each flex layer, and use pins to hold each layer in the correct position relative to all the other layers. We put a cavity in the lamination plates where the "hump" goes. Often, rather than ordering new tooling which is custom, dedicated, and has a relatively long lead time, we will take an existing set and have it re-tooled to build the part. This saves some money and lead time, and then we order replacements the next time we order plates. The expense is the hassle. Almost all PWB processes are planar, so introducing a 3D object into a planar process is a pain. We have to protect the hump from process chemistry, and remove rollers on conveyorized equipment, etc.
       
    5. Mark Stapleton

      Mark Stapleton Active Member

      Joined:
      Jan 2012
      Posts:
      38
      Likes Received:
      0
      Very interesting, Bob (really!). Confirms the bottom line in my own last two posts, but with more detail and rationale. Thanks. It seems to me that Adih - or Adih's superiors - should think twice again before going ahead with differing lengths for flex portions of the two connected rigid boards.
       
    6. adih

      adih Member

      Joined:
      Mar 2013
      Posts:
      5
      Likes Received:
      0
      Mark and Bob
      Thanks for the intersting discution, but both of you miss the point. (at least, my point). all advices that given here are nice, but obvious.
      I asked for advice regarding the location of the two rigid PCBs (the X and Y dimension in the picures I uploaded) after installing the card. I havn't got here any lead thinking about it. All conclusion we arrived here already descibes in my original post...(I especially liked Mark's emphatically about making decisions less than 24 hours after he discovered 'bookbinder' flex...)
       
    7. Mark Stapleton

      Mark Stapleton Active Member

      Joined:
      Jan 2012
      Posts:
      38
      Likes Received:
      0
      Adih, looking back at your last post it seems we did indeed miss the point you mentioned of the X and Y locations, but unfortunately that part of the question is not very clear to me. Obviously, the farther you can spread the two rigid portions of the board apart, the less stress will be on the flex portion, but I don't know if that is your primary concern. If so, then it would be better to have equal X and Y distances in order to prevent a sharper bend radius at one end or the other of the flex portion. If not then I think we need more info on your design considerations and concerns. Are you limited in how far you can spread these two rigid sections apart? If so, why? Are you simply trying to aim for a happy medium of package envelope and bend radii? As far as I understand right now there is no "right" and "wrong" answer to your query ... it all depends upon what you value and what you don't (at least don't quite so much). Your fabricator can give you a minimum bend radius for the thickness of the flex and the copper in the flex. The copper thickness depends most heavily on the highest power amperage passing through the copper at its narrowest point. There is a curve graph I can give you that will show you the temperature rise above ambient per amp per cross section of conductor, but of course there will also be the consideration of whether there is any air cooling in the package and what kind of ambient temperature in operation your module will encounter.

      I can also lay out your boards and flex in the orientation you need and figure the probable bend radii which would be necessary resulting from keeping all flex sections the same length. I'll be happy to do that for you, but first please determine the minimum bend radius for the flex thickness you need -- and to do that, as I mentioned, you need to contact your fabricator.
       
      Last edited: Apr 4, 2013

    Share This Page

    By using this website you agree to our Cookies usage. We and our partners operate globally and use cookies, including for analytics, personalisation, ads and Newsletters