Drilling my PCBs detaches pads

Discussion in 'General Electronics Chat' started by Envergure, Jun 28, 2014.

  1. Envergure

    Thread Starter Member

    Jan 14, 2010
    10
    0
    I'm trying to make my own PCBs using a toner transfer process. I've made several copies of a 1"x1" test pattern with some isolated pads. When I try to drill the holes, the pads often detach from the board, and I end up with a bunch of little copper rings on my drill bit! :mad:

    The drill bits I'm using are sharpened bits of piano wire, a technique I've been trying to perfect in order to avoid buying a drill press and carbide bits. The piano wire bits have worked great on a few boards I've made at school using the UV photolithography technique, but it seems to be clashing with the toner transfer method.

    The details of my process follow.
    I use 1/16" FR4, 1-oz, single-clad. The mask is printed on Pulsar Toner Transfer System paper with a Samsung ML-2165W printer and transferred with a Scotch Thermal Laminator (it doesn't really have a model number). I use the laminator's "3-mil" temperature setting and slow the motor way down by manually pulsing the jam release lever. I etch the board by rubbing it with a sponge soaked in ferric chloride. The transfer and etch are fantastic! I drill the board with a Dremel 3000 and my homemade bits at 10000~15000 RPM, using a scrap of pine as a backstop. I drill from the copper side. The pads have hollows in them where the drill hole will be, so I'm not drilling directly into the copper. The bit gets hot enough to burn me, and the board is quite warm after I've drilled a cluster of holes.
     
  2. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,283
    1,250
    They make drill bit for drilling pc boards. You might have better luck with those. The will also fit your dremel tool.
     
  3. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,685
    900
    Very interesting. I had an elderly uncle, who was a jeweler/watchmaker, who taught me how to make similar drills. That was a long time ago. Of course, they work for drilling solid easily machined metal.

    I am not surprised they do not work for copper (a surprisingly difficult metal to machine) plated onto a glass fiber reinforced matrix. You should look at the tensile strength of hard wire vs. glass fiber. (NB: Not all music wire is the same, particularly after you grind a point on it.)

    HSS bits will work for a short time, but I suggest you consider re-sharpened carbide drills. Re-sharpened bits are considerably cheaper than new bits. I bought a 1000 on an auctions site for less than $0.02 each. You may not be that lucky, but the prices are still cheap relative to new.

    John

    PS: The heat may be what is causing the pads to release from the FRP matrix. The glass transition temperature for FRP (Tg) can be quite low relative to what you consider hot.
     
    Last edited: Jun 28, 2014
  4. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    The heat for sure, and also the sharpness. If the "drill wire" gets blunt then it puts huge rotational torque on the thing it is jammed against while spinning.

    I use standard 0.8mm and 1mm PCB drill bits, and they last for many hundreds of pads even in fibreglass PCBs and only cost a few bucks. I don't remember ever ripping up a pad, but then I try to make pads a bit oversize too whenever possible.

    Or like Jpanhalt said buy a set of resharpened PCB drill bits from ebay for pennies.
     
  5. Alberto

    Active Member

    Nov 7, 2008
    169
    36
    ..... Or you can try to drill the board as first process then the toner transfer and etching, but better drill bits is the best option.
     
    Last edited: Jun 29, 2014
  6. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,015
    3,789
    I think better tools exist for drilling holes for wires than other wires. Your method is akin to using a fence post to make a hole for a fence post - and then asking why the process doesn't work well. Drill bits are made for drilling. Wires are made to nicely fit into those drilled holes.

    Or, if you are already familiar with the photo-etch process, you can make some very nice surface-mount boards and forget about all of those nasty holes.
     
  7. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    hi,
    I use the same drill types and pad technique as Roman, never had a lifted pad.

    I also ensure the pad hole diameter is slightly larger than the drill diameter.

    E
     
  8. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    If you drill before etching, the etchant eats the centre of the holes slightly, enlarging (and scruffing up) the copper hole making it hard to solder.
     
    shortbus likes this.
  9. Metalmann

    Active Member

    Dec 8, 2012
    700
    223
    Drilling tiny holes, in any material; you need a more dense backing material.

    Try 1/8" or 1/4" hardboard as a backer material.

    A lot, depends on the tip geometry of your home made bits.




    "I drill the board with a Dremel 3000 and my homemade bits at 10000~15000 RPM, using a scrap of pine as a backstop."
     
  10. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    3,783
    943
    If you have some drill bits that are smaller than your final hole size, try this.

    Use a hand held pin vise and the smaller bit and spot drill the copper surface enough to break through the copper and form a small divet in the fiberglass. This pre-drilling will decrease the likelyhood of twisting up the pad when you use the larger bit to open up the hole.

    It is more work, but worth it if professional results are needed.
     
    PackratKing likes this.
  11. Alberto

    Active Member

    Nov 7, 2008
    169
    36
    Never had such a problem, and I do it all the time with my prototypes.
     
  12. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    how about hooking up a compressor and hose to blow air on the area you are drilling to cool it? shouldn't take much air. it looks like friction from the cone shaped tip of the wire is causing heat. might try a little different shape.
     
  13. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Thanks for the info! :)

    Please explain your technique. Do you drill before applying the etch mask, or after? And do you touch-up the etch mask on the holes before etching?
     
  14. Alberto

    Active Member

    Nov 7, 2008
    169
    36
    For prototyping I use commercial pre-treated PCB. They are sold in different size and are painted with a layer of photo sensitive lacquer. I don't use them in the conventional way since I drill them, as first thing then the pcb is CNC contoured to remove the paint where electric isolation is wanted and finally the board is etched. An example of the final result can be seen with the attached picture.

    Cheers

    Alberto
     
  15. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Wow what an interesting and unusual technique! :)

    I assume your CNC process has to cut through the lacquer and at least partially into the copper? That probably reduces etching time and reduces the problem of etching into the hole areas.

    But don't you still get the other problem where drilling makes swirling debris that scratches the lacquer?

    And I have to ask, if you are already CNC cutting through the lacquer and partially into the copper, why not use standard PCB and cut all the way into the copper? That way would save $$ on PCB stock price and means no etching is needed?
     
    absf and djsfantasi like this.
  16. Alberto

    Active Member

    Nov 7, 2008
    169
    36
    I began cutting the copper, few years ago but since copper is one of the worst metal to machine, with the additional problem that very rarely pcb are totaly flat, so you have to zero the Z axis to the lowest point and you end up in cutting to much in the highest point.
    Copper debrids where an additional concern (shorting tracks) and find them was not so easy.

    So I came to the conclusion that etching could be the solution, etching desolve easily copper debrids, but at the same time etching could resolve also the machining problem, if I could remove only the protective layer of lacquer.
    The solution was this home made tool (see pic 1) where a spring will compensate for the uneven Z plane of the pcb surface and remove just the paint layer.

    I post also an example of a board without components so you can see in details of the final work.

    Cheers

    Alberto
     
    absf and THE_RB like this.
  17. Alberto

    Active Member

    Nov 7, 2008
    169
    36
    Since I was unable to post two pictures with the same post, here I post the second picture.

    Alberto
     
  18. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Ahah! A spring loaded dremel bit, that gives constant downforce even on warped PCBs. Non-flat PCBs is one of the main reasons I don't try to machine PCBs on my CNC setup. That is very cool, thanks for posting that!

    Re your second PCB photo, the drilling before etching does seem to have damaged and scruffed up the copper around the holes a bit, but not enough to cause any significant problems with tracks that size.

    I really appreciate your posting this info, it makes me want to go and machine some PCBs! :)

    Care to share which PCB software you are using? I know EaglePCB has a ULP plugin for PCB routing, but your boards look different to what I remember Eagle producing?
     
  19. Alberto

    Active Member

    Nov 7, 2008
    169
    36
    No problem! I use a common CAD program. The final dxf file is then converted into ISO file for driving the cnc (there are several free share on internet). I actualy end up with two iso files one for drill and one for contouring.
    I post a zipped dxf file, as an example, that you can open with any cad program.

    Cheers

    Alberto
     
    THE_RB and absf like this.
Loading...