Connect with us

Photo-resist PCB help

Discussion in 'PCB Layout, Design and Manufacture' started by BobK, May 11, 2014.

  1. BobK

    BobK

    7,609
    1,648
    Jan 5, 2010
    I have just tried my first attempt at a photo-resist PCB. The results, shown here are not quite ideal.

    All of the traces should be the same size at 0.12 in. Obviously some of them have bled out to the point where clearances are not met, while others look quite sharp. The image on the left is the laser artwork which I used,. it looks quite a bit sharper than the resulting board on the right.

    Can any of you who use this method tell what what I did wrong?

    I am using the MG Chemicals kit with their exposure light a the recommended 8 min exposure. The artwork was held down with a clear acrylic sheet. The image on the developed board before etching looked sharper than the final result, though I don't have a picture to prove it.

    photo-pcb.JPG
     
  2. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,178
    2,690
    Jan 21, 2010
    I notice toy have a broken track and some incomplete etching.

    It looks like the thicker traces are in one particular area. This may have been caused by some agitation issue, or perhaps some smudging of the image, since the widening of traces seems to be in one direction only.

    Was this positive or negative material that you used? (i.e. did the image placed on the board look like the one to the left, or was it an inverse of that?)
     
  3. BobK

    BobK

    7,609
    1,648
    Jan 5, 2010
    It was positive. The image on the left is exactly what was printed onto the transparency.

    If find it hard to think of any mechanism to widen the traces, it seems like most mistakes would do the opposite. The thing I can think of is underdevelopment, where not all of the resist was taken off. I guess on my next attempt I will make sure the developed image is good. The other thing that is funny is that on the good tracks I can still see the green color of the resist. On the tracks that widened, it looks like bare copper after etching, which again sounds like the opposite of what one would expect.

    Bob
     
  4. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,178
    2,690
    Jan 21, 2010
    Your thoughts are along the same lines as mine.

    I was hoping you'd say it was negative material, but even then it's an unusual pattern,

    Did you coat the board yourself or was it pre-sensitised?
     
  5. BobK

    BobK

    7,609
    1,648
    Jan 5, 2010
    Pre-sensitized. MG Chemicals brand, not suspect Chinese stuff.

    Bob
     
  6. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,178
    2,690
    Jan 21, 2010
    I have a collegue who has used both positive and negative material. I will get him to take a look and comment.
     
  7. shumifan50

    shumifan50

    548
    56
    Jan 16, 2014
    A very stupid question, but when you did the exposure, was the printed side of the transparency against the PCB or facing the UV light? I had this problem when the printed side was not against the PCB and light bled in during exposure.
     
  8. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,178
    2,690
    Jan 21, 2010
    The advice I have received is as follows:

    "The lack of Intimate contact of artwork with resist is the most likely problem. Unless the guy is using vacuum this is what he can expect.

    I forgot to mention that 6mm glass is the best hold-down medium. I have tried using gravitational force alone, and never got really acceptable results for fine work. Acrylic does warp, and it doesn't take much of a gap with a diffuse light source to produce a pcb as per photo. Pity he didn't take a shot pre-etch. I used to use 3mm glass, but it is too easy to break it, even under mild vacuum. Unfortunately, glass is a heavy UV absorber and the extra thickness extends exposure time.
    "​

    I will confirm he's aware you were using positive artwork, but Ross is a firm believer in glass and a vacuum.
     
  9. shumifan50

    shumifan50

    548
    56
    Jan 16, 2014
    I use an exposure unit that has the UV lights in the box underneath with a glass bed above it. You lay the transparency on the glass with printed side up and then put the PCB on top. When the lid is closed a piece of sponge in the lid applies pressure to the PCB pressing it tight against the transparency and the glass. This works quite well - it also protects eyes against exposure to the UV lights. The glass bed is 3mm so must be treated with respect.
     
  10. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,178
    2,690
    Jan 21, 2010
    An update. After reminding Ross that you were using positive material he suggested:

    "Valid point!

    The only other possibility I can see is incomplete resist removal, a problem I have experienced in the past.

    I use encouragement from a 2 inch paint brush, which was very effective using the negative stuff, and a technique I have carried over to the positive. The resist over the tracks is very resilient, so there is no danger of compromising its integrity.
    "​
     
  11. BobK

    BobK

    7,609
    1,648
    Jan 5, 2010
    The toner was against the PCB. I know this since the image is mirrored as it should be.

    Bob
     
  12. BobK

    BobK

    7,609
    1,648
    Jan 5, 2010
    Yes, it seems to me that with positive art, a gap would undercut the traces since light would creep in from the sides rather than widen them, this is why I did not think that was the problem.

    I did brush with a foam brush that came with the exposure kit during development, but perhaps I was not doing it vigorously enough.

    For my next attempt, which will probably be next weekend, I will try to brush it more and check it more carefully for full development.

    Thanks all for you suggestions!

    Bob
     
  13. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,178
    2,690
    Jan 21, 2010
    Perhaps you should make a test board and see how hard you actually have to scrub it to remove any of the resist.

    It will give you the confidence that you're not applying too much vigor when you;re removing the exposed material.
     
  14. shumifan50

    shumifan50

    548
    56
    Jan 16, 2014
    One last suggestion: Make a test strip with some narrow tracks. Expose an inch or so at a time by covering the rest. Move along until you cover quite a wide range of exposures, then develop and etch and see which bit comes out the best. UV tubes deteriorate and the distance between light and pcb is obviously important. By doing the simple test it is possible to determine the best exposure time for your setup. I found the best etching time by visual inspection after etching for 30 seconds at a time and adding the times together.
    I agitate the etching solution using a fish tank air pump with some tubing with tiny holes punched in it with a pin. I have also used my airbrush compressor to bubble the etching solution. Lots of posts advise against this method, but I have had the best results doing this as opposed to brushing and shaking and the result is fairly consistent.
     
  15. timbell

    timbell

    1
    0
    Apr 12, 2014
    I've had this before, time and time again it's because the pressure pushing the board against the glass platen is uneven.

    My old work exposure box was vacuum pressurised using some rubberised canvas but my home-made rig uses some sponge foam and a set of weights on top of the lid, when I get the weights off centre I get exactly what you're seeing here.

    edit: I note you are using acrylic sheet, perhaps there is some slight bowing on one end? Something I've seen others try - insert the PCB and transparency into a picture frame and try exposing it that way.
     
    Last edited: May 15, 2014
  16. BobK

    BobK

    7,609
    1,648
    Jan 5, 2010
    Success!

    It was NOT a problem with exposure, I did that exactly the same.

    I looks like it was a combination of underdevelopment, too little brushing and spent ferric chloride solution. I developed it to the point where I did last time, and I noticed that there was still a green tint over much of the board. So I developed it some more until that was gone. Then I noticed my ferric chloride looked pretty bad, so I used fresh. And voila!

    This is a much better result than I ever got with toner transfer, so it is my new process. For scale, the traces are 0.012 in, and I believe from this result that I could probably go to 0.008. The 6-pin thingy in the upper right is a 6-pin 0.1 in header.

    Thanks to all who had suggestions.

    board.jpg
     
    Last edited: May 18, 2014
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-