3

I tried to etch a PCB with the photoresist method, using sodium hydroxid as developer and ferric chlorid as etcher. I think the problem is in the development phase. Although reading a lot about etching, I seem to make a big mistake.

Here is what I did:

  1. I printed my circuit on a transparency twice, overlaid both
  2. Turned the lights off, red light on
  3. Put transparency on the photosensitive side of the PCB (Edit: I used a sheet of glass to keep transparency and PCB together)
  4. Using a 500W halogen lamp, I exposed it for 2 minutes (note: the PCB vendor says

1000-W halogen bulbs: 45 - 70 s

as I have a 500W lamp, I doubled the time to two minutes. I have no idea how to recognize when the exposure is finished, so I followed the instruction blindly.

  1. Put the PCB in Sodium Hydroxid developer, turned ceiling light (not the halogen) again on.

On my first trial, after 10s, I could see the traced of my printed circuit. I read that the development takes about 60s, so waited the remaining 50s, but then, no traces were visible any more. I continued etching and as I expected, everything was etched, no copper remained anymore.

So, second try. My thoughts were: I had seen the traces after developing for 10s, so the concentration of developer may be to high, so I added more water. Also, everything was etched, maybe the exposure time was too long. My second try was exposure for only 80s, but I never saw any traced during the development of this PCB. Again, all copper gone after etching.

Again, as everything was gone I thought the exposure may be too long. Repeating with 10s, 40s, 60s all resulted in the same as 80s.

I have no idea what is wrong. I only found people having over exposed boards with blurry edges but never found anybody having this problem: all copper is etched.

Do you have any idea what I might do wrong?

Edit: Is it important to keep the lights off during the whole development phase? The didn't do this here

The wired thing is: as everything is etched, it looks as if the exposure was too long. But when reducing the exposure time, I didn't see any traces even not in the beginning of the development phase, where I could see traces with 2 minutes exposure.

Edit2 Is it possible, that this black layer is the photo resist layer and should stay on the board when peeling the blue off? It is impossible to separate the blue and black layer, so this could be a fabrication mistake. enter image description here

turn

black
  • 141
  • 1
  • 7
  • If you hold up your transparencies to a light source, can you see though the black part at all? – winny Sep 05 '19 at 08:47
  • "Put the PCB in Sodium Hydroxid developer, turned lights again on." Just develop the image, don't turn the UV back on. The traces must be clearly visible after developing, don't waste PCBs if you can't see the traces. – Marko Buršič Sep 05 '19 at 08:51
  • @MarkoBuršič Thanks, I edited: turning ceiling light on, not the bright halogen lights I used to develop – black Sep 05 '19 at 08:56
  • @winny Nearly nothing. It looks like https://youtu.be/tWnfnt2rNO0?t=118 – black Sep 05 '19 at 08:57
  • 1
    You don't have to turn off lights or illuminate with red light, you can freely inspect and develop with ceiling lights on. How did you apply the photoresistive? Have you cleaned the surface with sanding? – Marko Buršič Sep 05 '19 at 09:07
  • @MarkoBuršič That is great to know! Even with usual LEDs? (As LEDs often have a peak in the blue region of the spectrum) I bought [these already coated with photoresist] (https://www.reichelt.de/photo-pcb-epoxy-single-sided-100x75-mm-1-5-mm-35-fpe1-100x75-p159495.html?GROUPID=7781&START=0&OFFSET=16&SID=96XVFN7KwQAT0AAAE3xjg5baf56ea58e93557caab22c4928d6a7c&LANGUAGE=EN&&r=1) and only peeled the protection foil off – black Sep 05 '19 at 09:09
  • 2
    I think you developing solution is too strong. You see the traces in the beginning, but then they disappear. It's been a while since I etched PCBs, but as far as I remember, we used a weak solution for developing the PCB, and a much stronger solution after etching to remove the remaining film from the copper traces. – Peter Karlsen Sep 05 '19 at 09:10
  • Good, then it's not that. Back in the day I had little luck with laser printers and ended up with an inkjet with some tweaked settings to make it non-transparent. – winny Sep 05 '19 at 09:23
  • Are you sure your blank pcb is photosensitive? Seems like it is not. You just have a plain normal blank copper pcb. Also with 500w the vendor says 6-7 minutes, not 2 minutes. – Soldersmoke Sep 05 '19 at 10:31
  • @Soldersmoke I use a 500W halogen lamp, which has more UV than usual bulbs. The vendor says 1000-W halogen bulbs: 45 - 70s – black Sep 05 '19 at 11:13
  • I'm voting to close this question as off-topic because it has been abandoned by the asker – Chris Stratton Oct 13 '19 at 21:03

1 Answers1

0

As already described in the comments, the develop solution appears to be too strong.

The traces should be nicely visible when developed. The fact that you saw them appear suggests that your exposure was at adequate. Too strong developer will eventually destroy even unexposed areas.

I am using similar boards, with granulated NaOH (sink cleaner) in concentration of 7-15g NaOH per 1 liter of water as suggested here.

akwky
  • 1,584
  • 1
  • 14
  • 21
  • I’m trying different concentrations now. Which part of the pcb should change color - the exposed or unexposed? I would try to dump one pcb part that changes definitely the color (I.e. exposed / unexposed) to eliminate any problem with the exposure. – black Sep 05 '19 at 12:39
  • I'd say both change color a bit. First, the exposed area gets dark, and progressively goes away. So at the end, the exposed area is bright. See photos 13-17 here http://www.tajned.cz/2017/06/vyroba-dps-fotocestou/ . If you stir the solution a little, you can see as the reddish laquer is "melting" away. The process takes few minutes to complete at the concentration I wrote above. – akwky Sep 05 '19 at 13:30
  • 1
    Thank you. The development solution was indeed too strong. – black Oct 14 '19 at 07:24