Homebrew PCBs

On Wed, 20 Aug 2014 13:55:56 +0100, you wrote:

>Hi, thanks for adding me to the group.
>
>
>I've just done a quick trial etch, with fair results for my first attempt
>in years.
>
>
>A scan of this board is at https://www.sendspace.com/file/5yy3qw , or if
>that's broken I can repost it again or somewhere else.
>
>
>I used a 600dpi inkjet transparency (ok I know not great), Mega UV exposure
>box for 2.5 mins after a 5min pre-heat, sodium metasilicate developer at
>20'C for about 3 mins, rinsed, etched with fresh sodium peroxidisulfate at
>50'C for about 8 mins, rinsed, then wiped clean with resist stripper.

The numbers don't mean all that much, generally (you can see major
overexposure and major underexposure).  What is important is what
works for your setup.

>
>
>The main problem I have is that there are fine vertical streaks of copper
>remaining on the board, and some track pitting. I can only think that I
>should have exposed the board for longer, as perhaps there was some resist
>still remaining when I etched the board.

Go back and look at your transparency.  See if those pits and streaks
match up at all with the transparency itself, if so, that's an answer
(transparency not dense enough)

Generally, I think that red ink seems to be better, but I haven't done
photoetching in a long while.  Multicolor to make black is not a good
idea either (IIRC).

You generally want to make an exposure strip, so take a piece of
board, and expose (say 1/5 of it) for one minute with the rest of the
board covered.  Move and record position of the mask, and expose for
another minute.  You now have 2/5 of the board exposed, 1/5 at 1
minute, 1/5 at 2 minutes.  Continue until you have the whole board
exposed at 1,2,3,4,5 minutes (or more if you wish).  Develop and look
to see what's best here.  That'll give you an idea of the proper
exposure.

>
>
>(The bottom of the TQFP footprint got accidentally rubbed away while I was
>trying to check if any resist was still clinging, oops!)
>
>
>I think that the streaking problem is from the resist clinging due to
>underexposure/underdevelopment, but tell me what you think.

See above to determine proper exposure.  If 5 minutes is too small,
then arbitrarily add 3 minutes to the whole board so you run 3 to 8
(or whatever you want).

Hope this helps a bit,

Harvey

>
>
>In any case, please would you kindly take a look at the results and give me
>your opinions?
>
>
>Many thanks,
>
>
>Matt.