@Rick Sparber
Your last picture is most revealing.
Apparently, you are faced with a difficult etching process as described in the Electronics_101 thread you sent. I would recommend getting a good etching process first, then tackle that harder problem. I suspect that is what you are doing.
When I etch a PCB, I do not seek convection etching as a goal; although, I am sure there is some of that going on.
1) Why do you want convection and laminar flow? Considering the chemistry, even without considering the interaction of oxygen and oxidized forms of copper, I would think that turbulent flow would be better. With laminar flow, there will always be that surface layer of etchant in contact with the copper that is not well mixed.
2) Your experiment shows "edge effect," i.e., the edges etch faster than the center. I am sure there are several reasons for that to occur, but certainly having laminar flow across the board will make it worse. I do my boards either copper on top or inverted with continuos magnetic stirring and heated to 60 to 70°C. Copper down goes faster. I made some glass trivets from 3 mm glass stirring rods to rest my boards on. Depending on the severity of any edge effect, I remove the board and use a sponge with etchant to speed up the central etching. With 1/2 oz and 1 oz copper, that is generally not necessary to do.
3) You don't say what your resist is. If it is a photoresist, underdevelopment and/or underexposure (for positive resist)can lead to difficulty etching. With my photoresist (Injectorall), it is easy to tell when development is complete. When properly developed, a water rinse will show wetting of all exposed copper. If in doubt, a quick dip in ferric chloride will show if there are any areas that are to be etched that still have remaining resist on them. If so, it can be rinsed and put back in the developer for more time. That final picture, however, shows far more scratches and defects than I have ever seen with photoresist. That is not meant as criticism, but only as an observation that something is wrong. Now, if you are using the toner-transfer method, then such defects may be more common. I don't use toner transfer and can't comment on that further on that.
My recommendations for ferric chloride would be to:
1) Increase your etching temperature to 60°C
2) Improve stirring and/or agitation. Remember, of the etchants available to hobbyists, ferric chloride provides the least undercutting and best edges. You may get better results, if you dilute concentrated ferric chloride a little bit -- say 10 to 20% with water (i.e., 1 part water to 4 to 9 parts concentrated ferric chloride). The Edinburgh etch (
http://www.lawrence.co.uk/acatalog/etch.pdf), which adds citric acid to the ferric chloride has been reported to give sharper etching. I have tried it, but found the difference for what I do was negligible. It might be worth a try for you, though.
Regards, John
----- Original Message -----
From: Rick Sparber
To: Homebrew_PCBs@yahoogroups.com
Sent: Friday, November 23, 2012 7:15 AM
Subject: [Homebrew_PCBs] first step into convection etching
Although I have done analog circuit design for 50 years, I'm a relative
newbie when it comes to circuit board etching. Furthermore, my last
chemistry class was in 1967. Here is my first step into convection etching.
http://rick.sparber.org/electronics/ce.pdf
How do I improve this process? I understand that bubbling air into the acid
will speed it up but won't that spoil the laminar flow?
Thanks in advance,
Rick
[Non-text portions of this message have been removed]
[Non-text portions of this message have been removed]