We would wager that most of the home etched PCB projects we see around here use the toner transfer method. But the next most popular technique is to use photosensitive ink which resists the etching acid once it has been exposed to light. Most people buy what are called pre-sensitized boards, but you can also get ink to make your own. [Jardirx] does this, and uses an old hard drive to apply an even layer of the light-sensitive ink.
The narration and subtitles of the video found after the break are both in Portuguese, but it’s not hard to figure out what’s going on here. He begins by using double-sided foam tape to secure the piece of copper clad board to the hard drive platters. You’ll want to center it as best as you can to keep the vibrations to a minimum. From there [Jardirx] applies a coating of the ink using a brush. The image above is what results. So as not to get ink everywhere, he then lowers a soda bottle with the bottom cut off to catch the excess. Power up the drive for a few seconds and the board will have a nice even layer ready for a trip through a UV exposure box.
[Thanks Daniel]
I tired this once with a laser pointer, it flew off and almost broke a window. =D
That’s pretty inventive. I wonder if there’s a way to place the PCB so you don’t end up with an accumulation of resist ink in the middle of the PCB.
My uneducated guess is that the reason if there’s an accumulation of ink in the center, that is because the board isn’t centered to the motor shaft. Another thing to consider is that the foam base of the tape may be distorting. Unless a less than perfect thickness would affect the exposure & development of the coating, why be concerned? In event there is a concern a centering mounting method that works in a way similar to a four jaw metal lathe chuck could be devised. To that effect take the top platter to a lathe to scribe concentric circles on on it, along with scribing an X centered on the shaft with the right angle as perfect as you can make it. That leaves coming up with a way to mount the PCB that allows for multiple adjustments to get it centered. All assuming my initial guesses are correct
I just now looked it up and found this helpful link:
http://large.stanford.edu/courses/2007/ph210/hellstrom1/
Spin coating typically makes a layer of uniform thickness, although this wasn’t immediately obvious to me. The force on any infinitesimal mass of goop is F = M*(V^2)/R, so one would expect thicker goop closer to the center.
This is apparently not true in practice, for mathematically complex reasons. Spin coating makes a uniform layer, even at the very center of a disk.
A question to ask: is the etch resist actually thicker in the center, or is the coloration due to evaporation at the center being slower than at the edges?
The relation between the V and R is not immediately obvious. You should write the formula as:
F = M * ^2 * R. And Ohmega, the radial velocity can easily be seen to be constant as you are spinning a solid object…. Now it can be easily seen that goop is forced to the outside proportional to the distance from the center of spinning.
With a wafer, I can imagine that you’d botch the center two or four chips with a non-uniform coating. That might be acceptable for the chip-industry, but here you want a specific part of your circuit in that area, so it would not be a useful situation if you’d get way too much goop in the center. I’d like to know the “mathematically complex reasons”, because at the moment I don’t understand how it would work.
I did not know how to write the ohmega symbol in the above comment. So I wrote ohmega in angled-brackets. Turns out the whole word is than scrapped.. Sigh.
You can see a good explanation here: http://large.stanford.edu/courses/2007/ph210/hellstrom1/
This process works reliably over a wide range of substrate sizes. In the semiconductor industry 50cm diameter wafers are about the maximum. This process gives very uniform coatings (a few microns thick) over the entire wafer. If you are careful there is no thick part in the center.
This is exactly they way they do it when they manufacture wafers for ICs, I like it! :-)
I like how when it’s finnished he says “Que belleza” , which I figure is pt for “what a beauty” .
Very clever ! Somebody knows where to buy the blue photosensitive ink ? Thanks
I suppose here http://www.alibaba.com/product-gs/330482675/PCB_ink_for_KSM_S6189.html surely it’s something that could be cooked up at home. Not something I’m interested in researching any longer the 5 minutes I already spent.
If you search ebay for item
190741561459
you’ll find some, but you need to dilute it before use, they say to use “banana oil” (Isoamyl acetate).
I suspect, just going by the similarity of labels, appearance in photos, and the fact that they are both negative acting (black areas of art wash off), that this stuff is probably the same as the 100g tubs of solder mask ink you can buy (also on ebay and much easier to search for – “solder mask 100g” will give you a pile), and that diluting it simply makes it easily removable after etching and perhaps improves the resolution and conformity.
Unfortunately “banana oil” really doesn’t seem like something you can readily get your hands on. I’ve wondered if linseed (flaxseed) oil might have similar effect though, or maybe even a little turpentine.
in the video comments a guy says he uses aceton (nail polish)
one important detail he says is that it has to be spinning at greater than 4500 rpm according to the manual for the paint, and says hes using a 5200 rpm hdd
Interesting project. Will be great when we really see translation of the spoken word available. Particularly in light how makers using video to share their projects, probably because it’s less time consuming.
Copper board, goo, glass plate. Angle between board and glass determines thickness of goo. Draw plate held at consistent angle and at consistent speed across board – voila, a thin and evenly spread layer of goo.
Works with much larger boards then you can spin on a HD platter.
Setup and cleanup is a fraction of the time.
A few practice swipes and it’s a simple procedure.
Would this help me evenly distribute the mustard on my sandwiches?
Are those motors powerful enough to spin a hoagie?
The mustard doesn’t move evenly across the surface of the bread because of it’s porous nature. You get a bunch trapped where you placed it, and whatever hasn’t soaked into the holes kind of skips across the surface.
You could mustardize slices of cheese, but that’s an affront to nature.
I think Western Digital makes hoagie rated drives.
Seagate obviously only expects theirs to be used for subs.
While this is a cool use of an old hard drive, is there a practical advantage to the home board etcher to have such a uniform coating? A commercial board shop needs them to be uniform so they can standardize exposure time and minimize their costs. A home user just bakes the hell out of the board during the exposure, so perfect evenness is not as relevant.
The guy said that he tried to spread using several techniques, brush, visiting cards, credit cards. It didn’t work.
@HaD He leaves an important note in the middle of the video: use a 4500 RPM hard-drive.
I would have expected the terms “spin casting” or “spin coating” in this post.
Spin Art!
Is there any need to translate any part of the video?
I’m interested to see of more research has been done in to the field of centripetally-enhanced condiment spreading.
Yeah, the Tesla Coil folks learned this the hard way.
Seems that the trick is to coat your unwound secondary first, preferably keeping the former at a lower temperature than the wire being wound.
This results in a nice smooth winding rather than a mess.
I did wonder if the acrylic rollers available from craft shops in .gsy would work as a TC former, these are sold as clay forming devices.