design of a UV exposure box / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

design of a UV exposure box

previous 1 2

Move Thread

LAN_403

Steve Conner

Sun Feb 28 2010, 09:20AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

hboy007 wrote ...

I have the impression that I lose about 50Âµm around the edges depending on the distrance from the lamp and that is far from being satisfactory.

Well, Mike's "10 mil" is 10 thousandths of an inch, which is 250 microns. It seems to me that 50 microns is within the error budget for that. Ship it!

As an intellectual exercise in ray-tracing though, it's very interesting. I think you said this was ray-tracing code that you wrote yourself, in which case, maybe you should tell us about it in the Projects or Computer Science forum.

Here's another optics head-scratcher: Would honeycomb really help, if the insides of the holes are reflective? Wouldn't it need to be black?

hboy007

Sun Feb 28 2010, 02:59PM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

Now that's better (35mm distance from the tube axis, curvature 0.065 1/mm) yields a very nice result. I had to fix some minor errors so now I am pretty sure the frist order reflections are correct. I neglected photons that bounce back to the tube walls because direct reflection of the glass surface is high only for small angles toward the surface and in case of absorption the re-emission characteristics are the same as with primary generation).

I added some box blur to the scatter plot to make it more viewable. Photons within 20deg full angle are allowed. The tube seems to be well in focus and it creates a central overshoot as expected.

As I seem to have reached the point where the geometry would get really complex in order to compensate for the significant loss of light, I consider this a good point to worry about the practical results.

Yes, maybe this thread should be moved to "Computer Science", you are right.

hboy007

ps. I made my thoughts about the coated honeycomb idea. Unfortunately, with the walls aligned vertically, the stray angle for each ray is preserved and the entire structure acts as a transfer optics.
I thought, well maybe it is possible to collect the light that enters a channel with a cone and shape it as it leaves the tip of the sawed-off cone. It doesn't work, because the light leaves the cone at the wrong end.

hboy007

Thu Mar 18 2010, 10:25PM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

"Other limitations..."

I'm back with some new impressions. Today I've produced a 0.5mm pitch 20pin FFC connector board (with a yield rate of 25%). I use a 600 dpi laser printer and transparencies. Here is a picture illustrating the feature size of 100Âµm

this seems the only way to make the board single sided. Tomorrow I will have a closer look at the traces. They don't look too good and I get the impression that bubbles that form during the etching process cause denting about 50Âµm deep which is more than half of the width of the remaining wire.
I used the center sharpness directly above the tube and all the limitations make me doubt if it is worth tweaking the illumination uniformity to perfection whereas the resolution I achieve in the process is mediocre. (pictures will follow)

Has anyone observed horizontal-to-vertical resolution anisotropy when printing artworks? Maybe rotation of the traces to lign up with the scanning direction of the printer can improve the edge smoothness.

MinorityCarrier

Thu Mar 18 2010, 10:33PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

The effect you call "denting" may be what in IC photolithography is called "mouse bites". That problem is one of photoresist adhesion to the substrate. Just a thought...

hboy007

Fri Mar 19 2010, 12:38AM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

Thank you for pointing me in the right direction, MinorityCarrier. My sodium hydroxide solution needs to be temperature controlled, the right moment to transfer the board to the stopper bath is still just my guess based on the appearance of the photoresist layer. But there are tiny holes within larger pads, too. I shall blame the lousy toner coverage on the foil for that

Conundrum

Fri Mar 19 2010, 07:44AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

NaOH is a pain guys, use silicate instead (sodium metabisilicate)

if you ask me nicely i have a bottle sitting here unused, was waiting till my etching tank was ready.

FWIW, I also found a nice trick using a spare power supply fan- spin coating. Rotate the PCB at hundreds of RPM and the coating is nearly perfect.

For free I have 36 48V (!) massive fans if someone has a use.
the chips in them alone are worth a fortune
:)

regards, -A

hboy007

Fri Mar 19 2010, 11:29AM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

as promised here is a picture of the result (I couldn't resist trying to tin the traces, the width didn't change).
Notice the little hole at the edge. Skin ridges are for size comparison

and here is another one showing even more bites:

let's see if I get these issues under control. I was so unbelievably distracted that I forgot to set the "mirror ouput" checkbox in the cam processor and now the layout is flipped *cough*... I'll redo them when noone is watching

ps @ Conundrum: thank you, unfortunately I'm from Germany and even shipping will probably be more than what they charge at the local home store (they sell potassium silicate, will that work, too?)

ragnar

Fri Mar 19 2010, 03:08PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Hhboy007, bites aside, these are exceptionally good results!

Observe this 24mil trace suffering a similar effect on a larger scale.

The damage was specific only to the second identical batch of boards I had done. Steve McConner proposed I may have let fly a booger onto the transfer surface, and I supposed him correct, considering this sheet had been exposed in the room for hours, potentially collecting 'fluff'.

When the fluff causes lift of the toner, forming a "bite", I imagine it's only a matter of time for the undercutting etchant to wreak havoc, the same way "shaped charges" affect explosives. But considering you are using UV-photosensitized copper, I have no idea where or what your boogers are :P

hboy007

Thu Mar 25 2010, 12:34AM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

I've managed to build the connector board, at last, using 0.3mm soldering wire and lots of flux:

I have also acquired a UVA face tanner for a very reasonable price. Now I have four tubes and with a small, low profile reflector and a honeycomb segment I will be fine.

I tried different orientations in respect to the printer scanning direction and parallel lines seem to be worse than lines that run perpendicular to the scanner (I can observe some speckles along the borders of the lines that might look like aliasing).

On the photo there are some flux residues which I do not dare to remove because the 100Âµm traces break soo easily :/
I'll be back with the on-topic content as soon as I find the time to work on my exposure box again. Thanks for your interest

previous 1 2

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob