Making PCBs with the UV-Exposure Method

What is it?

For reproduceable results when making PCBs and super-tiny structures, there is a UV-Device for making PCBs now. Making PCBs with UV Exposure requires that some things are done in a different way to when making them with the printer. This document should at some point give an overview what steps are needed to make a pcb and what things you should take care of.

Disclaimer / Who may use the hardware

For a test, it was decided that there is an off limits project box behind the fridge. Only few people may access it. It contains information on *who* that is. If you ask one of those people, they will explain the process to you. This is for safety because you will have to use even more chemicals than for pcb printing. If you access the box without permission, rage will come upon you.

Please help to fix spelling / grammar and expression errors.

What is needed

Theese are the things needed:

1. A foil to print on with the laserprinter (works good: Conrad / Laser-Spezial-Folie/matt Best.-Nr.: 519570 - 62) 2. foto-positive base pcb-material (works good: Conrad / Bungard Platinen-Basismaterial 120306E50 Best.-Nr.: 529184 - 62, 529273 - 62, 523683 - 62

alternatively you can use Butterbrotpapier and UV-Positiv-Lack (we even have that in the box with the uv stuff) but it seems like a mess.

Safety hints / precautions

You'll be working with quite some chemicals. Most of them are unhealthy for you. You should at all times when touching anything wear gloves. Latex-Gloves stand in the electronic lab.

Especially FeCl_3 will leave ugly spots on everything it touches. We have some cleaner but it's a hazle and best is to not let anything get dirty. If you've got anything on your gloves don't touch anything else with them. Pull them off, throw them away and take new gloves. They're cheap you know?

If you touch chemicals by accident, don't panic. Go to the restroom immediately and wash them off. It shouldn't be that bad.

If you've got injuries on your hands you should refrain from etching pcbs as you might get the chemicals in the wounds.

The etching process takes some time, you should at least plan 2 hours (though it shouldn't take more than 1.5 hours)

How to do it

The first thing to do is printing the PCB layout to the foil. This can be done with the foil named in (1) using the shack printer. Put the foil into the printer's manual feed face up (the side that is marked to be sensitive to the toner facing up). It's recommended to fix a small part of the foil to an A4 paper. Do not use gaffer tape or anything that melts because it will do so in the printer and break the printer. Then print. You'll probably notice that the print looks much more black compared to when you used *normal* paper. That is actually good. The foil with the print on it will be called PRINTFOIL.

Prepare the UV-Exposure-Device

* Open the device (it has two locks on the short sides). * Apply the PRINTFOIL to the glass surface. You might want to fix it with some tape. Don't use Gaffa or normal TESA but something you can remove with as little residue as possible. The PRINTFOIL side that contains the toner (and is thus darker) should face up. * If the device seems dirty, don't do anything to clean itbut contact Armin / Leave him a message. If you fail to comply, you're likely to be breaking the device.

Prepare the PCB

* The PCBs come with some protective foil on them (the blue stuff in 2). Take it off on the side you want to print on. * Place the PCB on the PRINTFOIL so that the uv-coating points towards the lightsource. * Close the lid, fix it with the locks

UV-Exposure for real

  • Set the device to the time you want to expose your PCB to light. The time should be 1 to 2 Minutes. If you have very fine structures, it's likely closer to 1 Minute than to 2 Minutes. For very large structure sizes, 3 Minutes is ok. It's a bit tricky to find out the optimum time. I usually try 70 seconds at first. Works for SMD stuff like an atmega 168 smd uC.
  • You can find some values that worked below.
  • If you don't use the things (1) and (2) times may vary. Someone should make a lookup table for that.

Fixing the PCB

Fixing is the process of hardening the part of the PCB that was not exposed to the UV light and washing off the rest from the PCB. Use the white shell from the UV-PCB Kit. Do never use this shell for anything else. The substance for fixing is Water and Natriumhydroxide (NaOH) which makes Natronlauge. The amount to use for 1 liter is 7 grams of Natriumhydroxide. Since we use only about a quarter to an eigth of a liter of water, you should use 0.6 to 0.9 gram of Natriumhydroxide.

If you use too much, the fixing will completely wash off even the parts of the coating that have not been in the light. If you don't use enough, nothing will happen.

It's not recommendet to put more Natriumhydroxide into the bath while the PCB is still in because it will result in places with ultra-high concentrations of Natriumhydroxide that will instantly attach the coating that is supposed to harden.

The fixing should not take longer than 10 seconds to a minute. Sometimes you already see the stuff after 2 seconds.

You'll see that your PCB structures that stay will become golden, while the rest washes of and shows plain copper.

etching the pcb

use FeCl_3 like ever. I think it doesn't work with Natriumpersulfate with most coatings…

Washing off the coating

Put a bit of Iso-propanol into a shelf (not the white one for the fixing substance, remember?) and drop the pcb in it. After a few seconds you should be able to wash the coating off. a greenish residue should stay in the fluid. Take out the pcb and wash it off with water.

Finishing the PCB

You might want to finish the PCB by applying tin to it. Use the chemical tinn substance for that as you're used to.

How long should i expose my PCB

PCB Type Contents Time to expose Color of coating after fixing How long did fixing take? Results
Proma Fotobeschichtetes Basismaterial (positiv), Conrad 528668 & 536202 SMD, < 0.5mm 70 seconds black ~ 30 seconds very nice
Proma Fotobeschichtetes Basismaterial (positiv), Conrad 528668 & 536202 SMD, < 0.5mm 180 seconds black ~ 30 seconds fine traces (< 0.7) and restrings vanished
Bungard Platinen-Basismaterial, Conrad 529184 et al smd 0.1mm to 0.4mm 70 seconds golden 1-2 seconds perfect.
Bungard Platinen-Basismaterial, Conrad 529184 et al throughhole, > 1mm 180 seconds golden 2 seconds finest parts (about 0.5 mm between two traces) worked nicely
Bungard Platinen-Basismaterial, Conrad 529184 et al SMD, <0.5mm 180 seconds golden 2 seconds fine parts vanished though not as bad as with proma, still inacceptible
Bungard Platinen-Basismaterial, Conrad 529184 et al SMD, <0.5mm 700 seconds golden 5 seconds nice!

What to do with residue

wip

project/uv-exposure-based-pcb.txt · Zuletzt geändert: 2012/07/29 18:42 von 93.231.163.123