Making SMD PCBs at Home (Photoresist Method)

17,710

179

28

Posted

Introduction: Making SMD PCBs at Home (Photoresist Method)

Making PCBs at home is probably a dying art, since more and more PCB manufacturing companies will print your circuit board and have them delivered to your home for a reasonable price. Nevertheless, knowing how to make PCBs will still prove useful when making prototypes or replacing that damaged circuit which will take weeks to ship. Also, the skills required to etch a PCB can be useful for etching many other materials with creative designs.

In this instructable I'll show you how to etch a one sided board, apply a solder mask and solder some SMD components to make a test board for some WS2812B RGB LEDs.

Step 1: Materials:

Here's a list of all the materials you will need to etch and populate your circuit board.

*Glass fiber copper clad boards are discouraged since they can be a bit difficult to cut.

**I used a 2:1 Hydrogen Peroxide and HCl solution, at 3% and 30% respectively.

Step 2: Design Your Circuit

There are are many programs available to design your circuit. The one I use the most is Eagle, which might not as advanced as some other PCB software available, but it is free and easy to use. It allows you to create boards of up to two layers. Eagle has been recently purchased by Autodesk, so it might gain more relevance in the professional field in the future.

If you're not familiar with PCB design I suggest you take a look at other instructables, there are also great tutorials on youtube, like the one by Jeremy Blum.

In the picture you can see how a circuit looks like once its finished. In this case, this circuit will be used to test some WS2812B LEDs.

I have attached the Eagle files in case you want to save or modify the circuit.

Step 3: Export a Mirrored PCB Negative

Sadly, Eagle doesn't have the option to invert the colors of the circuit before printing, thus some extra steps are required.

Once you've finished your circuit, select the traces, pads and other stuff you want to be etched, then export it in a PDF file. Do the same thing for the pads only, this will be used to expose for the solder mask.

Eagle outputs the PDF at a 600dpi resolution, which is usually the same resolution of your printer, so no adjustments in size are required.

To invert the colors of the PDF file I used Gimp. I opened the file with a resolution of 600dpi* and proceeded to invert the colors by selecting the circuit and using the invert tool. I also pasted the pads alongside to print it all at once.

I have attached the PDF file in case you want to print this circuit, no resizing should be required.

* Warning: Setting any resolution in the editing software other than 600dpi will result in size changes when printing that will be hard to correct. Also, make sure your circuit is mirrored so that it can be flipped over, with the toner or ink pressing against the board.

Step 4: Prepare the Transparencies.

After making sure all the components fit nicely on the circuit on paper, the circuit and pads are printed onto a transparency.

Quite often, one transparency is not enough, some holes might appear which will let light pass through, leaving exposed areas where we don't want them. It can be a major problem that is easily fixed by stacking another transparency on top.

The transparencies are glued with epoxy and lined up and the epoxy is left to cure. Placing the transparency under a flat surface is a good idea, so that the transparency is totally flat once the epoxy cures. The excess is trimmed before proceeding with the next step.

Step 5: Prepare the Board

The copper clad board is scored on both sides and broken to size. It's also cleaned with a scoring pad and some soap to leave a shiny and clean surface. Alcohol can also be used to clean any residue left.

Make sure there are no dust or fingerprints on the board, as any residues will determine the success of the next step.

Step 6: Stick the Photoresist Film

Cut a small piece of photoresist film. The film is composed of three layers, a plastic on top and bottom and the photoresist chemichal sandwiched in the middle. To separate the film I put two pieces of sellotape on an edge and then pull apart, a clear film will come off and will be discarded, the sticky side (the one with exposed photoresist) sticks to the board easily without any air bubbles.

To fully stick the film to the board some heat needs to be applied. A modified laminator is often the preferred method, but an iron or an electric stove can also work. In my case, I'm using the electric stove in my kitchen. A piece of paper is used in between to prevent the plastic film from heating up too much.

Step 7: Expose the Photoresist

The transparency sticks to the photoresist with the help of a drop of water. It is then exposed to UV light.

It is important that the ink or toner on the transparency is in contact with the board, otherwise light might be able to sip in between the film and the board and cause some defects.

The exposure time will depend on the strength of your UV source. In my case, 30 seconds in my homemade PCB exposer is enough. Under a normal CFL lightbulb several minutes might be required.

Step 8: Remove the Unexposed Photoresist.

Once the PCB has been exposed the transparency is removed and the plastic film which held the photoresist is peeled away. The photoresist will have hardened and will be stuck to the board.

Unexposed photoresist is removed in a solution of 1% Sodium Carbonate*. A brush is required to rapidly dissolve the unexposed areas. The copper board should be visible after this has been done.

Once this has been done, the board is exposed to UV light again to fully harden the photoresist and prepare it for the etching.

*Sodium Carbonate is also known as Washing Soda. If you don't have access to Sodium Carbonate you can make some by heating Sodium Bicarbonate (Baking Soda) in an oven at 200ºC for a couple of hours.

Step 9: Etching the Board

The board is etched in a solution of 2:1 Hydrogen Peroxide and HCl (Muriatic Acid), it usually takes around 10 minutes. Then, the photoresist is removed by submerging the board in acetone until it peels away.

The edges of the board are finally trimmed and sanded. After the board has been etched, continuity checks are carried out to ensure there are no short circuits.

Step 10: Applying the Solder Mask

A big blob of solder mask is applied in the centre of the board. The transparency is stuck on top and the solder mask blob is squeezed evenly. The same process as before is used to expose the PCB. The unexposed solder mask is removed with tap water and a brush.

In some cases the solder mask can stick to the transparency, so a piece of polypropylene film is sandwiched in between. Cleaning the board with alcohol and a paper towel will make it quite shiny.

Applying solder mask can be a bit tricky, as you can see, some spots have peeled off, but that's acceptable.

Step 11: Soldering the Components.

A bit of solder paste is applied to each pad. The SMD components are placed on the pads. Don't worry if it looks like a mess, after heat is applied the surface tension of the molten tin will pull the components straight. The solder mask will prevent solder from sticking anywhere else but the pads.

The components are heated up with a heat gun at a temperature of 300ºC for a couple of seconds until the solder paste melts. It is then allowed to cool down slowly for several minutes. Other methods might be used (reflow, hot plate...).

Some pin headers are soldered to the circuit before testing it on a breadboard.

Step 12: Testing the LEDs

A simple Arduino program with the FastLED library is used to control the LEDs. The good thing about the WS2812B LEDs is that they don't need resistors, since there's an internal IC which regulates the current. An entire array can also be driven with one single input.

This little test board has been completed successfully and it's time to move on with bigger projects.

If you liked this project and want to see more in-depth instructables about some of the topics discussed here, please like this instructable and consider giving it a vote for the LED contest.

Thanks for watching!

Share

Recommendations

  • Epilog Challenge 9

    Epilog Challenge 9
  • Paper Contest 2018

    Paper Contest 2018
  • First Time Author Contest 2018

    First Time Author Contest 2018
user

We have a be nice policy.
Please be positive and constructive.

Tips

Questions

21 Comments

Hi, great tutorial. Thank you so much!

"Unexposed photoresist is removed in a solution of 1% Sodium Carbonate*. A
brush is required to rapidly dissolve the unexposed areas. The copper
board should be visible after this has been done."

Could you say how much time take for you to dissolve the unexposed areas?

It takes a couple of minutes using a small brush, but it can take a bit more if your board is larger because you have to brush a larger area. The sodium carbonate softens the unexposed areas, but they don't dissolve easily without help, so a small brush helps a lot.

Time is not too critical in that step, I haven't had problems with exposed areas dissolving or chipping even after letting the board sit for a couple extra minutes. The exposed photoresist sticks to the board extremely well, and only dissolves after some minutes in a bath of acetone. The acid and the sodium carbonate don't seem to affect the exposed areas at all.

I definitely would like to see more in-depth Instructables like this. Keep up the great work!

Build_it_Bob

I have finals next month so I'll be busy studying, but after that I plan to finish some projects involving more homemade circuit boards, which will be bigger and probably double sided.

Firstly: excellent instructable, thanks.
Just a couple of useful tips:
Cupric chlorate (hydrochloric acid / hydrogen peroxide) is an awesome etchanrt but a few things are useful to know: 1. This stuff BOILS on mixing if given chance. Do this outdoors.
2. Its REALLY poisonous so wash your hands.
3. It gets better with use so on first mix add a bit of copper pipe until it goes green.
4. When it starts to get slow, dont bin it, just bubble air through it for 24 hours and it regenerates.

As per etching:
To make an etch tank you just need a tall but thin plastic container, a fish tank air pump and tube and a fish tank glass heater!

For exposure:
Quite a few ultraviolet LEDs are available as well as Ccfl tubes.
The 410nm range is ideal as they cover quite a span. Insect killer or pond cleaner tubes are a really poor choice.
The closer to 410nm the faster things go.
UV is REALLY BAD for your eyes. Be warned. Dont stare at the UV for long as you will cause either welders flash or worse still catteracts.

Apart from that, good advice all.

The concentrations I'm using are too low and the reaction is not too strong, it doesn't seems to heat up much. The Peroxide has a concentration of around 3%, and the HCl is around 25%.

I tried to keep it before, but it quickly lost its etching power. I have an aquarium pump, so I'll try pumping some air into it as you said. Thanks for the tip! As foor the etch tank, I only etch 1 PCB a month, so I don't care about loosing some time by using a simple plastic tray.

I used 405nm LEDs, they work pretty well, exposing the photoresist in 30 seconds or less. My exposure box can be closed so little UV light scapes from it, it is relatively safe. I would make an instructable on how I made it, but the SD card on my phone got corrupted and I lost all the pictures of that ;_;

Also i forgot to mention: though covered before, instead of using film you can use glossy photo paper in a laser printer and iron the toner onto the board. Be aware this is poor at anything with close spacing and tends to require mending broken tracks. Great for heavy connections though.

I've been using the toner transfer method for a couple of years now and when I switched to photoresist film I instantly noticed the difference. Toner works fine for thick traces, but it often leaves lots of holes, specially if the toner cartridge isn't new. I would highly encourage to do the change, it's almost as cheap and easier to do with less defects.

good to know. i must confess im old school. i like good old fashioned exposed board methods. I do like the newer etchants though. ferric was foul.

Yeah, the reason I don't use Ferric Chloride is because it seems to stain everything it comes in contact with. The etchant I use can be neutralized with some sodium hydroxide and the copper precipitates and can be extracted.

I've even used the copper oxide precipitate as a wood stain, since it has a nice dark brown color.