This is useful tip for anyone making or using surface mount components on PCBs. 

It relies on access to a laser cutter, so if you don't have one, look away now.....

I cut a plastic stencil on the laser, align it with the PCB pads and then squeegee a thin layer of solder paste through the holes. 

Peeling away the stencil will leave a perfect pattern of pads with the correct amount of solder on each one. 

I have just successfully used this process on components with a 0.025" pitch (0.65mm) - that's VERY small,

The  prototype in the pictures has been made by my good friends at Spirit Circuits in the UK, and is a full plated through, double sided board. with immersion silver coating.

Step 1: Prepare the Artwork

ALL PCB design programs these days will let you produce solder masks for surface mount boards. We take the solder mask output from the program, and direct it to plain, simple HPGL "plotter" files. 

Here, I'm using Number One Systems "Easy-PC", which may look very familiar to anyone using DesignSpark's free offering. 

Turn OFF all the other layers, turn ON the mask layer, change the output device to "plotter" and plot to file. I mirrored the mask, but I am not sure its important. Technically, the holes benefit the process if they are smaller on the TOP of the stencil than the BOTTOM. This ensures a clean peel at the later stages
Uncanny, I'm just about to do something similar myself once the mylar film I ordered turns up. (You seem to be UK based too, so if you're interested this is what I found: http://www.stencilwarehouse.com/acatalog/75-Micron-Polyester-Film--Mylar--1581Q.html)<br> <br> A couple of comments for you. Firstly, there is usually a difference between the soldermask layer and the paste layer - vias for instance. In Eagle you should be using tCream/bCream rather than tStop/bStop layer. Not sure if Easy-PC has the same thing.<br> <br> Secondly, I was hoping I could use my existing transparencies rather than buying mylar, but they totally failed the flame test. Please check yours to make sure you're not producing poisonous Chlorine fumes whilst lasering. http://www.nycresistor.com/2008/08/28/how-to-identify-polymers-with-burnination/<br> <br> My laser uses gcode so I was hoping to vector cut rather than raster etch. I'll see how I go but I'll bear your results in mind.<br> <br> Nice instructable by the way.
<p>Hello Fred, I've just finished building a laser cutting system (well almost - still need to put in the extremely important extraction) which uses G code. I saw your information about mylar, very useful. I was just wondering if you could share some information about your system (laser power and colour, spot size, cutting/etching speed) and if you managed to get decent results?</p><p>BTW, great instructable Steve, a good read.</p><p> Thanks,</p><p> Tim</p>
Thank you. <br> <br>The solder mask layer would work, but wouldn't look as pretty for sure. EPC actually says &quot;Top copper (paste)&quot; and omits the vias and pads - as you see from the tick box screenshot in the Instructable. EPC spots the difference between a through pad and a paste pad automatically <br> <br>Its not the coding that matters, its the feature size, and the length of time the beam dwells, from my direct observation. Fast scannning ablates the film without causing distortion in the features. I tried piddling around with cut parameters for ages before realising it. <br> <br>You may rest assured, I don't need to be told to identify my plastics before cutting them....These are premium grade PET/mylar
hey guys. im new here but it seems like this side is authum. im sorry if my english is worse. but i had to write you! this project u made is exactly what im doing now. i study electonical engerniering and got all exams. now im starting my master thesis. the theme is &quot;laserstcrutering of metallic and non-metallic materials&quot;. and im doing it with the diodes laser of the university. it can &quot;cut&quot; all materials till 150&micro;m in fact of corse high grade steel. in german company&acute;s they mostley use this so its heavy to find some options of other materials. another problem is i don&acute;t need 100 of them its more about different materials then examine one. the work does also inclueed creating a testlayout and bring the soldering on it etc. testing the structure later with an electron microscop. <br> <br> <br>so ur part is a very helpfull come in. i got now an idea what to do. next month im starting but i allready searching the whole day in the internet for tips how to create a good layout ( i need one perfect to use it on all different materials) and for different materials. im completely without any experience in this work so im very thankfull for any kind of ur advices. maybe u know a nice company where i can order some non-lasered stencils to work with them for low money^^? thanks al ot for ur work :)
wow its awesome.............but geeky way is to grab jwellers lens (optional)and use insulin injection for solder paste wouldnt even cost a dollar
That's how I used to do it. The pins on one of the devices I needed to do this for are on 0.25mm pitch.
very interesting!
Also, your boards look nice. I can thoroughly recommend this Instructable if you'd like a solder mask. I tried it and it works brilliantly. <br>https://www.instructables.com/id/Dry-Film-Solder-Mask/
This is one made by Spirit's very fast very cheap Go-Naked service - you don't get masks with it. The full production boards will have the works, obviously.
I'm going to build a laser cutter just to do this
SWEET! Gonna try this on my next SMD board!

About This Instructable



Bio: I'm an Engineer, who originally inherited the family business (Thanks Dad (RIP JC Taylor, 1938-2011)) after working in it for 25 years, designing and ... More »
More by steveastrouk:Pi-ece on Earth How to burn sawdust cleanly Building up a flight case 
Add instructable to: