We use a lot of solder stencils where I work (during the day). We usually buy stainless steel framed stencils for about $300 each. For prototyping we usually hand paste each pad. We have an semi-automated dispenser, but it is still tedious work. I see several places like Pololu selling low cost mylar solder stencils. I wondered if my Buildlog.net 2.x home made laser cutter could do it.
I researched a few blogs and pulled some information from Pololu. Pololu sells 3mil and 4mil mylar stencils and recommended 3mil for fine pitch work. I decided to buy the 3 mil mylar. I picked it up on my way home from McMaster Carr. It was a life time supply for about $15.
I found a bunch of old small SMT PCBs that I could play with. I got the top side paste mask gerber file for it. I imported the file into a Gerber tool called ViewMate from Pentalogix. This is a great program that I have be using for years. A partially disabled version is free. I have found that it has plenty of useful features.
Most PCB layout software has layers for the solder stencil. There are industry standards (IPC) for the size of these pads, but they are generally a little smaller than the pad. Often large pads, like thermal pads under big power ICs are divided into smaller windows or dots. This prevents excess solder from causing problems. With this in mind you probably need to shrink the pads even further to deal with the kerf of the laser.
ViewMate has a nice feature that allows you to shrink the apertures. Apertures are a somewhat archaic term from when artworks were done optically on film. They basically mean the shapes. To use this feature select the Setup…D Codes menus.
Select all the shapes in the list and select the Operations…Swell menus.
Enter a negative value to shrink the shapes.
I then printed 1:1 to a PDF.
ViewMate has a lot of export options, but most of them are not available in the free version. PDF is fine for what I needed to do. I then imported the PDF into Corel. I cleaned up a few extra lines and text in Corel and moved it over near the origin.
If you were always going to make your own stencils, you could probably skip a few of these steps by defining your stencils layers with the right values. Pololu actually shrinks it differently in X than Y for even better performance. Many CAD programs could print straight to PDF or other formats then.
Corel is a front end for my DSP laser software, so I was ready to try making the stencil. The PDF has vector information so you could cut it or engrave it. Everyone seems to recommend engraving, so I gave that a try.
I was not sure what to put the mylar on. I decided to hang it in the air. I taped inside a wooden frame. I tried different power levels and speeds and looked for my best result. I onlt tried about 3 settings combinations before I ran out room. I looked closely and they all looked pretty good. I think I got the best at 200mm/s and about 60% power. The power was not too much of an issue. It seemed better to cut it with more power than it need. It tended not to heat the surrounding area. The step over was 0.15mm. That probably could have been smaller for more accuracy. There was a slight smoky haze after cutting that I rinsed off with water.
I saw this post over at the Make Magazine Blog. This is an interesting method of containing the exhaust gas on a particularly nasty material. It basically puts the work in a box with an IR transparent window. The box has a fitting to suck out the fumes. It uses a polished silicon wafer as the laser window that was bought off eBay.
I probably would have put it in a Tupperware box, because the seal might be better and it is probably more resistant to the nasty fumes. I could have used this when I tried my vinyl record cutting.
I saw this while checking up on the progress of his selective laser sintering (SLS) project. Basically his idea was smaller power lasers might be able to cut through thicker material by “sawing” through it. The required power density might only be high enough at the exact focal point of the laser. Therefore, if you move the focal point through the material vertically, you might be able to cut it. He did it using a reciprocating motion, but you could also simply run the same outline at multiple levels, which is what us mill and router people have done for years.
The machine is actually a beautiful work of hacker art. He build it out of CD/DVD drives. He only gets about 1.5″ of travel on each axis, but that is all he needed.
Thomas Kumlehn of Pixel Partner sent me a nice email regarding the iPad Chair. He did a roughly similar Instructable to mine on an iPad 3-D viewer. This is another laser cut flat-pack iPad holder, but this holds the iPad in the perfect location for viewing stereoscopic 3D images and video. He uses a readily available lens the redirects your vision to over under pairs. The iPads portrait mode works very well for this.
I remember as a child, a friend had a vintage side by side stereo viewer with a bunch of black and white pictures for it. I remember how amazed I was by them.
He has files available for cutting on his web site.
I am working on a fun, geeky project and I needed a little stand for an iPad. I am not a real fan of the Apple company, but my wife got this though work and it will be perfect for my project. I love this flat-pack style of construction and I was inspired by my recent post on SketchChair. So I decided to make a little chair as my stand.
I started out by downloading the SketchChair software. I was able to create a lot of cool chairs, but I could never get exactly what I wanted. The software is a great concept, but it was not quite ready for what I needed. I switched over to my 3D CAD package (Pro/E). I stole a lot of the construction techniques in SketchChair, but modeled it in exact scale in Pro/E. The design only uses four distinct pieces: 2 Sides, 2 Center Pieces, 6 Slats and 1 rear leg brace.
I exported the the drawing of the pieces via DXF (DXF Is here) to my Vectric Aspire CAM program. Here I duplicated the parts to what I needed and used the nesting feature to fill the cardboard size I had. Normally I would restrain the amount of rotation allowed to keep a consistent grain direction, but this time I let it rotate any any angle (screen says 45deg, but I actually used 5deg). I wanted the finished product to look sort of cattywampus.
Everywhere I look these days I see a blog article about someone laser cutting old vinyl LPs. I think the first place I saw it was on the Make Magazine Blog. It seems like a great idea, especially if you choose an image that fits the record you are cutting. It also fits the recycle/upcycle trends.
I decided to give it a try. I had an old Zydeco LP on the Alligator records label. I found a simple alligator profile and decided to give it a try. I don’t remember the exact speed and power settings, but the material cut like butter with a nice clean edge. The only strange side effect was huge amounts of brown smoke. I had a decent assist nozzle, but my old evacuation blower was not the greatest.
There is a saying amongst DIY CNC router builders that goes something like this… “You only need to build your first router good enough to build you second one“. In my case that turned out to be true.
I built a wood, conduit and skate bearing Solsylva router. I painstakingly layed out the various pieces using calipers, t-squares and compasses. I cut them out using hand held tools like jigsaws and drills. It worked remarkably well, but every time I routed out a perfect CAD drawn piece, I always thought “Gee, I wish I had this thing when I built the router“.
It wasn’t too long before I built my bigger, better, more accurate router. I was able to use tougher materials, hold tighter tolerances and cut more exotic shapes. It works much better. A few months ago I finally pulled off all the good bits and Sawzall’ed the old one apart to get more room in the shop…a bitter sweet day.
Most of today’s designs develop inside a computer. Resolution and accuracy are infinite in this realm. We expect our fabrication machines to output similar accuracies, but how does one construct a machine with this accuracy with common (analog) tools.
Today’s open source machine are addressing this head on. There is a big push towards self replication. Struggle past the first one and the rest will be easy. It is not just an accidental bonus it is initial design requirement. It is a lot more work, but I think it builds the strong communities behind these projects that help insure their success.
Here are three examples of self replicating machines…
I am just amazed, and I am not easily amazed. I saw this SketchChair application at the Ponoko Blog. This is just plain amazing. I love Flat-pack and this is like Flat-pack crack. This program is a collaboration between Greg Saul and the JST ERATO Design UI Project in Tokyo. This is a Processing based program that allows anyone to simply sketch a chair and the program generates the parts required to build it in real time. The parts are ready to be cut on a CNC router or laser cutter.
You really need to watch the video. It is really fun to watch. The program even allows real world physics to be applied to the chair, to see how well it will stand up on it’s own and when a human sits in it.
Virtually all CAM systems allow you to target different CNC machines through the use of a Post Processors (post). Post Processors tailor the ouptut to the target machine. While most machines can handle G-Code, each machine has it’s own requirements. Many Post Processors are so flexible that you can even get them to output HPLG, DXF, and other vector based file formats.
Post Processor formats and rules vary between manufacturers, but the general procedure of creating a custom one is basically the same. I will detail how I created a basic Mach3 E1P1 style post processor for my VectricAspire CAM program. Aspire is way overkill for a laser, but this Post Processor will work for all the Vectric products including the more affordable CUT2D which is perfect for a laser. Most CAM vendors will produce a guide for the post processor format. I got the one for Vectric from their forum.
Most programs ship with many Post Processors. Start with one that is closest to what you want. In my case I picked a Mach2/3 Arcs (inch) post. This targets Mach3, and uses arcs for curves instead of multiple line segments. You should also create a simple CAM file to test. I created one with a 1″ x 2″ rectangle and a 1″ diameter circle. Both items are offset from home. Locate the items on a simple grid so it is easy to find the items in the resulting G-Code.