A few years ago, [Mitch Altman] from Noisebridge came up with the idea of a Hackerspace Passport. The idea behind it was not to hinder or monitor travels but to encourage visiting other hackerspaces. These passports can be purchased for just a few dollars or, in true open source fashion, be made with nothing more than a computer printer… the Hackerspace Passport design files are totally free and available here.

So next time you’re visiting a new hackerspace, bring your passport and get it stamped to document the trip…. and that brings us to the point of this post: The Stamp. At around $25, having a custom ink stamp made at an office supply store isn’t that much money, but buying a stamp is not as fun as making one! That is what we are going to do today; make a stamp… or more specifically, several stamps using different techniques. Then we’ll compare the performance of each method.

DESIGN

Stamp 3D Model

Inkscape solid HaD logo

Inkscape Lined HaD Logo

Printed Stamp Mold

Printed Stamp STL

Since this is Hackaday, we will be making a Hackaday Logo stamp. Back a couple years ago we ran a contest asking folks to make unique things with the Hackaday logo. To make it easy for the entrants, the Hackaday logo was made available in SVG format. We’ll start with that, since it is available, and make a minor change by adding some lettering, as most soon-to-be stamp makers will probably want letters on their stamps too. This is easily done in the FOSS vector graphic editor software: Inkscape.

The stamp size is important. A Hackerspace Passport page has room for 4 stamps up to 41 x 47mm and we’ll try to keep our stamp within those limits.

MATERIALS

As stated before, we are going to use several methods of making stamps and compare them. Certainly, there are countless ways to make a stamp and we will experiment with methods and tools that a hackerspace would typically have access to.

Lasered Line Stamp

Replication

Lasered Wood

Milled Solid Stamp

ABS printed stamp

Milled Line Stamp

Hobby Foam Stamp

Flexible PLA

Stamp Rubber

This can be bought at a hobby or craft store. It is specifically made for stamping and we’ll use it as a base line. We’ll be making 4 total stamps out of Stamp Rubber. The first two will be a standard Hackaday logo but each will be made with a different tool; Laser Cutter and a Mini CNC Router.

The 3rd and 4th stamps, also made from stamp rubber, will have ‘lined’ shapes. Large flat solid shapes have caused problems in the past by not transferring ink well, leaving an inconsistent splotchy pattern. Again, we’ll do one on the Laser Cutter and one on the CNC Router.

Hobby Foam

This material will be cut easily with a laser cutter. It is softer and more flexible than the rubber and comes in 2mm thick sheets. It is extremely inexpensive material available at any craft store.

Wood

This will be cut out using a laser cutter. Everyone has access to a small piece of wood, let’s see if it is a reasonable material for making a stamp.

3D Printed Plastic

The 3D print will be used as the stamp itself. Printed plastic is hard and slippery. Our past experience is that this method won’t look good but 3D printers are a very common tool found at hackerspaces, so let’s try it! Our stamp will printed with ABS.

3D Printed Flexible PLA

This 3D print will be printed in the same manner as the plastic stamp but the material is significantly more rubber-like and flexible. It will be interesting to see how well this material works out. This stamp will be printed using Form Futura Flexible PLA filament.

3D Printed Mold with Caulk Casting

3D Printers are a great tool but we expect that the 3D printed stamps may not perform well due to the material properties. For this test, a mold will be printed and filled with silicone caulk. The cured caulk will then be removed and hopefully result in a usable stamp. Like the 3D printed plastic stamp, this mold was also printed with ABS.

MAKING WITH A LASER CUTTER

That burnt look is on there for good

Burning Rubber!

lasered rubber, smells bad, looks good

Hobby foam. Where’d the letters go?

We started with the Hackaday Logo SVG file and modified it in Inkscape. It was then opened in Corel Draw for sending to the Laser Cutter, which in this case was an Epilog. Lucky for us the Epilog print driver has a stamp feature. After opening up the SVG file, verify the stamp is the correct size. Then draw a rectangle around the stamp geometry and make the line type ‘hairline’. The software’s stamp feature will need this rectangle to bound the relieved (lasered) portion of the stamp. To turn on the stamp feature, in the Corel Draw print dialog box click on ‘print properties’ to get to the Epilog driver dialog box. Click the ‘advanced’ tab to show the stamp options. Checking ‘mirror’ will reverse the image so the image on the physical stamp is reversed but the ink stamping will come out in the correct orientation. The stamp feature also will put what Epilog calls shoulders around the artwork, stiffening thin areas of the physical stamp. We used 30% speed and 100% power and had to do 3 passes to get a deep enough cut.

Physically, the lasered stamp rubber came out well but it also became burned. Even the non-lasered area became discolored from the fumes. Scrubbing with soap and water didn’t make any perceivable difference. This discoloration had no effect on the stamp’s performance.

We then moved on to the Hobby Foam which cut like butter. Typically that is a good thing but it also melted the foam and left a wide kerf, so wide in fact that the letters end up just being holes.

MAKING WITH A CNC ROUTER

all cuts are done

there’s a stamp under there somewhere

after the v-bit pass

In process!

LinuxCNC 0.060″ endmill tool path

LinuxCNC, 30 deg v-bit tool path

Starting with the SVG file from Inkscape, we need to save it in a format for use in a CAM (Computer Aided Manufacturing) Software which will take the image geometry, generate tool paths to cut out the stamp and save a g-code file. Saving as a DXF file resulted in some compatibility errors so we went with PDF and it worked great. Before exporting the file from Inkscape, the image should be reversed. Nothing is worse than cutting a nice stamp then finding out the stamping is backwards! The CAM software requires you to indicate what type of tool is being used as well as how deep to cut. We will be using a 30 degree pointed engraving bit and cutting 0.060 inches deep. The 30 degree bit has 2 benefits over a straight end mill:

1. The angle will provide some support to thin stamp features (similar to the Epilog’s shoulder feature).
2. The point of the bit can be used to make very fine details in the stamp.

We are using a program called VCarve to generate our g-code. It is an easy to use program but not cheap, starting at $350. A free CAM Software alternative is CAMBAM Free. Even though our opinion is that CAMBAM Free is the best free CAM Software available, it is reasonable to expect a long learning curve and many ‘net searches for how to use it.

Machine Control Software such as LinuxCNC or Mach3/4 will be able to load the g-code file created by the CAM Software and subsequently control the CNC Machine to cut the stamp. We are using LinuxCNC, which is free and open, for this project but Windows-only Mach3 ($175) is equally as functional.

The CNC Machine used for making the stamps is a generic unbranded machine typically referred to as a CNC3020. A search of that term will return a plethora of vendors on eBay selling the machine for about $600.

We only expected to do two cuts: remove most of the material with a 0.060 inch straight end mill and move in for the detail cuts with a 30 degree v-bit. After finishing the first two passes there was quite a few ‘rubber burs’ leftover. These first passes were done in a climb-milling direction. Two more programs were then created and run to do the same cuts but in a conventional-milling direction. This cleaned up a lot of the unwanted rubber but it still needed a good rubbing and scrubbing with a tooth brush to clean up the stamp to a satisfactory point. Watching the v-bit closely as it cut, the rubber was clearly moving out of the way as the v-bit passed by, then popping back to its original position. This behavior wasn’t expected and is the reason for some unclear features, such as the nose on the solid skull or the gap between the skull and wrenches.

MAKING WITH THE MOLD

Caulk didn’t work so well

impression material tube

mold filled up with 2 part impression material

The replication looks good

Using a 3D printed mold with silicone caulk did not work… at all. The mold printed great but the caulk did not cure in the center of the ‘stamp’ and it was not able to be removed in one piece. This method was a complete failure. It just happened that we did have some Impression Material kicking around. It’s not common to have this type of material so readily available but we did so we’re going to use it! The impression material is a two-part mixture that will cure regardless of how thick the application is. The impression material worked well filling most of the voids and was very easy to remove from the mold. One problem we noticed is that it took a very good replication of the top surface of the 3D printed mold, meaning the stamping part of the stamp had ridges from the textured 3D print top surface.

FINISHING UP THE STAMPS

The soft stamps were glued to a backer plate to prevent the stamp from flopping around in the breeze and to equalize force across the entire stamp while stamping. Anything can be used for a handle; we are mostly using some old wide-based drawer knobs and short pieces of small-diameter PVC pipe. These handles were hot glued to the backer plates.

RESULTS

The stampings are in and they certainly exhibit a wide range of quality. Click on the photo to check them out in detail.

All of the rubber stamps came out very well, as you would expect since this material is made for ink stamping. They all look good, are very clear and the letters are easily readable. Both the Laser Cutter and CNC Router produced a very usable stamp with the stamp rubber.

The hobby foam transferred ink well but the laser melted away fine details, thin sections and all of the lettering. Plus, each of the 5 pieces had to be glued onto a backer plate individually which made it difficult to align the pieces correctly.

The wood and printed hard ABS are the worst at transferring ink even though the geometry came out well. The flexible PLA worked much better and would be a great option for hackerspaces that do not have access to a CNC Router or Laser Cutter.

Don’t try molding with caulk. Even though this was a complete failure, we also learned something in the process: don’t do it. The impression material worked okay but the textured stamping surface and the fact that impression material is not common make this not the best choice for stamp fabrication.

When making ink stamps (or anything!) experimentation and trying substitute materials is totally reasonable. We hope this post has shown there isn’t one way to get where you’re going. Feel free to mix and match stamping materials, methods and techniques. Let us know your experiences down in the comments. If we missed a method that has worked well for you, let everyone know!

Got Stamps?

Handle Types

Hot off the press!