Tag Archives: cyclone

Rebuilding my cyclone – part 3

If you missed part 2, please find it here.

My new cyclone is finished

I’ve finally finished building my new cyclone. In part 2 I tested a new cyclone but it seemed that the air flow was too big, so I decided to build another cyclone – a double cyclone so the air flow would be around half for each cyclone. And here it is:

If you would like to build one yourself, you can find the OpenSCAD file and the stl files for the top here. You’ll need to make two copies of the “inlet full” stl file to have a complete top.

Rebuilding my cyclone – part 2

If you missed the first post, please find it here.

Assembling my cyclone

After milling and gluing the parts for my new cyclone I’m ready for assembling it. The plastic bottle is cut to the right length and I’ve milled a block that will fit both the bottleneck and the lid of a jar I’m planning to use for collecting the dust.

Parts for my new cyclone.

Parts for my new cyclone.

First I install the pipes in the top part of the cyclone. I’ve taken extra care to make every hole a snug fit so no glue is required. Both the vortex finder and the inlet pipe are 32 mm plastic pipes.

Top of cyclone with pipes installed.

Top of cyclone with pipes installed.

As you see in the photo there is a recess for the bottle itself (bottle not installed yet). Again it’s a precise fit so no glue is required.

Test run

Here is a short video of my test run:

I did some experimenting with the vortex finder length and the air flow while testing. Difficulties emptying the cone seemed to grow while testing – probably due to the seal between the bottleneck and the jar getting worse as I took it apart several times.

Furthermore, a lower air flow seemed to improve emptying the cone when no dust was added. I have no idea how much dust the vacuum cleaner actually took during the test but it seemed to be a lot. In the video it is easy to see a column of dust in the center of the cone indicating that the vacuum cleaner at least took some of the finer dust, which wasn’t supposed to happen…

I think I’ll make another test with two cyclones in parallel to keep the overall air flow up and to make the combined cyclone more effective. Back to the drawing board… 🙂

Part 3.

Another rebuild for my ShapeOko 2

Rebuilding the whole thing

As you might have seen in this post, my grbl shield is broken and I’m in the process of replacing it with a gaups shield from ASL. When my new shield is assembled it is my plan to build some sort of casing for it (mostly for protection). Furthermore, I’ve decided to make my ShapeOko 2 larger so I’ve ordered some new parts for it (longer MakerSlides among other things). I then realized that my dust collecting cyclone (made by a friend of mine) has to be remade as well to support the larger ShapeOko 2 in my workshop and to get it out of the way, when I use the ShapeOko. Any project has a tendency to grow on it’s own… 🙂

Cyclone

My old cyclone has been working very well so I’m not going to discard the basic idea: My friend who is a chemical engineer discovered that an empty bottle from a particular brand of soft drink has a cone shape, that fits the theory behind a cyclone very well (actually nearly perfect). Furthermore, it’s my plan to mount it on the wall and use pipes to get the dust collection where I need it, so I went looking for pipes that will fit the hose on my vacuum cleaner. As it turned out a 32 mm plastic pipe fits perfectly.

Bottle and plastic pipe.

Bottle and plastic pipe.

The theory of cyclones

I’m not an expert in cyclones so I went looking for information about cyclones on the Internet and Thayer School of Engineering at Dartmouth turned out to have a pdf from a course on their homepage, that explains everything very well. Especially the drawing below was helpful:

Standard cyclone dimensions

Standard cyclone dimensions.

Calculations

Since the cyclone diameter D, the dust outlet diameter Dd, the gas outlet diameter De and the cone length Lc all are defined either by my bottle or by the vacuum cleaner hose, the remaining calculations are easy. The values measured on my bottle are:

  • D = 75 mm
  • Dd = 26 mm
  • Dc = 29 mm
  • Lc = 160 mm

This gives the following values:

  • D/D = 1
  • De /D = 0,39
  • Dd /D = 0,35
  • Lc /D = 2,13

These values lies somewhere between (2) and (3) in the standard cyclone dimensions, so I decided to aim at a cyclone with dimensions as an average of the values of (2) and (3). I know it won’t be perfect but hopefully it will work just fine. My calculations are:

Cyclone calculations

Cyclone calculations.

Drawing in OpenSCAD

Then it’s time to make a drawing in OpenSCAD (I love that program). The drawing below is what will become the inside of my new cyclone. Here is the OpenSCAD file: Cyclone and the STL files.

It looks like this:

Cyclone in OpenSCAD

Cyclone in OpenSCAD.

The cone itself and part of the body is made out of my bottle and the top will be milled in wood on my ShapeOko. The top will be four slices 20 mm thick: One with a 32 mm hole for the vacuum cleaner hose, one with a 80 mm (the outer diameter of my bottle) hole holding the bottle and two pieces that glued together will form the inlet. The last two parts look like this:

 

Inlet top part.

Inlet top part (upside down).

Inlet bottom part.

Inlet bottom part.

Milling the parts

I used ArtCAM Express to generate the g-code for milling the parts. The finished result is here:

The four pieces for the cyclone inlet part.

The four pieces for the cyclone inlet part.

Stacked on top of each other.

Stacked on top of each other.

Notice that the inner diameter of the cyclone is 75 mm and the fourth piece has an 80 mm recess for holding the bottle. Furthermore, my calculations above suggested that the value of Lb should be 128 mm but the height of the inlet is about 40 mm, so I cut my bottle at 88 mm from the top of the cone in order to make the entire height 128 mm.

More will follow… 🙂

Part 2.