SO3 Dust Collection System - a work in progress

Moderator… feel free to move this if you feel it would be more appropriate inside another thread or in another forum…

I’ve had people ask to see more details about my dust shoe prototype(s). I’ve tried several designs and found that they do well for certain types of milling and not so well for others. So, I still have more work to do. Here’s what I have so far. I’ll update this thread as I progress…

Big picture… the dust shoe is currently fixed to the stationary portion of the Z gantry (I.E. it does not move up and down with the spindle). That seems to work better for 90+% of the milling that I do. However, I designed it so that it can be secured to the stationary part or the moving part of the gantry, so it can be used either way (with a couple more mods)

The dust is pulled into a cyclone separator then goes to a shop vac with a good filter.

The hose is from of a Bissell (yard sale) household vacuum cleaner. I like this hose because it has a double helix spring steel winding that allows it to stretch the full travel of the SO3 XXL without putting much tension on the supports. That way the Machine can move its full extents without the hose binding or getting caught under the rails even though it does’t have any supports between the two ends.

The shoe is a 1/4" polycarbonate plate milled out to accept the brush material that I bought from
Part number 7900T1 $2.28 per foot.

The hose connection is a 3D printed part that I solvent bonded to the Polycarbonate plate.

I added a cardboard plate (not shown) to channel the air closer to the cutter. That helped quite a bit. I’m going to make a 3D printed one soon.

Right now the shoe is held by one 1/4" steel rod that is clamped to a custom aluminum plate using a delrin block and two screws. The custom plate replaces the Z limit switch bracket plate. The plate doesn’t move with the router. If I want the shoe to move with the router I’ll add a Delrin clamp block to the router mounting ring so I can clamp the rod relative to that ring and it will just slide in the Delrin block I currently use to clamp the rod.

I had to bend the aluminum mounting plate to clear a boss on the moving plate that the router moves with. I just bent it in a vice. This all happened pretty quickly so its all very rough and prototypish… Don’t judge me :wink:

The 1/4" steel rod is screwed to a Polycarbonate block that is solvent bonded to the dust shoe plate.

I elevate the hose so it doesn’t drag when the gantry moves close to this corner. The red bracket is a 3D printed part and I’m using a 1/4" steel rod here also. The bottom black bracket is also 3D printed and is pressed into the enclosure floor. There’s another adapter on the bottom side that the shop vac plugs into. Its screwed to the bottom of the enclosure. That hose leads into the top of the cyclone separator (99.9+% efficient at removing dust so the shop vac never fills up and the filter lasts much longer between cleanings. I have another thread that shows more about that design.).

I don’t mind posting the .STL files if anyone wants them. Just let me know. It may be better to wait until I get to the final design though. Just post here if you want me to add them.

I’ll update this thread as the design progresses. Feel free to post constructive criticism or your design ideas and implementations. I don’t have any special love for this design and its still very crude. I find its better to just make something than to try to make a perfect design in CAD first. Its always better to make a crappy version then iterate based on lessons learned. This is about the 4th round of this design and its still has a long way to go.


Looks good, I am pondering how to retrofit my dust boot when my XXL expansion pack arrives. How rigid is your fixed design while milling?

I want a fixed height option since I get deflection on the Z using the 1" brush you are referencing when using 1/8" shank bits when the boot is mounted to the router.

I like to see what is being machined so I am thinking I may mount the hose/pickup tube on the back of the Z carriage now that the controller will be relocated off of the X axis. This would keep the front of the dust boot unobstructed and may help with counterbalancing the weight from the router on the opposite side of the X axis. Overall design still under consideration.

I like the idea of mounting the hose in the back. I would have done that but I made my enclosure a little too short.

I get too much deflection at the base of the boot with just 1 steel guide rod and a 1/8" aluminum top mounting plate. My next design will have two mounting features.

Do you have a fan or a vent to let in air?

No. I’m only drawing 90 CFM and the door of my enclosure is not sealed so it allows plenty of air in through the joints.

If you use a 600-1000 CFM dust collector and a big hose then you would need an inlet vent to allow the flow rate to remain high. That would be way overkill for this system though. Without an enclosure a high flow system makes more sense because you don’t want any dust to escape into the ambient air. With an enclosure you are always drawing in fresh ambient air so lower flow rates work just fine while keeping ambient air uncontaminated. I’ve tested the air inside and outside the enclosure while the system is running and most of the time air inside the enclosure remains very close to ambient particle counts. Occasionally it goes WAY up if I’m losing dust over the edge of a tall board or something like that, but none of the dust ever makes it out of the enclosure. I usually let the vacuum run a minute or so after a job is over to make sure I’m not breathing a concentrated cloud of dust when I open the door.

Most full size shop vacs will pull 120-180 CFM with no air restrictions. That drops quickly as you start running long, corrugated hoses, smaller diameter hoses and a cyclone. However 90 CFM is more than enough even when I’m tramming my spoilboard which uses near the full power of the router and creates a LOT of dust to vacuum.

1 Like

I attached a big dust collector with 4" hose to my enclosure and have been trying to figure out the best size/location/type of intake vent to use. Mine is apparently sealed pretty well because with the dust collector running, I can’t open the door because of the vacuum created. I don’t want it to implode, so I haven’t been running it closed until I figure out the vent.

I appreciate the info. Do you have a cabinet for the shop vac as well?

I would suggest throttling the vacuum until its just enough to do the job. That will save power and you won’t have to add more parts. Your blower draws considerably less power when the air is restricted. In fact, most blowers are not designed to run without at least some restriction or the motors can draw too many amps and burn up. Some are like that and some aren’t. Most dust collectors are designed to have at least 10’ of hose attached to restrict flow some.

If you want the high flow then how about just putting a thin plastic flap on the inside covering a 1 or 2" diam hole? The flap would open when the negative pressure got high enough and it would remain closed otherwise. Or add a baffle chamber so the noise level doesn’t rise because of the extra hole. They sell mufflers to quiet the exhaust noise for large house vacuums. They are basically just a plastic box or tube with a open cell foam core. A piece of PVC pipe with a bend or two in it and some 1/2 open cell foam lining the ID would prevent most of the high frequency noise from escaping.

Not yet but I’m working on making my own blower. Right now the shop vac is much louder than the router and its really annoying after just a few minutes so it has to go.

Round one of the new design will just be a yard sale vacuum motor in a MDF box lined with drywall attached using acoustical caulk, drawing through a shop vac filter from the top of my cyclone. I can get that pretty quiet and it will be relatively easy to make.

A little at a time I’m working through the design of a vacuum system that will be quiet and small that will be able to run thousands of hours instead of hundreds of hours like a shop vac system. It will have a higher pressure drop than a dust collector but more flow than a shop vac when turned all the way up. That’s at least a couple months out though. I hope to have Round One mentioned above knocked out in a couple weeks or less.

Yeah that’s why I asked lol
You should carve the blades for the blower fan out of wood with your shapeoko! :smiley:

You mean like this? :slight_smile: Nomad: Custom designed blower impeller

So if you did it again, you’d mount the shoe and the hose oriented to the back of the enclosure?

Since I’m still in enclosure design stage, I think this works best for me.

Good question. I like the idea of everything being tucked out of the way. I do most of my parts near the front. The hose tension is very low there. The hose tension would be highest there if the hose exited the rear of the enclosure. If the hose is right that probably wouldn’t be a problem though.

The only downside I can think of is that it would be harder to remove the hose to sweep the enclosure after a job. I could see that being an issue but maybe not significant.

On my SO3 w/o an enclosure, I have my Cyclone on the floor to the right of the workbench, when I need to clean the area I disconnect the DC hose to the dust boot at the cyclone and hook up my other hose. I will be getting a 2 1/2" “Y” fitting and blast gates soon, I am sure there is a better way to do what I am doing, but that’s what I’m used to in the shop.

Could a manifold be made so that when you close the dust boot hose, it opens the port for the other hose and vice versa?

Here’s a concept that would work as a splitter and it could be made using your SO3. The green part could be made out of wood and sandwiched in a housing made of wood. You could use felt seals to prevent leaks. It wouldn’t be air tight but it doesn’t need to be for dust collection as long as your leaks are small enough to be insignificant compared to the overall flow rate.

Just rotate the green part to switch between Inlet 1 and inlet 2. A pin in a slot could keep it from rotating too far either direction.