SA-4 Spark Arrester for Vacuum Systems

 SA-4 Spark Arrester

Warning! This information is presented for educational purposes only.  The publisher assumes no responsibility for the usage of this information. There are many risks that must be assessed. It is necessary to know and understand these risks before experimenting with a spark arrester in a dust collection system. 
Working with metals, woods and synthetic materials poses a big problem for knifemakers. Dust, dust and more dust. Everything we do seems to makes dust! This year I set out to build a dust collection system to replace my two venerable shop vacs. When designing the dust collection system, it became apparent that sawdust and sparks are probably not a good thing to be shooting into a cloth bag at 40 miles per hour. 

I quickly came to the realization that there was no handy way to catch the sparks that entered the vacuum system. Granted, there are some DIY solutions involving metal buckets, hoses, baffles and even water to catch sparks. By all accounts these seem to fit the bill. However, there are more frightening accounts of shop vacs catching on fire, melted hoses and so on. 

But, wouldn't it be cool to have an in-line spark arrester like the big boy's shops?

After some study, a simple concept came up. Pull the particles through a static impeller to creating a spinning (centrifugal) effect and force them to the outside. This coupled with an increase in area and correlating pressure drop should stop the sparks and contain them in the arrester body. Bernoulli's Principle at work. We've seen similar effects in conical dust separators and within Dyson style vacuum cleaners.

The device needs to be fairly easy to fabricate with basic tools, be made from easy to get parts and be able to catch close to 100% of the sparks. To measure the effectiveness of the design, I made the prototype labeled SA-4 (Spark Arrester, 4") as I have 4" diameter hoses on my 600 CFM dust collection system. It is important to note that higher CFM systems may require larger components. The SA-4 is intended for use on 4" dust collection systems in the 600 CFM range. These are generally sold as 1 horsepower units. 

Massive amounts of sparks going it. None coming out!


1 - Steel oven pan (Dollar Store)

1 - 6" to 3" duct reducer (Home Depot)

1 - 6" duct connector (Home Depot)

1 - 6" duct cap (Home Depot)

1 -  4" flanged dust port (Amazon)

4 -  #10 x 3/4" sheet metal screw

1 - 10-24 threaded rod, approximately 5" long. (Home Depot)

3 -  10-24 nuts.  (Home Depot)

3 - 1/8" rivets.

1 - Short length of 3/16" ID tubing (I used a stainless steel drinking straw. Dollar store)

21" of 1" wide strip light gauge steel (for sealing body) 

1 - #10-24  x 1" machine screw.

21" of 1" x 1/8" closed cell foam, weather stripping etc. For seal around body.

Miscellaneous: Foil tape, fibreglass automotive body filler, high-heat paint, silicone.

Links for items here:


Drill, with bit set
Permanent marker
Aviation snips
Rivet tool
Vise-grip pliers
Leather gloves
Hacksaw & vise
A few files for rough edges


What we're going to do is fabricate a cone, put some vanes on it and make what I'm calling a static impeller. Then we'll house the static impeller in some ducting and make it usable in a dust collection system.

Making the cone

A cone with a height of 6" and a diameter at the base of 4" is needed. I used a piece of stiff paper to form the basic shape then taped that to the oven pan as a template. 

Make sure that you have about 3/8" overlap on the seam. I cut about 1/2" off the top of the template so I could fit my finger in when making the shape in paper. So technically, this is a conical frustum, but that's way too nerdy. This hole at the top will be patched over later with foil tape. 

Mark and cut carefully with snips. 

While wearing leather gloves, gently bend the shape into a cone. 

Using the vise-grip, hold the seam when the diameter of the base is 4". This diameter is important as you'll see later. Drill a 1/8" hole. Rivet in place. Add a couple more rivets along the seam. Make the top of the cone rivet about 1" down from the top. 

You should now have a basic cone shape. 

Mark the base every 90°. It doesn't matter where you start. 

Making the vanes

The vanes (fin-like things) direct the air moving over the cone into a vortex. I prototyped these on paper to get the arc close. Mine weren't perfect, but they get the job done. 

The arc that I used had a radius of 3.75"

If you placed the cone on the oven pan conservatively, you should be able to get 4 vanes from the remains. The following downloadble PDF template can be used to make the vanes.

I didn't have the luxury of fancy template cut-outs, so I winged the shape with an arc between 3.75" and 4.75". These "sort of" fit the surface of the cone, but not as well as I hoped. I patched the gaps with foil tape.  

It's imperative that the vanes stick out 1" from the cone base. At the base, two vanes and the diameter of the code must be very close to 6". The vanes are about to be about the height of the cone when they are formed onto the cone. Make them about 7" long to start and trim them in place. 

Attaching the vanes

The vanes are placed every 90° around the base of the cone. I used foil tape for this. 

Start the vane any mark on the bottom and give it a 90º twist as you bring it to the top of the cone. Apply small pieces of foil tape to hold the vane in place as you go. Repeat this three more times. From both the top and the base of the cone should have four equally spaced vanes at 90° to each other. 

Inside the tip of the cone I epoxied in a small wooden cone. This gives the tip more strength so we'll be able to secure it later. Make a small cone from wood about 1" long and epoxy it inside. 

Form the top of the cone into a little dome with foil tape. This will allow the air to flow easily over it. Drill a 3/16" hole through the cone about 1/2" down from the top. This hole needs to be parallel with the base of the cone so that the impeller will hang squarely. 

I used some automotive fiberglass body filler to reinforce the vanes and to add roughness to the surfaces to create turbulence in the airflow. Put some wire through the hole for hanging the impeller when working on it. I don't know how much the body filler affects the finished product, but it really solidified the vanes to the cone. 

If you wanted to paint the separator after the resin has set you can. I choose red high-heat paint. 

Spark arrester body

The body consists of a few different pieces of ducting accessories and a dust port. Fortunately, these are generally easy to find. All the ducting parts were in stock at my local Home Depot. The plastic dust port I bought from Amazon.  

Instead of buying a length of 4" duct, some cans like peanut cans are 4" and can be repurposed.

Use the dust port to mark out a 4" hole on the 6" duct cap. Drill a couple of holes and work around the hole with snips. If it's tough to get close to the edge with the snips, the metal can be segmented and bent back to the line with pliers. 

Cut the 4" duct section down to 2.5" and fit into the hole. Press it through the hole until it flush with the rim of the cap. Wearing gloves is a good idea here. The edges can be very sharp. Once in, seal the joint with foil tape. 

Place the dust port over the duct that's sticking out of the smooth side of the cap. Run a small bead of silicone on the flange and mount the dust port to the cap with the flange mounting holes. I used some #10 sheet metal screws. 

This side of the body is pretty much done. If you want to go around any joints with foil tape, this will help seal the duct joints. They are for HVAC and not very well sealed for vacuum use. 

Place the separator into the reducer. Mark and drill two 3/16" holes to mount the separator. 

Once the fit is good and the cone of the separator is as dead center as you can get it, measure and cut two pieces of 3/16" tubing. These will be used as spacers. 

With the spacers in place. Now put the nuts on and secure the impeller. It shouldn't deform the circular shape of the 4" side of the reducer. 

The 6" duct connector should be able to slide into the 6" side of the reducer. The vanes should go right to the connector. 

You can seal the joint at the reducer and the connector with foil tape. In the future we'll disassemble with by sliding the cap off the connector, so we don't seal that joint. 

The cap with dust port can now be slid on to 6" connector. Slide the cap  until the gap is 1" between the bottom of the impeller cone and the 4" duct. Why 1" you ask? The area of a 4" duct is pi times radius (2) squared or 12.57". The when the gap is set to 1" the area of the gap is also 12.57". This matches the area of the 4" duct. Too small of a gap it will restrict the  airflow.

I put some foil tape around the step of the connector to better fit the cap on. 

The last step involve making a removable band that can be used to squeeze the cap rim on to the connector. This will need to be removable for cleaning out the unit. 

For the band I used some 16 gauge sheet metal. Any kind of banding that is about 1" wide should work  too. This band needs to seal the joint between the cap and the connector, so some kind of rubber or weather stripping is recommended. I couldn't find any neoprene or closed cell foam handy, so I used some toolbox drawer liner which is a grippy kind of rubbery stuff. Ideally some adhesive weather strip, about 1" x 1/8" would be best. 

A screw and nut us used to hold the band on and tighten it up creating a seal and fixing the cap in place. Remember to check the 1" gap around the base of the impeller cone before tightening.


Your SA-4 spark arrester is done! Label it as you see fit. 

Note: I recommend at least 1 yard/metre of hose between the tool side boot (scoop) and the spark arrester. If you use clear hose like this shown below, you can monitor the effectiveness of the spark arrester. 

My testing bed. Video can be seen on YouTube.

Future tweaks

I am looking into a trap that can be easily cleaned to remove the filings out of the body. Perhaps a small transparent jar or a vacuum line and a small blast gate to use a shop-vac to clean the metals caught in the arrester body. 

I can see no reason why the same principles couldn't be scaled up or down for different systems. Larger vacuum systems could have more vanes to offset the increased air velocity. Perhaps a smaller version specifically for shop vacs could be made as well. 

If you make this device, I hope that it works well for you. If you have any issues, let me know. 



UPDATE 2020-08-28: I used an anemometer to measure the change in air speed with and without the arrester on my vacuum system.

There appears to be approximately a 15% drop in velocity with the SA-4 connected into the system. This was taken at the end of about 25 feet of 4" hose. 


unidisco said...

Exactly what I was looking for. Thanks so much for the thorough tutorial and sharing your knowledge/discoveries. Much appreciated.

Silver Forge Studio said...

Hey DC- thanks for putting this out there- I found you on the Youtube first- So my question is what do you have for a dust collector system? 2hp?

-Interested in your CFM and static pressure measurements as I want to combine this into a 2 stage system with a thein and cyclonic as well

D. Comeau said...

Mine is a 1 hp, 600 CFM, 8" static pressure and 4" inlet.