Box dust collector

I was experimenting with using a motor from a dishwasher to make a blower for another mini dust collector. The motor was more powerful than the furnace exhaust blower motors I used here, so I could build a bigger blower with more suction.

The first blower I built didn't use all the available power of the motor, so I built a slightly bigger one. I made a bigger impeller, using the same method as this one. But this time I designed the housing in CAD. That way I could print out templates for the spiral housing, which made making the housing much quicker if not counting design time.


The nice thing about cutting out the segments like this was that I could sand them one at a time on my belt sander. I don't have a spindle sander.


Spiral housing laid out.


Testing it, the motor consumed 222 watts with air free flowing. That's more than its rating, so it was overloaded. But if I constricted the air flow, the motor's power consumption dropped below its rating and RPM was above, so it was no longer overloaded.


I designed the spiral with a fairly generous exit path, but if I can't put that much air through it because of motor power limits, I might as well tighten up the spiral so it's closer to circular. This makes the blower slightly smaller.


So I just angled the segments together slightly differently. This change increased static suction from 170 mm to 175 mm (a more circular housing tends to produce slightly more static pressure).

The motor was still overloaded with free air flow, but if it has to pull air through a dust collector with hoses and filters, air flow will be more restricted, and hopefully the motor won't be overloaded.


Having decided on the housing geometry, I screwed it all together.


My initial idea had been to just mount the blower onto a cardboard box with a filter in it for a test. But at 175 mm static suction, I would have had to reinforce the cardboard box. And then there is the question of how to open the box to look inside and also seal it. And the box wasn't quite the right size for my filter. So I figured it would be simpler to just make a wooden box.


I used some scraps of plywood from the garbage, and two cabinet doors from a yard sale. Here I'm jigging things up for screwing the pieces together. (The door that's horizontal is just there to help jig things up temporarily)


Screwing the top plywood onto the doors. The plywood only overlaps the doors by 8 mm. Ideally, that plywood would have been wider, but this piece was just about the right size. In fact, cutting all the pieces to size for the box, I ended up with almost no scraps at all.

At right, screwing on what will be the "back" of the box.


The bottom of the box is some particle board from some kitchen cabinets.

I glued most of the joints, so I cut away the melamine coating along the edges where it joins with the kitchen cabinet doors.


Here applying a generous amount of glue before screwing on the bottom.

The glue's primary function here is to act as a sealant.


The filters I'm using are designed for use in a furnace. Furnace blowers produce about a tenth of the static suction as my blower, so I made a lattice to help support the filter against the pressure (for when the filter becomes very caked with dust).


A big 24x20" filter goes against the lattice.


I wanted two layers of filtering, with the first filter folded against the sides of the box. The first filters are actually the ones that will get most caked up, so they need support even more.

I sanded the ends of some pieces of wood round and pounded them into angled holes on bigger pieces of wood.


These pieces in turn form brackets that fit into the box, resting against the inside edge of the frame of the doors.


I cut a series of notches into a piece of wood, then cut that into two strips. These strips fit around the ends of the brackets I just put into the box, and screw into the sides to hold the brackets in place.


Then the two filters, which I kinked to make them fit. They are already dirty from some other experiments.

In these photos, the box is upside down from how it will be in the dust collector.


A hole cut into the top of the box allows the blower to suck air out.


With the blower sucking air out of the box, it pulled the lid on very tight.


I cut two holes into the lid for a window.

My initial idea was one tall window so I could better see how high the dust was in the box, but then decided to make it two square holes instead, so I wouldn't be weakening the lid as much.


I used duct tape to attach and seal the Plexiglas to the box.


I then cut a hole in the back for the dust inlet.

I didn't have an appropriate flange so I just put a 4" ducting elbow in the hole and screwed it on from the inside. I put some duct tape around it to help seal it.


Then installing the first filter. I'm using masking tape to seal into the box. I had some problems with filter blow-by on my mini dust collector, so I figured tape should solve that. I don't anticipate changing the filters that often, and if I do, the tape is a small cost compared to the filter.

This filter is more of a back-up for the other filters. Furnace filters, when not caked with dust, have very little air resistance, so it's not doing any harm.


Starting to install the main filters.


I taped these filters in as well. I had to make some pieces of cardboard to seal the end of the filters.


I wanted some sort of compressible seal for the main lid. Wood chips really stick to felt, so I figured I'd try some strips of corrugated cardboard. I'm not sure how that will hold up over time as the lid is removed from time to time.


The lid just screws onto the box with wood screws. The screw holes may wear out eventually from opening and closing the box, at which point I might switch to gluing some threaded rod into the box and using wing nuts or something like that.

But right now, I'm not sure how practical this thing is, and wood screws are quick and don't preclude using a better solution later.


Tipping the box onto a dolly. With the hardwood side panels and particle board bottom, it's not very light.


Testing it again. With this set-up, it's 222 watts, so the motor is still overloaded, but putting just a hose adapter on the elbow cuts air flow down enough to put the motor within its rated load.


I figured a rather overloading test for this set-up was to hook it up to my Dewalt 735 thickness planer. The blower on this planer blows so much air out that it overwhelms a lot of dust collectors.

I hooked it up with some cheap dryer hose. Not the best hose for this job, but it has the advantage that you can see if the hose has positive or negative pressure in it.


The hose expanded. This means positive pressure. The planer was pushing air into the dust collector.

But as soon as I ran a board through the planer, it slowed it down enough that its internal blower put out less air, and the hose started shrinking again. So while planing, the dust collector was barely keeping up.

The planer also didn't drop a lot of chips on the table. When there isn't enough air flow, a lot of shavings end up falling onto the table.


Looking in the window, while planing.

I checked the air with my Dylos particle counter. It showed that some of the very fine dust was making it through the filters. I expect this to improve as the filters become caked with dust (a layer of fine dust is an excellent filter). Though once it's caked with dust, air flow will be reduced and might not be enough for the planer.

At any rate, I think this is a useful dust collector. And with a motor that is at most 1/8 hp, it's doing pretty good. I would say, performance is very comparable to my cheap dust collector. Static suction is about the same, and air flow on that unit is not that great now that the filter bag is caked with dust.

I may end up making a mount for a thickness planer to go on top of this box, probably my cheap one (this one), which doesn't need quite as much air flow.


Two weeks later, particle counts when using it were down substantially, so the filters were sufficiently caked with dust. The sawdust itself actually helps to filter (this is called "Seasoning" the filter).

Power consumption also dropped to 1.74 amperes (or about 200 watts), which tells me air flow is down a little from extra air resistance in the filter.

It still works ok with my thickness planer and also sucks all the shavings out of my jointer. So overall, I'll call this experiment a success. Though if I was building it again, I'd try to find a slightly bigger motor. If you want to build one, use a filter with wide pleats. Tightly spaced pleats end up getting the space between the pleats filled with dust, causing unnecessary air resistance.


See also:


Back to my Woodworking website