Building a Spray Booth

 

I am fortunate in that I have a basement area to work on models.  While this gives me real estate to set up my workbench, there is one serious flaw: this (though large) enclosed area has no airflow.  Any time I wanted to spray primer or any paint from a can I had to don a respirator mask, spray and then leave the basement.  If I did some serious spraying, such as with a primer, I would stink the whole house up—the furnace and A/C are also in the basement, and invariably sucked up the smell.  I have wanted to build a spray booth for some time, and finally got off my duff to do it.

 

A WORD OR TWO OF CAUTION:

 

I cannot guarantee my solution meets safety standards.  It is loosely based on U.S. OSHA standards, but I make no claims that it is safe.  It passes a common sense check, which means it may still be unsafe.  Any electrical wiring should be done by someone qualified to do so.  This page is not intended to be used as instructions; to do so is at your own risk.

 

Design Choices

 

 

The biggest hurdles I had were where to exhaust the fumes and what to use for a fan.  My choices are based upon the situation, so I’ll take a minute to describe the basement.  In my basement, there are two windows that open to the outside, but below ground level.  The basement windows are in casement wells a couple feet deep, and covered over with plastic rain guards.  To run ducting outside, I either had to crack the windows open and run the duct out, or completely remove the window and permanently mount the duct in place.  With my house, the option of cutting glass and mounting the exhaust duct through the window was not an option.  I suspected that the crack-a-window solution would result in filling the window wells with toxic fumes that would simply spill back into the room.  I ended up replacing the window with a heavy plywood sheet I could cut up and mount the ducting to.  The tight fitting sheet, sealed with weather-stripping around the edges, gave me an airtight barrier between the room and exhaust fumes.  With the “where” solved, I went on to “how”.

 

I have looked at a number of fans as possible solutions.  I planned on building a sheet wood spraybooth to enclose the fan, making a sort of “wind tunnel” to suck fumes out the back and through the duct.  As an engineer and former fluid mechanics instructor at a 4-year college, I knew a house fan under 18 inches just wouldn’t move enough air, and was a poor choice.   Even larger house fans (18 inch and up) would not perform well either—they are designed for free flow, not ducted flow behind obstructions.  Since I was planning to enclose the fan, a larger fan meant a larger enclosure, which meant a bigger hassle to make.  How did I establish how much airflow was enough?  Well, this is actually the first question you gotta ask.  Which means the zero’th question is “How big a spraybooth do I need”?  The answer for me was in another question:  What exactly does a spraybooth do?  It catches the overspray of paint, sucks it out of the ambient air and vents it.  Therefore, a spray booth only really needs to be big enough to catch the spray pattern of the largest thing you spray—usually a large spray can, and contain the fan/blower motor.  That is the minimum requirement; anything else is gravy.  I added the requirement that the booth should be big enough to park sprayed pieces in so the can be vented as they dry (again, keep the fumes down).   I do not reasonably expect to put a 1/72 B-52 or 1/48 B-1b in the booth (the latter would require a booth over 3 ft wide…), but I would like to fit in subassemblies or whole smaller kits.  The fan solution was one I kind of stumbled on at Home Depot.  I ran across the oven/range hoods used to vent cooking fumes, and encountered a $69.99 model that has an unducted max flow rate of 192 cfm, and also a light built in.  I have read of other folks using the range hood solution, and thought it made sense.  The range hood is designed to operate in an environment filled with flammable particulates (cooking grease fumes in the air), and has a powerful enough motor to work when ducted.   A recommendation based on OSHA guidance is that a vent fan should exhaust 100-200 linear feet of air per minute.  The linear feet number is found by dividing the fan flow rate in cfm by the spray booth intake area in square feet.  So, for this 192 cfm fan, the largest opening the spray booth could have was 1.92 ft2.  Realistically, this is too big.  There will be flow losses in the ducting and filters that will decrease the cfm the fan can push.  Additionally, the range hood has a 7” diameter circular exhaust, which I wanted to decrease to 6” to fit the largest flexible aluminum duct I could find.  This decrease in flow diameter will incur a minor loss in the cfm, which is probably negligible.  The real problem from decreasing the flow area is the increase in airflow velocity.  In simple terms, velocity increases with the square of the diameter decrease; this simple decrease in area of 1 inch will result in a 36% increase in airflow velocity.  Losses due to the duct increase with the square of the velocity increase, so with this 1-inch diameter change, I just increased the losses by 85%!  This is a very simplified example—actually, as the velocity increases, the losses increase too, which means the total loss would not reach 85%.  The point is there are no free lunches.   I choose the flexible 6” duct over a rigid 7” duct, which made mounting easier, but resulted in a performance hit. .  At some point, I’ll solve the pipe flow energy equation or momentum equation to see how much I killed the flow rate with the installation.

 

Assuming the losses incurred by the duct, the area reducer, and the exhaust vent kill ~20% of the max rated cfm, I have a working number of ~154 cfm to base my booth on.  For a 100 linear ft/min venting, I could make the max opening 14.8”x14.8”.  This is a conservative estimate, as 20% losses are probably too high.  The range hood is 30 inches at the back, and narrows down to 24 inches at the front.  My intent was to have the booth side walls smaller than the hood so I could hang tools and my airbrush off the outside.  The opening area I planned for was 17” wide by 13“ tall, which is more than larger enough to contain the spray of a large can of automotive primer.  Having bought the hood, the reducer, the exhaust vent, clamps and the flexible ducting, I was ready to assemble the exhaust system for my booth.

 

Construction:  The Exhaust System

 

I wanted the air path from the fan to the window to be as airtight as possible.  Since this is the pressurized part of the system, any leaks mean you are pumping the exhaust right back into your work area.  I started by cutting a 6” diameter hole in the plywood sheet and then screwing the exhaust vent in place.  To make the exhaust vent seal to the plywood, I poured spare (i.e. too crappy to cast with) resin into the space between the vent tube and the plywood, then sealed it further with tub caulking.  This effectively sealed the vent in place, as well as permanently bonding the metal and wood together.  I did a similar thing with the range exhaust and the 7” to 6” reducer pipe, except here it was a bigger issue.  The fan exhaust is a simple ring of sheet metal that is buried inside the body of the hood.  There was no good way to mechanically attach (bolts, rivets, etc) the reducer to the hood without cutting a ton of sheet metal, so I used resin again to glue/seal the pieces together.  After the resin had cured, I was able to connect the range hood/reducer to the window-mounted vent with the flexible hosing.  In the picture to the right, you can see where some of the white resin spilled out from underneath the reducer ring.  The spillage was minimized by caulking where I could around the ring to hold the liquid resin in place.  Yes, those are a bunch of 1/10 scale Rei Ayanamis on the hood.  One other modification I made was removing the plain metal mesh filter from the hood and replacing it with an interior range hood filter.  For places where you can’t exhaust the air, the range hood has to recirculate the air it sucks in.  These types of installation require a combination filter that has wire mesh, fabric and a charcoal filter.  But, wait—won’t that mean your filter will end up trapping all the junk you want exhausted outside?  Kind of.  What the filter doesn’t trap will go outside (the vapors).  What the filter does trap (paint particulates and heavier stuff) will stay in the filter and still keep out of my lungs.  When the filter gets too clogged, it gets tossed—and in the mean time it keeps my fan cleaner so it will work longer/better.

 

Construction:  The Booth

 

OK, I cheated at first.  Another reason I took so long on building a booth was I couldn’t decide on the design.  Should it be collapsible, since I move every 2-3 years?  Should it be a simple box or should it converge to the rear?  I settled on a collapsible box shape.  I had purchased 3 2’x4’ 1/8” plywood sheets for ~$12 to build the box.  In order to complete the basic booth super quick, I used the cardboard box the range hood came in to make the booth.  The obvious advantage here is speed.  A not-so-obvious advantage is once I had the booth shape I liked, I now also have templates to cut the wood with!  One serious disadvantage is the cardboard booth is very flimsy compared to wood, and is hard to seal up airtight.  Another disadvantage is when you want to use one of these cardboard templates to make the wooden walls--you have to pull it out.  The cardboard walls are slightly more stable than a house of cards when all three were in place; pulling one out causes instability and collapse.  On more than one occasion the weight of the hood was supported entirely by the aluminum ducting as the whole thing fell apart.  I was reassured that the resin and duct tape I used to attach the range fan outlet to the duct was very solid;).

 

Since I wanted a booth that could be broken down, I made slots from 1”x1” strip pine on the back wall for the sidewalls to slide into.  The back is slides into guides made with the pine strips on a 2’x2’ plywood sheet.  Since the strips are glued and screwed into place, they not only lock the walls in place but provide an airtight seal around the booth.  To disassemble:  Pull the sidewalls straight out and the back wall straight up.  Despite being designed to break down the walls and floor are very solid, especially when capped with the hood.  The completed booth is shown on the right with cardboard front flaps (wooden ones to follow).  Pegboard hooks were used to make an airbrush holder, and in the very upper right of the picture you can see the regulator for the air supply.  The top of the range hood will have a tray to hold airbrushing supplies such as the bottles and color cups, respirator, latex gloves, etc.

 

Performance

 

Proof is in the pudding.  I fired up the fan a got one of the two nastier smelling full size primer spraycans I normally use.  I blasted it into the booth, and the results were in.  I sprayed for about 30 seconds, enough to get the back wall of the booth showing primer.  At the immediate vicinity of the booth, there was a faint odor of primer.  This odor though was not as strong as the smell of fresh duct tape I had used to seal up the ducts!  From about 5 feet away on either side of the booth, there was no paint smell at all.   With the amount I sprayed, I normally would have had to leave the basement for several hours so the fumes would die down.  With the booth, in a few minutes there was no smell at all.  I have since used it to spray Testor’s Dull Coat lacquer.   The largest source of odor now is if I leave the spray can out of the booth after I finish spraying—the drips around the nozzle cause the odor.  If I park the spraycan in the booth with the drying parts, there is no odor.   So, for about $115 total, I now have a spraybooth that is larger and can be broken down more easily than any commercially made booth costing twice as much or more.

 

If you found this page useful, e-mail me with you comments!