Archive for November 20th, 2010
Now, having seen what we’ve been living through, you might ask yourself
Wouldn’t It Be Nice If there was some way to be absolutely sure that mumble does not happen to me?
There isn’t, but you can stack the odds in your favor by disinsecting everything that enters your house. In particular, when you return from a trip, you must treat your luggage with the same casual regard as you apply to any lump of highly radioactive waste.
Because all bed bug stages die when exposed to temperatures over 45°C (113°F, which I round to 120°F), the simplest way to ensure that you’re not bringing any passengers home is to heat your luggage / packages / clothing / whatever to an internal temperature around 120°F, then let it soak for maybe an hour to ensure all the occupants get the message.
What you need is a box that gets hot on the inside, but not hot enough to set your luggage on fire. As with all things sold for bed bug problems, the commercial solution seems grossly overpriced for what looks like an uninsulated ripstop nylon bag containing a rack, a heater, and a fan.
It should come as no surprise that I built something that’s bigger, uglier, and harder to use… but it produces data and you can do science. And, with liberal use of my parts heap, the overall price is maybe 10 dB down from the commercial version…
I figured that this widget is going to be a major part of our lives from now on, so a foldable / storable heater wasn’t particularly useful. In point of fact, we’ve been using it heavily and I don’t expect that to stop any time soon.
It’s a rigid box made of Dow Tuff-R rigid polyisocyanurate foam insulating board, held together with 4-inch wide aluminum HVAC tape. The rim around the top is sealed with opposing strips of felt weatherstripping, held on with double-stick tape.
Inside, I used lengths of wire shelving to support the thing-to-be-baked. After we’ve used it a bit more, I’ll conjure up permanent supports for the second level shelving (stacked on the right of the exterior picture); right now, they’re supported on wood blocks as needed.
The interior dimensions work out to 34x22x24 inches: it’s made from a single 4×8 foot sheet of insulating board. Here’s my working sketch showing how the parts lay out and fit together. (clicky the pic for more dots).
The only waste is the 1-inch strip along the right edge; the slab I bought came with a molding imperfection, so discarding that edge was OK.
I cut the sheet into four 2×4 foot strips, cut a 13-inch strip off each plank, then trimmed the 1 inch waste. That seemed less prone to catastrophic blundering than (trying to) make a pair of 8-foot cuts and whack each resulting strip in quarters. An ordinary razor utility knife worked fine, although I found that making two passes along each cut produced cleaner results than trying to do it all in one.
I assembled it with the heavy / shiny aluminum foil side inward, although I doubt it makes any difference. Cover all the edges with tape, tape all the joints both inside and outside, and it becomes a nice rigid box when you’re done. Pay attention to getting the sides at right angles; I used a framing square.
The board allegedly has an insulating mojo of:
R = 6.5 ft2 • h • °F/Btu
Figuring a surface area of 32 ft2 and a temperature differential of 120 – 60 = 60°F, the box should require 295 BTU/hr = 87 W to maintain that temperature.
Which, as it turns out, is pretty close to how it worked out:
The lower curve shows a 60 W bulb with a 10 W 120 VAC fan heats the interior to a bit over 100°F in 100 minutes, where it looks to be stabilizing. That was the first test and showed that I was on the right track.
The second test, with a pair of 60 W bulbs and the fan produced the two upper curves: one for air, the other inside some cloth jammed inside a plastic bucket to simulate a (tiny) suitcase. The combined 130 W heats the box over 150°F in two hours, with the somewhat insulated bucket trailing neatly behind as you’d expect.
Without opening the box, I connected the bulbs and fan to a Variac plugged into my Kill-A-Watt meter and dialed it for 100 W total dissipation. The temperature fell to slightly over 130°F in 80 minutes and looks like it would stabilize near there.
Ambient temperature was 67°F, so
R = 32 ft2 • 67°F / (341 BTU/hr) = 6.3
Close enough, I’d say. Given those few data points, it looks like the temperature sensitivity around 130°F is 0.7°F / W. [Update: typo in the equation. Doesn’t change the answer much at all.]
I swapped in a 100 W bulb, removed the Variac, and heated the cushions from my office chair.
One thermocouple is hanging in mid-air, the other is wedged inside one of the cushions. After nearly 5 hours the cushion is up to killing temperature and I turned the heater off. The air temperature drops rapidly, but the cushion stays over 120°F for another two hours.
The light bulb is just a proof of concept, because it’s entirely too hot: if the fan fails, your luggage ignites. I plan to build a rather subdued heater with a surface temperature around 140°F and a controller that monitors several sensors to ensure the contents reach killing temperatures and stay there long enough.
But that’s a project for another day…
[Update: If you’re arriving from a link, start at the overview to get The Whole Story.]