Bird Houses Aren't Real Houses
November 8, 2011 6:25 PM   Subscribe

How can my kid best test an experiment regarding roofing materials and their ability to block heat?

My kid wants to do a project very similar to this:

With parental guidance our child wants to test terracotta tile, aluminum, tar shingle, and slate for his roof materials.

What is the best way to go about this? Spouse and kid are proposing to make four birdhouse type structures with different roofing materials and one birdhouse type structure with just a wood roof without additional roofing material. They will then measure the temperature inside each house.

Are bird houses a good way to go about testing roofing materials and how they block heat?

We cannot go inside or test different houses. We want to make our own structures with our chosen roofing materials or test roofing materials in another way that makes sense. How should we do this?

What is this kid doing on the right hand side of this project? I think this is the experiment part of his project but I cannot figure out what he is doing with the sand and such.

Any ideas or thoughts on what this student is doing on the right hand side of the project board and what the best way to go about testing is greatly appreciated. I think this particular student tested pebbles, PVC pipe, and shingles. See link here (wrong photo but correct description and title of project)

Thanks so much.
posted by Fairchild to Education (10 answers total) 1 user marked this as a favorite
What this project needs is a blowtorch.

I'm not sure what's going on in the picture, but I would set the project up like so:

"Build" a few "houses" out of regular old brick. Basically, just make a hollow cube of bricks with the top open, like a chimney with a bottom. Probably about four bricks high. And leave one brick out of it so you have a viewing window. (Does that make sense?)

Then, find something (preferably something amusing, since it's a kid's project) that will melt or otherwise be affected by heat. Get enough of those somethings to have enough for each test. Put one in each of the brick houses. (Alternatively, you could just make one brick house and make sure that it's fully cool before doing the next test.)

Once you've got your victim set up in the brick house, place the roofing material over the opening at the top. Set up a flashlight and have someone watching through the window with a timer. Have someone else blowtorch the roofing material on top. Time how long it takes for the victim to melt. Then just repeat for different roofing materials.

And make sure to take lots of pictures!
posted by phunniemee at 6:50 PM on November 8, 2011 [1 favorite]

Ha! This will be right up my kid's alley and the brick houses totally make sense. Thanks for your answer phunniemee.
posted by Fairchild at 7:20 PM on November 8, 2011

For a more realistic setting, I'd replace the blowtorch with a heat lamp, placed a set distance from each roof, with a thermometer along with the little melty thing. The thermometer will give you an accurate reading, alongside the visuals. And take a picture of the setting at set intervals, to provide something more for the presentation.
posted by filthy light thief at 7:44 PM on November 8, 2011

If you live somewhere where it might snow, you could observe heat lost through the roof of real houses by gauging the amount of snow melted after a snowfall ie if it snows a foot and the next day there is a foot of snow on a roof, there is very little heat escaping from that house, whereas if a foot of snow is reduced to an inch of snow on a roof, the house is losing lots of heat via the roof. Kind of requires the weather to cooperate, though.
posted by lulu68 at 8:10 PM on November 8, 2011

Blowtorch is cool, with a heatlamp you might be more scientific but it would take a lot longer.

Stick an oven/candy thermometer inside the brick structure. Apply heat1. Take readings of the temperature every 30 seconds, to make a graph of internal temperature/time, and take pictures of melting visual. Two experiments in one! Given estimate of sampling frequency is for a blowtorch. With a heatlamp, you'll want to do the experiment indoors in a controlled temperature and draft condition, with sample (readings) frequency in the once/hour range.

Make sure to mention that you let the apparatus2 return to steady state ambient temperature (and note what the ambient temperature is at the time of experiment. Equilibrate all roof samples to the ambient temperature so they all start pretty equal. Report all of these safeguards and measurements.) between experiments with different roofing material/visual-aid-sacrifice. Also, make notes on how you are applying the heat; like distance from roofing material, whether you were moving the heat source around (and if so, in what kind of pattern, and how quickly), and whether there was any wind, &c.

1 I think a blowtorch will be able to achieve temperatures through even an insulated roof to melt green army men and whatnot. Maybe compare army men made from different plastics or a different material during the repetition of the experiment. The best thing about this is that you also get the opportunity to discuss analyzing the data; means, medians, standard deviation, standard error, &c&c.

2 Maybe replace one side of the brick "house" with glass that will withstand high temperature?

posted by porpoise at 8:12 PM on November 8, 2011

I think what I would do is instead of building a brick "house" and adding a roof, I would make a box out of the roofing material. Keep in mind that your roof has a plywood substrate underneath the actual roofing material, so build a box of plywood and then cover that with your roofing. Don't use a blowtorch, because the fire part of the torch isn't very realistic. Instead, put a thermometer inside the box and put the whole system inside your oven, set on low (200 degrees). Let each box sit for the same amount of time - let's say half an hour - and then measure the temperature of each. If you have one of those probe-style kitchen thermometers with the digital readouts that sit outside the oven, you're in even better shape - just place the probe inside the box and you can watch the readout and record how long it takes for the temperature inside the box to equal the temperature outside the box.

The reason I'm suggesting building the whole box out of roofing material is because, in this case anyway, making a realistic model of a house is not going to make much sense from an experimental standpoint. In fact, if you do use the oven suggestion, you won't see any difference between the materials at all, unless the rest of your "house" is very highly insulative.

Actually, the more I think about this, the more I think you need two separate tests. My understanding of home construction is that the insulation in your attic is usually what provides most of the heat retention to the home; the attic itself is left to "breathe" to prevent moisture buildup. The roof may provide some insulation against radiative heat from the sun, but probably little protection against the ambient temperature. So bake the boxes in the oven, and then take a plank from each and subject it to a heat lamp or other point source of heat (blowtorch might work if you are careful not to get too close and melt things). You'll want to measure the temperature on the underside of the plywood substrate, so either a thermocouple or an IR thermometer would be best.

If you have neither, what you want to do is make some identical ice cubes. WEIGH the water and try to make at least one side as consistent as possible - having them all be the same size is important. Then, set up your experiment in the following way: heat source on the bottom, roofing tile facing down towards the heat, and then rest an ice cube on the plywood backing. Apply heat, measure how long it takes to melt.
posted by backseatpilot at 9:15 PM on November 8, 2011

So, this was basically my EXACT project for my third grade science fair. I had an unfair advantage as a distant relative was a science teacher and loaned my a ring stand to do the project, and after that things were pretty easy.
posted by midmarch snowman at 12:13 AM on November 9, 2011

Wow, not sure how I posted that, I was aiming for the "link" button which is nothing like the "post" button.

Anyways, with the ring stand I simply hung the materials I was testing between a thermometer and an old desk lamp, with the shade removed, and a 100w incandescent in place. I measured the change in temp over a certain period of time. I used insulating foam, play-doh, wood and metal as insulation materials and had white and black cardboard that I put over the materials to control for color and test black versus white in terms of temp conductance.

In retrospect, there is a LOT more I could have done to evaluate the materials and the experimental design, but that's probably why I didn't win any prizes at that science fair... but I did get a good grade. I think a blow torch would be cooler... but my project was done in a single evening after school.
posted by midmarch snowman at 12:19 AM on November 9, 2011

He could also ground it in a discussion of U values, the U value of a material is a measure of how rapidly heat will pass through it. He won't be able to get an accurate measure of the U values with the rough equipment but he could make a comparative discussion.
posted by biffa at 2:55 AM on November 9, 2011

Thanks for the answers so far. I really appreciate them.

We were hoping we could leave the structures outside and measure temperature. We live in FL and it's still fairly warm here. We may have to use an alternative heat source other than the sun.
posted by Fairchild at 3:35 AM on November 9, 2011

« Older She never gets fucked hard enough..   |   It's either this or they burn their rent checks Newer »
This thread is closed to new comments.