New construction. The buyer says that the builder replaced the lower sash earlier in the week because it was cracked. I did see visually that both the lower and upper sashes of this window were different from the other two, so I’m guessing that they replaced both.
I’m curious to hear theories, as to what would cause this. I have my own,which I will share and my reasoning.
This is a straight forward image. There is no special manipulation, nor is this intended to be a gotcha. Just straight-up thermal observation and analysis.
This is in Texas. It is a cold (by Houston standards), clear morning. Glass is opaque to long wave IR and has a typical emissivity of between .87 and .94. We are only doing qualitative analysis so the imager Emissivity value is set to 1.0.
I think it is a highly reflective coating or tinfoil behind the window reflecting the outside temperature. If it was a lost seal the center would be warmer as the window bowed in. Not a defect in my opinion.
Opps, Yes tinfoil would not be seen. My bad.So deduction tells me that the cold temperature is seen on only the one sash but not on the other. What is making it cold. The interior temp if it is occupied would be between 60-70 degrees and the exterior temp is 40-50 degrees. There must be something either within the window(argon gas) or on the interior of the window blocking heat transfer. You may not be able to see into the window through the imager but you can see something blocking the heat from getting onto the window and warming it up.
Since the glass is opaque to our imager, we wouldn’t actually see heat transfer through the “glass”, we would be seeing a combination of heat emitted by the glass and being reflected off of the glass. If there was a bright hot incandescent light bulb behind the glass, we would not be able to see it with our long wave imager (Scott Gilligan posted a great demonstration of this a while back showing the difference in glass transparency between long and short wave imagers here: http://www.nachi.org/forum/f58/fun-thermal-imaging-61162/). Linus also has short wave imager that can “see” through glass.
Normally, when we look at a window, we see mostly the infrared that is being emitted from the glass itself. In Texas the low-e coating on a window is usually applied to the 2nd surface (the back side of the outer pane). This keeps us from “seeing” it with our imagers. In the case of the top left sash here, the individual pane appears to have been installed inside out so almost all of the incident radiation we are seeing is being reflected from the clear blue (i.e., very cold) sky. This causes us to see an apparent temperature of -24F vs. 55-60F for the other windows. There is very little difference in actual temperature between these windows (we verified this from inside).
A low-E coating applied to surface 1 is considered a defect as the coating will not perform its job properly (heat will pass through due to conduction and radiate off of the interior glass surface) and it will not be durable exposed to weather.
This is a good example of why understanding emissivity and the effect that it has on your thermal images is important, even for qualitative inspections.
Anyone think that there is really an ~80F difference between these panes of glass?
I’ll report the finding as suspected because it really should be verified with an ETEKT type device that I don’t carry.
Barry obviously knew what was going on with the window. Edit: Credits also to Marcel and Linas who posted while I was typing.
Emissivity matters when doing infrared thermography, even for home inspectors.
BTW: these images are not public domain, If you wish to use one, please contact me outside of the forum.
Nice post, Chuck. It reminds us once again that regardless of the abilities of the imager IR scan on its own is never definitive and we need to use other devices to verify our findings. BTW, ETEKT is not expensive at all. I own one because I need to verify Low-E coatings on the windows installed in ENERGY STAR homes, and I think it was about $150 or so (although I bought it almost 6 years ago). Thanks for the post.