GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Audio Play Icon Headphones Icon Plus Icon Minus Icon Check Icon Print Icon Picture icon Single Arrow Icon Double Arrow Icon Hamburger Icon TV Icon Close Icon Sorted Hamburger/Search Icon
Musings of an Energy Nerd

An Introduction to Thermal Imaging

An infrared camera can help diagnose a variety of home-performance problems, including air leaks, missing insulation, or thermal bridges

Making heat leaks visible. This thermal image shows a poorly insulated brick house. The warm chimney flue on the the gable end shows up as a vertical red stripe. [Photo credit: Tom Barbour - Thermal Image UK -]
Image Credit: Tom Barbour - Thermal Image UK -
View Gallery 8 images
Making heat leaks visible. This thermal image shows a poorly insulated brick house. The warm chimney flue on the the gable end shows up as a vertical red stripe. [Photo credit: Tom Barbour - Thermal Image UK -]
Image Credit: Tom Barbour - Thermal Image UK -
This thermal image clearly shows studs as vertical stripes. Each stud is a thermal bridge that leaks more heat than the insulation between the studs. [Photo credit: Gerald Brady - Infrared New England -]
Image Credit: Infrared New England -
This thermal image of a house in Houston shows heat loss at an uninsulated rim joist. The vertical stripe at the lower left is heat loss from copper tubing conveying refrigerant from a heat pump to a fan-coil unit. [Photo credit: Chuck Evans -]
Image Credit: Chuck Evans -
This thermal image reveals which stud bays have no insulation. [Photo credit: Gerald Brady - Infrared New England -]
Image Credit: Infrared New England -
This infrared image shows evidence of heat leaking into an attic. [Photo credit: David Valley - Massachusetts Infrared -]
Image Credit: David Valley - Massachusetts Infrared -
This thermal image was taken from the interior of a timber-frame home that was depressurized with a blower door. Air leaks at the eaves show up as cold spots. [Photo credit: Gerald Brady - Infrared New England -]
Image Credit: Infrared New England -
This thermal image shows air leaking into a home from behind the baseboard. [Photo credit: Gerald Brady - Infrared New England -]
Image Credit: Infrared New England -
Thermal imaging can reveal leaks in low-slope commercial roofs. When such roofs are inspected after a sunny day, damp insulation (which is warm under these conditions) is revealed as a relatively warm spot on an otherwise cool roof. [Photo credit: InfraMation newsletter]
Image Credit: InfraMation newsletter - November 2006

Energy auditors and insulation contractors have been using infrared cameras to diagnose home-performance problems for over 30 years. Without opening up your walls or ceilings for inspection, a trained specialist can use one of these cameras to locate insulation voids, air leaks, moisture intrusion, thermal bypasses, and thermal bridges. It’s even possible to use an infrared camera to locate leaks in hydronic tubing embedded in a slab.

These tools are known by a variety of names, including infrared (IR) cameras, thermographic scanners, and thermal imaging devices. An image produced by such a camera is called a thermogram, and a trained user of the device is called a thermographer.

Although many people assume that infrared cameras measure surface temperatures, that’s not really how the tools work. An IR camera actually measures the intensity of infrared radiation (radiant energy) being emitted by the surface it is aimed at.

The history of infrared cameras

Academic researchers began using infrared cameras to diagnose thermal envelope defects in the late 1970s, using cameras that cost more than $25,000 each. The cameras were cumbersome devices that were cooled by liquid nitrogen.

As new models of IR cameras were developed, prices began to drop. Gautam Dutt, one of the original “house doctors” at Princeton University, recalls, “For an infrared camera, we used a Magnavox unit marketed by Aga, the Aga 110. It cost about $13,000. It was expensive equipment, but it was cost-effective. The first thermal bypasses that I discovered took me weeks to discover, crawling through attics. Another one of our teams, working with infrared, could find thermal bypasses right away by using an infrared camera in combination with a blower door. That was the innovation we developed at Princeton — the combination of a blower door and infrared.”

Prices for infrared cameras continue to drop. It’s now possible to buy a useful IR camera for $1,500, and a high quality one for $5,000. Lower prices mean that more contractors can afford the devices; however, since an infrared camera in untrained hands isn’t very useful, anyone starting out in the field should budget some money and time for training.

John Snell — admittedly a biased source, since he owns a company that provides thermography training — explained in a recent JLC article that “a two-day training session in IR-imaging basics is worth about a year of flopping around on your own.”

Wait for appropriate weather

This article is intended as an introduction to the topic of thermal imaging. However, I am not a trained thermographer, so readers should not depend on this article for technical advice concerning the use of an infrared camera.

If the outdoor temperature is 70°F, you usually won’t get any useful information with your IR camera. That’s because the indoor temperature and the outdoor temperature are about the same; there usually won’t be any heat flowing through a building’s walls or ceilings under those conditions.

To conduct a thermographic inspection of a building, you need a minimum delta T (temperature difference) of 18F° between the interior and the exterior for several hours before the inspection begins. If the interior is at 70°F, you usually need an outdoor temperature that’s below 52°F or above 88°F.

If the weather isn’t cooperating, you’ll have to be patient and wait for the weather to change — or you’ll have to adjust the thermostat so that the heating system or cooling system makes the interior of the house unusually hot or cold. To be sure of an adequate delta-T, most thermographic inspections are scheduled for early in the morning.

Early morning inspections are best for another reason: sunlight tends to warm exterior walls and roofs, complicating thermographic inspections. Ideally, you don’t want any sun shining on the building for at least three hours before the inspection begins. If the house has brick or stone veneer, the sun-free interval should be at least eight hours long. According to Snell, “Inside and out, in both summer and winter, too much sun can wreak havoc on otherwise acceptable conditions.”

Very windy days aren’t good for thermography. Ideally, the wind speed should be 8 mph or less. The roof and walls should be dry.

For an inspection of an occupied house, it’s a good idea to tell the homeowners to move furniture and pictures away from exterior walls 12 hours or more before the scheduled inspection.

Start on the outside of the building

Most IR cameras need to be adjusted by the user each time images are made. Typical user-adjusted settings include:

  • A choice between a color image or a grayscale image.
  • Settings for the upper and lower temperature limits (the span setting and the thermal level setting — sometimes referred to as “gain” and “contrast”).

Assuming that the home’s walls have conventional wood framing without a layer of rigid foam insulation, a thermographer hopes for weather conditions that produce images showing details of the building’s framing. For example, if the camera is looking at the outside of a house during the winter, vertical stripes should be visible on the walls. These are the studs, which act as thermal bridges, leaking more heat than the insulation between the studs.

If the framing members aren’t visible, that usually means that either (a) the weather conditions aren’t appropriate for a thermographic inspection, or (b) the house is well built, with details designed to minimize thermal bridging. Examples of such walls are SIP walls, ICF walls, or walls that include exterior rigid foam sheathing.

If the wall framing is visible (or if the lack of visible framing is explained by the presence of rigid foam insulation), a thermographer inspects the outside of the building, looking for anomalies. An IR camera will easily reveal common thermal bridges like studs or poorly insulated rim joists.

The camera will also reveal insulation voids. In some cases, it’s possible to see stud bays that were left uninsulated. In other cases, blown-in insulation may have settled, leaving a void at the top of each stud bay. Stud bays with hot-air ductwork show up very clearly.

Get out your blower door and move inside

After inspecting the exterior of the building, a thermographer usually moves indoors. As with the outdoor inspection, the idea is to scan surfaces for anomalies. After an initial inspection, most thermographers set up a blower door and depressurize the building.

The blower door increases the flow rate of air leaking into the building through cracks in the walls, floors, and ceiling. Since the outdoor air is at a different temperature from the indoor air, the air cools (or warms) the surfaces it encounters in the area of the crack, creating visible plumes in thermal images.

According to “Guidelines for Thermographic Inspections of Buildings,” a standard produced by RESNET, “The thermal image for air leakage will appear as ‘fingers’ or ‘streaking’ showing as dark when cold air is observed and lighter colors when warm air is viewed. The thermal images will produce irregular shapes with uneven boundaries and large temperature variations. These air leakage sites are often at joints, junctions or penetrations in the enclosure. There is often a temperature gradient within a finger or streaking area. … Take care to discriminate between thermal bridging sites and thermal bypass or air leakage sites. Thermal bridging sites will not change size or shape during the inspection.”

Low-slope roofs

An IR camera can be a useful tool for pinpointing leaks in a low-slope roof. Such an inspection is best performed at night after a sunny day. For a useful thermographic inspection, the roof must be dry.

When the sun shines on a low-slope roof, it heats up the roofing and the top of the insulation below. Once the sun sets, the insulation begins to cool; however, damp insulation cools at a much slower rate than dry insulation. That’s why leaky areas of the roof show up as warm spots when viewed at night.

Pitfalls for the unwary

The best thermographers are those who understand building construction, understand building science, and have been conducting thermographic inspections for many years.

Almost anyone can point an IR camera at a building and fiddle with the knobs until the colors are pretty, but it takes a trained and experienced expert to tell the difference between evidence of thermal bridging, air leakage, and moisture.

Last week’s blog: “Energy Predictions for 2012.”


  1. Doug McEvers | | #1

    IR Camera Sensitivity
    The best cameras are really amazing, the thermal bridging through a 14" TJI wall was evident during a home tour with an outside temperature of 50F. Insulating contractors that have and use IR cameras are way ahead of the game, couple this with a blower door and you will stand out.

  2. David Meiland | | #2

    One quibble
    Martin, good piece.

    You wrote "Indoor inspections are usually conducted after a blower door has been set up to depressurize the building". My suggestion is that you conduct an interior inspection before starting the blower door, and then make another pass after the door has been running a short while. If there's a decent delta T--I can easily work with 10F or even a bit less--the air leakage problems show up quickly. If you run the door before scanning anything, you will have a harder time deciding what's what.

    I find that exterior inspections are a lot harder to conduct reliably. Wind and sun erase a lot of thermal signatures. If I'm going to do the exterior I usually need to go early in the morning and hope for calm conditions.

  3. Deniz Bilge | | #3

    This is an excellent blog
    I never thought about how thermographic images can be misconstrued; nor have I considered weather in the role of taking accurate measurements. This is excellent.

    On that note, I can say the first image showing the poorly insulated brick house can mean either of the following: 1 -- This is a poorly insulated house with the IR camera readings taken first thing in the morning, or 2 -- This is a decently (but not perfectly) insulated house with the IR camera measuring the surface heat dissipating off the bricks after a long sunny afternoon in the western sky.....

    Doug, great point about insulation contractors using IR cameras--sounds so obvious, but I never thought of it, and apparently, nor has any insulation contractor in my area.

  4. User avater GBA Editor
    Martin Holladay | | #4

    Response to David Meiland
    Thanks for your comment; I appreciate the feedback.

    I have edited the text of the blog to reflect your point.

  5. User avater GBA Editor
    Martin Holladay | | #5

    Response to Deniz Bilge
    Like you, I'm curious about the circumstances under which the first image was taken. I'm especially curious to hear an explanation of why the roof is blue, the front wall is red, and the gable wall is yellow. I have e-mailed questions to Tom Barbour in the U.K., the thermographer who took the image, and will post more information after he responds.

  6. Lucas Durand - 7A | | #6

    Not sure if this applies...
    Do you know if "angle of incidence" has any effect on how an image is recorded by a thermal imaging camera?

    Would that explain the subtle difference in colour for the two faces indicated in the snip below?

    Would "cranking up the gain" exagerate that suble difference until it resembled something different?

  7. John Brooks | | #7

    Image 1
    My guess is early in the morning...
    The Red wall is facing East
    Yellow wall facing North
    Blue roof was facing cold sky overnight

  8. David Meiland | | #8

    ... maybe early in the morning, but has the wall been heated yet? A brick wall would probably take some time to make a thermal swing.

  9. David Meiland | | #9

    Discussing infrared images
    It's hard to talk about what's happening in an infrared image without knowing the circumstances--time of day, interior and exterior temps, sunny/cloudy/rainy, makeup of building assemblies, etc. None of the images have a temperature scale or any temperature markers included, which would help.

    In the image of the brick house, the roof is blue because the camera is seeing the sky reflected on the roof. You can also see some sky reflection in the windows. It's best to keep the lens axis as close to perpendicular to the surface being measured as possible--less (a lot less...) than 45 degrees is ideal--and with sloped roofs that's very hard to do from the ground.

    I wouldn't draw many conclusions at all from this image--it doesn't really tell me that the house is poorly insulated.. It seems likely to me that the wall on the left was heated by the sun and is now cooling, where the wall on the right didn't get as much sun. I would tentatively say that it's a cold day or night (because the flue is hot, probably from a fire burning), the red wall has or had the sun on it, the yellow wall less sun. Really, this is a probably a good example of why outdoor inspections are hard--the inspector was probably there are mid-day or afternoon and was seeing a lot of solar loading on the exterior, rather than heat loss. I wonder what he would see if he went there on a cold, calm morning before dawn to image these walls, with the house well heated all night.

    Here's an image taken when interior and exterior temps were about 5 degrees apart. The sun has been on this wall a couple of hours, no blower door.

  10. David Meiland | | #10

    Sky reflection
    Here's an image with lots of reflected sky. Calm evening, air temp about 30 degrees.

  11. Lucas Durand - 7A | | #11

    Did you "crank up the gain" in that image?
    Is that why the colours look so dramatic and exagerated?

    Edit to say:
    Maybe I'm not using the right term...
    I'm thinking of "gain" in the same way as for a radio antenna...
    As in signal resolution or sensitivity...
    "Cranking up the gain" = increasing resolution = broader colour scale?

  12. David Meiland | | #12

    Lucas, three main factors control the intensity of the image. First is the color palette itself. There are several on my imager, and that one is called "high contrast". Basically, it attempts to produce the most exaggerated image. Others are milder, for lack of a better description. Second is the "span" of the image--the upper and lower limit of temperatures rendered by the color palette. The tighter the span, the more colors are used to render a given range. Third, and related to span, is the level, i.e. the center point of the span. Here's another view of the same image. Different color palette, wider span. Notice that a lot of detail disappears. For that image, a span of ~26 degrees is to wide to be of much use.

  13. John Brooks | | #13

    IR Camera Envy
    David, I am jealous ....
    thanks for posting more photos and explanations

  14. Doug McEvers | | #14

    21st Century Man
    With what we know today as a building community and the diagnostic tools available to us, there is no excuse for building a crappy house.

  15. User avater GBA Editor
    Martin Holladay | | #15

    A question for David Meiland
    When you use the phrase "sky reflection," what exactly do you mean?

    Since the thermal camera is registering the intensity of the radiant energy being emitted from the surface it is pointed at, I assume that you mean that the low intensity of radiant energy emitted by the roof is due, in part, to sky radiation (the fact that the roof is radiating thermal energy to the sky and therefore cooling).

  16. David Meiland | | #16

    I won't try to give a technical explanation, but it is very common to see the sky temperature reflected by objects. It often happens when you point the camera up at a roof or a window that's a couple of stories up from the ground. Objects can often be colder than ambient temperature due to night sky radiation, and that may be part of what is seen in the blue areas of roof in the images above, but another part of what is seen is reflected, not emitted. My imager and many others allow the user to adjust emissivity and background temperature, so that in theory you can compensate for reflection, but in practice it is very tricky to do accurately. Anyway, infrared energy is emitted, reflected, and in some cases even passes through objects, so the imager is combining all of those into one reading. Bottom line for the building science person trying to assess a building without a PhD in physics... point the imager straight at the surface you want to inspect, have a good understanding of emissivity, and all bets are off when imaging low-e materials like shiny metal.

  17. User avater GBA Editor
    Martin Holladay | | #17

    More information on the first image on this page
    [Tom Barbour, the thermographer who took the first image on this page, has e-mailed me with more information on the circumstances at the time the image was captured:]

    The image you selected was taken during the daytime. It's a traditional Scottish sandstone house with no wall insulation (600 mm [24 inches] stone, approx. 30 mm [1.2 inch] gap, then lathe and plaster internal lining). The loft [attic] has 270 mm [10.6 inches] insulation. However the red elevation (front) is due to solar loading, the gable shows a chimney with the fire burning, and the blue roof is reflection from the cold sky due to the low angle. This image does not show any heat loss due to missing insulation or otherwise - I've included it on my website as visually it's eye catching, and it leads in to interesting questions about what you can and can't tell from a thermal image. I hadn't appreciated the context in which you wanted to use the image - however if it's got a debate going for you then perhaps it's served it's purpose!

    Hope this helps.

  18. Lucas Durand - 7A | | #18

    Angle of incidence?
    Your explanation with the extra photo are very helpful, thanks.

    A couple of times now you have alluded to the importance of pointing the camera so that it is as close to perpendicular as possible to the surface being inspected...

    What type of error do you "see" in an image as the surface being inspected (or the camera) is tilted away from the perpendicular?

    Does the surface appear to get "colder"?

    In the snips below, is the surface indicated by the arrow really that much "colder" than the surface facing the camera?

  19. David Meiland | | #19

    Assuming your arrow is pointing to part of the wood siding, I would expect that to be about the same temperature as other wood siding on the building, assuming there is no wind or rain, and the building is uniformly insulated and heated inside. The low angle puts a lot of reflection into the reading, whereas a view closer to perpendicular minimizes reflection. When you go to the lake and skip stones, you have to release them low to the water or they just dive in. Same sort of idea applies here... you try to get yourself squared up to the surface you want to image.

    Here's another shot, taken July 29 at 7:30 in the evening. Probably about 60 degrees out. No way is the roof as cold as the camera says it is. Also little to no chance that it's accurate re the sky temp. If I really wanted to know the roof temp, or even get an accurate qualitative image of it, I would need to get up there and point straight at it.

  20. Lucas Durand - 7A | | #20

    Thanks David.
    That makes sense.

  21. Deniz Bilge | | #21

    This gets one to wonder...
    I've learned through this blog that thermography is, after the mechanics are learned, really an art in which the camera operator, not necessarily the physical subject material, is responsible for the results of the final image.
    Taking this thought one step further, I can't help but wonder about the future of this field. If anyone with $5,000 can buy a camera, I see three situations: 1-- The true professional who knows his/her instruments and the sciences behind thermographic imaging, and can interpret the data most accurately in terms of what the image is truly showing at the time of capture; 2-- The professional who does NOT know his/her instrument and the science behind it, and who may not be able to accurately interpret the image, which can mislead clients; and 3-- Any combination of the above who would intentionally produce images which may be used to instead manipulate the client into making a decision beneficial solely to the purpose of the camera operator.

    I see lots of confusion. Should people have multiple imaging sessions under similar circumstances to compare? Should different sessions be taken under similar circumstances but at different times of the year? of the day?

  22. User avater GBA Editor
    Martin Holladay | | #22

    Response to Deniz Bilge
    The problem you describe -- poorly trained contractors who give undependable advice -- is not limited to thermographers. (In fact, thermographers are probably better trained than workers in many other trades.)

    If you are a homeowner with an ice dam problem, and you call up four contractors, what happens? One wants to sell you heating cables for your eaves; one says you need new roofing; one says you need more attic ventilation; and one advises air sealing your thermal envelope.

  23. John Brooks | | #23

    Infrared Thermometers
    How about a blog on IR thermometers....
    How to use them (and not to use them)
    I have really found my $55 Raytek to be "helpful" for measuring surface temperatures.

  24. User avater GBA Editor
    Martin Holladay | | #24

    Infrared thermometers
    Good idea -- I'll put it in the hopper of ideas for future blogs.

  25. David Meiland | | #25

    You can buy IR for less than $2000, and lots of folks in the industry think we'll soon see $500 models at WalMart. Lots of folks already have IR thermometers, which probably go for less than $100 now in some cases. Some people will be able to make good use of these tools, and others will make lots of mistakes. There is a lot involved in doing quality non-destructive investigation.

  26. Nick Grant | | #26

    This looks like thermal bridges but is thermal mass
    Great post and comments.

    Agree powerful tool but needs good user experience.

    Attached shows underside of roof in my home. Looks like solid rafters causing thermal bridge but I know they are 400mm deep (home made) I trusses and should be OK. What can't be known from the single image is that the heating has just gone on and that the 'rafters' are not visible on an image taken a few hours later despite same temperature difference as the extra mass of the timber flange has had time to warm up. I have seen this a lot with I joists and service void battens. Need to let temperatures stabilise.

    Similarly the apparently dramatic cold corner at the ridge also almost vanishes (a little air leakage visible as whisps in windy weather). Corners will always be colder though as are a geometric thermal bridge.

    I mostly find myself working from inside as usually looking for subtle faults in super-insulated, airtight Passivhaus Buildings that can be harder to see from outside except under ideal conditions. Also we use a lot of vented rain screens which limit what can be seen from outside.


  27. David Meiland | | #27

    There weren't any images attached to your post. I'd like to see 'em!

  28. Douglas Horgan | | #28

    Still pretty useful if you understand a few things
    I recently bought a $1700 camera and have found it very useful for building diagnostics.
    There are some important points in the blog and comments--the camera only shows the amount of IR radiation coming off a surface, no more no less, and different types of materials show differently under different conditions.
    My opinion is, if you know how buildings are constructed, and a few of the limitations of the cameras, you are in pretty good shape even without $2500 in training.
    If you're preparing expert-witness level reports instead of finding things to fix, it's probably different.
    For a typical experienced builder these are great tools even without elaborate training.

  29. Eric Rasmussen | | #29

    It is very exciting to see
    It is very exciting to see this technology more readily available to the market, and to see the large adoption by contractors and professionals. This is a great step in the right direction to making our homes more energy efficient. These IR cameras combined with a reasonable knowledge of building construction allow us to more readily take the first step of accurately identifying the issues that must be tackled, and precisely knowing the areas of susceptibility allows us to efficiently and correctly rectify problems.

    I also think using this technology on newly constructed homes would prove to be quite interesting. It will definitely keep contractors honest and give homeowners a reasonable means to verify they are receiving the quality they are paying for.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |