GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Cheap thermal imaging camera

Aaron Birkland | Posted in Green Products and Materials on

It appears that a startup company called Seek Thermal recently has been offering a $199 thermal imaging camera that attaches to Android or IOS smartphones. What little specifications there are look impressive, for example:

  • Reasonably high resolution, 206×156
  • -40C to 330C range
  • Automatic calibration
  • Nearly 10 frames/sec video

A similar device, the FLIR One was described on GBA nearly a year ago, but this one has a few significant differences:

  • Much higher resolution, 206×156 vs 80×60
  • Compatible with many Android phones with a micro USB port, as well as Apple IOS-based phones with some other funny looking connector that I can’t recall the name of. The FLIR One only compatible with the Apple Iphone 5.
  • The FLIR One compensates for its very low thermal resolution by overlaying details from the visible light spectrum (combining thermal and traditional images), whereas the Seek produces pure thermal images
  • Price: $350 for FLIR One vs $200 for Seek

The Seek thermal camera has been reviewed by several mainstream tech sites, including Wired (with an excellent comparison image that illustrates the different approaches of the FLIR One vs the Seek near the end of the article), CNet, and others

So my question is: What’s wrong with it? Despite its specs, it this merely a toy, or is thermal imaging really starting to get cheap and useful? Does it seem worth purchasing from a no-name startup company with no track record?

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.

Replies

  1. wjrobinson | | #1

    Flir is $250 on sale right now. Both interesting, the Flir pic looks better to me even though it has less resolution.

    This post looks like and ad for Seek....

  2. Aaron Birkland | | #2

    AJ - I noticed that too. Of the comparison images in the articles, the FLIR looked better to my eye. At least one of the articles mentioned that the additional context (i.e. physical surfaces overlayed over heat map) made it easier to understand what was going on in the image. I'd really like to know what difference it makes in practice, when attempting to use either one as a tool in diagnosing heat loss from a home.

    But I guess my question would be - can these cheap smart phone thermal cameras serve as useful tools or homeowners who wish to diagnose visible heat loss characteristics of buidings? If not, then what's the catch? If so, then is there a significant difference in usability or data quality between the FLIR One and the Seek? Does anybody have any experience?

    Sorry if this question seemed like an ad - I just discovered it last night, and seemed enthusiastic. I had known about the FLIR One, but wrote it off as "not applicable" as it only works for a specific Apple iphone model. The seek seemed to occupy a sweet spot for me, almost too good to be true. Specs can be deceiving, so I'm still waiting to find out what's wrong with it.

  3. davidmeiland | | #3

    I'd have to try one out to know how it performs, compared to a "real" thermal imager. Seems to me that only a tiny percentage of homeowners and other non-construction-types are going to be interested in IR, at least as it related to buildings. Folks prowling around in the dark are a more likely customer.

  4. wjrobinson | | #4

    I'm thinking Flir just got started and is lowering the price of the iphone 5 unit and will most likely announce new units for newer Iphones... just guessing.

    I don't think the pixel number makes a huge difference for what us builders would use one for.

    My thought was to buy a used iphone cheap someday and pop a FLIR on it.

  5. Tim C | | #5

    I'm the proud owner of a Seek Thermal. Short version: it's adequate for DIY "energy audit" usages. It's also cheap crap that's cut as many corners as possible to get the price down, but if you were expecting anything else for $200 retail, you're delusional.
    Some sample images: http://i.imgur.com/KOazJkn.jpg shows my porch door. Variations in temperature & emissivity make it pretty easy to see all the details here. You can obviously see the weatherstripping on the bottom right isn't sealing effectively.
    http://i.imgur.com/GbojYnc.jpg shows my fireplace. FLIR's MSX would make it a lot easier to understand what you're looking at here, but it's not really necessary - while using the device, I knew it was my fireplace because I was pointing it at my fireplace.
    I've got other pictures, and can take more, if anyone is interested in seeing something specific.

    As far as "catches", making it seem too good to be true, there are basically two. 1. The Seek Thermal microbolometer uses 12 micron pixels, rather than the 17 micron pixel the FLIR One and other lower end thermal imagers use. Without getting into complicated physics that I don't really understand, that means the Seek's pixels are lower quality than the FLIR One's (although it does still have an effective resolution advantage, it's smaller than it appears). 2. The Seek Thermal suffers from a design flaw that causes it to see a thermal gradient of 5-9 degrees F (seems to vary by device) after it's warmed up, with the lower left corner measuring hotter than the upper right. This can theoretically be corrected for in software, but currently it is not. (That's why my pictures are in grayscale; it's easier for me to see the local contrast and ignore the gradient than with the typical hot body false color for thermal images)

  6. Aaron Birkland | | #6

    Tim C: Thanks for the first-hand account! Can you see thermal bridging from the outside, looking at the exterior of the house?

    As far as the pixel size, could it be that it's diffraction limited? It claims to be sensitive to infrared in the 7.2-13 μm range. If I'm doing the math right, at the midpoint wavelength of 10μm a lens at f/2.0 would create an image with an Airy diameter of 48.8μm, meaning that the image would be diffraction limited for pixels less than 24.4 μm. If so, you're right - I don't see much advantage to the smaller pixels

  7. Tim C | | #7

    Sure, here's an exterior image: http://i.imgur.com/PqSJiSV.jpg

    I didn't save a picture, but if you get up close, you can see the nails in the siding, which I thought was pretty cool.

  8. DIYJester | | #8

    One of the big box stores will rent out one of the $2000 FLIR cameras for $49 a day. If you get them at the end of the day, you will get a bit more time. For $49 I don't think you can beat it unless you are going to be doing IR regularly. If that is the case I don't think these $350 and $200 models compare.

    Make sure if you're going to rent one you do some research to understand how to operate the camera and the theories behind IR such as emissivity and reflections.

  9. charlie_sullivan | | #9

    I think there is value in being able to get a thermal image more routinely and casually. It can be a good way to spot problems that might have gone unnoticed. This looks like a great option for that. Obviously not what you want if your business is audits with a thermal camera, but better to have some indication of what is going on than none.

    Thermal imagers have been getting cheaper in general--for $1k you can get the capability that used to cost $10k. And you can get a basic but good quality FLIR 80x60 resolution model for $500 http://www.tequipment.net/FLIR/TG165/Building-and-Industrial-Thermal-Imagers/?v=7463, or the fluke model with lower but undisclosed resolution for as little as $333 http://www.tequipment.net/FlukeVT02.html?b=y&v=7749

    Personally, I might wait for the FLIR One to be regularly priced at $250 and Android compatible.

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |