Keeping your world up and running.®

A test: wing with frost

Frost thermal image analysis of airplane wing from an aviation safety expert

John Horrigan discusses an image of frost on an airplane wing


Active Frost: Radiation Cooling

I have changed the color palette to better show what was going on with this wing and tightened up the span to eliminate irrelevant background readings. This version shows what I’m after more easily.

One of the benefits that Smartview® [Thermal Imaging Analysis and Reporting Software] provides is very easy data export, so I can choose the palette that shows what I need. Especially when merging images, I can instantly get them all on exactly the same temperature span. Since we use the imager to expand our field of view, the ability to merge images and then sync the data is a force multiplier. Thanks to the support of Michael [Stuart] and Jordan [Schlicting], Smartview’s data export features have been improved significantly in the time since this image was new, allowing sharper rendering and post-process analysis.

Specific data about this image:

  • This is a calibration wing, originally from a business jet, now instrumented for use in de-icing methods and fluid proving.
  • The image was taken with a Ti25 in April of 2008.
  • The outside air temperature was hovering just above freezing, with dewpoint just below 0°C. This was a trial for a frost removal method. The wing was instrumented with contact temperature data loggers. We were waiting for the temperature to drop and the wing to cool to the point of frost formation. This image was taken ahead of the expected frost accretion time in an effort to establish a correlation between the imager and the probes ahead of the test. We time-synched the imager to the clock used for the data loggers. It was during this very first run that I thought I had an error to deal with.
  • The black zones are bare aluminum patches. The imager reported their temps as off-scale low. I knew emissivity would be poor and that I would be reading mostly reflected temps from these patches. That said, as the image shows, the values coming from the patches during a springtime event made me realize that the sky was a much colder background than we had expected. As a result, the radiation transfer was probably more aggressive than we had counted on, which meant frost formation would be occurring earlier than expected.

We didn’t initially see frost and the contact probes were not yet indicating a temperature for frost formation, yet the imager also showed an area on the wing that looked colder than what the probes were reporting. Again an assumed error, but just to be sure, I visually checked the spot and, voilà - the area had a very light haze that could have been mistaken for dew by a human observer, but was in fact the very first sign of frost sublimation. The curvature of the wing in that area provided a nearly vertical exposure to the clear night sky and the paint was weathered. The paint color also made it difficult to see this stage of frost formation. How many times had this occurred in industry without detection? That was just one of the many questions that have defined the time since that shot.

The rest of the night revealed a lot more to us, some of it specific to the technology we were testing and how it actually interfaced with the surface. Unfortunately I can’t share this info as easily.

Learn more

Go to the main article »
Find out more about the collaboration »