Basics Reviews

How we test with thermal imagers: Infrared technology explained and how to use it correctly

Different surfaces are the death of every standup measurement

Every surface (different materials and structures) has a certain emissivity, which must first be known in order to be able to measure anything accurately at all. Better handhelds can be calibrated manually or you can use a suitable software of the manufacturer. Unfortunately, a value of 1.0 is almost always preset by default, which is never or only very rarely encountered in reality. Thus, at least all measurements where the emission level was set to 1.0 can be confidently chalked up to expensive entertainment that is of no use to anyone. Because:

Without prior input or indication of the respective emissivity of a measuring point, any measurement carried out is completely worthless!

A nice example of this is the following shot. The different colors actually visually recreate the temperature curve, but what happened to the heatpipe and VRM heatsink? Although they are extremely hot, they appear rather dark and thus cold in the picture! Only by applying our special varnish from the “poison cupboard” (orange measuring points) usable measuring results are obtained”. Of course, one could also store the emissivity of the copper composite material and the aluminium for each of the measuring points, but would then encounter the problem that these emissivities in particular change strongly with increasing temperatures.

Therefore, we use a very temperature-resistant, somewhat matt clearcoat, whose emissivity is known to us and remains almost constant over the entire temperature range. Such special and unfortunately also very expensive lacquers are used by the industry to protect electrical components and even entire circuit boards from environmental influences such as high humidity. One speaks therefore also colloquially of a tropicalization. Unfortunately, you won’t find such lacquers in normal shops, especially as they have to be tested (or had tested) for our purpose in elaborate series of measurements beforehand.

Without special varnish and Messtape nothing works (except crooked)

For heatpipes (GPU-, CPU-cooler) up to 100°C we can however also fall back on our matte black Messtape, whose purchase is also a little cost-intensive, but without which in the end all the measurements would turn out rather silly. Let’s just take a look at the illustration in our Fundamentals article from the year before last. While the measurement of the 80°C hot heatpipe with the factory default of 1.0 fails grandiosely, the measurement with the measured tape (0.656) as well as the measurement with the correctly entered emissivity for the material (0.12) are correct. Unfortunately, the latter changes by a factor of three over the temperature spectrum that occurs and is thus completely useless in the event of possible temperature fluctuations.

The varnish is ready: We measure pins correctly

It is important, not only for high-resolution cameras, that no disturbing temperature differences due to different emissivities are recorded in the measuring range, as they have a negative influence on the output mean value. The next two illustrations show how extreme something like this can be. First, we measure a VRM pin (which is very hot in itself), which gives off heat directly to the copper traces and thus to the immediate vicinity of the PCB. Nevertheless, due to the much lower emissivity of the solder joint without correction for our measurement with the (incorrectly) read approx. 50°C, this area is first a kind of cold spot.

Now we paint part of the pins and find that suddenly the entire surface has the emissivity we know and can use.  A difference of 23°C between the two measurements at the identical measuring point is an order of magnitude that can decide about the success or failure of such a measurement!

Thus, without the most mundane basic knowledge of measurement technology as well as the necessary basic requirements for the measurement setup and procedure, much too low measurement results are almost always the regrettable rule.

We keep the perspective!

But what good is the best infrared camera if you can’t see inside closed systems? We have also come up with a trick for this, but the transmittance of a material must be known. In order to be able to “see inside” closed computer housings, for example, we need a foil that is as thin as possible and whose transmittance (thermal transmittance) is still high enough to be able to use it for plausible measurements.

In this way the original airflow in the case is preserved and also the waste heat cannot escape so easily through an open side panel. The following figure shows a system measurement with a total of nine different measuring points, which were previously defined individually with the correct emissivity and transmittance.

Of course, we don’t want to withhold what a housing measurement with a “simple” handheld camera looks like. For this purpose, we have used our own archives:

Nowadays you find in handhelds a simple black/white VGA camera with 640 x 480 pixels, on whose picture the extremely coarse measured values of a mostly only 80 x 60 pixels large cheap bolometer are interpolated. This certainly looks PR-like crisp, but is at most suitable for finding coarser hotspots, e.g. in electrical lines, PV elements or raw lines. These parts are pure indicators, but not real measuring devices. If you think you can determine anything more precisely with this, you should go straight back to page 2 and look up the reasons again. pure self-delusion.

Already years ago I developed together with Lian Li a suitable benchtable, which also works closed and which can de facto simulate a closed case well:

The story with the calibrated foil I included at that time, so that for the first time also measurements into the closed housing became possible:

Danke für die Spende



Du fandest, der Beitrag war interessant und möchtest uns unterstützen? Klasse!

Hier erfährst Du, wie: Hier spenden.

Hier kannst Du per PayPal spenden.

About the author

Igor Wallossek

Editor-in-chief and name-giver of igor'sLAB as the content successor of Tom's Hardware Germany, whose license was returned in June 2019 in order to better meet the qualitative demands of web content and challenges of new media such as YouTube with its own channel.

Computer nerd since 1983, audio freak since 1979 and pretty much open to anything with a plug or battery for over 50 years.

Follow Igor:
YouTube Facebook Instagram Twitter

Werbung

Werbung