How accurate is non contact thermometer for electronic component temperature measure?

[treez]

How accurate is this thermometer for getting temperature of ....

a ferrite inductor.
a heatsink thats bare aluminium
a heatsink thats anodised aluminium (black)
an electrolytic capacitor.
a mosfet case (the black body of a to220)

Thermometer.
**broken link removed**

(Maplin N92FX)

 

[chuckey]

I have used an industrial grade infra red camera and found that on a large aluminium structure, its was not accurate at all because it was effected greatly by the surface finish. Actual temperature of structure ~ 40 C, cast attachments (dark grey and rough) ~ 50 read, shiney polished bits ~30 C. No reason why it can not be compensated for with tables for specific material finishes. These gizmos basically indicate how well a surface "launches" infra red waves.
Frank

 

[FvM]

To answer the question yourself, you need to study optical properties of materials in far IR range. Or perform some empirical tests.

Metallic surfaces will primarly reflect the temperature of other objects in the vincinity rather than showing their own temperature. Sticking a piece of self-adhesive paper or plastic tape to it will give reliable temperature readings.

Ferrite parts, IC packages, any plastic covered part are O.K.

I'm not completely sure for anodized aluminium, because the coating thickness and exact composition matters. The basic coating matter Al2O3 should be absorbing in the wavelength range of interest, thus anodized aluminium will probably show the right temperature.

 

[treez]

the ferrite surface looks highly polished, so do you still think it will be accurately measured?

 

[FvM]

the ferrite surface looks highly polished, so do you still think it will be accurately measured?

Yes according to my observations (with a thermo camera). Ferrites are metal oxides, not pure metal. Reflectance for visble light doesn't tell about IR properties. White lacquer, white paper or glass have e.g. high emissivity in the thermal wavelength range.