Reflected temperature with thermal camera.

[treez]

Hi,
We are using Flir E8 thermal cam.

A diode that is obviously much hotter to touch then a different diode, is showing up on the thermal cam as being 20degC cooler.
Could this be due to the fact that the hotter-to-touch diode is next to a silvered SMD electro cap which is reflecting its temperature onto the diode, and THUS making it seem cooler than it is?

Flir E8

https://www.flir.com/globalassets/imported-assets/document/flir-ex-series-user-manual.pdf

Support for FLIR E8 | Teledyne FLIR

Get support for FLIR E8. Find contact information, downloads, and other resources.

www.flir.co.uk

 

[FvM]

No. The measured temperature of packages with sufficient emissivity isn't affected by reflection effects. Either different power dissipation or thermal resistance due to board layout.

 

[treez]

Thanks, I see your point...the thing is, they are both SMC diodes...one is far hotter to touch then the other...and yet the hotter_to_touch one shows up 20degC cooler on the thermal cam.

I am wondering about the distance the thermal cam is held away from the components aswell.
Flir E8 only has options for integer numbers of feet (!!) away from the target.....0,1,2,3,4....

......but i am holding it just 10cm away from the components........and so i have to select "zero feet" because thats the nearest option to what i am doing...but its obviously incorrect...i wonder how to get round this?

 

[FvM]

I know that you have a favour for making simple things complicated.

Distance correction is taking care for the rather small effect of atmospheric absorption and emission and mainly relevant for far distance measurements, e.g. of buildings. 10 cm is simply 0 feet in this respect.
You may want to check yourself how much the distance parameter affects the measurement.

I have no idea how you manage get the apparently conflicting results. If you're unsure about the IR measurement, stick thermocouples to the components. If you doubt the material emissivity, use black lacquer, paper stickers or adhesive tape to get a surface of defined emissivity. If you suspect airflow or radition of other objects to affect the measurement, use a shield.

 

[BradtheRad]

Do the diodes show equal temperature when everything is shut off and cooled down?

How about 10 seconds after turn-on? Do the readings make sense?
A minute?
Ten minutes?

What amount of time elapses until the readings don't make sense?

 

[treez]

I suspect if you are holding the cam 10 cm away, but have selected "zero distance", then the shutters are receiving radiation from a wide input angle and so perhaps this is causing the problem? I have bought some matt grey paint (couldnt find matt black in a small tin), so will try that , but the SMC diode packages are matt black anyway.

 

[FvM]

Read the manual. Distance setting has nothing to do with "shutters". Just a minor correction of temperature calculation.