Is there a way to calibrate a FLIR C5 at home?

[finetuned]

I'm fairly new to thermal imaging. I have a FLIR C5 that we use around the house. The temperature calibration for low temperatures is way off: the FLIR is reporting about 10-15 °C too low, even with sensible settings for emissivity and distance.

I'm wondering if there is a way to change the calibration yourself (without sending it back to a service provider that is). I don't need a very accurate calibration, but I'd at least like to anchor the calibration at 0 °C (pointing at melting ice in water) and 100 °C (pointing at boiling water), to get it in the ballpark of being correct.

Is there special software that only FLIR has that can be used to calibrate these devices, and / or has anyone found a way to do this?

I apologize if this has been asked before; I tried searching the forums but nothing relevant seemed to come up.

Thanks!

[markb1980]

You need a thermal calibration target aka black body calibartor aka black body calibration target aka ... i guess you get it!
The 1st problem is that these devices cost more then the C5. - e.g. https://thermometer.co.uk/temperature-and-thermometer-calibration-equipment/1404-ir-500-black-body-calibrator.html
The 2nd problem you can just test if the camera is within spec but you can't adjust it!

Flir is charging more for calibrated camera and they charge also for the re-calibration or testing of the calibration each year.

To be hornest - I guess your test was not senceful. First you should learn how to setup the camera correctly. You have to understand what does which setting, etc. Then you should use emisivity stickers (1000 pcs cost a few Euros / USD) so that you do the test with a know emissivity!

Then you need to know the exact temperature of the object and you have to check if the reading from the camera is correct. Noch each device is ideal for that e.g. setting my soldering iron to 390°C results in 340-350°C on the tip.

Your water / ice experiment has far to much variables!

You need something where you can controll the temperature presice (e.g. a black body calibrator) or you need 1-2 contact thermomethers / sensors to test the temperature and when both devices show basically the same value then you know the object temp. In the 2nd case you can use basically everything what can output heat on a constant level. Best way something what holds the heat for a while and need time to cool down.

After that you can measure the sticker and I am pretty sure the camera will be within the +/-2% or +/-2°C range. If it's then still so far off as you told I would RMA it!

[finetuned]

I understand the parameters for emissivity and environmental temperature, and I understand their effect on the measurement. In fact, adjusting these parameters to different values makes the measurement error worse.

My question was strictly: is it possible to change the calibration of the device, or is this somehow something only FLIR or an authorized party can perform?

[IR_Geek]

Generally only FLIR has the software to build and load new calibration tables into the camera.  Not aware of any third party but there may be someone who is authorized?

big problems with these types of cameras is lack of access to the RAW pixel count.  of course even with access you would still need good (calibrated) calibration sources.   

Another thing to consider is that the camera was likely factory calibrated only at 'ambient' temperature and there are adjustments that need to be made when at high and low air temperatures.   

Radiometry is tricky and not very friendly - especially when stuck with a factory calibrated instrument.

[IR_Geek]

How low of temp's are you looking at?  The spec's are vague on stuff below 0°C.   They only mention ambient air of 15 to 35 and the object needs to be above 0°C ... even that is +/- 3 degrees.   Starting in lower air temp, unknown optics temp, guess at emissivity of object and I could see easily 5 to 8 off.  The 10 to 15 degrees off is worrisome. 

https://www.flir.com/products/c5/?vertical=condition+monitoring&segment=solutions

At ambient temp. 15 to 35°C (59 to 95°F) and object temp. above 0°C (32°F), 0 to 100°C (32 to 212°F): ±3°C
(±5.5°F), 100 to 400°C (212 to 752°F): ±3%

[Fraser]

Finetuned,

Earlier generations of FLIR thermal camera contained Service Menu’s that offered both dead pixel map and calibration table updating. Those service menus were protected by a password that is now well known. FLIR removed the Service Menu from cameras Circa 2014, after the Ex series upgrade became known. They did not want the Service Menu to be accessible by users so removed it rather than improved its protection. When we look at the FLIR C5 and its kin, what we have is a LEPTON core with a host electronics package. The Radiometric version of the LEPTON is calibrated at factory and defects in calibration are normally detected at that point. The Lepton  does not offer any option for a user, or 3rd party manufacturer to update its calibration table. As with many thermal imaging cameras, the accuracy of measurement remains pretty close to factory unless a component fails or ages over time. Most 3rd party Calibration checks are just that, a check, not a re-calibration as they do not hold the required FLIR Factory calibration utilities.

The LEPTON radiometric core can suffer from issues with the ambient temperature thermistor that is attached to the front of it and connected to the electronics via conductive paint. Damage or corrosion to the conductive paint can cause wildly inaccurate measurements. The Lepton is not really intended to be a core that will be re-calibrated… if the core fails measurement calibration checks, a new Lepton will be fitted into the host. The host electronics, in this case, have no involvement in calibration errors.

LEPTON is an inexpensive core. It can suffer quality issues.

Fraser