Interesting.
The results of your test show that the camera is ‘aware’ of the temperature surrounding its microbolometer and the temperature of the microbolometer die. Sadly this is not great news as it eliminates the easily diagnosed issues that could be causing the symptoms that you are seeing
Some background on what happens when a microbolometer based thermal camera first starts. It is simplified but will give you an idea of what is happening and why.
1. The camera carries out a self test
2. The contents of the Flash memory are loaded into RAM
3. The camera starts its thermal imaging program and sub systems.
4. The temperature of the microbolometer die, and any Delta T associated with it, is measured
5. The ambient temperature within the camera is measured.
6. An FFC event is initiated to create a flat field image correction and to measure the FFC flag in order to self calibrate the measurement system.
7. A Delta T in the microbolometer die temperature is detected and once it crosses a set limit, another FFC is initiated to both create a flat field correction and reset the measurement calibration offset using the FFC flag temperature. (using the ambient temperature as the source of the FFC flag temperature)
8. Another Delta T in the microbolometer die temperature is detected and the above process is repeated. This process repeats until the microbolometer die temperature reaches thermal equilibrium (in a non-temperature controlled microbolometer system). Once the microbolometer die reaches thermal equilibrium (internally generated heat energy is equal to that lost to ambient via conduction, radiation and convection (though convection is limited by the vacuum within the microbolometer module) the Delta T reduces to a point where no FFC event is required for some time. Temperature drift in the microbolometer die will occur and an FFC event will be triggered based on a detected excessive drift in the pixel temperatures, die temperature or change in ambient temperature. The camera will also carry out a routine FFC event to correct for minor pixel drift and measurement calibration. The routine FFC event is usually set to around 120 second intervals to ensure a decent image and measurement accuracy. The well known exception to this is the SEEK Thermal thermal imaging core that needs to carry out an FFC every few seconds in order to combat thermal instability in its microbolometer die. I know of no other thermal imaging core that requires such regular FFC events, except upon initial power on whilst the microbolometer is warming up.
So from the above you will see that the bahaviour of your camera is anything but normal. Thermal equilibrium in a modern non-temperature stabilised microbolometer is normally achieved within 2 to 5 minutes and the FFC events become much less frequent after less than a minute. Your camera appears to be behaving as though it is suffering from a significant Delta T in the microbolometer die that continues long after it should have decreased due to thermal equilibrium being achieved.
If we consider the thermal stability of a modern microbolometer we may discover a clue as to what is happening in your camera. In most modern budget microbolometer based thermal cameras there is no Peltier temperature stabilisation of the microbolometer die, even if such is provisioned in the microbolometer module. A Peltier temperature stabiliser is a nice enhancement to a microbolometer thermal camera but it comes at the cost of increased system power consumption as Peltier modules are power hungry. Most budget cameras operate in thermal equilibrium mode, as detailed above. It is important to employ enough thermal mass in the microbolometer to provide decent short term thermal stability, even if not stabilised at a fixed temperature. To achieve adequate thermal mass in the system, it is not uncommon to mount a large metal heatsink to the rear of the microbolometer module. The addition of the heatsink also helps to dissipate the heat being produced within the Read Out IC that is integrated into the microbolometer module. This is important as the enemy of the microbolometer is excessive local heating that can thermally contaminate the pixel array. If the heatsink becomes separated from the microbolometer it could cause temperature instability in the microbolometer and potentially create a situation where the FFC sub system is having to cope with excessive Delta T in the microbolometer die. Whilst the regular FFC events will try to maintain measurement accuracy through recalibration using the FFC flag temperature, there will, of course, be frequent undesirable FFC events for the user.
I suggest you dismantle your camera and visually inspect it for problems around the microbolometer PCBA area.
If there is not a problem with the heatsink coupling to the microbolometer, you are faced with the unwelcome possibility of a fault causing localised heating of the microbolometer die.
Fraser
Home
Help
Search
Login
Register
