STHS34PF80 T_OBJECT interpretation

[tuanh]

Hello,

First time posting in this forum!

I'm evaluting the STHS34PF80 sensor and went through these materials: Datasheet & AN5867.

However, I could not find any information regarding the interpretation of the object temperature (after calculating the signed value of 16 bit (TOBJECT_L and TOBJECT_H), then divide by the sensitivity).

My very first thought was that this represents the so-called absolute object temperature value, as the sensor is T-MOS based technology and its sensitivity is calibrated with a black body covering the full field of view. However, the function of f(TOBJECT_L, TOBJECT_H, SENS_DATA), doesn't give me that.

Rather I have a feeling that it gives me the difference between the object temperature and the ambient temperature. I have attached an image to show you how the values changed when I have my hand approaching the sensor, to completely cover it, then get my hand off the sensor.

I have tried with both the ST driver: https://github.com/STMicroelectronics/sths34pf80-pid and implemented my own driver but the results are the same.

Also during the test, I noticed that the presence and motion flag are somewhat inconsistent. But that's another story as these flag depends on various factors such as threshold, algorithm, ambient temperature, object temperature, compensation, .etc. I haven't set anything else as the object temperature is bugging me. 

Now here are the questions:

1. Could you please provide me guidance about this?

2. Is there any other materials that I can go through to have a better understanding at this IC?

3. I'm planning to commercialise products with this IC, is there any other source of contact I can do beside this Community Forum?

Thank you and looking forward to hearing from you,

Tu

[Federica Bossi · Accepted solution]

Hi @tuanh ,

Welcome to ST Community!

The TOBJECT data does not represents an actual temperature value. It represents the net amount of radiation (radiant power per unit of area) in the infrared spectrum that is received by the sensor.
The net amount of radiation is described by a simple formula derived from the Stefan-Boltzmann law (which links radiation emitted by a body with its absolute temperature): TOBJECT [LSB] = τk(T_BODY^4 - T_ENV^4).
τ [-] is the transmittance of the optical system in use with the sensor.
k [LSB/K^4] is a constant of proportionality related to the Stefan-Boltzmann constant and to the sensitivity coefficient (which is calibrated for each unit with a blackbody at T_BODY = 297.15 K covering the full FoV of the sensor); this means that the formula is the more valid the more T_BODY is near the value of 297.15 K.
T_ENV [K] is the temperature in the immediate surroundings of the sensor. 
As for the T_BODY [K], the issue is more complex:
1. if one single body covers the full FoV of the sensor, T_BODY represents the value of temperature of such body;
2. if multiple bodies are inside the FoV of the sensor, T_BODY is related to the values of temperature of the various bodies inside the field of view depending on the amount of FoV that each of them covers in a way which is not possible to know a priori.
Please note that everything is complicated by the fact that the body/bodies inside the FoV may not be black bodies (i.e., their emissivity may be lower than 1) so the formula above may not be a true mirror of reality.