This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Could the internal temperature sensor on MSP430G2 be so sensitive?

Other Parts Discussed in Thread: ENERGIA

Hi everyone,

I am puzzled about something which I am observing on MSP430 G2 LaunchPad:

I found some code on the Internet for Energia that supposedly reads the internal temperature sensor. The code itself was no more than reading analog 0:

int value = analogRead(A0); 

I then post the results to Ubidots. The Lauch Pad is literally 10 feet away from my desk yet I have noticed credible cause and affect relationship between the verifiable fact when human is present in the room and the readings produced by that int value = analogRead(A0); 

This is beyond coincidence as the relationship is concrete: I have tested it over time. So how could that sensor be so sensitive to human presence? Could it be that the sensor has nothing to do with temperature but the thing is sensing something else? Like air capacitance, infrared body emission (like the PIR sensors), erc

Here's the Ubidots graph (the human entered the room around 9pm -- no joke):

  • Hi Boyan!

    I doubt that the sensor recognizes someone coming into the room by the raise of room temperature. There are code examples for the processor which can be downloaded on the product website. There will also be one that uses the temperature sensor. I would have a look if the large scattering also appears with the example. Do you have some sort of filter applied to your data? A moving average or just the average over the last x values?

    Dennis
  • The sensor measures nothing but temperature. However, things like the building's heating, or air currents, might be (indirectly) affected by a human's presence.
  • I think that you are reading a pin A0 that is not connected to anything, which is why I think you are seeing all this noise - see the note here energia.nu/AnalogRead.html. As you can see in the pin diagram here: energia.nu/Guide_MSP430LaunchPad.html A0 is simply P1.0 on a pin rather than being connected to the internal temp sensor.

    On MSP430G2xx parts, the temp sensor is on ADC input channel 10 (see MSP430x2xx user's guide www.ti.com/.../slau144), so try analogRead(A10) instead.

    Regards,
    Katie

  • Oh, yeah, totally missed that. Somehow I read A10 not A0.
  • Thanks everyone, A10 it is so this is just an empty pin but here's the kicker: after spending some time today "testing lol" it appears that it is NOT human presence but light. When the lights are ON I see the higher reading and then OFF it goes back to the baseline of noise.

    Anyone has any clue how that could be? There clearly isn't any light sensor connected to the pin as all of you have pointed out so we are talking a floating pin on the Launch Pad.

    I've read the Raspberry Pi's sensitivity to flash due to dye exposure or poop packaging so I am wondering whether this maybe something similar. 

  • Hi Boyan,

    A lot of lights in buildings give off electrical noise so you may be seeing some effect from that. I knew someone once that connected a really long wire (like 1 meter) to a pin and they could see all sorts of noise glitches on the pin when they would turn on/off the overhead fluorescent lighting tubes.

    Regards,
    Katie

**Attention** This is a public forum