• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 31 subscribers
  • Views 3803 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TMP007 IR Sensor - Minimum object distance

Miriam Savage
Prodigy 80 points
Other Parts Discussed in Thread: TMP006

Hello, 

I've been attempting to calibrate a TMP007 for skin temperature measurement using the EasyCal software. What I've found is that when the skin is too near, the voltage readout drops off. I cannot find any information on the minimum distance an object should be placed from the sensor. Is there some information you could provide me with on this? The way the application is set up, the sensor is 1mm from the skin. 

Regards, 

Miriam

  • 0
    Prodigy 80 points
    Also, if you had any application notes or design guidelines for designing a lens for the TMP007, that would be helpful.

    Thanks,
    Miriam
  • 0
    TI__Mastermind 43715 points
    Hi Miriam,

    There isn't a minimum distance for TMP007. The voltage reported from the thermopile is the difference between object temperature and device temperature. If temperatures are the same, then the voltage should be zero.

    I apologize, but we don't have any guidelines for lens designs at this time. App notes and guidelines are always available at the product landing page. www.ti.com/.../technicaldocuments Documents for TMP006 may also be applicable. www.ti.com/.../technicaldocuments

    Ren
  • 0 in reply to Ren Schackmann
    Prodigy 80 points

    Hi Ren, 

    Thanks for the information. The TMP007 chip I'm using is enclosed in a housing, with an opening in the housing over the FOV of the sensor. The housing is very thin and thus the object (skin) is placed at only about 1mm from the sensor.

    What I'm seeing is that when the skin is placed slightly further away (maybe 2mm), I get an expected reading from the sensor - an initially positive voltage that approaches 0 as the sensor heats up towards the object temp. However, when the skin is placed up against the housing, the initially positive voltage reading quickly turns to a negative reading. This is throwing off my calibration, and I am finding it very difficult to get coefficients that give accurate output temperatures across the range I need. 

    Would you have any ideas as to how I could avoid the negative voltage reading? It is not an option to have the sensor further from the object in this instance. I've read that the negative reading occurs when energy inflow < energy outflow - if you could give me some more insight into this, that would be great. 

    Kind regards, 

    Miriam

  • 0 in reply to Miriam Savage
    TI__Mastermind 43715 points
    Hi Miriam,

    The negative voltages are normal and can't be avoided. In fact, it is necessary to report negative voltages in order to measure an object that is colder than the sensor.

    Are you only experiencing the negative voltages momentarily when the sensor is experiencing a change in temperature? This should be compensated by adjusting transient correction. See page 7 of TMP007 Calibration Guide. www.ti.com/.../sbou142.pdf

    Ren
  • 0 in reply to Miriam Savage
    TI__Mastermind 36940 points
    Hi Miriam,

    Just a thought. When measuring the skin temperature especially a very close contact, the skin might create a bump at the opening of the FOV. This will shorten your distance may be less than 1mm.

    Aaron
  • 0 in reply to Ren Schackmann
    Prodigy 80 points

    Thanks Ren & Aaron, that makes sense. However in this case the skin temperature is warmer than the device temperature... Tdie heats up with the skin but slowly. 

    The negative voltage is not necessarily a problem but what is happening is that I am seeing an inverted temperature response. I can calibrate fine for one temperature, but when I put warmer skin in front of the sensor I see an inverted response - it gives a cooler temperature readout instead. What I would have to do to make the read outs accurate is to use a negative S0, which is outside the maths engine limits for that coefficient. 

    So this is why I'm wondering about trying to ensure the voltage stays positive. When skin is held a little further away, the voltage stays positive and warmer skin results in warmer readouts. 

    Any suggestions for improving this would be appreciated. 

    Miriam

  • 0 in reply to Miriam Savage
    TI__Mastermind 43715 points
    Is it possible that the skin is being cooled locally by contact with the housing? You said that Tdie does gradually heat up from skin contact.

    Ren