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 - very narrow temperature range despite calibration

Other Parts Discussed in Thread: TMP007

The TMP007 is being used in an application to measure skin temperature. The EasyCal kit was used and the calibration guide was followed. Temperatures were inconsistent before calibration and afterwards the measurements are more accurate. However the accuracy only exists in the narrow band of 35C to 37C. Outside of this range the accuracy drops. The target temperature range is not much larger 32C to 42C and accuracy is +/- 1C. There is a Dr Madre IR Thermometer that is being used as reference. 

This is the calibration procedure that was used:

  1. The calibrated temperature range is determined to be 32 degrees C to 42 degrees C. These temperature limits were converted to HEX in 2’s complement format.
  2. The high temperature limit was stored into the respective TMP007 registries, namely TOBJ object temperature high-limit (06h) and TDIE local temperature high-limit (08h).
  3. The low temperature limit was stored into the respective TMP007 registries, namely TOBJ object temperature low-limit (07h) and TDIE local temperature low-limit (09h).
  4. Obtained the calibration coefficients using the TMP007EASYCAL and store the results in the respective Coefficient Registers (0Bh-10h).
  5. Verified that the data was stored correctly. 

What could be done to increase the range of measurement? 

  • Hi Ahmad,

    Thanks for reaching out to us, our expert on this will be back next week. Thank you for your patience.

    Also what application is this for? And why are you interested in measuring human skin temperature?

    -Michael Wong
  • The application is a wearable body monitor in development by a company in Florida. I'm their FAE.
  • Hi Ahmad,

    Can you share the calibration data? We can move this conversation to the internal forum or email if you'd like.

    Thanks,
    Ren
  • The calibration data for the coefficient register values are:

    S0        0x4747

    B0        0xFF80

    B1        0x1B23

    B2        0x8290

    A1        0x01CB

    A2        0xFEE6

    C2        0x1190

     

    and the temperature limits:

    HIGH_LIMIT        0x3200

    LOW_LIMIT         0x0040  

  • Hi Ackie,

    Can you share the data that was collected using EasyCal? Also, if you could also share any data collected after calibration, and details of your calibration environment.

    Thanks,
    Ren
  • The calibration data is attached. The calibration was done on the arm (human skin) with the surrounding room temperature set at 74 degrees Fahrenheit.

    Thanks, 

    Ackie

    Calibration_Data.txt

  • Ackie,

    Thanks for the data. It looks like your TMP007 is getting very little IR signal. Have you placed it behind a window? Many plastics do not pass medium wavelength IR energy that is needed for TMP007. If you are using a cover material, please verify that it is transmissive in the 4 to 16um range. We recommend polypropylene.

    Ren
  • Ren,

    When using the TMP007EasyCal to obtain the calibration data, the device is not placed behind a window.  Also our device which uses the TMP007 is enclosed in a case. However, there is cutout in the case for the TMP007. The dimension of the cutout is equivalent to that of the TMP007. As a result the medium wavelength IR energy need for the TMP007 does not have to pass through the plastic.

    What is an efficient setup in which I can utilize the TMP007EasyCal device that will ensure that I am getting maximum IR signal to the TMP007 and therefore accurate calibration data ?

    Thanks

    Ackie

  • Can you verify that you get positive thermopile voltages when you point the device at a hot object? A cup of coffee should be OK if you don't have anything better handy. I'm still concerned that your device isn't working correctly. If it is working correctly, then your skin temp is always slightly cooler than device temp.

    The industry standard for IR source is a Black Body Calibrator. Omega makes a small, affordable instrument that we have used, but we use large (6" diameter) models for our characterization work.
    www.omega.com/.../BB703.html

    Ren
  • I am using the TMP007EasyCal Kit.   When I point the device at a hot object ( a cup of hot water) I get negative thermopile voltages. A copy of the data collected is attached. Does that mean that the TMP007EasyCal device is in error ?

    Ackie

    Hot Object.txt
    TMP007 Object Voltage (uV),TMP007 Local Temp (C),TMP112 Temp (C),Date,Time,Sample Number
    -625.781250,28.031250,24.500000,7/19/2016,2:04:41 PM,0
    -625.781250,28.031250,24.500000,7/19/2016,2:04:43 PM,1
    -625.781250,28.031250,24.500000,7/19/2016,2:04:44 PM,2
    -352.500000,29.062500,24.500000,7/19/2016,2:04:45 PM,3
    -352.500000,29.062500,24.562500,7/19/2016,2:04:46 PM,4
    -352.500000,29.062500,24.625000,7/19/2016,2:04:47 PM,5
    -352.500000,29.062500,24.625000,7/19/2016,2:04:49 PM,6
    -380.312500,29.718750,24.750000,7/19/2016,2:04:50 PM,7
    -380.312500,29.718750,24.687500,7/19/2016,2:04:51 PM,8
    -380.312500,29.718750,24.750000,7/19/2016,2:04:52 PM,9
    -383.593750,30.218750,24.875000,7/19/2016,2:04:53 PM,10
    -383.593750,30.218750,24.875000,7/19/2016,2:04:54 PM,11
    -383.593750,30.218750,24.875000,7/19/2016,2:04:56 PM,12
    -383.593750,30.218750,25.062500,7/19/2016,2:04:57 PM,13
    -387.343750,30.656250,25.062500,7/19/2016,2:04:58 PM,14
    -387.343750,30.656250,25.125000,7/19/2016,2:04:59 PM,15
    -387.343750,30.656250,25.187500,7/19/2016,2:05:00 PM,16
    -390.625000,31.156250,25.187500,7/19/2016,2:05:02 PM,17
    -390.625000,31.156250,25.312500,7/19/2016,2:05:03 PM,18
    -390.625000,31.156250,25.312500,7/19/2016,2:05:04 PM,19
    -454.687500,31.687500,25.437500,7/19/2016,2:05:05 PM,20
    -454.687500,31.687500,25.437500,7/19/2016,2:05:07 PM,21
    -454.687500,31.687500,25.500000,7/19/2016,2:05:08 PM,22
    -454.687500,31.687500,25.562500,7/19/2016,2:05:09 PM,23
    -464.843750,32.250000,25.500000,7/19/2016,2:05:10 PM,24
    -464.843750,32.250000,25.500000,7/19/2016,2:05:11 PM,25
    -464.843750,32.250000,25.562500,7/19/2016,2:05:13 PM,26
    -541.562500,32.875000,25.562500,7/19/2016,2:05:14 PM,27
    -541.562500,32.875000,25.625000,7/19/2016,2:05:15 PM,28
    -541.562500,32.875000,25.625000,7/19/2016,2:05:16 PM,29
    -514.531250,33.343750,25.625000,7/19/2016,2:05:18 PM,30
    -514.531250,33.343750,25.687500,7/19/2016,2:05:19 PM,31
    -514.531250,33.343750,25.687500,7/19/2016,2:05:20 PM,32
    -567.343750,33.968750,26.625000,7/19/2016,2:05:21 PM,33
    -567.343750,33.968750,29.562500,7/19/2016,2:05:23 PM,34
    -567.343750,33.968750,34.312500,7/19/2016,2:05:24 PM,35
    -567.343750,34.593750,42.750000,7/19/2016,2:05:25 PM,36
    -519.218750,34.593750,45.750000,7/19/2016,2:05:26 PM,37
    -519.218750,34.593750,48.562500,7/19/2016,2:05:28 PM,38
    -519.218750,34.593750,52.187500,7/19/2016,2:05:29 PM,39
    -577.968750,35.343750,53.437500,7/19/2016,2:05:30 PM,40
    -577.968750,35.343750,54.687500,7/19/2016,2:05:31 PM,41
    -577.968750,35.343750,56.500000,7/19/2016,2:05:33 PM,42
    -509.375000,35.656250,57.000000,7/19/2016,2:05:34 PM,43
    -509.375000,35.656250,57.812500,7/19/2016,2:05:35 PM,44
    -509.375000,35.656250,58.562500,7/19/2016,2:05:36 PM,45
    -554.375000,36.218750,59.062500,7/19/2016,2:05:38 PM,46
    -554.375000,36.218750,59.437500,7/19/2016,2:05:39 PM,47
    -554.375000,36.218750,59.875000,7/19/2016,2:05:40 PM,48
    -565.000000,36.718750,60.187500,7/19/2016,2:05:41 PM,49
    -565.000000,36.718750,60.250000,7/19/2016,2:05:43 PM,50
    -565.000000,36.718750,61.062500,7/19/2016,2:05:44 PM,51
    -539.843750,37.156250,61.187500,7/19/2016,2:05:45 PM,52
    -539.843750,37.156250,61.437500,7/19/2016,2:05:46 PM,53
    -539.843750,37.156250,62.187500,7/19/2016,2:05:48 PM,54
    -539.843750,37.156250,62.562500,7/19/2016,2:05:49 PM,55
    -490.625000,37.343750,63.250000,7/19/2016,2:05:50 PM,56
    -490.625000,37.343750,64.250000,7/19/2016,2:05:52 PM,57
    -490.625000,37.343750,64.562500,7/19/2016,2:05:53 PM,58
    -611.406250,37.937500,64.625000,7/19/2016,2:05:54 PM,59
    -611.406250,37.937500,64.937500,7/19/2016,2:05:55 PM,60
    -611.406250,37.937500,65.996094,7/19/2016,2:05:57 PM,61
    -551.093750,38.093750,65.996094,7/19/2016,2:05:58 PM,62
    -551.093750,38.093750,65.996094,7/19/2016,2:05:59 PM,63
    -551.093750,38.093750,65.996094,7/19/2016,2:06:00 PM,64
    -545.625000,38.312500,65.996094,7/19/2016,2:06:02 PM,65
    -545.625000,38.312500,65.996094,7/19/2016,2:06:03 PM,66
    -545.625000,38.312500,65.996094,7/19/2016,2:06:04 PM,67
    -504.687500,38.500000,65.996094,7/19/2016,2:06:06 PM,68
    -504.687500,38.500000,65.996094,7/19/2016,2:06:07 PM,69
    -504.687500,38.500000,65.996094,7/19/2016,2:06:08 PM,70
    -585.625000,38.937500,65.996094,7/19/2016,2:06:09 PM,71
    -585.625000,38.937500,65.996094,7/19/2016,2:06:11 PM,72
    -585.625000,38.937500,65.996094,7/19/2016,2:06:12 PM,73
    -535.468750,39.187500,65.996094,7/19/2016,2:06:14 PM,74
    -535.468750,39.187500,65.996094,7/19/2016,2:06:15 PM,75
    -535.468750,39.187500,65.996094,7/19/2016,2:06:16 PM,76
    -517.968750,39.343750,66.996094,7/19/2016,2:06:18 PM,77
    -517.968750,39.343750,66.996094,7/19/2016,2:06:19 PM,78
    -517.968750,39.343750,66.996094,7/19/2016,2:06:20 PM,79
    -451.406250,39.562500,66.996094,7/19/2016,2:06:22 PM,80
    -451.406250,39.562500,66.996094,7/19/2016,2:06:23 PM,81
    -451.406250,39.562500,66.996094,7/19/2016,2:06:24 PM,82
    -608.437500,39.968750,66.996094,7/19/2016,2:06:26 PM,83
    -608.437500,39.968750,66.996094,7/19/2016,2:06:27 PM,84
    -608.437500,39.968750,66.996094,7/19/2016,2:06:28 PM,85
    -542.968750,40.031250,66.996094,7/19/2016,2:06:30 PM,86
    -542.968750,40.031250,65.996094,7/19/2016,2:06:31 PM,87
    -542.968750,40.031250,65.996094,7/19/2016,2:06:33 PM,88
    -518.593750,40.218750,65.996094,7/19/2016,2:06:34 PM,89
    -518.593750,40.218750,32.996094,7/19/2016,2:06:35 PM,90
    -518.593750,40.218750,127.746094,7/19/2016,2:06:37 PM,91
    -504.687500,40.281250,127.746094,7/19/2016,2:06:38 PM,92
    -504.687500,40.281250,127.746094,7/19/2016,2:06:39 PM,93
    -504.687500,40.281250,127.746094,7/19/2016,2:06:41 PM,94
    -553.437500,40.468750,127.746094,7/19/2016,2:06:42 PM,95
    -553.437500,40.468750,127.746094,7/19/2016,2:06:43 PM,96
    -553.437500,40.468750,127.746094,7/19/2016,2:06:45 PM,97
    -562.031250,40.593750,127.746094,7/19/2016,2:06:46 PM,98
    -562.031250,40.593750,127.746094,7/19/2016,2:06:48 PM,99
    -562.031250,40.593750,127.746094,7/19/2016,2:06:49 PM,100
    -590.312500,40.781250,127.746094,7/19/2016,2:06:50 PM,101
    -590.312500,40.781250,16.496094,7/19/2016,2:06:52 PM,102
    -590.312500,40.781250,127.746094,7/19/2016,2:06:53 PM,103
    -542.031250,40.718750,127.746094,7/19/2016,2:06:54 PM,104
    -542.031250,40.718750,127.746094,7/19/2016,2:06:56 PM,105
    -542.031250,40.718750,127.746094,7/19/2016,2:06:57 PM,106
    -572.812500,40.968750,127.746094,7/19/2016,2:06:58 PM,107
    -572.812500,40.968750,127.746094,7/19/2016,2:07:00 PM,108
    -572.812500,40.968750,127.746094,7/19/2016,2:07:01 PM,109
    -536.562500,41.093750,127.746094,7/19/2016,2:07:02 PM,110
    -536.562500,41.093750,127.746094,7/19/2016,2:07:04 PM,111
    -536.562500,41.093750,32.441406,7/19/2016,2:07:05 PM,112
    -500.781250,41.250000,32.441406,7/19/2016,2:07:06 PM,113
    

  • Ackie,

    I'm attaching a picture of me using the EasyCal EVM on a styrofoam cup containing hot water from our coffee machine. Can you verify that you are testing this way? Styrofoam has a poor emissivity of 0.6, so it does not produce as much IR as a black object. It's not ideal for this test. Still, I was able to produce 150-200uV on my thermopile. If you hold the thermopile above the water, it will produce unstable readings due to condensation collecting on the device.

    If you can't get a positive voltage from your thermopile, then it must be damaged, obstructed, or thermally shorted in some way.

    When you look at the unmounted TMP007, the thermopile is a cavity in the middle of the device. This cavity is sealed with a thin membrane on the bottom, solder side of the device. Handling the solder side of the device during assembly can damage this membrane. Additionally, anything left beneath the device by the solder process, such as flux residue, can have the effect of thermally shorting the thermopile to the PCB or itself.

    If you have not already, please test other assemblies to see if you get the same results.

    Ren

  • We received another TMP007 EasyCal device. Using the test setup that you demonstrate, I obtain positive voltages from the thermopile when measuring the temperature of Hot water and negative voltages for ice cold water. Copies of the data are attached. Whereas I received negative voltages in both cases when using the previous easycal device. 

    I will go through the calibration process and will let you if any issues occur. Thanks for your help troubleshooting this problem.

    A-Bon 

    Cold.txt
    TMP007 Object Voltage (uV),TMP007 Local Temp (C),TMP112 Temp (C),Date,Time,Sample Number
    29.687500,25.531250,25.312500,7/26/2016,10:38:22 AM,0
    18.125000,25.531250,25.312500,7/26/2016,10:38:24 AM,1
    -22.031250,25.500000,25.187500,7/26/2016,10:38:25 AM,2
    -34.531250,25.468750,25.125000,7/26/2016,10:38:26 AM,3
    -20.468750,25.437500,25.062500,7/26/2016,10:38:27 AM,4
    -49.531250,25.375000,24.875000,7/26/2016,10:38:28 AM,5
    -65.312500,25.375000,24.187500,7/26/2016,10:38:29 AM,6
    -67.187500,25.343750,23.125000,7/26/2016,10:38:30 AM,7
    -75.312500,25.312500,22.750000,7/26/2016,10:38:31 AM,8
    -76.562500,25.281250,22.500000,7/26/2016,10:38:33 AM,9
    -73.437500,25.250000,22.437500,7/26/2016,10:38:34 AM,10
    -71.406250,25.218750,22.000000,7/26/2016,10:38:35 AM,11
    -80.000000,25.187500,21.750000,7/26/2016,10:38:36 AM,12
    -72.343750,25.156250,20.937500,7/26/2016,10:38:37 AM,13
    -75.468750,25.125000,20.562500,7/26/2016,10:38:38 AM,14
    -72.187500,25.125000,20.437500,7/26/2016,10:38:39 AM,15
    -69.218750,25.125000,20.000000,7/26/2016,10:38:40 AM,16
    -80.781250,25.125000,19.750000,7/26/2016,10:38:41 AM,17
    -77.812500,25.093750,19.500000,7/26/2016,10:38:43 AM,18
    -75.781250,25.062500,18.437500,7/26/2016,10:38:44 AM,19
    -78.593750,25.062500,18.125000,7/26/2016,10:38:45 AM,20
    -72.656250,25.062500,17.750000,7/26/2016,10:38:46 AM,21
    -57.031250,25.031250,17.500000,7/26/2016,10:38:47 AM,22
    -80.937500,25.031250,17.187500,7/26/2016,10:38:48 AM,23
    -75.468750,25.031250,16.625000,7/26/2016,10:38:49 AM,24
    -69.843750,25.031250,16.437500,7/26/2016,10:38:50 AM,25
    -72.343750,25.000000,16.062500,7/26/2016,10:38:51 AM,26
    -78.281250,24.968750,15.875000,7/26/2016,10:38:53 AM,27
    -62.656250,24.968750,15.687500,7/26/2016,10:38:54 AM,28
    -81.718750,24.937500,15.000000,7/26/2016,10:38:55 AM,29
    -87.968750,24.937500,14.750000,7/26/2016,10:38:56 AM,30
    -70.937500,24.937500,14.625000,7/26/2016,10:38:57 AM,31
    -70.000000,24.937500,14.312500,7/26/2016,10:38:58 AM,32
    -57.031250,24.875000,14.187500,7/26/2016,10:38:59 AM,33
    -63.125000,24.875000,14.000000,7/26/2016,10:39:00 AM,34
    -69.375000,24.875000,13.750000,7/26/2016,10:39:02 AM,35
    -62.187500,24.875000,13.500000,7/26/2016,10:39:03 AM,36
    -66.718750,24.875000,13.312500,7/26/2016,10:39:04 AM,37
    -67.500000,24.843750,13.187500,7/26/2016,10:39:05 AM,38
    -66.093750,24.843750,13.000000,7/26/2016,10:39:06 AM,39
    -19.531250,24.875000,13.562500,7/26/2016,10:39:07 AM,40
    11.406250,24.906250,13.812500,7/26/2016,10:39:08 AM,41
    5.937500,24.906250,14.062500,7/26/2016,10:39:09 AM,42
    -21.718750,24.906250,14.437500,7/26/2016,10:39:11 AM,43
    3.593750,24.937500,14.750000,7/26/2016,10:39:12 AM,44
    

    Hot.txt
    TMP007 Object Voltage (uV),TMP007 Local Temp (C),TMP112 Temp (C),Date,Time,Sample Number
    -0.312500,24.406250,23.562500,7/26/2016,10:35:52 AM,0
    -1.562500,24.406250,23.562500,7/26/2016,10:35:54 AM,1
    -1.406250,24.437500,23.437500,7/26/2016,10:35:55 AM,2
    38.750000,24.468750,23.625000,7/26/2016,10:35:56 AM,3
    158.437500,24.531250,23.625000,7/26/2016,10:35:57 AM,4
    228.281250,24.687500,25.250000,7/26/2016,10:35:59 AM,5
    257.656250,24.781250,29.687500,7/26/2016,10:36:00 AM,6
    261.718750,24.843750,32.750000,7/26/2016,10:36:01 AM,7
    261.406250,24.937500,36.375000,7/26/2016,10:36:02 AM,8
    255.781250,25.000000,38.125000,7/26/2016,10:36:03 AM,9
    255.000000,25.062500,40.125000,7/26/2016,10:36:04 AM,10
    255.937500,25.218750,42.312500,7/26/2016,10:36:06 AM,11
    260.000000,25.281250,45.812500,7/26/2016,10:36:07 AM,12
    253.750000,25.375000,47.375000,7/26/2016,10:36:08 AM,13
    282.656250,25.500000,48.750000,7/26/2016,10:36:09 AM,14
    256.093750,25.593750,50.812500,7/26/2016,10:36:10 AM,15
    242.031250,25.750000,51.687500,7/26/2016,10:36:12 AM,16
    242.187500,25.812500,52.375000,7/26/2016,10:36:13 AM,17
    256.562500,25.906250,54.062500,7/26/2016,10:36:14 AM,18
    258.906250,25.937500,54.750000,7/26/2016,10:36:15 AM,19
    250.937500,26.031250,55.250000,7/26/2016,10:36:17 AM,20
    287.812500,26.187500,55.687500,7/26/2016,10:36:18 AM,21
    260.312500,26.281250,56.437500,7/26/2016,10:36:19 AM,22
    223.906250,26.343750,56.812500,7/26/2016,10:36:20 AM,23
    230.156250,26.437500,57.312500,7/26/2016,10:36:21 AM,24
    251.718750,26.531250,57.937500,7/26/2016,10:36:23 AM,25
    247.500000,26.593750,58.125000,7/26/2016,10:36:24 AM,26
    250.625000,26.625000,58.437500,7/26/2016,10:36:25 AM,27
    244.375000,26.687500,58.875000,7/26/2016,10:36:26 AM,28
    234.843750,26.750000,59.000000,7/26/2016,10:36:28 AM,29
    248.125000,26.875000,59.312500,7/26/2016,10:36:29 AM,30
    200.625000,26.875000,59.187500,7/26/2016,10:36:30 AM,31
    30.000000,26.843750,58.062500,7/26/2016,10:36:31 AM,32
    40.468750,26.781250,56.937500,7/26/2016,10:36:32 AM,33
    36.250000,26.781250,56.375000,7/26/2016,10:36:34 AM,34
    

  • Thanks for updating me. I'm glad you've solved the issue. I'm very sorry that you encountered problems with our EVM. We'll be looking into ways to improve our assembly partner's process.

    Ren