We did some expriments with TMP006. In one test setup, I got quite large negative values for Vobj. The register says: 0xEC30. TDie = 0x10F0. The die temperature is 33 °C and the object seen has quite the same temperature. The reaction of "vobj" seems good. Applying an object more warm increases the value in Vobj register.
Using your equations in SBOU107 document, the Result of (TDIE4 + (F(Vobj) / s) ) is negative. (I have no callibration value S0 yet. I took 6E-14 for default) With having a negative value, I cannot calculate a 4th root of it.
What may be wrong? Did we damage the sensor while sodering?
Thanks for help.
I take it you are doing these experiments using your own board. You said you got these negative Vobj results with one test setup. Did all your other test setups work properly?
Have you followed the layout guidelines given in the TMP006 User Guide? Do you have a solder mask layer over the copper "dot" which is directly underneath the sensor? What soldering method was used? Did you properly clean off all solder flux?
Ian WilliamsLinear Applications EngineerPrecision Analog and Sensing Products
We were not able to follow the exact layout guidelines. I setup the TMP006 in a laboratory environment. We etched a single sided pcb on our own and applied the whole thing to an existing µC Board. The Sensor was soldered by hand with hot air. I must admit, that this is quite a lot stress to the device. I got 5 samples of TMP006. We allready soldered 3 of them. 2 of 3 are working quite well. One is "out of range" as described before.
What I am wondering: Your equations in sbou07.pdf don't work with the full range of "VObj" and "TDie", which can be read from the registers. The "damaged" sensor shows a change on "Vobj" Register, when I apply different temperatured surfaces to measure. So the sensor seems to work. But it seems to be "out of range".
If I now decide to use your tmp006, can I be sure, that (with correct handling and soldering) no sensor will be "out of range?" Or is it possible to apply an initial correction factor (early calibration value) so that there will be no spare root of a negative value?
What are the specified limits of the values of registers Vobj and Tdie, which are allowed? Do you ship only sensors which are in that limit?
Thanks for your help
Soldering the TMP006 by hand will produce unpredictable results as you have observed. You must follow the assembly procedure given in this document: TMP006 Layout and Assembly Guidelines. We provide recommendations for proper handling as well as the thermal profile for reflow. Following our recommended layout is also important as deviations from this will introduce errors that require more work to calibrate out on the software/processing side.
A TMP006 which has been properly installed on a PCB with the correct layout will not return values for VOBJ and TDIE that result in negative numbers under the 4th root. All devices which are shipped have passed extensive testing. The cause of your failed board is damage to the TMP006's thermopile sensor due either to improper handling or excessive heat during hand soldering.
Thanks for your help.One of the sensors, we soldered by hand seems to work and we did some long-term measurement. We additionally added some conventional temperatur sensors to the objects and logged all data.
I am not able to calculate a TObj which makes sense.
I detected, that Vobj decreases, when the temperature of the measured object increases. If I understood it right, Vobj is a 2-compement of register vobj. That means: 0xFFFF is -156.25nV. 0x0001 is +156.25nV.
If I multiply vobj by -1 (change the direction), the result's are much better.
Is it possible, that I damaged my sensor that way, that it has swapped gain into negative?(By the way: The measured TDie-Value is absolutely correct)Or have I done a stupid mistake while interpreting your equations..... (may be possible)
I added an xls file, which contains all our measured data.
I've attached an Excel calculator tool which does the Tobj and Tdie calculations for you. With this calculator you can do a quick check on some of your data to make sure that your calculations are correct. However based on your description it seems like you have the correct understanding of how to handle the data from the Tobj and Tdie registers.
Again, it is possible that the sensor of the TMP006 could have become damaged and have the response which you are seeing. Another possibility is that if Tdie is heating up quickly enough, the sensor voltage will undershoot due to conductive heat transfer between the die and the PCB while the temperature stabilizes. If the temperature continues to heat up quicker than the system can stabilize, you will get a response like you show in the "Diagramm2" tab.
All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with respect to these materials. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.
TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs andembedded processors, along with software, tools and the industry’s largest sales/support staff.