• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 1248 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.

TMP006 - Field of view restriction

Rob Wheeler
Prodigy 30 points
Other Parts Discussed in Thread: TMP006

We have an application in which we would like to replace our existing thermopile sensor with a TMP006.

 

However, the existing sensor has a field of view of only 35°. The distance from the sensor to the target is about 250mm.

 

We do not make sufficient units to justify bespoke semiconductor solutions and even manufacturing / supplying lenses for the quantities required, may prove prohibitive.

 

The target has a high emissivity (>0.8) and can run up to 160 °C. However, the environment of the sensor can also reach 140 °C.

 

We can think of three possible ways to limit the field of view:-

 

1)      Use a lens.

2)      Use a mask made of a low emissivity material (polished metal), thermally grounded to the sensor.

3)      Use a mask made of a low emissivity material thermally isolated from the sensor.

 

In our experience lenses are expensive and difficult to manage, especially in environments with high varying temperature.

 

Thermally connecting a mask to the sensor is also prone to difficulty, especially in trying to match the rate of flow of thermal energy with those of the sensor and PCB under the sensor. The only way we could think of to do this would be to fit a very thin optical shield, probably made in polished copper, extruded to fit as closely as possible to the profile of the sensor fitted to the PCB, over the sensor and connect it directly to the vias on the PCB.

 

Much easier would be to fit a thermally isolated mask, relying on the fact that, due to it’s, much lower, emissivity, it would have an insignificant effect on the overall measured result.

 

What, in your view, would be the best way in which to limit the FOV of the sensor, in this type of application?

  • 0
    TI__Mastermind 27715 points

    Hello Rob,

    My recommendation is to use a mask made of polished metal which is thermally grounded to the sensor. Your reasoning on using a low-emissivity material is 100% correct, but the mask must be coupled to the sensor GND as the temperature of the GND plane under the sensor is used as the cold junction/absolute temperature reference. Any temperature delta between sensor GND and the mask will result in a offset that causes measurement errors.

    For installation, simply place a footprint on your PCB around the TMP006, connected to GND, onto which you can solder the mask. 

    Best regards,

    Ian Williams
    Linear Applications Engineer
    High Performance Linear and Sensing Products 

  • 0 in reply to Ian Williams
    Prodigy 30 points

    Ian Williams said:

    Hello Rob,

    My recommendation is to use a mask made of polished metal which is thermally grounded to the sensor. Your reasoning on using a low-emissivity material is 100% correct, but the mask must be coupled to the sensor GND as the temperature of the GND plane under the sensor is used as the cold junction/absolute temperature reference. Any temperature delta between sensor GND and the mask will result in a offset that causes measurement errors.

    For installation, simply place a footprint on your PCB around the TMP006, connected to GND, onto which you can solder the mask. 

    Best regards,

    Ian Williams
    Linear Applications Engineer
    High Performance Linear and Sensing Products 

    Hi Ian,

     

    Many thanks for your previous answer.

     

    I have a few, subsequent, questions that I would be pleased if you could answer:-

     

    1. 1.       In our application, the sensor would be isolated from any other temperature generating circuitry (at the end of wiring or a flexi-rigid PCB).  Can you confirm that the isolating region around the copper sensor ‘island’ would be unnecessary in this case and that we could confine the PCB layout design for the sensor  to the copper ‘island’?

    The copper island would be connected to the system ground via a 10 mil trace.

     

    1. 2.       I would propose to shield the sensor from the unwanted FOV by use of a cylinder of polished aluminium, set around the sensor and connected to the copper sensor ‘island’ at both the AGND and DGND via holes (obviously  the bottom of the shield would have to be electrically isolated from the PCB (to prevent shorts to other traces). Would this seem to be a good plan to you?

     

    1. 3.       Would you agree that the shield should have as good a thermal conductivity as possible whilst having as low a thermal mass as possible? Therefore the cylinder should be made from a highly thermally conductive material, but as small and thin as possible whilst being capable of retaining its structural integrity.

     

    1. 4.       I have read somewhere else on the forum that the sensitive area of the dye is 330um x 330um. I need to know the exact position and dimensions of the sensitive area in order to be able to calculate the height of the shield. Can you just confirm for me that this is centred on the dye?

      

    Rob Wheeler

  • 0 in reply to Ian Williams
    Prodigy 30 points

    Hi Ian,

     

    Many thanks for your previous answer.

     

    I have a few, subsequent, questions that I would be pleased if you could answer:-

     

    1. 1.       In our application, the sensor would be isolated from any other temperature generating circuitry (at the end of wiring or a flexi-rigid PCB).  Can you confirm that the isolating region around the copper sensor ‘island’ would be unnecessary in this case and that we could confine the PCB layout design for the sensor  to the copper ‘island’?

    The copper island would be connected to the system ground via a 10 mil trace.

     

    1. 2.       I would propose to shield the sensor from the unwanted FOV by use of a cylinder of polished aluminium, set around the sensor and connected to the copper sensor ‘island’ at both the AGND and DGND via holes (obviously  the bottom of the shield would have to be electrically isolated from the PCB (to prevent shorts to other traces). Would this seem to be a good plan to you?

     

    1. 3.       Would you agree that the shield should have as good a thermal conductivity as possible whilst having as low a thermal mass as possible? Therefore the cylinder should be made from a highly thermally conductive material, but as small and thin as possible whilst being capable of retaining its structural integrity.

     

    1. 4.       I have read somewhere else on the forum that the sensitive area of the dye is 330um x 330um. I need to know the exact position and dimensions of the sensitive area in order to be able to calculate the height of the shield. Can you just confirm for me that this is centred on the dye?

      

    Rob Wheeler

  • 0 in reply to Rob Wheeler
    TI__Mastermind 27715 points

    Hi Rob,

    Sorry for the delay in my reply.

    1. If your design already has sufficient thermal isolation (and it sounds like it does), then the additional isolation region is not needed.

    2. Your implementation of a metal cylinder should do a fine job of restricting the FOV. Typically we use a square metal cover with a small aperture to do this, and I've attached some slides in the link below that give a design example. Just make sure that you calculate what your new effective angle of view is with the cylinder installed and that it's sufficient for your application. 

    Metal cover FOV limiting example: 4572.TMP006 Field of View Limiting.pdf

    3. You are spot-on about the properties of the metal cover. It should have as small of a thermal mass as possible, so the thinner the better. The metal covers we use are formed from very thin sheets, but they're sufficiently rigid to be practical for normal use.

    4. The thermopile sensor is indeed the 330μm x 330μm area directly in the center of the package. I've pasted a photo below which shows this relatively well.

    Best regards,

    Ian Williams