• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 31 subscribers
  • Views 92 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Original question:

TMP117: possible sources for measurement non-linearity

TMP117: guaranteed performance within 0.2°C @130-140°C?

Stefan Lintner
Expert 4255 points
Part Number: TMP117

Hi Team,

a customer is looking for a temperature sensor with guaranteed performance within 0.2°C @130-140°C.

Given the plot form the DS below, they are asking what makes the difference between the “average temp accuracy” and the min/max lines in red. Is it simply the variation between silicon, and does this mean that for a given sensor, the black curve is the same, but simply moved up or down? 

Or to put it differently: if the linearity of the sensor is specified, they might be able to use that to compensate and still get within our required accuracy?

BR,
Stefan

  • 0
    TI__Mastermind 35895 points

    Hi Stefan,

    As you know, TMP117 is specified as +/-0.3C at 130-140C. The linearity is quite good, but there is no specification for linearity. 

    The data shown in Figure 6-2 and 6-3 comes from a sample of production units. The data represents silicon variation between devices, but also represents variation in our ability to trim them to zero error in production. Since they are trimmed to zero error at ambient temperature in production, our tightest accuracy is found there. As you move away from ambient temperature, the black line does not "simply move up or down." You can see that the 3sigma shape opens slightly wider. This reflects the fact that some units are trending slightly higher while others are trending slightly lower. The 3sigma values are being calculated across the population of units at each temperature set point. If there was no silicon variation, and the error came solely from our ability to trim at production, you could expect a set of black lines that "simply move up or down." Let me know if this makes sense.

    thanks,

    ren