• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 31 subscribers
  • Views 781 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.

TMP451 EVM with NPN and PNP transistors

Ken Dillinger
Expert 1220 points
Other Parts Discussed in Thread: TMP451

I have been playing around with the TMP451 EVM and I have noticed that there can be up to a 1°C difference between an NPN and PNP. It seems rather consistent where the PNP always reads higher in temperature: 0.5°C up to 1°C normally.

It's has been an awfully long time since I cracked open a devices book, but should there be this difference? Do I not know how to solder or read the datasheet?

-Ken

  • 0
    TI__Mastermind 30960 points

    Hi Ken,

     

    What you see is probably normal. TMP451 is optimized for third-party 2N3906 and 2N3904 transistors. However there are subtle differences between these two types. Especially the ideality factor has a pronounced effect on temperature accuracy.

     

    What complicates matter even more is that ideality factor is not constant over temperature. Besides, devices from different manufacturers tend to have different ideality factors. But overall the recommended transistors are the best characterized and provide most reliable performance. It is exactly due to this reason, TMP451 provides the ability for ideality (N) factor calibration.

     

    You’ll find that TMP451 also provides an option for OFFSET calibration. Either one or both can be used in order to achieve high accuracy.

     

    Regards,

    Guang-Apps Sensing Products

  • 0 in reply to Guang Zhou
    Expert 1220 points
    Hello Guang,

    I had to go back to old devices books, and it is my understanding that ideality is a constant. Are you saying ideality itself is not constant over temperature or the *error* from ideality is not constant over temperature?

    If ideality was constant then I would hope manufactures would specify it in their datasheets and it is an easy register value to program. If it is not, that's rough. If changes over temperature then that would suggest a multi point calibration over the range of interest.

    The TMP451 datasheet does mention that ideality is trimmed for 1.008, but makes no mention of differences of PNP vs NPN even though both may be used.

    -Ken
  • 0 in reply to Ken Dillinger
    TI__Mastermind 30960 points

    Hi Ken,

     

    Yes, TMP451 is trimmed assuming ideality factor of 1.008.

     

    Ideality factor is not constant over temperature, although the two recommended transistors have a relatively flat characteristic. A single point calibration should produce relatively high accuracy.

     

    We recently did a comparison of three options of NPN transistors, I’ll post the results here for your reference.

    As you can see 2N3904 gives the best overall performance. And please keep in mind this data was collected with NO calibration at all, with N-factor and/or offset calibration, very high accuracy can be achieved.

     

    Regards,

    Guang-Apps Sensing Products

  • 0 in reply to Guang Zhou
    Expert 1220 points
    Hello Guang,

    That is very interesting! I spent some time looking up ideality and references regarding its behavior over temperature are scant; however, there were references to ideality changing over very low or very high currents.

    Now I am interested in why different NPNs (let alone PNPs) vary wildly. That's something I will put on the backburner and determine later. "I'm an engineer, not a physicist" as they say.

    It looks like the 3904 is the way to go for now. Otherwise, the TMP451 suits my needs for my hobby project.

    -Ken
  • 0 in reply to Ken Dillinger
    TI__Mastermind 30960 points

    Hi Ken,

     

    Thank you very much for your interest in TMP451 and any other TI product. Please let us know whenever you have a question.

     

    My guess as to why ideality factor varies wildly is that many of these devices are optimized for higher operating current (compared to what TMP451 uses), doping levels etc. play a role in the variability.

     

    Have fun with your project!

     

    Regards,

    Guang-Apps Sensing Products