• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 901 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.

TMP107: 150C digital temperature sensors

Louis Dupont
Intellectual 480 points
Part Number: TMP107
Other Parts Discussed in Thread: LMT01, , LM95071, TMP117, TMP116

We’re evaluating 2 digital temperature sensors from TI in order to measure temperatures up to 150C (302F) at the end of 5 meters (16.4’) cables:

  • LMT01 (~90khz pulses)
  • TMP107 (one-wire type 4800bps UART compatible)

The TMP107 is a better fit for us (can be daisy-chained, lower bandwidth and easier to condition the signal for long cables, less stringent real-time processing requirements, overall lower power consumption) but its measurement saturates at 128C although it keeps operating up to 150C. The 128C limit appears to be an ADC binary count limit. We were wondering if TI has a TMP107 style sensor in the pipeline that would allow us to measure up to 150C (either by shifting the measurement range or by increasing the ADC resolution). Thanks.

  • 0
    TI__Mastermind 43785 points
    Hi Louis,

    We're aware that this is a feature that many customers are interested in, but we don't have anything to announce at this time.

    May I ask if you would consider an SPI interface instead? Do you need to perform any write operations to TMP107 to configure it, or is it sufficient to only read back temperature? I ask because an SPI device wired for read-only would only require one more conductor in the cable, and it would be very robust over long cables.

    Thanks,
    Ren
  • 0 in reply to Ren Schackmann
    Intellectual 480 points

    Thanks Ren,

    A sensor that would simply stream data over a read-only SPI interface data upon chip-select assertion would be great if it can match the accuracy of the TMP107. I believe that 4 conductors are available. If I remember correctly we considered the LM95071 at one point but it was put aside due to its +-2C accuracy.

    Louis

  • 0 in reply to Louis Dupont
    TI__Mastermind 43785 points
    Louis,

    I'm sorry; I counted wrong.

    LMT01 requires 2 conductor cabling. VP/VN.
    TMP107 requires 3 conductor cabling. V+/GND/Bidirectional Comms
    LM95071 requires 5 conductor cabling. V+/GND/SCK/SO/CS

    We could potentially make an SPI interface where CS is not needed or could be tied off, but currently it's required in LM95071 and our other SPI products to provide packet framing. I'm not aware of a clock-less SPI; it would make sense to just do UART instead.

    At any rate, we are working on a replacement for LM95071, and it is a high priority for us.

    Thanks,
    Ren
  • 0 in reply to Ren Schackmann
    Intellectual 480 points

    Ren Schackmann said:
    At any rate, we are working on a replacement for LM95071, and it is a high priority for us.

    Thanks Ren,

    I believe that down the road we could accomodate 5 conductors. Can you disclose the expected accuracy of the LM95071 replacement?

  • 0 in reply to Louis Dupont
    TI__Mastermind 43785 points
    Louis,

    It's expected to be the same or better than TMP107.

    Thanks,
    Ren
  • 0
    TI__Prodigy 355 points

    Hello Louis.

    I am sure the I2C bus with 100KHz communication speed with proper valued and located pull-up resistors can reliable cover 5 meter distance. Than you can use TMP116 or TMP117 parts which could go above +150C. 4 wired needed.

    Regards