• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 31 subscribers
  • Views 1175 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.

LMT70: Searching for cabled sensor probe making use of LMT70

Robert Mohr
Prodigy 40 points
Part Number: LMT70
Other Parts Discussed in Thread: LMT01, TMP107, TMP116

Hello,

I am searching for a sensor probe making use of the LMT70 which employs a cable and attachment method similar to the following: https://www.omega.com/googlebase/product.html?pn=SA1-RTD-120&gclid=Cj0KCQjwtOLVBRCZARIsADPLtJ1BH5FEWLBiLXkZ6_cQOm0xvVHaGIFRwGWVibDSzo8mTyQ2_ERPfsYaAituEALw_wcB.  A cable with a stainless steel overbraid option to reduce electrical noise and protect the lead wire from abrasion would be ideal.  I'd like a cable with a lead length of ~6ft if possible, however that is not critical.

I've done many, many internet searches and have not been able to locate a cabled probe making use of the LMT70 yet.  I am hoping someone can point me to a third-party supplier who has implemented this in a "stick-on" format which would be most conducive to the requirements of my monitoring project.

 

Thanks in advance for any information you can provide.

  • 0
    TI__Expert 7855 points
    Robert,

    Please tell me more about your application requirements (operating temperature, accuracy, analog/digital interface) so that I can help you find the right sensor. LMT70 is in a chip scale package and might be difficult for you to mount on a probe. Have you looked at the LMT01 or TMP107?

    -Kelvin
  • 0 in reply to Kelvin Le THS
    Prodigy 40 points
    Kelvin,

    Operating temperature will be in the range of ~20-60°C. I am planning to interface to the sensor using a Labjack T7 (analog, however I2C,SPI, etc. are also supported). I would like to achieve an accuracy of ~+/- 0.1C, however I could possibly live with up to perhaps +/- 0.3C. I provided the link to the Omega RTD "stick-on" probe as an example of probe design, as this would be the most conducive type of arrangement to use in the test environment we have, in which the probes will be affixed to copper pipes (1-3/8" and 3/4" diameters). Standard stainless steel tube probe type layouts are not ideal for this. The 'stick-on" design also eliminates the need for a clamp, which must be non-conductive in our testing environment, and also acts as a heat-sink affecting the measurement. This type of probe design also allows for easy relocation of the probe to another location on the pipe.

    Yes, I have looked at the LMT01 and TMP107 sensors. Putting aside the lower accuracy specifications, the same issue remains. I need a cabled probe, not simply a chip. The LMT01 would certainly provide a package which would make it easier to roll my own DIY probe, however I'd prefer to just buy an assembled probe and be done with it. That is why I was hoping that someone at TI would know of a manufacturer making use of the LMT70 in a pre-assembled probe assembly similar to what is available from Omega as an RTD.

    Rob
  • 0 in reply to Robert Mohr
    TI__Mastermind 37370 points
    Hi Robert,

    Thank you for choosing TI products for your design. LMT70 and TMP116 is the best choice for your applications. We have put these devices into probe form factor. Please refer to this apps note below.

    www.ti.com/.../snia021a.pdf

    Our lab is still under construction. I will try to see if I can enter the lab and get the wires information for you.

    Aaron
  • 0 in reply to Aaron Heng
    Prodigy 40 points

    Aaron,

    Thank you very much for the reply and the link to the application note.  The probe depicted in Figure 27 looks very promising for our application.  The copper pipes are connected in a loop into which current is injected with a CT.  The current will be between 0-800A, however the voltage is negligible.  If the back side of the sensor PCB is in contact with the copper pipe, will the current flowing in the pipe have any effect on the device?

    I assume that I could solder a cable to the left hand side of the flex PCB to run to the Labjack.  What length of the 40 AWG nickle wire could be used without impacting the measurements?  I would imagine 6 ft would be the upper limit of what might be needed.

    Rob

  • 0 in reply to Robert Mohr
    TI__Expert 7855 points
    Rob,

    You are targeting a system accuracy of 0.3C so you will need to consider many factors that can affect the analog signal chain including the ADC. Make sure you don't have the large signal ground mixing with the small signal ground. The current noise can overwhelm the LMT70. The large pipe current may cause EMI so you will need to pay close attention to the layout (loops, couplings tc). The long cable poses a challenge when designing with an analog temperature sensor. Proper shielding is needed. The longer the cable the more offset error you will have due to the increased resistance. Perhaps calibration is needed. You will need to calculate the cable resistance to determine the limit of the system. Use the TAO leakage current spec to determine the offset from the cable resistance.

    My advice is, if you can...use digital temperature sensor. It has guaranteed system accuracy.

    -kelvin
  • 0 in reply to Kelvin Le THS
    Prodigy 40 points
    Kelvin,

    Yes, these concerns are being taken into account. I plan to incorporate a stainless steel-over braid if I am forced to go with a longer cable. I believe I can achieve a much shorter lead length however. The copper pipe is not grounded. The pipe loop is supported by a non-conductive stand, and current is injected into it using a CT as mentioned above. This will generate an EMF I need to be concerned with, however I do not have a large signal ground.

    I am open to other options that can meet the specifics of the application, which have been outlined above. Primarily: high accuracy, easy application to copper pipe, temperature tange of 15-60C, and compatible with labjack as an ADC. The Labjack can support I2C as mentioned previously, so this is an option. I'm not convinced that a digital output option would be as easy to integrate as an analog output with Labjack however.