• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 142 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.

TPS25751: TPS25751S liquid detection resistor selection

Schroedinger Karl
Prodigy 160 points
Part Number: TPS25751

Tool/software:

Hi, 

You propose a pull down resistor of 1MOhm (NMOS connected) and a pull up of 100k (in another app (EVM) 300k, PMOS). What is the reason for the big difference?

When I calculate the thresholds for the liquid based resistors (the resistors based on the humidity itself)  I came up with 10MOHM (PMOS off) and 500k (NMOS off) which is a factor of 20!

Thank you. 

  • 0
    TI__Mastermind 34826 points

    Hi, 

    Schroedinger Karl said:
    When I calculate the thresholds for the liquid based resistors (the resistors based on the humidity itself)  I came up with 10MOHM (PMOS off) and 500k (NMOS off) which is a factor of 20!

    Can you provide more details about your test setup and how you calculated the humidity resistance? Is the humidity resistance shorting the SBU pins to another pin(s) within the Type-C connector? 

    Thanks and Regards,

    Raymond Lin

  • 0 in reply to Raymond Lin
    Prodigy 160 points

    Hi Raimond, 

    I use the circuit in fig. 9-8 and table 9-3 of the datasheet tps25751(SLVSH93A). I use thresholds of 0.3V and 2V for the liquid detector. I assume the range between is the forbidden range as in worst case (short) we have 0 at PMOS on and VDD at NMOS on and  with no liquid (good case) 0 if NMOS is on and VDD if PMOS is on. 

    For the bad case (liquid): 
    If the NMOS is on we have a resistor divider from VBUS to GND with Rliq-up (resistance of the water) and Rdwn (1M)  for the test point (SBU). At the threshold of 0.3V we can assume a Rliq-up of about 15M (e.g. VBUS = 5V). 

    If the PMOS is on we have a resistor divider VDD to GND with Rliq-dwn (resistance of the water) and Rup (100k) for the test point (SBU). At the threshold of 2V we can assume a Rliq-dwn of around 150k. 

    The proposed values in the GUI are a bit higher, but not very much different. This are only theoretical calculations, no test with real liqid at the moment. I tested with short and it works. 

    That was my question: why are these liquid impedances such different. 

    Best regards

    Karl Schroedinger

  • 0 in reply to Schroedinger Karl
    TI__Mastermind 34826 points

    Hi Karl, 

    Schroedinger Karl said:

    The proposed values in the GUI are a bit higher, but not very much different. This are only theoretical calculations, no test with real liqid at the moment. I tested with short and it works. 

    That was my question: why are these liquid impedances such different. 

    The pull-down resistance (Rdwn of 1M) was chosen to adhere with the USB Type0C specification of the minimum required resistance on SBU pins (minimum of 950k). Also when the Rdwn resistor is enabled (NMMOS is on), TPS25751 is checking for any possible shorts from the SBU pin to voltage pin such as CCx (can be up to 5V if Vconn is present) and VBUS (can range from 5V to 20V within SPR range and maximum up to 48V for EPR capable devices). 

    TPS25751 measures the voltage on its ADC pin in order to detect liquid, it doesn't calculate the liquid (if present) resistivity. Let us know if this clears up your question or if you need further clarification! 

    Thanks and Regards,

    Raymond Lin

  • 0 in reply to Raymond Lin
    Prodigy 160 points

    Hi Raimond, 

    thanks. 

    I understeand that the ADC measures voltages. But these are generated by (liquid) resistances (SBU to GND or CC or VBUS). 

    You say, that the minimum resistance for SBU pins is 950k. Does this mean that the USB interface may have such a resistance undependent from liquid at the connector, i.e. the electronics (at the other side) may have such a resistance?

    What about the other case when the PMOS is enabled: in this case the threshold referes to a resistance of 150k (SBU to GND). 

    Best regards

    Karl Schroedinge

  • 0 in reply to Schroedinger Karl
    TI__Mastermind 34826 points

    Hi Karl, 

    How are you applying "liquid" resistances on the SBU to GND/CC/VBUS? Is the resistance constant or does it dynamically change?

    Thanks and Regards,

    Raymond Lin

  • 0 in reply to Raymond Lin
    Prodigy 160 points

    Hi Raimond, 

    this are assumed resistors. If you have contacts (e.g. inside the connector) and you put water on them there is a current flowing. The assumed resistor caculates from applied voltage and this current. It is not a real resistor. Therefore it changes with humidity, if everything is dry this resistance is infinite. 

    Karl Schroedinger

  • 0 in reply to Schroedinger Karl
    TI__Mastermind 34826 points

    HI Karl, 

    When liquid/moisture/humidity is shorting the SBU (measurement) pin to something else within the Type-C connector (i.e. VBUS, CC, GND), there is some resistance between the pin(s) to SBUx. During this time, TPS25751 will continuously measure the SBU pin voltage and checking for shorts to VBUS (when NMOS and Rdown are enabled) and Ground (when PMOS and Rup are enabled). If any of the high or low voltage measurement exceeded the set threshold, then TPS25751 determines liquid is present in the port. 

    As the liquid/short slowly dries up, the resistance across the SBU pin to pin(s) will slowly increase to infinite, bringing the voltage measurement back down to below the set liquid threshold. Once the voltage reading on the SBU pins return to normal, TPS25751 exits the liquid detection state and will resume normal operation. 

    Hopefully this helps clear up your question, if you need further clarification please let us know! 

    Thanks and Regards,

    Raymond Lin