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.

TLC5924 Junction Tempurature

Other Parts Discussed in Thread: TLC5924
I have a customer with the following question:
 
We instrumented up a TLC5924DAP with a temp sensor on the case of the part.  We have roughly the recommended copper cooling area on the board, and bond to the back solder tab.
 
At room (23C) temperature, with one of our sign settings, we are seeing 94C on the case and in another situation 81C.  The  circuit board should be our primary cooling path.  In that situation, what is the expected thermal rise from case (our measurement) to Junction.  …  In other words are we getting the junction too hot when we run the sign up to 70C or higher?
Can you help?
  • Hello A Train,

    The junction to case thermal resistance of this package is below 5°C if you have a highK board. So in general I expect that the part will work in your configuration without a thermal error setting with 70°C ambient temperature. I always recommend to test this at the highst ambient temperature you have because the influence of the board layout is relatively high. If you do not have a thermal chamber, I would try to reduce the power dissipation to a value that generates a lower case temperature to be on the save side.

    Nevertheless, I wonder if you could reduce the power dissipation in the IC in general. The biggest power dissipation will happen in the OUTx stages if the LED voltage is higher than necessary. Could you reduce the VLED to a lower value to reduce the power dissipation inside the IC? Maybe you could reduce this power dissipation by adding a series resistor to the LEDs?

    Best regards,
    Brigitte

  • First, unfortunately in this version of the design, we can not implement any of the power savings suggestions given.  That is why we have a condern.  We had to increase the voltage to 15V to allow for the Green LEDs.

     

    Specifically, we are measuring the temperature on the top case, of the part with the part properly mounted to the board.  How close is the temperature on that top surface of the case to the junction temperature? Any way to tell what that delta is? Presumably the back cooling plate will be at a lower temperature than the top of the case due to the mass of the board.  The given junction to case thermal conductance presumably is from Junction to back solder tab.  I am measuring temperature on the top side.

     

    Or maybe I just need to re-instrument to measure temperature on the backplane.

  • The following link provides very useful information to help you with your themal measurements.  www.ti.com/thermal  The thermal parameter you need for your measurement is psi-JT.  Knowing psi-JT and the power dissipation, you can calculate your ICs junction temperature.  Psi-JT for the TLC5924 is less than 1deg C/W, so your measured temperature is very close to the actual junction temperature.

  • We already are doing what we can in regards to your suggestions.

     

    But I still have a question.  Looking at the temperatures listed on the data sheet, I don’t really see a maximum Junction temperature specified.

    http://focus.ti.com/lit/ds/symlink/tlc5924.pdf

     

    But on that same datasheet, when I look at the temperature at which the Thermal error flag is set, it would seem that the maximum Junction temperature is over 150C.

     

    In new testing last night, at our highest ambient temperature, +70C, I am seeing +140C measured on the opposite side of the circuit board from where the Driver chip is mounted.  We have sizable vias from one side of the board to the other under the Thermal pad, so I am guessing this temperature is pretty close to the Thermal pad temperature.  My concern is that this is pretty close to 150C.(I had previously thought this was the maximum junction temperature)  But what is the absolute max on the Junction temperature?  I don’t see it in the data sheet.

    Can you advise?

  • Hello A Train,

    The recommended operating junction temperature is maximum 125°C and the absolute maximum rating for the junction temperature is 150°C. So with 140°C you are already above the recommended operating junction temperature and I would recommend to get to a lower power dissipation per string.

    I wonder if you could lower the voltage on the switch to a lower value e.g. by adding a resistor at the top end of the LED string if you cannot adjust the LED voltage.

    Best regards,
    Brigitte