• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 48 subscribers
  • Views 532 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.

LM98620: Estimation of junction temperature

az
Prodigy 130 points

Part Number: LM98620

Hi,

Let me know how to estimate junction temperatue, Tj.
In my proto-type. the device surface temperature showed over 90 C-deg. It would be serious problem.

The former qustioner asked you about heated LM98620. And th answer said it consumes 1.1W and then TJ increase 35 C-deg.
I understood this calculation, but the asnswer and the datasheet didn't resolve this heat problem clearly.Because thereis no way to have TJ even a heat-sink or other external thermal part is used.

The datasheet refers another document "SPRA953", however, the document doesn't show clearly how to estimate TJ from R-theta, Power-consumption and any measureable temperature like ambient, device body or surface. I really need the equation of them.

regards,

az

  • 0
    TI__Expert 6330 points

    Hi az,

    The datasheet states that the junction-to-ambient thermal resistance is 32 C/W. What conditions are you operating the device under to get 90C? If operating at room temp I would expect there to be about 2W of power being dissipated based on the junction-to-ambient figure in the datasheet.

    Thanks,

    Aaron

  • 0 in reply to Aaron Goodson
    TI__Guru 51415 points

    Hi,

    LM98620 power consumption is ~1W with typ operation condition (Datasheet page8). Since Theta j-a is 32/C, temperature delta from ambient should be ~32C range.

    Generally Case top temperature is ~5C lower than Tj (It depends on package etc, but rough estimation). If you see 90C on case top, die temp could be high as 90-100C. 

    -What is the ambient temp?

    -What is the LM98620 operation condition for 90C?  Does temp change if goes to Power-Down Mode(0x00 bit4) and stop MCLK?

    -Datasheet page 8 shows each current consumption of IDDA/IDDD/IDDLVDS. Is it possible to check if one of them is extremely high? Or can we check supply current to LM98620?

    -Do you provide CLP voltage externally at VCLPEXT pin? If yes, better to check it voltage, register settings is correct. Any possibility high current supply on VCLPEXT pin wrongly?

     regards

    Shinya

  • 0 in reply to Shinya Morita
    Prodigy 130 points

    Hi Aaron and Shinya,

    Thank you for your answer.

    As Aaron wrote,

    > What conditions are you operating the device under to get 90C?

    The 90C measurement was NOT done in the room temperature, but higher temperature which simulates the user's worst case.

    I showed the measured values from 5 samples.

    Power(all) I(VDDD) I(VDDA) I(VDDLVDS) V(VCLPEXT)
    Mean 879.3 29.5 196.4 41.8 1.608
    Max 899.0 31.5 199.0 42.8 1.613
    Min 845.0 24.5 192.0 41.0 1.603
    Unit (mA) (mA) (mA) (mA) (V)

    I think my system is going well because the device power consumption looks ok and I got scanned images from my proto-type system with the device too.

    As Shinya wrote,
    > Generally Case top temperature is ~5C lower than Tj (It depends on package etc, but rough estimation). If you see 90C on case top, die temp could be high as 90-100C.

    From this mention, I think it is possible to estimate the thermal parameter "JT" is about 5C/W, as the device power dissipation is assumed ~1W .
    Moreover, with two reference documents below, this estimation seems to be enough good for a typical QFP package.


    (1) www.ti.com/.../spra953c.pdf
    In page 8, "For plastic packages, ΨJT is typically 0.5C/W - 2C/W compared to RθJC values of 4C/W - 15C/W."

    (2) www.nxp.com/.../AN4388.pdf
    In page 24, "Table 4. Thermal resistance example, RθJA = 35C/W and, ΨJT = 2C/W."


    Therefore, I consider that my TJ is calculated as 85C and meets the maximum rating of 110C TJmax.

    Please tell me your opinion.

    regards,

    az

  • 0 in reply to az
    TI__Guru 51415 points

    Hi,

    Thank you for sharing information. Let me investigate them and feedback to you by Dec 18th.

    regards

    Shinya

  • 0 in reply to Shinya Morita
    TI__Guru 51415 points

    Hi,

    Thank you for sharing power consumption data. Your power consumption looks similar to datasheet and no problem. And I understand your ambient temp is higher than room temp.

    Your estimation JT=5C/W is good enough including margin. Good to be used for estimation.

    However if your package top temp is 90C, Junction temp is not  90-5 =85C as you commented on your post. We need to add the calculation JT*Power on TC.

    Let me use JT= 2C/W based on page 8 of TI application note (It said typically 0.5C/W - 2C/W). 

    Then if measured TC is 90C.

    Tj = 90 + 2*1W = 92C

    Junction temp 92C is still lower than abs max 110C. No problem.

    regards

    Shinya