• State Resolved
  • Locked Locked
  • Replies 13 replies
  • Subscribers 62 subscribers
  • Views 2102 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPS82130: Thermal Caracteristics

user4948849
Prodigy 180 points
Part Number: TPS82130
Other Parts Discussed in Thread: LMZ20502, TPS62130, TPS62135, TPS62136, LMZM33603, TPSM84824

Hi support,

I'm actually using a TPS82130 on my design to provide 2V5@1A max. from 12V. It gives me a Total Power Dissipation of 420mW max.

Thermal simulations show limitations because of power module reach approximatively 122°C in the worst case.

Datasheet specifies:

  • A 10^5 hours lifetime for a Tj_max of 110°C.
  • A 5*10^4 hours lifetime for a Tj_max of 125°C!

In comparison with the LMZ20502 (electrical caracteristics are not the same and it cannot be used in my application but package is a uSIP module as TPS82130):

I observe for LMZ20502, that thermal caracteristics seem better :

  • Rjc_Top : 20.8°C/W
  • RjC_Bot : 1.8°C/W
  • Tjmax =125°C

For TPS82130:

  • Rjc_Top : 9,4°C/W
  • RjC_Bot : 21,3°C/W => It's pretty high for a QFN-like package? In comparison with LMZ205xx series.
  • Tjmax =110°C (for better reliability)

What does it explain a such difference between these 2 components?

What is the expected lifetime for LMZ20502 at Tjmax=125°C?

What power module can I use to replace TPS82130 in my application? I'm initially searching for an uSIP component (can't find anyone) or any other regulator with a low footprint area and good thermal performance.

Thank you for your help.

Mickaël.

  • 0
    TI__Guru*** 145650 points
    Hi Mickael,

    Thanks for your detailed description. There's also good reading in this app note I wrote and in the thermal paper linked under each device's thermal table:
    www.ti.com/.../slyt724.pdf
    www.ti.com/.../spra953c.pdf

    While they look similar, these 2 devices are different internally. As well, the LMZ is larger (in all 3 dimensions). So, they will give different thermal performance.

    I'm unclear what Vin and maximum ambient temperature you have in your system. Knowing this would help me double check the math.

    Also, if your system draws 1A peak, it sounds like it draws lower currents on average (typically). This reduces any lifetime reduction that occurs above 110 C.

    Finally, does your system have any cooling other than natural convection? Do you have an estimated PCB temperature?
  • 0 in reply to Chris Glaser
    Prodigy 180 points
    Hi Chris,

    Thanks for your reply.

    I'm actually using a 12V input voltage to generate a 2V5 output voltage @ 1A.
    Ambiant temperature can rise up to 78°C in worst case scenario.
    We have different kinds of cooling solution: Air cooled or conduction cooled.
    Limitations are already observed in air cooled solution (with an Air flow of 25m^3/h) and Ta=55°C: junction temperature of the component is 104°C. Board temperature around the TPS82130 is approximatively 100°C according to thermal simulation.

    According to datasheet, TPS82130 is design to work in industrial applications. Even with a 125°C junction temperature, lifetime of regulator is only 100000 hours and it is not compatible of our products. Why the MTBF is so low? Is it due to on-chip inductance or something else?

    We are searching for a regulator with a small footprint, a high MTBF, good thermal caracteristics and a input voltage between 5V and 12V. Output voltage must at least reach 6V (as the TPS82130)and a current load up to 3A.

    Is there others regulators smaller than the TPS62130 or TPS62135 (with a tiny inductance such as IHLP-1212)?

    Mickaël.
  • 0 in reply to user4948849
    TI__Guru*** 145650 points
    Thanks for explaining.

    I see one possible discrepancy: the 100,000 hours (>11 years) operating time is TI's standard reliability rating for most all our products. It is different than MTBF. I just answered an MTBF inquiry earlier this week: e2e.ti.com/.../660416

    What application does your product fall under?

    The TPS821 and TPS621 are the smallest options for your input voltage and power levels. The TPS62136 is a little more efficient but requires a larger inductance, due to its lower switching frequency.

    I'll check with others to see what the next step up in efficiency/thermal performance would be. It will be larger.
  • 0 in reply to Chris Glaser
    TI__Guru*** 145650 points
    The TPSM84824 and LMZM33603 are the next step up in power modules. Samples are available for both. How do these look, besides being larger?
  • 0 in reply to Chris Glaser
    Prodigy 180 points

    Hi Chris,

    Thanks for replying.

    What is the FIT/MTBF of the TPS82130? I cannot find this information on website? Is there a significant difference compare to a LMZ20502 or another uSIP package module? what is the difference between lifetime and MTBF/FIT information? Is there any relation?  

    I have 7 TPS82130 on my product. I'm worried about this impact my product MTBF.

    My product will fall under industrial application.

    Do you think TPS62130 or TPS62135 (VQFN ) + IHLP-1212 inductor  are interesting in order to replace the TPS82130? It will be larger with the association of the inductor but probably more suitable (and reliable) to my requirments?

    Thanks.

    Mickaël.

  • 0 in reply to Chris Glaser
    Prodigy 180 points
    These power modules are pretty interesting but TPSM84824 seems to be a little high for my application.
  • 0 in reply to user4948849
    TI__Guru*** 145650 points
    Hi Mickael,

    Here is an explanation of the reliability terms: www.ti.com/.../reliability-terminology.html

    I'll check on the MTBF for those 2 devices.

    The 100k hour lifetime judges the intrinsic capability of the device, as measured at its worst case rated junction temperature during qual.

    The MTBF is looking at actual failures during operation of a much larger sample size.
  • 0 in reply to Chris Glaser
    Prodigy 180 points

    Hi Chris, 

    Thanks for explanations.

    Did you check the MTBF for the TPS82130? There is no information on website.

    Mickaël

  • 0 in reply to user4948849
    TI__Guru*** 145650 points
    Hi Mickael,

    The person who normally does this is on holiday. I have asked his replacement for this MTBF information and will provide when available.
  • 0 in reply to Chris Glaser
    TI__Guru*** 145650 points
    Hi Mickael,

    The MTBF of TPS82130 is calculated as 5.24 x 10^7 hours.

    LMZ20502 is 4.37 x 10^7 hours.
  • 0 in reply to Chris Glaser
    Prodigy 180 points

    Hi Chris,

    Thanks for the MTBF.
    However, you don't give me the MTBF supporting data such as Usage Temp, Conf Level, Activation Energy, etc.

    Few months ago, I had a MTBF of 2.33*10^8 for LMZ20502SILT with:

    Usage Temp=55°C

    Conf Level =60%

    EA=0.7

    Test Temp=125°C

    Test duration 1000 hours

    Sample Size=2724

    Mickaël

  • 0 in reply to user4948849
    TI__Guru*** 145650 points
    Hi Mickael,

    On these modules, the overall FIT/MTBF is a combination of the inductor's FIT and IC (silicon) FIT. It seems like the above values you previously received were just giving the silicon FIT and not including the inductor.

    For both TPS82130 and LMZ20502, the above supporting data for the silicon is the same as what you wrote. However, the sample sizes have changed/increased.

    The LMZ20502 now has a sample size 2963. The TPS82130 is designed on a much more popular process and therefore has a sample size of 150,108.

    If you have specific reliability concerns, perhaps we can discuss these further directly. Please start a conversation with me in the top right of the screen for this.
  • 0 in reply to Chris Glaser
    Prodigy 180 points
    Thanks for your support. It help us a lot!

    We do not have specific reliability concerns about TPS82130.
    It was more a misunderstanding with "Lifetime vs. MTBF" topic.

    Best Regards.

    Mickaël.