• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 184 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.

AMC3336: Insulation lifetime, working voltage and derating

Zeki Sisman
Intellectual 320 points
Part Number: AMC3336
Other Parts Discussed in Thread: AMC1336, AMC1411

Hello,

I have couple questions on Figure 6.5 from the datasheet.:

  1. As far as I know DIN VDE V 0884-11 defines reinforced isolation working voltage as where insulation has 37.5 years lifetime with 1ppm failure rate under ([working voltage] x 1.2) stress. Then why is the lifetime graph on the datasheet is given for 76 years and not 37.5 years? It looks like at 37.5 years the rating would be higher than 1.2kVrms.
  2. I see 143 years at ~1.5kVrms. Then At 1.2kVrms I see 76 years lifetime. I understand the safety factor on the voltage, but I would expect the lifetime to be greater at the lower stress. Where does 76 years value come from with respect to 143 years?
  3. If I need 10 years lifetime instead of 76 years, can I simply assume that the device can handle higher than the specified "Maximum-rated isolation working voltage" for my lifetime requirement? Or should I never exceed "Maximum-rated isolation working voltage"?

Kind regards,

Zeki

  • 0
    TI__Mastermind 49051 points

    Hi Zeki,

    1. Our devices offer performance that is beyond the minimum requirements for reinforced isolation.

    2. Derating is associated with the specification as described in this document: https://www.ti.com/lit/wp/sszy024/sszy024.pdf

    3. The strength of the isolation barrier is essentially defined as applied voltage vs time. Working voltage is a steady state application, while transient and surge are specified for smaller time durations. Yes, you can apply higher voltage for a shorter period of time as shown in the curve. 

  • 0 in reply to Alexander Smith
    Intellectual 320 points

    Hi Alexander,

    1. I don't fully get it. Is 76 years an arbitrary choice by Texas Instruments and it happens to coincide with 1.2kVrms? So it doesn't come from any source etc? Why not select 40 years (as long as >37.5 years) and advertise a higher working voltage rating for example?

    2. Thanks for the link! Derating is clear to me now.

    3. Thanks!

  • 0 in reply to Alexander Smith
    Intellectual 320 points

    Hi Alexander,

    I suspect that the 1.2kVrms is limited by package creepage and that is the driving factor for the reported 76 years. If package creepage was not a problem, I think the datasheet would report the VDE required 20 year - 1ppm failure lifetime (37.5 / 1.875 with margin) at >1.2kVrms voltage. What do you think?

    If this is the case, then if I want to use this part at >1.2kVrms working voltage, I need potting or some tracking prevention measures I guess.

  • +1 in reply to Zeki Sisman
    TI__Mastermind 49051 points

    Hi Zeki,

    Apologies for the delay, I have been traveling. 

    We apply much higher voltages than 1.2kV and track performance vs time then derive the expected lifetime; which yields the curve you see in the datasheet. These devices have been released for a much shorter time than 76 years. 

    In general if the isolation barrier is strong enough, 1kV applied across the isolation barrier corresponds to 1mm of clearance (from pin to pin). So with a 8.5mm package you should be able to apply <8.5kV across the isolation barrier without arcing. This of course is in air, where many of the specifications are measured when the device is in oil as per IEC specification. As this device has an integrated DC/DC transformer, the isolation specifications are slightly lower than the same product without the integrated DC/DC converter, the AMC1336. However pollution degree and altitude of the application must be considered as well. There is a table in the standard that will determine actual creepage/clearance requirements of the isolation barrier. You may want to consider the AMC1411 if higher creepage/clearance is necessary. 

  • 0 in reply to Alexander Smith
    Intellectual 320 points

    Hi Alexander,

    I understand the time-dependent dielectric breakdown measurement. Only thing I don't get is why 1.2kVrms is put as the maximum working voltage. Is there no reason for it? 76 years is not a requirement, then spec could've been 1.1kV or 1.3kV and still comply with reinforced insulation per DIN VDE V 0884-11. 

    Thanks for the details on clearance and creepage, I will check AMC1411 too!

  • 0 in reply to Zeki Sisman
    TI__Mastermind 49051 points

    Hi Zeki,

    I see now. Thank you for clarifying. This is a fair point and I'm not positive why this number of years was settled upon. As we pride ourselves on no field failures for our isolation devices, this number of years was likely determined to be well beyond the lifetime of any expected application that the devices would function in. I will seek clarification on this and let you know if I am incorrect. 

  • +1 in reply to Alexander Smith
    TI__Mastermind 49051 points

    Hi Zeki,

    Following up, increasing the working voltage such that lifetime = 37.5years would yield ~1280Vrms. Selecting 1200Vrms was done for simplicity.