Compliance of IEC61010-1 and VDE0884-10

Hello,

 

I am confused about the safety ratings of the part ADuM4120. Its datasheet states reinforced insulation for a working voltage of 400Vrms to meet IEC61010-1, while the rating for reinforced insulation with VDE0884-10 is 849Vpeak.

 

According to the section 14 of IEC61010-1, this norm accepts any part meeting another recognized safety standard. This means that ADuM4120 would meet reinforced insulation for 849Vpeak because it complies with VDE0884-10. So, why the rating specified for IEC61010-1 in the datasheet is lower (400Vrms)?

 

Thank you in advance.

  • 0
    •  Analog Employees 
    on Dec 8, 2017 12:57 AM

    Hello,

    The confusion here is that you are comparing two types of certification.  VDE-0884-10 is a component level certification, and to start, it quotes peak working voltage, so there is a factor of 1.41 to be accounted for when making comparisons.  The philosophy behind a component cert is to give each parameter its maximum possible value under the test conditions of the cert.  So Working voltage gets quoted independent of the limitations of package size for example.

    System level certs like IEC61010 ties all of the requirements into a coherent bundle depending on the installation, pollution degree, mold compound category etc..  So they will not quote a working voltage that is larger than the package and application can support.

    The component certification has to tell what is possible, the system level cert sets the limits for that system.  When we ask for a system level certification, the agency makes some general assumptions about systems that get applied to everyone, such as Pollution degree 2 and Material Group III and installation class 2 in order to quote numbers that can be compared between products.  Those assumptions may not reflect a real system, so the system numbers can be adjusted if the assumptions do reflect the actual usage or component capabilities.  A good example is that our parts actually support at least Material Group II, so the working voltages can be higher than what is written in the certification directly.  If we imposed those limitations at the component certification level, it would exclude those types of adjustments.  This leads to questions like yours when the numbers do not seem to line up.

    Best Regards,

    MSCantrell

  • Hello MSCantrell,

     

    Thank you for your explanation. I found it very useful.

     

    If I understood correctly, in this case the rating concerning IEC61010-1 is mainly based on the creepage constraints of the package (besides assuming the worst case of Material Group III). If I consider that the package belongs to Material Group II, the working voltage can be increased up to something more than 500Vrms while meeting reinforced insulation for IEC61010-1.

     

    Kind regards.

  • 0
    •  Analog Employees 
    on Dec 8, 2017 9:59 PM

    Hello

    If there are no other limits, then yes that is true.  However we have a maximum working voltage to guarantee lifetime, Please check the applications section of the data sheet for that value.

    Mark Cantrell

    781-937-1486

  • Hello Mark,

    Thank you for this information. I see that the maximum recommended voltage is 600Vrms at 60Hz (lifetime of 20 years with 0.1% of failure rate).

     

    These data lead me to question the effect of the frequency of the AC signal across the barrier on the lifetime of the isolator. This is important for my design because it should deal with frequencies up to 5MHz. After spending several hours reading about this topic, it is clear that the barrier can degrade faster at higher frequencies. However, I have found no models relating the lifetime with the frequency. Do you know any data or bibliographic reference where I can find information about this topic?

     

    Best regards,

    Victor M.

  • 0
    •  Analog Employees 
    on Dec 9, 2017 2:15 AM

    Hello Victor,

    This is a complex topic.  In general you are correct, that higher frequencies will degrade the insulation faster.  However there is not a simple predictive derating formula.  Many things can affect this and we can engineer insulation that has better performance at high frequency, for example in a gate driver product.

    In general we know the performance for line frequency and for fast slewing switching applications up to about 20kHz, but not to the kind of frequencies that you are asking about.  If you want to discuss this in greater detail, we should look at your application in detail off line.

    Best Regards,

    MSCantrell