T
treez
Guest
Hello,
We are using the IPP90R1K2C3 FET in our offline SMPS.
Its junction temperature is 105 degC.
We wish to know how long it will last.
Infineon sent us the equations which we need to use to calculate the probable number of failures that will occur over a given amount of time.
This document is attached.
Do you know what we should put in the equation for the “confidence level, PA”.
Presumably the equation supplied doesn’t tell us what failure rate we’ll get, but just gives us parameters which we then insert in to the chi^2 distribution?
IPP90R1K2C3 FET
https://www.infineon.com/dgdl/Infin...n.pdf?fileId=db3a30432313ff5e0123a89fe8085c04
+++++++++++++++++++++++++++++++++++++
All we want is a figure of how many failures we’ll get over an inserted period of time for a quantity of n FETs under investigation.
Am I right in saying that a full knowledge of the chi^2 distribution is going to be needed here, and a dusting off of the old probability books is going to be required? I remember doing the “normal” distribution, but its all many years ago now.
We are using the IPP90R1K2C3 FET in our offline SMPS.
Its junction temperature is 105 degC.
We wish to know how long it will last.
Infineon sent us the equations which we need to use to calculate the probable number of failures that will occur over a given amount of time.
This document is attached.
Do you know what we should put in the equation for the “confidence level, PA”.
Presumably the equation supplied doesn’t tell us what failure rate we’ll get, but just gives us parameters which we then insert in to the chi^2 distribution?
IPP90R1K2C3 FET
https://www.infineon.com/dgdl/Infin...n.pdf?fileId=db3a30432313ff5e0123a89fe8085c04
+++++++++++++++++++++++++++++++++++++
All we want is a figure of how many failures we’ll get over an inserted period of time for a quantity of n FETs under investigation.
Am I right in saying that a full knowledge of the chi^2 distribution is going to be needed here, and a dusting off of the old probability books is going to be required? I remember doing the “normal” distribution, but its all many years ago now.