Risknowlogy / Conocimiento / Foros / Functional safe... / Hardware Fault...

Forum Topic

Hardware Fault Tolerance

Tienes que estar conectado para tener acceso a los foros. Puedes hacerlo aquí

 Functional safety forum Asunto anterior | Asunto siguiente  
 Autor Mensaje

George Azzopardi

30. abril 2008 12:39:47

Hardware Fault Tolerance

I am conducting a preliminary SIL capability assessment of a product during quotation stage. One of the important aspects to consider is the concept of architectural constraints. I am using table 3 IEC 61508-2 part 2 which gives you a limit of the maxium SIL that can be claimed in terms of SFF and HFT. Now my question is this : at an early stage of the project how can you jugde the HFT of your system without doing an in depth analysis e.g. FMEDA ? As far as I understood in deciding the HFT level, diagnostics cannot be accounted for. So HFT is pure hardware redundancy.


Michel Houtermans

05. mayo 2008 09:00:37

Re: Hardware Fault Tolerance

You are right, HFT is a measure of redundancy. Redundancy can be done on architecture level or on device level. When on device level it means that inside the device there is redundancy.

If you have 1oo2 transmitters then you need to determine the SFF with an FMEDA on each transmitter.

If you have a transmitter with internal redundancy, e.g., 1oo2, then you need to perform an FMEDA on each leg/path of this transmitter.

In other words you do it the other way around. First you determine the SFF of each leg/path and then you determine whether you have redundancy or not. If you have redundancy (internal or external) you can then apply the architectural constraints table to see which SIL you achieved for the hardware.