© exida.com GmbH berthold 0408-10 r003 v1r3.doc, Apr. 12, 2007
Rainer Faller Page 13 of 18
5 Results of the assessment
exida did the FMEDAs together with Berthold Technology.
For the calculation of the Safe Failure Fraction (SFF) the following has to be noted:
O
total
consists of the sum of all component failure rates. This means:
O
total
= O
safe
+ O
dangerous
+ O
no effect
+ O
annunciation
O
du total
= O
du
+ O
annunciation undetected
SFF = 1 – O
du total
/ O
total
For the FMEDAs failure modes and distributions were used based on information gained from
[N3] to [N5].
5.1 Level Transmitter LB490 Uni-Probe
The FMEDA carried out on the Level Transmitter LB490 Uni-Probe leads under the
assumptions described in section 4.2.3 to the following failure rates:
Failure category Failure rates [FIT:=10
-9
/h]
Pre-amplifier, voltage
divider and digital board
Fail Safe Detected 752
Fail Safe Undetected 97
Fail Dangerous Detected 261
Fail Dangerous Undetected 45
Annunciation Detected 7
Annunciation Undetected 5
No Effect 313
Safe Failure Fraction 96,6%
4..20mA Output - new Fail Safe Detected 31
Fail Safe Undetected 22
Fail Dangerous Detected 92
Fail Dangerous Undetected 1,3
Annunciation Detected 43
Annunciation Undetected 17
No Effect 31
Safe Failure Fraction 92,3%
AC/DC-DC converter Fail Safe Detected 0
Fail Safe Undetected 55
Fail Dangerous Detected 3
Fail Dangerous Undetected 0
Annunciation Undetected 3
No Effect 18
© exida.com GmbH berthold 0408-10 r003 v1r3.doc, Apr. 12, 2007
Rainer Faller Page 14 of 18
Failure category Failure rates [FIT:=10
-9
/h]
Safe Failure Fraction 95,9%
Photomultiplier Fail Dangerous Detected 21
Fail Dangerous Undetected 2,4
Safe Failure Fraction 90%
The failure rates for the photo multiplier were evaluated from the field experience report of the
manufacturer Photonis.
Remark: Due to the isolated mounting of the contacts in the Ex-D housing, the PG inlet and the
O-ring to the connector room of the photo multiplier were not considered.
Under the assumptions described in section 4.2.3 the following table shows the failure rates
according to IEC 61508:
O
sd
O
su
Total
=
O
su
+ O
no effect
O
dd
Total
=
O
dd
+ O
ad
O
du
Total
=
O
du
+ O
au
SFF DC
S
DC
D
783 fit 535 fit 427 fit 74 fit 96% 59% 85%
These failure rates are valid for operating stress conditions typical of an industrial field
environment similar to IEC 60654-1, class C (sheltered location) with an average temperature
over a long period of time of 40ºC (see chapter 4.2.3). For a higher average temperature of
60°C, the failure rates should be multiplied with an experience-based factor of 2.5. A similar
multiplier should be used if frequent temperature fluctuation must be assumed.
Note that the “no effect” and “annunciation” failures on its own will not affect system reliability or
safety, and should not be included in spurious trip calculations.
The PFD
AVG
was calculated for three different proof test intervals.
T[Proof] = 1 year T[Proof] = 2 years T[Proof] = 5 years
PFD
AVG
= 3,2E-04 PFD
AVG
= 6,4E-04 PFD
AVG
= 1,6E-03
The boxes marked in green mean that the calculated PFD
AVG
values are within the allowed
range for SIL 2 according to table 2 of IEC 61508-1 and table 3.1 of ANSI/ISA–84.01–1996 and
do fulfill the requirement to not claim more than 35% of this range, i.e. to be better than or equal
to 3,5E-03.
To Next Page
