Appendix
B:
Nonmandatory guidelines for certification/validation of safety systems
for presence sensing device initiation of mechanical power presses
Objectives
This Appendix provides employers, manufacturers, and their representatives,
with nonmandatory guidelines for use in developing certification documents.
Employers and manufacturers are encouraged to recommend other approaches
if there is a potential for improving safety and reducing cost. The
guidelines apply to certification/validation activity from design
evaluation through the completion of the installation test and the
annual recertification/revalidation tests.
General
Guidelines
A. The certification/validation process should confirm that hazards
identified by hazard analysis, (HA), failure mode effect analysis
(FMEA), and other system analyses have been eliminated by design or
reduced to an acceptable level through the use of appropriate design
features, safety devices, warning devices, or special procedures.
The certification/validation process should also confirm that residual
hazards identified by operational analysis are addressed by warning,
labeling safety instructions or other appropriate means.
B. The objective of the certification/validation program is to demonstrate
and document that the system satisfies specification and operational
requirements for safe operations.
Quality
Control
The safety attributes of a certified/validated PSDI safety system
are more likely to be maintained if the quality of the system and
its parts, components and subsystem is consistently controlled. Each
manufacturer supplying parts, components, subsystems, and assemblies
needs to maintain the quality of the product, and each employer needs
to maintain the system in a non-degraded condition.
Analysis
Guidelines
A. Certification/validation of hardware design below the system level
should be accomplished by test and/or analysis.
B. Analytical methods may be used in lieu of, in combination with,
or in support of tests to satisfy specification requirements.
C. Analyses may be used for certification/validation when existing
data are available or when test is not feasible.
D. Similarity analysis may be used in lieu of tests where it can be
shown that the article is similar in design, manufacturing process,
and quality control to another article that was previously certified/validated
in accordance with equivalent or more stringent criteria. If previous
design, history and application are considered to be similar, but
not equal to or more exacting than earlier experiences, the additional
or partial certification/validation tests should concentrate on the
areas of changed or increased requirements.
Analysis
Reports
The analysis reports should identify: (1) The basis for the analysis;
(2) the hardware or software items analyzed; (3) conclusions; (4)
safety factors; and (5) limit of the analysis. The assumptions made
during the analysis should be clearly stated and a description of
the effects of these assumptions on the conclusions and limits should
be included.
Certification/validation by similarity analysis reports should identify,
in addition to the above, application of the part, component or subsystem
for which certification/validation is being sought as well as data
from previous usage establishing adequacy of the item. Similarity
analysis should not be accepted when the internal and external stresses
on the item being certified/validated are not defined.
Usage experience should also include failure data supporting adequacy
of the design.
[53 FR 8360, Mar. 14, 1988]
