Chapter 7 of the NEC can be confusing unless you understand the difference between emergency systems, legally required standby systems, and optional standby systems. Once you sort these out, you may still be scratching your head if you have systems vital to public safety (including systems that protect against environmental releases of toxins or waste).
This is where Annex F becomes tremendously helpful, despite being only two pages long. The first thing it does is walk you through how to determine:
- Availability. The percentage of time a system is available to perform its functions
- Reliability. The inverse of the probability and frequency of failures.
- MTBF. Mean time between failures.
- MTTF. Mean time to failure.
- MTTR. Mean time to repair.
A common way of expressing availability is the number of point-nines. Annex F provides a chart showing hours of downtime expressed on the scale of point-nines. For example, an availability of 0.9 translates into 876 hours of downtime per year (based on 8,760 hours, which is simply 365 days times 24 hours). But 0.999 means 8.76 hours of downtime per year — basically down for an entire shift. If your availability is only 0.9, you're looking at 100 shifts or roughly an entire month of no operation.
Annex F also discusses how to improve availability. Almost anyone at 0.9 would want to do this, but so would many whose facilities are currently at 0.999. That's Part I of Annex F. In Part II, it provides a methodology for developing and implementing functional performance tests (FTP) for critical operations. Do you have FTPs for any of your operations?