For details of the Proof of Performance test requirements, refer to the Code of Federal Regulations (CFR) 47, part 76. This book is available from the US Government Printing Office, Mail Stop: SSOP, Washington, DC 20402-9328. Many cities in the US have US Government Printing Office stores that sell Government publications directly to the public.
Who must run the tests?
All systems with 1000 or more subscribers. Smaller systems are required to comply with the Cumulative Leakage Index requirements (including quarterly measurements) but are not normally required to demonstrate proof of performance of the other technical standards. The technical standards are expected to be met by all systems -- but small systems don't need to prove it.
When do the tests need to be run?
Most tests need to be conducted "at least twice each calendar year (at intervals not to exceed seven months)". The "24 hour" tests "shall be measured and recorded , along with the date and time of the measurement, once every six hours (at intervals of not less than five hours nor more than seven hours after the previous measurement), to include the warmest and the coldest times, during a 24 hour period in January or February and in July or August".
The 24 hour test is the only one required during those particular months.
The color tests (Chrominance to Luminance Delay Inequality, Differential Gain, and Differential Phase), need to be conducted at least triennially (once every three years).
How many channels need to be tested?
Many tests are required on all "NTSC or similar video channels" of the system. Other tests are required "a minimum of four (4) channels plus one additional channel for every 100 MHz, or fraction thereof, of cable distribution system upper frequency limit (e.g. 5 channels for cable system upper frequency limit of 101 to 216 MHz; 6 channels for cable television systems with a cable distribution system upper frequency limit of 217-300 MHz; 7 channels for cable television systems with a cable distribution upper frequency to 300 to 400 MHz, etc.)."
All channels are required to meet all of the technical standards. However, for certain of the technical standards, you only need to demonstrate "proof of performance" on a sampling of the channels..
How are the test channels selected?
"The channels selected must be representative of all the channels within the cable television system." Generally, systems select channels from across the system's spectrum.
How many test sites do we need?
For systems from 1000 to 12,500 subscribers, tests "...... shall include measurements taken at six (6) widely separated points. However, within each cable system, one additional test point shall be added for every additional 12,500 subscribers or fraction thereof (e.g., 7 test points if 12,501 to 25,000 subscribers, .... etc.). "
Also, there must be at least one test point "for each portion of the cable system served by a technically integrated microwave hub".
Note that this does not mean a test point for each fiber node.
Where, in the system, are the tests conducted?
Test points "shall be balanced to represent all geographic areas served by the cable system. At least one-third of the test points shall be representative of subscriber terminals most distant from the system input and from each microwave receiver (if microwave transmissions are employed), in terms of cable length".
Measurements may be made at convenient monitoring points in the system so long as the data reflects system performance for nearby subscribers.
Some measurements (visual signal level and 24 hour tests) are required to be measured "at the end of a 30 meter (100 foot) cable drop that is connected to the subscriber tap, ..." Usually, all tests in the field are conducted at the end of a 100 foot drop cable. The 100 foot drop cable simulates typical drops provide to homes. For tests at the output of a set-top converter, connect the converter to the end of the drop cable.
Are there specified procedures for each test?
In general, no. However, good engineering practices are to be used for all tests. For some measurements, the rules refer to documents such as the "NCTA Recommended Practices for Measurements on Cable Television Systems, 2nd edition, November 1989". A review of that document is an excellent starting point.
Of course, measurement equipment and techniques are continually evolving. There is a lot of test equipment available today that make the tests much easier than some of the procedures described in the NCTA documents. Before you decide to use any procedure, you should of course, have a thorough understanding of the test requirements and the algorithms used by your test gear. Do not blindly trust automated test gear simply because it provides "numbers"!
Where do we send the test results?
Unless told otherwise by the FCC or your franchise authorities, you simply keep the data on file for at least five years. The test data is required to be made available to the FCC or local franchise authorities upon request.
What happens if we don't pass?
I've run tests on over 50 systems, many of them multiple times. So far, one system has "passed" all the requirements (it passed one of the two times I ran tests on it).
Given the complexity of today's cable systems, the thousands of mechanical connections involved, the number of back hoes and other rodents in our society; something, somewhere, is likely to be broken at any given point in time.
The most important thing is to run the tests, find the problems, fix them, and keep a log of the whole process.
Of course, carrier leakage problems should be addressed as soon as they are discovered. Authorities tend to frown when aircraft start honing in on bad splices.
Useful Definitions (from part 76):
Subscriber Terminal: "The cable television system terminal to which a subscriber's equipment is connected." Basically, where your subscriber's equipment (TV, VCR, etc.) connects to your equipment (at the output of the set-top box you provide, at the end of the drop cable, etc.).
System Noise: "That combination of undesired and fluctuating disturbances within a cable television channel that degrades the transmission of the desired signal and that is due to modulation processes or thermal or other noise-producing effects, but does not include hum and other undesired signals of discrete frequency. System noise is specified in terms of its rms voltage or its mean power level as measured in the 4 MHz bandwidth between 1.25 and 5.25 MHz above the lower channel boundary of a cable television channel."
Back to Proof-of-Performance Summary page