Field Acceptance Tests
Once the vendor has fully installed and optimized your system, they will perform three types of tests:
- Functional tests
- Coverage tests
- Burn-in test (sometimes skipped)
These tests are expanded versions of the previously-conducted Factory Acceptance Tests. They should include thorough examination and measurement of installation quality, especially the antenna systems, lightning protection, power systems and power supply in accordance with industry standards. Any tests that were only simulated during factory testing (if conducted with limited set-up) should now be performed live across the whole network, in their entirety.
It is especially important to verify system’s resilience. Automatic or manual switchovers (power, backhaul, and controllers) need to be forced into failure mode and tested for reliable switching.
The crucial parts of the field acceptance testing are the coverage test procedures. Make sure these are performed in full accordance with the agreed test plans, and the results meticulously recorded.
All coverage tests should be conducted with participation of your representatives. It is essential that you have high level of expertise available to you, which, in most cases, will mean that you need an experienced consultant to oversee this process.
Coverage tests are typically a combination of automatic, computerized signal measurements and voice quality tests. Measurements include signal strength (RSSI), Bit Error Ratio (BER) and Message Error Rate (MER) that can be automated and archived. Because voice quality testing is highly subjective, we recommend that the test participants’ ears are “calibrated” by the entire group listening to audio samples just before the testing commences – each vendor can provide the samples in advance. To minimize subjectivity, test crews, whether in field or stationary, should consist of at least three people, so that a reasonable level of consensus can be achieved.
There is no consensus as to which type of coverage test is best. Signal strength and Bit-Error-Rate tests are run automatically and thus cheaper than voice quality (DAQ) tests which require significant amount of manual labor and time. However, DAQ test is seen by many as the ultimate proof of system’s performance. Others cite a strong correlation between DAQ and BER measurements and so are comfortable foregoing the DAQ testing.
Signal strength testing is not sufficient by itself for two reasons – the automated test cannot differentiate between the valid signal and interference and, in case of simulcast systems, cannot recognize areas with high level of Time Delay Interference.
If performed formally, burn–in tests should last 30-60 days and can start as soon as the vendor finishes system optimization. To conduct the tests, a limited number of trained users will start using the system and formally report any problems.
It is important to agree on a number of issues up front.
- What kinds of failures, if any, reset the test?
- What are the expected response times?
- What formal reporting of the undertaken actions and results is required when problems are reported?
It is very important to agree up front who can use the system during this time, and how it can be used. System usage during your burn-in tests must not trigger the “beneficial use” definition and start the warranty period prematurely.
All issues discovered during the burn-in test need to be carefully tracked until they are fully resolved.
Click "NEXT" to learn about System Cutover or select a topic from the menu below.