PDF of this tutorialFacts Affecting Test Head Accuracy
It is essential that a line test system provide accurate test results. The accuracy of the test head the system uses depends on the following factors:
- the ability of the test head to calibrate itself automatically
- the number of bits used in the analog-to-digital conversion
- the measurement algorithms used (three-terminal test technique)
- rejection of unwanted noise (power-line influence, etc.)
It is necessary to have self-calibration of the test device, because parts change with age and temperature. Over time, these changes in the parts add up and produce measurement errors in devices that are unable to compensate for these changes. An advanced line testing device can self-calibrate and adapt to these changes by using software algorithms. A high level of rejection of unwanted noise is an extremely important feature.
How Bypass Pairs Limit Test Accuracy
It is difficult to achieve test accuracy when a digitized bypass pair is involved. That is because there are two essential features of an accurate test that are in conflict. It is necessary to use many bits—fourteen or more—to obtain accurate measurements. But it is also necessary to get these bits to and from their destination quickly to eliminate a time skew that could affect the accuracy of the measurements. Using more bits generates greater accuracy but also creates a delay in getting the bits end-to-end. This delay affects the accuracy of the measurements. For example, more bits are needed to do finer analog/digital resolution. The slower sampling rate of the digitized pair makes the bits arrive at their destination too slowly for accurate measurement. If fewer bits are used, they arrive faster but result in larger granularity—a poor image, which makes test results difficult to interpret. The 56–kbps bandwidth typically used with digitized pairs makes it impossible to send enough bits fast enough to get an accurate test. Depending on the type of test and algorithm used, test results will be affected by the slower sampling rate or the use of fewer digitized bits. The double conversion process also robs the measurement of accuracy.
The Importance of Measurement Algorithms
Many switch-based testers only perform two-terminal tests, much like a handheld multimeter. These tests cannot resolve multiple faults correctly because the results cannot differentiate between the individual electrical components of the faults, such as resistance and capacitance. A remote test unit (RTU), however, performs three-terminal tests that isolate the various components of multiple faults between all three points in a telephone circuit so that the true readings are displayed. In addition, the RTU algorithm changes the sequence of subsequent testing steps according to the information it receives from initial tests, a process that also contributes to greater accuracy. These measurement algorithms also facilitate the reduction of unwanted noise, especially power line influence.
Limitations of Switch-Based or DLC–Based Devices
A number of limitations of switch-based and DLC–based test devices have been discussed in the tutorial thus far. The limitations of these devices often result from the economic model on which they are based. Typically, these devices are built under the same economic model as the transmission-path components of the switch or DLC. Therefore, they are less likely to use the more costly, higher-bit-count analog-to-digital converters and instrumentation amplifiers that a stand-alone line test system uses. In other words, in test devices built into switch and DLC products, the priority is controlling costs rather than promoting test accuracy.
Costs and Benefits of Various Test Devices
In doing a cost and benefit analysis of test devices, it is important to realize that a certain level of technology is required if testing is to be an effective tool in cutting costs. Less advanced converters and instrumentation amplifiers do not have the level of technology to provide the needed test results. The most obvious cost-cutting benefits of accurate test results—such as the results obtained from an advanced stand-alone test system—are in reducing or eliminating dispatches, especially in no-trouble TOK cases. It is clear that in evaluating various types of line-test equipment, it is important to focus not on the cost per line to let calls go through but on the cost per trouble to fix anything that has gone wrong. When viewed from this perspective, an advanced stand-alone line test system is invaluable.
