Advertisement
Articles
Advertisement

Preventing ingress in the return path

Mon, 09/30/1996 - 8:00pm
Kevin J. Oliver, Product Marketing Manager, CATV Division, Wavetek Corp

With new opportunities for revenue available and competitive threats building, the focus of the cable industry is on making cable networks quality, interactive, two-way communications pipelines. As such services as high-speed data, interactive TV applications, telephony, and even Internet TV are being considered or activated, one of the big questions being asked is how do I handle ingress encountered in the reverse path? To their dismay, many are finding that there is not an easy solution to this question. However, with proper investments in time, sweat, effort and equipment, they are succeeding.

i_9610c1.gif

The major problems that operators are having with the reverse path have to do primarily with ingress and noise, but also with alignment. Proper reverse path alignment and frequency response are critical to reverse path operation. As is true with the forward path, sweeping of the reverse path is widely regarded as the optimal approach to testing and alignment. Other methods such as using limited frequency carrier generators are cumbersome, and more importantly, miss testing portions of the reverse spectrum for frequency response.

Ingress, however, is the dominant problem that operators of two-way services are struggling with most today. This is because, in large part, of the noise funneling effect of the reverse path, but also because of the inexperience of dealing with ingress and the reverse spectrum on a daily basis. Many a technician is finding the reverse plant to be a strange and delicate network. In the forward path, proper carrier-to-noise (CNR) levels provide for clear picture quality with gradual, visible degradation as noise levels rise. However, in the digital world of the return path, the effect of high levels of noise and ingress, or low CNR, may not be noticed until they completely destroy communications or, in the case of spread spectrum technology, drastically slow down network performance.

Where does ingress originate? It is becoming widely accepted that 95 percent of ingress problems come from the drop system (drop and subscriber premise), and only five percent of ingress originates in the trunk or feeder network. According to sheer chance, one would expect more problems from the drop system of an HFC network because more coaxial cable and more connectors are actually in the drop system than in the trunk and feeder network.

Also, the hard line feeder network is typically well- shielded and better maintained. The drop system is entirely the opposite. It has been estimated by operators of advanced services that up to 50 percent of the cable in drop systems and in-home wiring is not capable of handling high-speed data. In addition, the drop system is the least accessible, least controllable and least maintained portion of the network.

Do I really have to deal with ingress, or can I just cover up the problem? In an effort to speed the activation of the reverse plant, many operators are installing high-pass filters or windowed filters on every home, except trial homes and two-way subs. Although fairly effective and affordable, heavy reliance on filtering can really only be a limited or temporary solution. Once two-way subscriber penetration increases, and more reverse spectrum is utilized, more and more filters will need to be removed, redesigned or replaced. Ingress will begin to grow unless properly handled, or if subscribers remove the filters, the ingress will immediately appear. Reliance on filtering could also lead to quite lengthy new advanced service hook-ups unless more proactive measures for ingress reduction are taken at the time of original install.

It has also been suggested that to improve the CNR to acceptable levels, that instead of lowering the noise, the carrier level could just be increased. This means that the reverse path modems, set-tops and such would have much higher outputs than the current typical 50 or 55 dB. One of the problems with this approach is that, according to studies, drop cable across the U.S. currently has an average of only 50 dB worth of shielding. Thus, 65 or 70 dB output devices distributed widely in homes could begin to produce significant leakage in the sub-band frequencies that would be interfering and easily noticed by CB and ham radio operators.

i_9610c2.gif
Figure 2: Installation technician using combination SLM, leakage and ingress meter to test the quality of an install.

Can I just wait until ingress and noise problems occur, then fix them? Operators are finding that troubleshooting for ingress sources is time-consuming and frustrating. Experienced technicians say it takes from two hours to two days or more to find a single ingress source, and weeks to clear an entire node. Meanwhile, this can be keeping one or more advanced service customers from operational service. Aggravating the troubleshooting process is that most of the interfering ingress problems are impulsive or intermittent. Just as a tech gets close to the problem, it goes away, and he has to wait for it to show up again, perhaps today, maybe tomorrow. In addition, when the source is narrowed down to a house, the tech must then wait for permission to enter, which may require an additional service call.

Will looking for ingress and leakage only when sweeping and maintaining the hard line feeder network help? Testing for noise and ingress when sweeping or aligning the reverse path is a good practice and very helpful at providing a summary indication of node/network performance and identifying when a problem exists. However, the reverse noise/ingress feature on most sweep and alignment systems only displays to the field technician the combined noise at the headend from a node or multiple nodes. From this alone, it is not known whether the point of ingress is up- or downstream from the trunk test point, or in the drop system or home, or even within a specific node. Additional troubleshooting methods, such as using a spectrum display on the sweep meter or portable spectrum analyzer, are required to locate the source. Furthermore, remember that only an estimated five percent of ingress points are in the hard line trunk, so even if all ingress and noise points were eliminated in the feeder network while sweeping, 95 percent of the problems would still be unresolved.

i_9610c3.gif

Testing for leaks while maintaining the network and while driving throughout the system is also a good practice. However, driving around in a truck does not often take one close enough, or is not always sensitive enough to find all the sources of ingress that can cause problems. Setting the typical alarm threshold of 20μV/m for leakage detection while driving in a vehicle may work for FCC proofs, but not for finding damaging ingress sources. Much more effort and sensitivity must be used. The leakage equipment must get closer to the majority of the ingress and noise sources.

The real answer is act now and attack ingress and noise where it originates—at the subscriber's home. This provides the best chance to prevent problems and prepare the communications pipeline for the future. Even if advanced services are not being activated in a system, proactive preparations now will be a big benefit later. With access to subscribers' homes very limited, operators must take advantage when access is given-during the installation.

i_9610c4.gif

Utilizing good installation practices and testing procedures can be one of the most important steps in minimizing noise and ingress and ensuring good performance of the reverse plant. This is opposite to conventional cable TV forward-path-only wisdom, which says: "take care of the feeder network, and the customer should have good service," or "no one else will suffer if one person has a bad drop." This is obviously not the case anymore.

After all, noise emanates from every home and, because of the noise funneling nature of a cable plant, noise from any one source can, and does affect many others. To be complete, installations for all customers, not just two-way subscribers, should be done carefully.

Quality installation devices need to be installed, and good installation practices followed. Using tri-shielded or better cable, quality weather-proof connectors, and proper grounding and bonding techniques are almost a necessity. Operators are learning that "little" things make a big difference. Potentially one piece of bad drop cable or one poor connection can wipe out an entire node for reverse services.

Diligent installation testing practices and comprehensive test equipment can help to ensure optimal forward and reverse path performance. The catch is that fairly comprehensive forward and reverse path testing is required to be effective, but to be practical for widespread use by installers, the test equipment must be affordable and easy to use. A spectrum analyzer, sweep receiver, signal analysis meter and leakage meter will help find and fix problems in a home; however, this combination is cost prohibitive and too complex for installers.

The good news is there are effective, comprehensive and cost-effective testing solutions available today. An easy-to-use installer's meter that combines simple signal level, leakage and ingress testing has the ability to verify forward path signals, identify and find leaks, and test the level of ingress and noise being generated by the drop system. This package of tests in one instrument helps ensure the installations are satisfactory and prevents future service calls (see Figure 2).

Video and audio level testing with a signal level meter provides forward path verification. This is the end result of proper alignment and forward path network performance. When properly performed, this quickly and clearly indicates if all channels are being received at the subscriber's drop at appropriate system design levels. The equipment should be capable of simple verification of pass or fail and, when desired, more extensive and detailed troubleshooting. Any level discrepancies that may affect picture quality should be made easily identifiable to the installer.

i_9610c5.gif
Figure 5: Ingress with a common path distortion component as displayed on a common PC-based analysis software package.

Leakage testing is no longer just to help systems comply with the FCC requirements. Testing for leakage, or egress, is an effective way to find sources of ingress. Loose connectors and poor shielding are typical places where signals can leak out of a system, and as such, they are also places for electronic noise and over-the-air signals to enter the cable system. Because a majority of sources for ingress and leakage are in the drop system, the best time to look for them is during the install while the technician is nearby and has access to the home. Leakage and ingress sources can be found and fixed in minutes if the tech is close to the source and has the right tools, preventing hours or days of troubleshooting later. In practice, after all drop cable is installed and connections are made, the technician should then walk through the subscriber's home with a leakage meter to identify any noticeable leaks. A simple near field signal strength leakage display will assist in finding leaks within a home (see Figure 3). There really should be no acceptable leakage level that is totally ignored. Where leaks are found, the tech should do his best to fix or improve the situation.

Testing the reverse path spectrum for sub-band signals being generated in the drop system improves the effectiveness of finding ingress sources and common path distortion. As stated above, leakage is a good way to find sources of ingress, but it is not 100 percent effective. Because of different propagation characteristics, different frequencies travel through materials and air differently. Also, in the case of very small leaks or ingress points, a lower level forward path signal may not leak out, but an external signal source (for example, a nearby CB or ham radio signal or electric heater) may produce a strong enough sub-band signal that ingress results. Thus, in some cases, a leakage meter alone may not find all the ingress points.

The ability to scan the reverse spectrum and display any signals found provides an additional method for identifying and finding ingress. A display of the reverse spectrum with clear pre-set limits allows the installer to easily identify if ingress exists (see Figure 4) in the drop system. Because this is available in an installer's meter, part of the standard installation practice should include a final test in which the installer views the reverse path before he hooks the subscriber to the tap. This will help to prevent hooking up a new drop that may immediately or in the future cause disruptions in reverse path services.

In addition to ingress detection, scanning the reverse spectrum can reveal other common reverse path problems such as common path distortion. Common path distortion can result from corrosion or oxidation on connections of dissimilar metals, as with many cable plant connections, causing a diode-like effect. When forward path signals pass through this diode, potentially harmful second- and third-order beats every 6 MHz can be created in the reverse path (see Figure 5 for an example showing major beats every 6 MHz). Common path distortion is being found often in drop systems, as well as feeder plants. Without looking at the actual spectrum, this problem is hard to identify.

In summary, many of the problems and heartaches currently encountered in operating a two-way cable plant result from ingress and noise. Because the generating source of a vast majority of the ingress and noise is in the drop system and the home, making strong efforts and investments during the installation will earn returns in the future. Combining quality installation practices and materials with comprehensive installation testing procedures is helping to optimize the operation of the reverse path and minimize future troubleshooting efforts.

Advertisement

Share This Story

X
You may login with either your assigned username or your e-mail address.
The password field is case sensitive.
Loading