cable tech facts issue 105

In This Issue

• Signal To Noise Ratio
• HUM
• Picture Effects
• Next Issue
• Industry News

Back To Cable Tech Facts main index.

In issue #104 of the Cable Tech Facts we covered two of the FCC mandated performance tests - In-Channel Frequency Response and Percent Modulation and how to document the tests results. In this issue we will cover more of the tests and also some tests that are not mandated by the FCC, but failure to test these critical areas can effect your system's performance.

[Top]

Signal To Noise Ratio

There is no S/N specification set by the FCC, however the S/N of the video signal is important to your overall picture quality. Similar the C/N or Carrier To Noise, the S/N can not be improved. If your signal at the headend starts with a poor S/N, you will deliver a poor quality picture to your customer. Since most systems measure C/N using a method that measures the noise outside the channel's video bandwidth, the actual picture quality with respect to noise is still unclear.

You can have a good C/N level and still have a poor quality picture due to poor S/N. Typically S/N is not a problem for most channels taken from a satellite using quality equipment. If a problem does occur, it is typically a result of an equipment failure or the result of a distant off-air signal being brought in by marginal means. While most system sources of noise are flat across the spectrum, noise that is generated by imaging devices, such as, microwave transmitters or satellite transponders is not flat. Generally, noise increases as the square of the frequency. This is a good reason to insert your own test signals using a VITS insertion generator.

Let's review the definition of signal to noise and its effect on picture quality

Definition: Signal To Noise is the ratio of the signal level to the level of the noise that is added to the signal. Since the effect of noise on picture quality depends on the level of the picture signal, the most meaningful measurement is the ratio between the noise and the signal level. S/N is the ratio of the luminance signal level from the black level to the white reference amplitude compared to the RMS amplitude of the noise signals. This is done after weighting the noise for perceived picture degradation according to EIA 250C or CCIR567. The recommended performance objective in the NCTA Recommended Practices suggest that a 53 dB S/N be maintained for good system operation. Signal To Noise can be computed by the following formula:

S/N = 20 log (V sync to White /Vnrms)

The effects of a poor S/N ratio on picture quality will be similar to that of a poor C/N problem. Pictures that contain high levels of noise will be grainy or snowy with "sparkles". Sharpness and resolution may also be poor at high noise levels. At the headend, a 55 dB S/N should be obtained.

To measure S/N at the headend, follow this procedure:

Measurement Procedure: Low level noise is very difficult to measure. Quality video equipment found in the typical headend will have S/N performance in the 55 to 60 dB weighted range. To compare this with an unweighted measurement, like C/N, subtract 6.8 dB from the weighted measurement. This 6.8 dB compensation factor is only accurate when the measured noise has a flat frequency response. Weighting refers to use of a weighting filter, such as the EIA 250C or CCIR 567 filter, to shape the noise frequency response similarly to the way a human eye would perceive noise. Higher frequency noise is less perceivable to the human eye. To measure a 55 dB weighted S/N, we should expect to measure ~ 4 mVRMS of noise in a 1V video signal.

S/N weighted = 20 log ( 4 mV / 1V) + 6.8 dB
S/N weighted = 55 dB

To make a S/N measurement, no test signal is required, any quiet line may be used. A quiet line is a horizontal line in the VBI with a continuous signal level between sync pulses as shown in Figure 1. Automated measurements are available on newer CATV instruments, and are more accurate than typical manual measurements. Manual measurements can be made by comparing the amount of noise riding on the blanking signal to the overall signal level of the video signal. The peak to peak noise measurement must be multiplied by .35 to get the RMS value of the noise. A test signal may be used that contains a black reference or significant blanking level signal.

Here are a few precautions when making S/N measurements

1. If permanently installing an insertion device into the channel's video path, be sure that only a single 75 ohm termination exists on each video line.
2. Before connecting a Video Signal Analyzer or Waveform Monitor/Vectorscope to any test point, be sure that it is in the high impedance mode.
3. Remember whether you looking for weighted or unweighted S/N and also remember that the 6.8 dB correction factor is only good if the noise measured is relatively flat across the frequency spectrum.
4. Distortions which fall in the video bandwidth may also be perceived as noise using this measuring technique. Be sure that your X-Mod, 2nd and 3rd order distortions, as well as any video linear distortions are lower than the noise level you are attempting to measure.

[Top]

HUM

The next performance test is Hum or Modulation Distortion Of Power Frequencies. Hum modulation is an FCC test that was reinstituted in July of 1992. It is a measurement of the low frequency AM distortion of an RF carrier. A DC restorer or AGC must be turned off in order to make HUM measurements. The following information provides detailed information on the FCC Rules and Regulations, a definition of the test, a measurement procedure and some helpful hints and precautions.

FCC Specifications: 76.605 (a) (10): The peak to peak variation in visual signal level caused by undesired low frequency disturbances (hum or repetitive transients) generated within the system, or by inadequate low frequency response, shall not exceed 3 percent of the visual carrier level. Measurements may be made on a single channel using a single un-modulated carrier...

76.601 (c)(1): For cable systems with 1,000 or more subscribers, but with 12,500 subscribers or less, proof of performance tests conducted pursuant to this section shall include measurements taken at six widely separated points with each mechanically continuous set of cables within the cable television system. Within the cable system, one additional test point shall be added for every additional 12,500 subscribers or fraction thereof. Such proof of performance 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 in terms of cable length.
... An identification of the instruments, including the make, model number, and most recent date of calibration, a description of the procedure utilized, and statement of the qualifications of the person performing the test shall be set forth.

Although the FCC only requires HUM to be measured on the distribution system and in a manner which will only measure HUM caused by problems in the distribution system, it is good engineering practice to measure the HUM contribution from the headend components, since this will represent the picture quality that will be delivered to your customer. This is also a good method of preventative maintenance, finding problems before they become noticeable to affect system performance.

The definition of HUM is the low frequency AM distortion of the video signal, generally considered < 400 Hz, and is the percentage of the peak to peak distortion to the peak signal level. This is approximately 2 times the percentage of the % Modulation of the low frequency distortion. Note the HUM illustrated in Figure 2.

[Top]

Picture Effect:

Hum distortion produces horizontal bars that slowly roll through the picture. One or two bars is common to HUM that is related to the AC power. The intensity of the bars will be proportional to the level of the HUM signal. Other low frequency distortions also can occur. Ground loops are the most common cause of small HUM distortions in the headend. The typical performance level in the headend is .5% to 1 %. HUM that is created in the headend will add to the HUM distortions caused by the distribution system. This addition may or may not be in phase, creating distortions that may not appear to be the expected sinewave or "clipped sinewave" form.

Measurement Procedure: For HUM measurements to meet the FCC performance tests, follow the procedures that are outlined in the FCC Recommended Practices or use a field strength meter or spectrum analyzer that has the built in measurement for HUM. To measure the HUM distortion that is generated in the headend, you may use any full field or test signal. Automated CATV video analyzers automatically switch off the DC restorer and AGCs and compare the amount of hum riding on the top of the sync pulses to the overall signal level of the video signal. HUM measurements cannot be made if the DC restorer or AGC can not be turned off.

[Top]

Next Issue

In the upcoming issue of Cable Tech Facts, we'll cover the remaining performance tests that include: Differential Gain, Differential Phase, and Chroma To Luma Delay. We will also touch on some additional tests that can help improve your signal quality. If you have any questions regarding the tests that were covered in this issue, or any related cable system questions, give your Area Sales Representative a call at 1-800-SENCORE (736-2673).

[Top]

Industry News

CableLabs Testing Digital Conformance CableLabs Inc. began digital video system conformance testing in Louisville, CO. The test will verify the compatibility of different vendor's compressed digital bit streams and decoders. The tests are being performed on MPEG-2 transport streams at the system layer using a common agreed upon interface in order to test the video, audio, data synchronization and multiplexing.

Ameritech Lands Cable Franchise

Ameritech has become the first regional Bell operating company to land a municipal cable franchise in the U.S. It will begin building a competitive system in Plymouth Township, MI. The telco is presently talking with as many as 30 other municipalities in its five-state region.

Sony Unveils Its First DSS Units

Sony Electronics unveiled is first DSS consumer hardware. Sony's introduction includes an LED signal seeker on the LNB; a 32-bit microprocessor-controlled on-screen menu, and a multiple-event program timer. Three Sony systems offer good, better and best selection. The product will be test marketed in Southern California during June. The step-up and advanced models will be available in August. The basic model will retail for $749 while the step-up model should retail at $849. The advanced model has a suggested retail price of $949.

[Top]

[Sencore Home Page]