Pe2bz.philpem.me.uk

be less than 47 dB for coherent channel cable systems,when measured with modulated carriers and time Measuring averaged.”
In other words, the visual carrier must be at least 51 dB above any interfering signals, except (there isalways an “except”) in a system with harmonically cable system related carriers (HRC). In an HRC system, the visual
carrier must be at least 47 dB above the distortion prod-uct that falls at the visual carrier frequency (composite How to comply with
triple beat or CTB). This higher level of distortion is FCC regulations distortion allowed because the synchronous nature of the CTB in
an HRC system is less objectionable to the viewer than Editor’s note: Last year, the NCTA revised its
the asynchronous beat in a non-HRC system.
Recommended Practices for Measurements on Cable Test requirements
Television Systems to reflect new technical standards imposed by the Federal Communications Commission. The intent of this measurement is to accurately mea- This is the fifth installment of a series of articles that sure the visual carrier level, and then just as accurately focus on specific test parameters to explain how and measure the average power of the distortion products in the visual bandwidth. The difficult part of the mea-surement is accurately determining the level of the dis- The distortion section of the NCTA Recommended
tortion products in the presence of noise and video Practices thoroughly discusses the process and philoso- phy of measuring distortions in a broadband network.
The FCC Technical Standards [76.605(a)(8)] specify The intent here is to focus on some key points of the the distortion performance at the output of the sub- NCTA document and hopefully provide additional scriber’s terminal. This makes good sense because poorly designed convertors or excessively high levelscan contribute to the distortion. One statement in part FCC requirement
76.601(c)(1) which often gets overlooked reads: Section 76.605 (a)(8) says that “the ratio of visual “The measurements may be taken at convenient signal level to the rms amplitude of any coherent distur- monitoring points in the cable network: Provided, that bances such as intermodulation products, second and data shall be included to relate the measured perfor- third order distortions or discrete-frequency interfering mance of the system as would be viewed from a near- signals not operating on proper offset assignments shall This statement leads to the understanding that “good engineering practice” will allow creative methods to be Figure 1: Distortion - test equipment setup
used to make these measurements, as long as the results can be related to the performance at the sub- scriber’s terminal. From a practical standpoint, what the FCC is trying to accomplish with these specifica- tions is the guarantee that your system exceeds the minimum performance standards. If the measurement technique is based on sound engineering principles, a repeat of the tests by more traditional test methods will It is extremely difficult to measure distortion through the convertor because of the low signal levels available and the signal processing that occurs in many convertors. In consideration of this, the NCTA engi- neering committee took the approach of measuring the set-top convertor’s performance independently andcombining it mathematically with the distortion mea-sured at the system test points. When done properly, Figure 2: Third order distortion alternate test equipment setup
this method will do an excellent job of relating themeasurement to the output of the subscriber’s terminal.
Interruptions vs. non-intrusive
The most accurate method for measuring distortion requires the removal of the video modulation for CSO and the removal of the visual carrier for CTB. Because CED: COMMUNICATIONS ENGINEERING & DESIGN OCTOBER 1994 Figure 3: Typical normalized system distortion vs. channel
carrier and measures the AM modulation component on the carrier. This AM component corresponds to the CTB which is at the carrier frequency. The advantageof this method is that it does not require removing the carrier at the headend, but it still requires the removal of video modulation on the channel under test.
Method 3 eliminates the need to disrupt an active channel by measuring the distortion in an unused portion of the band and extrapolating this reading to the portion of the frequency band with the worst performance. Theaccuracy of this approach is dependent upon the accura- cy of the system distortion characterization. It is less accurate than other methods, but when used with cush-ion for error, serves as an acceptable method for guaran- In all three methods, the NCTA Recommended of the necessity of eliminating service interruptions, the Practices provides a step-by-step approach which will approach of the NCTA engineering committee was to not be repeated here. Instead, this article will address a provide methods of measurement which will minimize, few additional items. A new method (Method 4) will be and if possible, eliminate these interruptions. It is high- presented which has recently been introduced in several ly recommended that every effort be made to become new pieces of automated test equipment. For more comfortable with a procedure for this test which will information on the details of the procedures, please not require signals to be taken off the air.
refer to the NCTA Recommended Practices.
Three measurement methods are discussed in the NCTA Recommended Practices. Method 1 is the tradi- tional process which requires removing the video mod- See Figure 1 to view the distortion test equipment ulation for CSO and the visual carrier for CTB. This is the most accurate method but requires a service inter- 1) Measure the peak level of the visual carrier level 2) Remove the video modulation for CTB or remove Method 2 is an alternate method which uses a CW Reader Service #
Reader Service #
Ad ABC Cable Products.39.70 Siecor.12, 45.25, 79
Index American Lightwave Systems.15.29 Sumitomo.2.6
Product Showcase
Fujitsu Network Transmission Sys. .27 .52-53 CED: COMMUNICATIONS ENGINEERING & DESIGN OCTOBER 1994 3) Measure the average value of the distor- Figure 4: Set-top convertor distortion - test equipment setup
accurate and repeatable results because the masking signals in the band of interest are eliminated. Unfortunately, it also requires that somebody or something at the headend dis- ables the modulation, and/or carrier, for the duration of the measurement after the carrier reference is stored. Because of the video aver- aging and slow sweep speed required with the 30 kHz IF bandwidth, this measurement can or at a 6 MHz increment in an unused portion of the band. If the system is well behaved and flat, an adjacent visual carrier may be used as The trade-off of accuracy in this method is equipment set-up for third-order distortion.
the visual carrier reference. If this approach is offset by the ability to constantly monitor the used, the result must be adjusted for the differ- performance without service interruptions. In 1) Record the voltage of the carrier refer- ence in system gain and tilt between the fre- order to maintain good engineering practices, quency of the measured carrier and the fre- the predicted error in this approach must be 2) Record the level of the CTB imposed on added to the FCC requirement. For instance, if The distortion measured in Step 2 (a nega- you suspect this may contain 3 dB of error, tive number) is adjusted by the correction fac- then the -51 dBc specification you normally tor from Figure 3 to represent the worst case test to becomes -54 dBc to guarantee compli- system distortion. For example, if the distor- ance. If your system is operating close to the tion measured at channel 37 in Step 2 is -58.3 limit, then this approach will not have enough The biggest advantage of this method is that dBc, the worst case distortion at channel 11 accuracy to keep you in a safe zone.
communication is no longer needed with theheadend to turn the carrier off for CTB if achannel is available for a CW carrier. It stillrequires an inactive channel, and in addition,hum and cross modulation will appear as AMcomponents, and are indistinguishable fromthe CTB.
Method 3 is a three-step process. The first step is to characterize the distortion of the sys-tem channel by channel. The second step is tomeasure the distortion at a frequency in anunused portion of the band. The third step is toextrapolate this measurement to the worst casechannel in the system by using the characteris-tic generated in the first step.
The system characterization is the key to the accuracy of this method. Either Method 1,2 or 4 may be used to make this initial set ofmeasurements. The characterization requiresinterruption of service, but only needs to bedone once unless the system configurationchanges. This must be done for each portion ofthe system with unique channel loading, sys-tem tilts, AML or fiber links, amplifier spac-ing, etc. The characterization will stay relative-ly constant until major changes such as hard-ware layout or signal level changes occur inthe system.
Figure 3 is a typical plot of the beat distrib- ution and will not vary significantly from sys-tem to system. The plot is normalized to the THE PREMIER MAGAZINE OF BROADBAND COMMUNICATIONS OCTOBER 1994 Figure 5: Beat-near-noise correction
of meeting the FCC requirements. Fortunately, the performance of the convertor is typically much better than the system itself and has littlecontribution to the overall system’s distortion performance. The first step is to know whether the convertors are volume control (baseband) or In the case of baseband convertors, because of the signal processing on the demodulated method to recommend. The safest approach is to use specifications provided by the manufac-turer. The specifications should be generated using conditions similar to the system operat- ing conditions seen at the subscriber’s drop, both in signal level and channel loading. Ifmeasurements are required, stick very close to channel, but the hardware set-up is the same.
the manufacturer’s recommended procedure.
For RF convertors, any procedure used for on all unscrambled channels can be monitored measuring these should have the approval of which recently became available in some auto- automatically without service interruptions. On the convertor manufacturer, because there are mated test equipment. It allows the measure- channels 5 and 6 in a standard channel plan, a wide variety of units available. The proce- the CTB component is offset from the visual dure outlined by the NCTA replaces the nor- interruption of service by measuring the distor- carrier, allowing CTB to be monitored on these mal visual carrier with a carrier offset 250 kHz tion during quiet lines in the vertical blanking two channels without turning off the carrier.
to 500 kHz below. This keeps the AGC of the interval (VBI). It is typically an automated Measuring the convertor distortion perfor- convertor operating at a normal level and measurement which requires a non-scrambled mance by itself is perhaps the most difficult part allows the CTB product to be measured nextto the substitute visual carrier. Because of thedynamic range of this measurement (typically>70 dBc), a preamp is necessary between theconvertor output and the analyzer input.
Using the set-up in Figure 4 and the visual carrier offset by 250 to 500 kHz, the measure-ment procedure is the same as Method 1described in the NCTA procedure.
Calculating set-top distortion
sured at the system test point by any of themethods described, and the performance of theconvertors has been established, the two num-bers need to be combined to arrive at the FCCrequirement. The easiest way to do this ismathematically with Equation 5. You can seefrom the example that a typical convertor willhave less than a 1 dB contribution to the over-all performance.
Convertor distortion: DISTCONV = -75 dBc (3) Figure 5 provides a graphical approach for CED: COMMUNICATIONS ENGINEERING & DESIGN OCTOBER 1994 Figure 6: Combining two distortion values
Difference between two distortion values (dB) Potential errors
linearity. This will usually be specified as x.x dB / 10 dB or a maximum error of x.x dB.
measure the distortion products near the noise One step that is often overlooked is the addi- floor of the analyzer. If additional attenuation tion of this error to the measurement limit cannot be removed to provide higher signal level to the analyzer, then a correction factor For instance, if the maximum error of the should be used to adjust the measurement. The analyzer is 1.5 dB, then the FCC target speci- easiest way to determine this correction factor fication becomes -52.5 dBc instead of -51 is to remove the signal from the input to the dBc. If the attenuator is changed between the analyzer and note the change in level at the measurement of the visual carrier and the beat frequency. Figure 6 plots the correction measurement of the distortion, then the attenu- ator accuracy also needs to be considered.
Another common error is driving the input This can add another 1.0 dB or more of uncer- of the preamp or analyzer into overload. If this tainty, depending on the quality of the test is the case, the system distortion will be masked by the preamp or analyzer distortion.
Because of this, it is important to under- The way to check for this is by increasing the stand the accuracy of the instrumentation. It is attenuation at the input to the preamp or ana- conceivable that the potential error could be 3 lyzer by 10 dB and verifying that the distor- dB or 4 dB with a lower cost piece of test tion changes by only 10 dB. If the distortion equipment. It also means that if a higher products drop by more than 10 dB, you need dynamic range analyzer is used (one that can to start with a lower signal level, or use a make the measurement without changing the bandpass filter at the input to limit the input attenuator), you can eliminate the attenuator uncertainty and lower your target specification.
If a bandpass filter is used to limit the input power, make sure the passband of the filter is (and minimize the uncertainty of the measure- flat across the band of use, from the visual ment) it’s a good idea to verify the log scale carrier to the frequency of the distortion. This fidelity of the analyzer with a precision attenu- can be tricky when using a tunable filter and ator. A good attenuator will allow the analyz- automatic test equipment. Prior to selecting er’s performance to be checked. Remember, the automatic measurement mode, adjust the this is a comparison measurement between a filter by using the noise floor of the analyzer high level and low level signal, so absolute to determine the filter’s location.
accuracy is not the concern for this particulartest. What is important is that when you step Test equipment errors
the attenuator 60 dB, the signal on the display In order to get accurate results in any of the changes 60 dB. Log scale fidelity is a perfor- mance criteria that will change with operating tributed by the test equipment need to be temperature, so this should be verified across understood. Because this is a high dynamic range measurement, the most important speci- Hopefully, this brief synopsis will make the fication is the log scale fidelity or log scale semi-annual proof tests a little easier. CED: COMMUNICATIONS ENGINEERING & DESIGN OCTOBER 1994

Source: http://pe2bz.philpem.me.uk/Comm/-%20-%20Reference-Design/-%20Noise/MeasuringCableInducedDistortion/E206CATVAB6.pdf