 |
 |
|
|
|
|
|
|
||||||||||||||||||
|
|
|||||||||||||||||||
|
|
|||||||||||||||||||
|
|||||
|
|
IV. Is It Really Power-Line Noise Or Something Else? (By Terry Rybak, automotive EMC engineer)
A good first step is to eliminate some obvious sources. Although not usually necessary, you can quickly determine if the problem is being generated within the affected device by a simple test. Have the customer remove the antenna connection to the radio in order to see whether the noise goes away. If no change or little change in the noise results, the problem with the receiver or its power supply, or the RFI source may is located near the receiver or connected to the same AC circuit. Typically this is not the case however. I mention it only for situations where you suspect a problem with the customer's radio or television.
Assuming you don't feel this test is necessary, or observe a significant noise drop while performing it, proceed to verify the source is not in the customer's house. The proliferation of electronic devices and electrical appliances can often result in a plethora of confusing and hard to identify sources. Many of these sources are actually the cause of harmful interference. Follow the steps below to determine if the source is in the home:
All steps should be performed while interference is active!
Here are some household items commonly found to cause interference:
|
|
These devices, when causing harmful interference, are in violation of Federal Communications Commission rules and regulations and can be a nuisance to the customer and their neighbors. It is important to have the offending device repaired or replaced to ensure normal safe operation. Many sources of radio and television interference are also caused by arcing. The arcing will generate heat and may signal a fire hazard.
If the noise source is not in the customer's home, check with the closest neighbors. The place where the interference is the most intense may indicate the source of the disturbance. If one of the neighbors has a similar problem, ask him, or her, to run the breaker test to try to locate the faulty equipment. A household appliance or electrical device rarely causes interference that extends beyond a few houses on a secondary system.
Note that, if the source is not in the customer's home or a neighbor's home, the noise is originating from a source that is beyond the customer's control. Direction finding techniques may then be used isolate the noise to a particular residence or an area of the utility's power-line system. We'll be exploring some of these techniques in a later section.
How To ID Power-Line Noise
Many electrical devices, such as electrical motors, tools and appliances, can cause interference. The types of interference differ greatly from one electrical device to another. Interference caused by a computer, for example, is not the same as that produced by a household appliance.
Noise that varies with the time of day is related to what people are doing, usually pointing to some electrical device or appliance. Noise from consumer type devices, as opposed to power-line noise, will often come and go with periods of human activity. It will frequently correlate with evenings and weekends. Unless it is associated with climate control or HVAC system, an indoor RFI source less likely to be affected by weather than power-line noise. The importance of maintaining a good and accurate interference log cannot be overstated. Ask the customer to record date, time and weather conditions. Correlating the presence of the noise with periods of human activity and/or weather often provides very important clues when trying to identify power-line noise.
Often Weather Related
If the interference appears and varies in intensity depending on weather conditions (dry or damp weather, or wind), and if the breaker test excludes a source inside the home, the interference may be caused by faulty components associated with the electrical power-lines near the home. Wet weather may temporarily reduce or eliminate the noise by shorting out spark gaps on the power-line. Windy weather may cause the noise to vary or even stop for a while, as loose hardware is affected.
Is There A Smoking Gun?
 Figure 3. The 60-Hz signal found on quiet power lines is almost a pure sine wave, as shown in A. If the line, or a device connected to it, is noisy, this will often put visible noise onto the power-line signal, as shown in B. This noise is usually strongest at the positive and negative peaks of the sine wave. If the radiated noise is observed on a 'scope, the noise will be present during the peaks, as shown in C. |
While there may not be a smoking gun, power-line noise often reveals itself with some important clues. As previously discussed in Section 1, virtually all radio noise originating from utility company equipment is caused by a spark or arcing. The radio noise is only generated during the times when a breakdown and ionization of air occurs, and current flows between two conductors in a gap.
Once an ionized path is established in the gap, current flows at all parts of the cycle where the voltage is higher than the breakdown voltage of the gap. This typically occurs only near the positive and negative voltage peaks -- the times of highest instantaneous voltage. Sometimes, the gap may break down only on one polarity of the waveform.
Because power-lines carry 60 Hz ac, the voltage on them passes through two peaks each cycle (one positive and one negative) and pass through zero twice each cycle. This gives 120 peaks and 120 zero crossings in each second. Power-line noise follows this pattern, generally occurring in bursts at a rate of 120 (sometimes 60) bursts per second. This gives power-line noise a characteristic sound that is often described as a harsh and raspy hum or buzz. Because the peaks can occur twice per cycle, true power-line noise usually has a strong 120-Hz modulation.
Noise occurring in bursts at a rate of 120 bursts per second, and the resulting characteristic raspy buzz or frying sound, is often the first and most obvious clue of power-line interference. It is typically a broad banded type of noise starting at the low end of the radio spectrum. Power-line noise is usually stronger on lower frequencies. It occurs continuously across each band, up through the spectrum to some upper frequency where it will taper off.
A good test for the 120 Hz burst rate for both indoor and power-line noise sources involves an oscilloscope. The oscilloscope should show the bursts occurring every 1/120 seconds, or 8 1/3 ms. Look at the suspect noise from a radio's audio output using the AM mode. Use the wide filter settings and tune to a frequency without a station. Power-line noise bursts should repeat every 8.33 ms. If this is not the case, you probably don't have power-line noise. See Figure 3.
Alternately, you can perform a similar test if the noise pattern is visible on a TV set. The noise occurs in two horizontal groups or bands. Typically these two bands drift slowly upward on the screen. One group is a result of arcing during the positive half of the 60 Hz sine wave. The other group is a result from the negative half of the sine wave.
Note: The slow drift upward is caused by a slight difference in the power-line noise burst rate and the rate at which the TV images are transmitted. The TV images are transmitted at a rate of 59.94 Hz. This is because when television was first developed, a 60 Hz vertical scan frequency was selected so that any power-line noise would remain stationary on the screen, and be less annoying. When the color burst signal was added some years later, the scan frequency had to change to accommodate the color burst signal. Power-line noise occurs at 120 bursts per second. Since the power-line noise burst rate is almost twice the TV rate, two synchronized bands of noise appear on the screen. The slight difference in frequency causes these two bands to slowly drift upward on a TV screen.
It is usually best to perform this test at
the lower VHF TV channels and with an antenna (as opposed to a cable
hook-up). In addition, the positive and
negative power-line noise burst may also have slightly different
characteristics. This can cause each
half of the cycle to have a slightly different pattern on the screen. As you turn the channel selector to higher
frequency channels, the interference should diminish. If the interference can be observed on UHF channels, the source
is probably relatively close by. See
Figure 4.
| 
| 
|
Figure 4. The power-line interference shown in A is not typical. In this case, the noise occurred as a result of two sources that were out of phase from each other. The interference from each source overlapped on the screen. Once one phase was corrected, the more typical upward drifting horizontal snowy bars can be seen. Both noise sources were corrected in C and the noise was gone! | ||
 Figure 5. Power-line noise can be generated by a variety of sources, many of which are not intuitively obvious to a beginning RFI investigator. In this case, arcing between a staple and a ground wire was the source. This is a frequent cause of power-line interference. |
What to Look For
As previously discussed in Section 1, corona typically does not cause radio noise. Radio noise is almost always caused by a spark or arc across an air gap. (There are also many other non-arcing sources, such as lights.) Any voltage across an air gap can cause radio noise -- even ground wires, neutral wires and wires not directly connected to a power-line.
Typical culprits include broken or loose hardware such as bolts on wood cross arm brackets, a broken lightning arrestor lead wire, inadequate hardware spacing such as a gap between a ground wire and a metal staple, metal tags left on hardware, or metal objects thrown on the power-line. Any metal parts that are not well insulated from, or well connected to, one another may form a spark gap. See Figure 5. We'll discuss more of the specifics later the next section.
