View Full Version : The ABC's of Bleed Over

01-31-2012, 09:46 AM
The ABC's of Bleed Over.
One of CB radio's worst annoyances explained in detail.

CB radio was a popular and fun hobby for many people in the 1970's and it still continues to this day, albeit at a much reduced level of popularity. The mass media-influenced popularity of the hobby (or fad depending on your perspective) back then attracted people from all different walks of life to the 23 (and then later 40) channel band where they clustered together with other like-minded people within their local communities. In more densely populated areas, the amount of people flooding the channels could exceed the capacity of the band to support the users in a somewhat orderly manner. But even in situations where not all the channels were used, many users were subject to interference from strong local stations on other channels. This interference was commonly known as "Bleed Over", "Splash", "Splatter". Bleed over became a frequent irritant and the source of many on-air conflicts and would often precipitate deliberate jamming of those seen as causing the problem. But what is "Bleed Over", what are its causes, and what could be done about it? Well radio fans, just read on to find out the answers to those nagging questions.

While most radios were comprised of many individual channels, which in theory should accommodate an equal number of individual groups of users, the RF "wall" between each channel was not as thick as one might think. If you were receiving a station close by and turned to another channel, you might still be able to hear the other station splashing over. It is this splash, or "bleeding" over of strong stations which can interfere with your ability to receive weaker signals on your desired channel. As might be suspected, some radios picked up more "bleed" than others. Some operators were able to carry on conversations with little interference from strong locals, while others were completely shut down during those prime time operating hours. But bleed over was not just a phenomenon of an individual radio's receiver. Some bleed over was caused or exacerbated by improperly tuned transmitters. Since the bleed over phenomenon has a few different faces, it's helpful at this point to try to identify and explain in greater detail these individual contributors to the whole bleed over picture. So now, drum roll please, here they are:

1. Adjacent Channel Rejection (ACR). Adjacent channel rejection is probably the most well known and talked about aspect of bleed over with respect to comparing one radio's immunity against another. ACR is usually given as a receiver performance specification. The higher then number (in db) the greater the rejection of adjacent channel signals. But what does this actually mean in technical terms? Well, "Radio 101" states that any radio receiver will have a given design bandwidth. This bandwidth is designed into the radio based on the requirements of the signal you are receiving. A simple CW signal can be received with as little as a 100 Hz bandwidth. A typical AM signal on the CB band has a bandwidth of about 6 Khz (+/- 3 Khz), so the receiver has to be at least this wide to be able to hear the signal at the full audio fidelity. Ok, achieving that should be easy, and it would be if the CB band was only one channel wide. The receiver could be as broad as a barn door, and still work well in this application. Anything falling above or below the necessary 6 Khz could be ignored. But since the CB band is divided up into many channels, there has to be a limit on the receiver's bandwidth. The CB band currently consists of 40 channels, the center of each channel being 10 Khz apart from the next. A proper signal needs to fit within the 10 Khz spacing but still be wide enough to let the 6 Khz signal through. Fitting a 6 Khz wide signal into a 10 Khz wide channel leaves 2 Khz on either side of the normal signal to act as a guard band between any given channel. With this in mind, the engineers designed the receiver to allow the full 6 Khz wide signal through with little attenuation, but sharply drop receiver gain beyond that. There are several ways to accomplish this. The easiest method is through the use of Hi-Q tuned circuits. These provide acceptable performance, but are not always the best at providing a sharp cutoff. There are also crystal, ceramic and mechanical I.F. filters which provide a much sharper cutoff (Or "skirt" in response terms). Radios with high quality or multiple filters will usually have a much higher degree of ACR and a much sharper response skirt.
So how do you actually measure Adjacent Channel Rejection, and how do these number translate into daily performance expectations? In basic terms, a radio which claims to have an ACR value of 50db @ 10 Khz will need a signal 50 db stronger than the desired signal at 10 Khz away to make the same signal reading on the radio's meter. So how much is 50db in real world terms? A signal 50db stronger than an S1 signal would be just shy of S9, which means that a station on an adjacent channel giving you an S9 signal will put about an S1 level of bleed on your current channel. But this example alone doesn't tell the whole story. Remember that each CB signal is 6 Khz wide (+/- 3 Khz from center frequency) with modulation, and that each receiver's bandwidth is 6 Khz wide within adjacent channels that are 10 Khz apart. This means that the difference between the edge of the receiver's bandwidth and the edge of an adjacent channel's modulated signal will only be 4 Khz apart, or starting at 7 Khz above and below the desired channel center frequency. So the filter not only has to have deep rejection at 10 Khz, it also has to be sharp enough to provide reasonable rejection at as little as 6 Khz above and below the center frequency. So an honest ACR spec should give the numbers at both 6 Khz and 10 Khz away.
As you can probably guess, there are great variations in design between the low priced radios and the higher cost and better designed rigs. Look for radios with crystal filters to improve your chances of overcoming splash from adjacent channels. Poorly designed radios will get more adjacent channel bleedover. Usually adjacent channel bleed over is only a problem one or two channels away. Bleed over that extends to many more channels is probably the result of Splatter, Front End Overload, Image intrusion, or some form of Intermodulation Distortion.

2. Front End Overload (FEO). While a lot of lip service is paid to Adjacent Channel Rejection as the principle indicator of a radio's bleed over rejection, in far more cases, it is Front End Overload that causes the most severe problems for stations very close to one another. Front End Overload usually manifests itself as a severe desensing of the receiver when a strong local signal is received on another channel, thereby preventing you from hearing anything weaker than other strong signals on your channel. FEO is much more prevalent on solid state radios than with tube rigs, and this has to do with the characteristics of the devices. Solid state radios run on low voltages and transistors can be turned on with as little as .6V of bias. An especially strong signal can effectively "Swamp" the front end transistor causing it to lose sensitivity on the desired channel. Tubes, on the other hand, run at high voltages and are far less affected by a strong R.F. signal. Since the receiver front end is exposed to all signals, there isn't a heck of a lot you can do to cure a bad case of FEO, without affecting your desired signal as well. A simple attenuation of signals might be enough to offset the effects of Front End Overload. Even though a signal attenuator will reduce the sensitivity on your desired channel, there will be a point where the attenuation will reduce the overload the front end stage and allow the receiver to process the desired signals again. While sensitivity will be reduced from maximum, it will still hear better than with the FEO in full bloom. For those with a little engineering prowess, it is also possible to change the front end device to a more overload resistant part. Some F.E.T. type transistors are somewhat better than bipolar types, but will require some circuit changes to properly implement.

3. Intermodulation Distortion (IMD). Going somewhat hand-in-hand with Front End Overload is Intermodulation distortion. IMD is a condition where a non-linear device will produce unwanted frequency mixing components. In a super heterodyne radio receiver, we have deliberate non-linear stages used for mixing signals for frequency conversion. These mixer stages take one frequency and "mix" it with another frequency to produce a sum or difference frequency for the I.F. stage. All mixers produce both sum and difference components and a properly designed receiver will filter out the unwanted mixer components so that only the desired signal passes through. With IMD, this "mixing" process occurs in a stage that is not normally supposed to be used as a mixer, most notably the receiver front end. If a signal enters the receiver front end of sufficient amplitude to drive the transistor into a non-linear range (Saturation), IMD will occur. A classic symptom of IMD is when a fairly strong CB signal is present, and riding in on the signal may be the audio of a nearby TV station or AM broadcast station, or if someone throws a dead carrier on the channel you're tuned to, and you hear the audio from another CB station several channels away riding in on the carrier. That's IMD. If you live in an "RF rich" environment, your radio may be inundated with all sorts of strange signals which come and go. That's because these signals are the sum or difference products of two or more signals which are mixing in your receiver's front end and these mix products are falling on-channel and being passed on through your radio. If you suspect IMD problems, there are a few things you can do to help your situation. The best thing to do is to find a narrow bandpass filter to put in front of the receiver. The bandpass filter will attenuate signals outside of your passband, and hopefully take them down well below the threshold of non-linearity. Finding a store-bought bandpass filter for the CB band may not be easy though, and you may need to construct your own. In lieu of the bandpass filter, you can also try to reduce the level of the strong signals by backing down the R.F, gain or attenuating the signals similar to what you would do for Front End Overload. Again, as in the case of FEO, there will be a reduction of sensitivity, but the point where the IMD becomes manageable will allow better overall sensitivity. IMD mix products tend to decrease at a level 2 to 3 times the level of the desired signal. So for every 1 db you drop desired signals, IMD products will drop at least 2 db. For IMD issues with users on other channels, there isn't much you can do, other than find a different radio that is less susceptible to IMD products. Some radio designs are far better than others.

4. Noise Blanker (NB). The pulse gate noise blanker circuit found on most CB radios, is a clever circuit which takes noise pulses (Which are by nature broad banded), and processes them through a separate receiver strip, and then uses them to switch off (blank) the desired receiver for the brief millisecond that the pulse is present, thereby preventing the noise from making it through to the audio stages. The duration that the receiver is "blanked" is so short that the operator barely notices it. While this is a great circuit for reducing pulse-type noise (like car ignitions), it can also contribute to the overall bleedover picture. The Noise Blanker circuit normally uses a separate receiver tuned to a frequency outside of the CB band, but close enough to be effective on the same noise frequencies. But the NB receiver is just as susceptible to IMD and FEO as the main receiver. Also if there are strong local stations running distorted transmitters and class "C" amplifiers, their spurious emission products can make it into the NB receiver as well. When these conditions occur, the receiver will "blank" for a longer duration than when a simple noise pulse is present. The NB may also be modulated by the interfering signal resulting in a "hashy" sounding receive which sounds much like adjacent channel bleed over. The difference is that this type of interference will usually disappear when the NB is switched off. Obviously, when you operate in an area with many strong local signals, it's probably best to leave the NB off.

5. Image Rejection (IR). Every super heterodyne receiver uses a mixer to convert from the desired variable receiver frequency, down to an easier to tune and manage fixed I.F. frequency. The mixer stage which performs this duty takes the incoming frequency and mixes it with a local oscillator frequency to produce a sum and difference frequency component. Since only one of those mixer products is desired and used, the other one is normally filtered out. But the math works backwards as well. There are also two input frequencies (sum and difference) which, when mixed with the L.O., will produce the desired I.F. frequency. Normally a radio is designed to use only one of those input frequencies. The other one is the "image" frequency. Clever radio design will use an I.F. frequency high enough that the image frequency will fall way out of band and can be filtered out of the I.F. But many older single conversion CB radios use a 455 Khz I.F. If one of these older 23 channel rigs were tuned to channel 1 (26.965), it was quite possible to receive the image of another (albeit illegal) freebander operating on 27.875 (2 times the I.F. frequency). While the signal will be somewhat weaker, if the station is strong enough, it will interfere with the desired channel. Better designed radios utilize dual conversion I.F. frequencies which better isolate image intrusion. Image rejection issues were one of the reasons why the original CB band expansion plan was trimmed back to 40 channels from the original proposal of 99.

6. Splatter and Spurious emissions. Up until now, the causes of bleed over were due to design flaws or shortcomings in receiver design, or performance. But Splatter is a different animal. Splatter is caused by an improperly tuned and/or operated transmitter. The usual causes are overmodulation (caused by "clipping" the modulation limiter), poorly implemented transmitter modifications, and running with a class "C" amplifier. Splatter manifests itself as a broader cone of energy than what would be expected under normal conditions. A properly operating CB transmitter has an audio bandwidth of 6 Khz (Upper and lower sidebands) modulated by an audio frequency range of 300 to 3000 hz. These conditions produce a signal that will stay within the 6 Khz bandwidth (and the 2 Khz upper and lower guard bands) of each channel. But if you remove the modulation limiter and allow overmodulation, the signal will flat top and be pinched at the zero power point and will no be longer linear and sinusoidal. A waveform with sharp edges will generate harmonics of the audio frequencies. which will fall outside of the design bandwidth and well into the adjacent channel(s). Of course, the audio is almost never a single tone, and the amount of audio harmonics of a complex voice pattern will spread all across the adjacent channel. Splatter is further compounded by any non-linearity in the R.F. amplification chain. The absolute worst is to run an overmodulated transmitter into a class "C" amplifier. A class "C" amp is NOT linear and will generate Intermodulation Distortion (IMD) products. Even normally "linear" amplifiers will produce high IMD products if they are overdriven into saturation (which is a non-linear region). The most offensive IMD distortions are the odd-order products which take fundamental audio frequencies, and mix them with the second harmonic of another frequency and produce even more energy further and further away from the desired channel.
The effects of splatter are hard to combat on the receive end. You can't filter it out by tighter bandwidth filters because the splatter energy is on-channel! The only practical defense against someone who splatters is to have a directional antenna and point it away from that station to reduce the amount of garbage you are forced to receive. An alternative is to try educating the locals on the perils of dirty transmitters. The affected people are not always receptive to criticism however, and you might end up causing more problems. Try to be diplomatic.

So there you have it. The major contributors to bleed over featured here for you to hopefully understand. While much of the how's and why's are a little too technical for the average chit-chatter to fully understand, suffice to say that not all bleed over is created equal and in many cases, it really is the fault of your receiver if you are subject to an excessive amount of regular bleed interference. In this case, it really is better to spend a little more for a better designed radio. A directional beam antenna can also do much toward reducing interference from directions other than where you're pointed. Lastly, it's every operator's responsibility to be a good neighbor. I know most of us strive to put out as strong a signal as humanly possible, but questionable mods done to radios can increase the interference to other stations (as well as generate RFI), while only giving slight increases in range, even if your signal is perceived as being "louder".

Find this and more great articles here