Home » Radio » Radio Frequency Interference » Does a Noise Canceller Really Work?

Does a Noise Canceller Really Work?

Yes, a noise canceller can work well. But success with a noise canceller really, really depends on how well you position the noise antenna, sometimes called a noise probe. Here’s what you need to know to set up your noise canceller and its noise probe. This is stuff I learned by experimenting and goes beyond what you find in the manual.

Listening to high frequency radio has become a real challenge. The challenge is RFI or radio frequency interference. RFI is is a form of noise pollution caused by power lines, power supplies and consumer or industrial devices. These devices can (and usually do) emit radio signals. The emissions may be intentional or unintentional. Mostly, ham radio operators or shortwave listeners face local interference that is unintentional but nevertheless very real.

There are three types of noise fighting techniques used in receiving systems.

  • Noise Blanker. These have been built into receivers for a long time. They are useful for reducing the effect of pulse-type noise, such as ignition noise from cars. A noise blanker works by detecting pulses of noise and switching off the radio amplifiers for a few milliseconds whenever the pulse is present. This happens so quickly that you cannot hear it, but the effect is that the pulses seem to disappear. Noise blanking is not very useful for modern, more complex RFI.
  • Signal Processing. Modern radios contain some form of Noise Reduction, where a radio or audio signal is processed digitally. Noise Reduction is useful for removing noise that is either statistically correlated (like an interfering tone which is notched away) or uncorrelated (like white noise or background noise). Signal processing can provide some benefit to reducing the effect of RFI.
  • Noise Canceller. This is a device which goes between your antenna and the receiver. It attempts to reduce interference by phase reversal. Noise cancellers use a separate noise antenna to pick up the noise signal without the desired signal, and then phase reverse the noise signal by 180 degrees. This phase shift and subsequent mixing with the combined signal plus noise effectively subtracts the noise, or at least tries to.

This article focuses on the third approach, the Noise Canceller. It’s main benefit is that it can remove or reduce the noise before it enters your receiver.

How a noise canceller works

The theory is simple but the implementation is not. If you sum two signals of equal strength and exactly opposite phase, they will cancel each other at the output. In the real world, components are imperfect and signals are hard to isolate. But even with these imperfections, a noise canceller can accomplish a lot.

Diagram of noise canceller signal flow

The key to successful use of a noise canceller are:

  1. Picking up the noise signal without the desired signal using a separate noise antenna or probe.
  2. Adjusting the noise signal strength so it is similar to the strength of the noise in the combined signal.
  3. Reversing the phase of the noise signal so that it subtracts from the combined signal, leaving only or mainly the desired signal to flow into the receiver.

Two popular Noise Cancellers are the Timewave ANC-4 and MFJ-1026. Personally, I use the ANC-4 which was originally made by JPS. This article applies to both.

Success keys 2 and 3, adjusting noise signal strength and phasing, are done by controls on the front panel of the ANC-4 and MFJ-1026. Figuring out success key #1 – setting up a noise antenna to best receive the noise signal as the dominant signal – is something you have to do yourself. So, let’s focus on setting up a good noise antenna, and the rest will take care of itself.

A noise antenna does not have to be a “great antenna”. In fact, it should be a poor antenna. It only needs to do one thing well: pick up much of the noise signal and little of the desired signal. This is the reverse of what you normally want an antenna to do. My suggestion would be to just use a piece of wire connected to a coax cable that runs to your noise canceller noise antenna input. At my location, I use a length of RG-59 coax (typically used for cable television) and an impedance matching network to connect the coax to the single wire antenna.  (My impedance matching device is what SWL would call a Magnetic Longwire Balun. You can buy one or just build your own MLB.) Typically, my wire for the noise probe might be 5-15 feet in length. All of what follows assumes your main receiving antenna is outdoors.

If the source of your noise is in your shack, hooking up a “poor” antenna to receive noise is pretty easy. This might work well for cancelling noises from power supplies, computers, etc. near your radio. If the source of your noise is within your house, one option might be to place the noise antenna near your power panel in the basement. RFI within your house tends to flow along electrical lines in your walls, and these all end up at the power panel. Having your noise antenna near the power panel gives you a strong collection of local noise without much signal. (Please note: this antenna should not be connected to your power panel or wiring in any way, either purposely or accidently. Just near enough to pick up the noise being radiated from the wiring.)

When the noise is coming from outside your house (typically a neighbor’s house or maybe power pole, sometimes hundreds of feet away) you need to experiment with different configurations. At my house, I have noise antenna coax running to two outside entry points. Then, I can walk around the house trying different noise antenna lengths and orientations until I find the best one. This experimentation is time consuming but necessary.

The symptom that you have not yet found the right noise antenna is that you will think your noise canceller is not working or not doing enough cancellation. When you do find the right noise antenna, you will know, because you will see at least 20-30 dB cancellation (3 – 6 S units). If you are persistent and I suppose lucky, you should get much more than this. If you doubt this, just see the videos listed below.

Two other points:

  • Noise Gain. Sometimes your noise antenna may pick up too much noise, so that your Noise Gain knob provides little control. You might want to attenuate the noise antenna. A simple way to do this if you are using RG-59 cable television coax is just to put a splitter in line, which will give you 6 dB or more attenuation.
  • Multiple Sources. The approach described above works best when there is one dominant noise source. If you have more than one, you might just have to focus on the worst one. If you get lucky, you might find a compromise noise antenna location that will work on both sources.

Here are a few examples of these noise cancellers in action: ANC 1, ANC2, MFJ 1, MFJ 2. You can see my personal experience with the ANC at 20:39 (near the end) of my video Adventures with RFI. If you are an advanced builder and want to build your own noise canceller, this page from G8JNJ might get you started.

If you are wondering about things like the type of coax to use, impedance matching your noise probe, or the loss you might experience with 75 ohm coax for a 50 ohm radio, don’t worry. We are talking about shortwave frequencies and receivers, not VHF/UHF or transmitting, where impedance matching is more critical. Also, both the ANC and MFJ units are designed to be used with transceivers, and will switch themselves out when you transmit.

Have fun and success reducing RFI with your noise canceller.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.