SM 5 BSZ - Noise Blankers
(May 05 1997)

Impulse noise

Some interference is in the form of short pulses of very high peak energy, but with a long time gap between each pulse. Typical, and most well known is the car ignition noise.

The inherent length of a typical noise pulse at VHF is very short and the length as measurable at the antenna is typically defined by the frequency response of the antenna.

A noise blanker is a device that detects pulses and blocks the signal during the time of each pulse. A noise blanker becomes more effective (very much) at large bandwidths as long as there are no strong signals present in the pass band.

The bandwidth is important

I have been experimenting with wide band noise blankers since 1968. The first design was described in Radio och Television june 1969
RT.s och Schlumbergers konstruktionstävling

And in more detail later the same year:
Störningseliminator för 2 m-amatörer (del 1)
Störningseliminator för 2 m-amatörer (del 2)

These articles are all in Swedish, but the figures give a general idea (and some nostalgia). This design operated directly on 144MHz with a bandwidth of about 5MHz, and it was extremely efficient. The antenna I use today has a bandwidth of only 0.5MHz so my present system has somewhat lower performance.

Presently I use several noise blankers that operate in different bandwidths. If the bandwidth is increased by a factor of 10, the width of the noise pulses decreases by a factor of 10 also. The amount of energy in the pulse becomes 10 times larger at the same time, so the power level during the time of the pulse increases by a factor of 100.

In a 10 times larger bandwidth the noise level increases by a factor of 10, but still the pulses have gained a factor of 10 over the white noise.

It is always advantageous to use the largest possible bandwidth for a noise blanker, and the reason I am not satisfied with one only is that some times strong signals make the use of large bandwidths impossible.

When trying to remove a noise spike, the receiver is switched off during the time of the pulse. Of course the strong signals within the passband will be switched off simultaneously, which causes keying clicks. If the strong signal is too strong these keying clicks become stronger than the noise spike one tries to remove - and then the noise blanker does not work.

Selecting a narrower bandwidth may move the strong station out of the passband, but the increased selectivity makes the keying click problems even worse for stations within the narrow passband.

Noise blankers and noise limiters

to be continued

To SM 5 BSZ Main Page