Brats club logo 


Syllabus Sections:-

Filtering and remedial measures

6D1 43   Understand the use of high, low, band pass and band stop (notch) filters of L, T and π configuration, including coaxial stubs as notch filters or traps, in improving the immunity of affected devices. 

Filters must be used in equipment at the earliest part of the circuits and this is carried out in the design of the equipment.

A low pass filter (LPF)

A low pass filter (LPF) is designed so that it passes frequencies that are lower than that of the stop band so in effect signals above the cut-off frequency (fc) are reduced.

low pass filter

A high pass filter (HPF)

A high pass filter (HPF) is designed so that it passes frequencies that are higher than that of the stop band so in effect signals below the cut-off frequency (fc) are reduced.

High pass filter

NOTE : the LPF and HPF are sometimes called pi filters as they resemble the Greek letter pi  Π


A band pass filter (BPF)

A band pass filter (BPF) is a combination of low pass and high pass filter that will passes a range of frequencies in the pass band, any frequencies above or below this range are reduced. Unlike the LPF and the HPF, the BPF has two stop bands and two cut-off frequencies (fc) at the meeting points of each of the stop and pass band.

Band pass filter

with the use of filters remember that if one type is not sufficient then you can also use another in what is called a "cascade" of filters.


For the band stop filter see page 91 Fig 13.16 and there is text to read.

Also check out page 90 Fig 13.13 and Fig 13.14 for different type of High pass and Low pass filters in the "L" (inverted) configuration but using resistors in place of capacitors.

-------------------------------------------------------------------------------------------------------------------------


Notch filters including coaxial stubs as notch filters or traps

A stub is connected to a feed line. (parallel to it) It can be a length of co-ax, (connected inner to inner, outer to outer when connecting to the feed line) or it can be a length of open wire feeder. ("left wire" of the stub to "left wire" of the feeder, right to right...)

Stubs are usually made to be an ELECTRICAL 1/4 wavelength at the frequency of interest. 1/2 and 3/4 wave stubs can be used, but these are a) longer, and less convenient, b) more lossy, c) less "Q" and hence have a wider bandwidth (Although this may be desirable, and hence is the reason such longer stubs are used).

What happens at the far end of the stub is interesting. For odd 1/4 wavelength stubs (1/4, 3/4, etc) an OPEN circuit at the end results in (AT THE FREQUENCY OF INTEREST) a short circuit at the end connected on the feed line. And vice versa...

For EVEN wavelengths (1/2, 1, 1/1/2 etc) what happens at the end of the stub is what happens at the end connected to the feed line. Again, ONLY AT THE FREQUENCY OF INTEREST.

You can now see that it is possible to arrange for a transmitter to have a short circuited 1/4 wave stub to be connected on its feed line. As it is shorted at the far end, then AT THE FREQUENCY OF INTEREST, the stub produces a very high impedance across the feed line, and has little/no effect. But at different frequencies, the stub can look like a short circuit, and cause a bad VSWR, causing those unwanted frequencies to be reflected back to the transmitter, and dissipated in the output stage. (And not sent to the antenna!).

The application for a receiver is that you have an OPEN end on the stub. This causes signals coming from the antenna (unwanted, strong, interfering signals, on a single frequency) to see a bad VSWR, and return to the antenna, and not be fed to the receiver. Typical application here is an open circuit stub cut for two metres, to stop, say 2M interference to an FM broadcast RX.

You have to be careful, because as stated above, stubs also work when 3/4 wavelength long.

This means that you may find you use a 1/4 wave stub on a 2M transmitter, hoping to stop the third harmonic on 70cms from being radiated. But that stub is a 3/4 wave stub on 70 cms :-(

The trick is to cut the stub short, and retune with a trimmer capacitor across the end. The stub in the transmitter feed line then does its job on 2M, allows that through, but is no longer a 3/4 wavelength on 70cms.

So it does not allow that through.

Practical application of shortening and tuning is to have a 2M 1/4 wave stub on a TV coax to keep 2M out of the TV. This will also stop signals on 720MHz (5 times 144) as it will be a 5/4 length stub (1 1/4 wave). If this happens to be the TV channel of interest, you shorten the stub, and retune with the trimmer to 2M, and then the "overtone" is no longer on 720MHz. It will be somewhere, but as long as it is not where the neighbour wants to watch a particular channel, all well and good. Of course, if he is trying to watch a channel on that frequency, you had better have no 5th harmonic from the transmitter anyhow!

A stub is intended to work on one set of frequencies. Hence if it is working at the frequency of interest, it will "kill" or allow. If you are not in this ballpark, the stub will cause the feed line to see either an inductance, or capacitance, and you will get a bad VSWR. So no, you cannot use stubs on a feed line unless they are cut for the frequency involved. This means that you need to have a "T" piece in the coax, and connect the appropriate stub. In the case of a receiver, one stub per band, which will exclude the nearby transmitter on another band, BUT remember the nearby transmitter will be producing unwanted harmonics, which will be where you are trying to receive, so the stub will allow these through :-( Also, if you have a stub on the receiver, at 7MHz, and the nearby TX is 21MHz, the stub won't help, as it will pass odd multiples of its design frequency.

You can use more than one stub on a feeder when they are all cut for the same band. You make each an electrical quarter wavelength long and you space them along the feeder an electrical quarter wavelength. This arrangement works on one frequency, as a single stub does, but the attenuation of unwanted products is much higher :-)

You MAY be able to use a 144MHz stub on an HF feed, but it will cause some VSWR. Remember, it will be short in length compared to the HF frequency, so effectively, you are connecting a capacitor across the HF feed line. A much better arrangement is to have a low pass filter for the HF RX. If a transceiver, you should have a low pass filter anyhow!

Another useful application for an open wire feeder/stub is to make a short circuited version to only allow the frequency of interest to pass. The far end from the feeder is shorted. And can be connected to earth. Useful lightning protection. Will work on 7MHz and 21MHz but that you can't do this with coax.

Stubs have other uses, mainly on open wire feeder. In this case, you can actually design the stub to tune the feed line, as it will add capacity/inductance. The trick there is to know how long it should be, and on what part of the feed line it should be connected!


6D1 43 continued Recall the use of ferrite beads or rings in internal and external filtering.

Sometime the filtering has to be carried out inside the equipment. Here the "tiny" ferrite bead (FB) is actually used on the "BASE" leg of a transistor or adjacent PCB track is broken and a bead and wire soldered in as a replacement.

See also page 94 Fig 13.22 which explains in more detail.

FB on Base lef of transistor

Note the FB is placed on the BASE leg of the Transistor

The ferrite bead is acting as an RF choke just like it did when using the ferrite ring on interconnecting cables externally, it is used internally in equipment over the BASE legs of the transistor.

-----------------------------------------------------------------------------------

6D2  43  Understand the construction and use of a typical mains filter.

For this see  page 92 Fig 13.19 which shows the circuit diagram of a mains filter.

The simplest mains filter to fit is the "ferrite ring" where several turns of the mains cable is wound onto a ring or several rings stacked up - with the assembly bring placed as close to the chassis of the equipment as possible.

The ends of the cable should be kept apart and the turns on the ferrite only occupy about 2/3 of the diameter.

See also page 92 Fig 13.20

This filter is know as a ferrite ring choke. 

Whilst here it is coaxial cable wound on the ring the principle is the same.

The diagram above is the equivalent circuit of the ferrite ring filter - note the earth core of the cable is not shown as it is only the live and neutral cores that would carry any interference into or away from the equipment.


-----------------------------------------------------------------------------------------------------------------------------

6D2  43 continued Identify a typical circuit of a braid-breaking filter and a combined high-pass/braid-breaking filter. Understand their use.

The section deals with filters used on antenna feeds and NOT mains filtering.

Ferrite Ring

The braid breaking filter can also be the simple ferrite ring as shown above.

The circuit is similar to that above.

The ferrite ring may be the usual one you can buy at rallies but sometimes the ferrite is not "correct" for the frequency of the interference and a different "mix" of ferrite is required.

More information on this is in page 93 Fig 13.21 

High pass/ braid breaking filter

The combined high pass/ braid breaking filter is more complex but is basically a high pass filter with a resistor across one of the capacitors to allow a continuous link for the braid through the resistor.

Each of the filters works to stop the unwanted signals and pass the wanted signals. They are placed as close to the chassis of the equipment as possible.

--------------------------------------------------------------------------------------------------------

6D2  43 continued Understand why a ferrite ring will attenuate common mode currents without affecting the differential mode wanted signal.

Many problems with stereos can be traced to common-mode propagation currents on long speaker leads and other interconnecting cables.

A common-mode current is an unbalanced current flowing in one cable which is not balanced by a similar current flowing in the return cable when making the circuit complete and is caused by unwanted external signals being applied.

The differential mode or signal current is the wanted current.

When a cable is wound on a ferrite ring it creates a series inductance in the lead which increases the inductance of the leads and is seen as a high impedance to RF but allows the differential mode or wanted signal to pass unhindered.






brats copyright logo