The superhet or superheterodyne radio receiver

when strong signals are present, but enables the signals to be amplified sufficiently toensure a good signal to noise ratio is achieved.The tuned and amplified signal then enters one port of the mixer. The local oscillatorsignal enters the other port. The local oscillator may consist of a variable frequencyoscillator that can be tuned by altering the setting on a variable capacitor. Alternatively itmay be a frequency synthesizer that will enable greater levels of stability and settingaccuracy.Once the signals leave the mixer they enter the IF stages. These stages contain most ofthe amplification in the receiver as well as the filtering that enables signals on onefrequency to be separated from those on the next. Filters may consist simply of LC tunedtransformers providing inter-stage coupling, or they may be much higher performanceceramic or even crystal filters, dependent upon what is required.Once the signals have passed through the IF stages of the superheterodyne receiver, theyneed to be demodulated. Different demodulators are required for different types oftransmission, and as a result some receivers may have a variety of demodulators that canbe switched in to accommodate the different types of transmission that are to beencountered. The output from the demodulator is the recovered audio. This is passed intothe audio stages where they are amplified and presented to the headphones orloudspeaker.Block diagram of a basic superheterodyne receiverThe diagram above shows a very basic version of the superhet or superheterodynereceiver. Many sets these days are far more complicated. Some superhet radios havemore than one frequency conversion, and other areas of additional circuitry to provide therequired levels of performance. However the basic superheterodyne concept remains thesame, using the idea of mixing the incoming signal with a locally generated oscillation toconvert the signals to a new frequency.Selectivity is one of the major specifications of any receiver. Whilst the sensitivity isimportant to ensure that it can pick up the signals and receive them at a sufficientstrength, the selectivity is also very important. It is this parameter that determineswhether the receiver is able to pick out the wanted signal from all the other ones aroundit. The filters used in receivers these days have very high levels of performance andenable receivers to select out individual signals even on today's crowded bands.Superhet principleMost of the receivers that are used today are superhet radios. In these sets the incomingsignal is converted down to a fixed intermediate frequency. It is within the IF stages thatthe main filters are to be found. It is the filter in the IF stages that defines the selectivityperformance of the whole set, and as a result the receiver selectivity specification isvirtually that of the filter itself.

Block diagram of a basic superhet receiverIn some receivers simple LC filters may be used, although ceramic filters are better andare used more widely nowadays. For the highest performance crystal or mechanicalfilters may be used, although they are naturally more costly and this means they are onlyfound in high performance sets.Filter parametersThere are two main areas of interest for a filter, the pass band where it accepts signals andallows them through, and the stop band where it rejects them. In an ideal world a filterwould have a response something like that shown in Figure 2. Here it can be seen thatthere is an immediate transition between the pass band and the stop band. Also in the passband the filter does not introduce any loss and in the stop band no signal is allowedthrough.The response of an ideal filterIn reality it is not possible to realise a filter with these characteristics and a typicalresponse more like that shown in Figure 3. It is fairly obvious from the diagram that thereare a number of differences. The first is that there is some loss in the pass band. Secondlythe response does not fall away infinitely fast. Thirdly the stop band attenuation is notinfinite, even though it is very large. Finally it will be noticed that there is some in bandripple.Typical response of a real filter

In most filters the attenuation in the pass band is normally relatively small. For a typicalcrystal filter figures of 2 - 3 dB are fairly typical. However it is found that very narrowband filters like those used for Morse reception may be higher than this. Fortunately it isquite easy to counteract this loss simply by adding a little extra amplification in theintermediate frequency stages and this factor is not quoted as part of the receiverspecification.It can be seen that the filter response does not fall away infinitely fast, and it is necessaryto define the points between which the pass band lies. For receivers the pass band is takento be the bandwidth between the points where the response has fallen by 6 dB, i.e. whereit is 6 dB down or -6 dB.A stop band is also defined. For most receiver filters this is taken to start at the pointwhere the response has fallen by 60 dB, although the specification for the filter should bechecked this as some filters may not be as good. Sometimes a filter may have the stopband defined for a 50 dB attenuation rather than 60 dB.Shape factorIt can be seen that it is very important for the filter to achieve its final level of rejection asquickly as possible once outside the pass band. In other words the response should fall asquickly as possible. To put a measure on this, a figure known as the shape factor is used.This is simply a ratio of the bandwidths of the pass band and the stop band. Thus a filterwith a pass band of 3 kHz at -6dB and a figure of 6 kHz at -60 dB for the stop bandwould have a shape factor of 2:1. For this figure to have real meaning the two attenuationfigures should also be quoted. As a result the full shape factor specification should be 2:1at 6/60 dB.Filter typesThere is a variety of different types of filter that can be used in a receiver. The olderbroadcast sets used LC filters. The IF transformers in the receiver were tuned and it waspossible to adjust the resonant frequency of each transformer using an adjustable ferritecore.Today ceramic filters are more widely used. Their operation is based on the piezoelectriceffect. The incoming electrical signal is converted into mechanical vibrations by thepiezoelectric effect. These vibrations are then affected by the mechanical resonances ofthe ceramic crystal. As the mechanical vibrations are then linked back to the electricsignal, the overall effect is that the mechanical resonances of the ceramic crystal affectthe electrical signal. The mechanical resonances of the ceramic exhibit a high level of Qand this is reflected in its performance as an electrical filter. In this way a high Q filtercan be manufactured very easily.Ceramic filters can be very cheap, some costing only a few cents. However higherperformance ones are also available, and these are likely to be found in scanners andmany other receivers.For really high levels of filter performance crystal filters are used. Crystals are madefrom quartz, a naturally occurring form of silicon, although today's components are madefrom synthetically grown quartz. These crystals also use the piezoelectric effect andoperate in the same way as ceramic filters but they exhibit much higher levels of Q andoffer far superior degrees of selectivity. Being a resonant element they are used in manyareas where an LC resonant element might be found. They are used in oscillators - manycomputers have crystal oscillators in them, but they are also widely used in highperformance filters.Normally crystal filters are made from a number of individual crystals. The mostcommonly used configuration is called the half lattice filter as shown in Figure 4. Furthersections can be added to the filter to improve the performance. Often a filter will bequoted as having a certain number of poles. There is one pole per crystal, so a six polecrystal filter would contain six crystals and so forth. Many filters used in amateurcommunications receivers will contain either six or eight poles.

A basic half lattice crystal filter sectionChoosing the right bandwidthIt is important to choose the correct bandwidth for a give type of signal. It is obviouslynecessary to ensure that it is not too wide, otherwise unwanted off-channel signals will beable to pass though the filter. Conversely if the filter is too narrow then some of thewanted signal will be rejected and distortion will occur. As different types oftransmission occupy different amounts of spectrum bandwidth it is necessary to tailor thefilter bandwidth to the type of transmission being received. As a result many receiversswitch in different filters for different types of transmission. This may be done eitherautomatically as part of a mode switch, or using a separate filter switch. Typically a filterfor AM reception on the short wave bands will have a bandwidth of around 6 kHz, andone for SSB will be approximately 2.5 kHz. For Morse reception 500 and 250 Hz filtersare often used.SummarySelectivity is particularly important on today's crowded bands, and it is necessary toensure that any receiver is able to select the wanted signal as well as it can. Obviouslywhen signals occupy the same frequency there is little that can be done, but by having agood filter it is possible to ensure that you have the best chance or receiving and beingable to copy the signal you want.he superhet radio receiver is one of the most widely used types of receiver available. Oneof the important specifications associated with its operation is image response or imagerejection. Along with this the IF breakthrough is also of importance, although less criticalin many applications.Image responseThe basic concept of the superhet radio means that it is possible for two signals to eneterthe intermediate frequency (IF) implifier. For example with the local oscillator set to 5MHz and with an IF of 1 MHz it can be seen that a signal at 6 MHz mixes with the localoscillator to produce a signal at 1 MHz that will pass through the IF filter. However is asignal at 4 MHz is also able to produce an output at 1 MHz. It is clearly unacceptable toreceive signals on two frequencies at the same time and it is possible to remove theunwanted one by the addition of a tuned circuit prior to the mixerFortunately this tuned circuit does not need to be excessively sharp. It does not need toreject signals on adjacent channels, but instead it needs to reject signals on the imagefrequency. These will be separated from the wanted channel by a frequency equal totwice the IF. In other words with an IG at 1 MHz, the image will be 2 MHz away fromthe wanted frequency.

Using a tuned circuit to remove the image signalImageIt is clearly important to specify the level of rejection of the image signal. Thespecification compares the levels of signals of equal strength on the wanted and imagefrequencies, quoting the level of rejection of the unwanted signal.The image rejection of a receiver will be specified as the ratio between the wanted andimage signals expressed in decibels (dB)at a certain operating frequency. For example itmay be 60 dB at 30 MHz. This means that if signals of the same strength were present onthe wanted frequency and the image frequency, then the image signal would be 60 dBlower than the wanted one, i.e. it would be 1/1000 lower in terms of voltage or 1/1000000lower in terms of power.The frequency at which the measurement is made also has to be included. This is becausethe level of rejection will vary according to the frequency in use. Typically it falls withincreasing frequency because the percentage frequency difference between the wantedand image signals is smaller.IF BreakthroughAnother problem which can occur with a superhet occurs when signals from the aerialbreak through the RF sections of the set and directly enter the IF stages. Normallyintermediate frequencies are chosen so that there are likely to be no very large signalspresent which might cause problems. However when the receiver has a fixed frequencyfirst local oscillator this is not easy to ensure as it will sweep over a band of frequencies.The specification for breakthrough is quoted in the same fashion as image rejection.Normally it is possible to achieve figures of 60 to 80 dB rejection, and on some receiversfigures of 100 dB have been quoted.Navigation:: Home >> Radio receiver technology >> this pageRadio receiver sensitivity- including the concept of noise and sensitivity, signal to noise ratio,SINAD, and noise figure.Receiver sensitivity is one of the key specifications of any radio. The two mainrequirements of any radio receiver are that it should be able to separate one station fromanother, i.e. selectivity, and signals should be amplified so that they can be brought to asufficient level to be heard. As a result receiver designers battle with many elements tomake sure that these requirements are fulfilledA number of methods of measuring and specifying the sensitivity performance of radioreceivers are used. Figures including signal to noise ratio, SINAD, noise factor and noisefigure are used. These all use the fact that the limiting factor of the sensitivity of a radioreceiver is not the level of amplification available, but the levels of noise that are present,whether they are generated within the radio receiver or outside it.

NoiseToday technology is such that there is little problem in being able to achieve very largelevels of amplification within a radio receiver. This is not the limiting factor. In anyreceiving station the limiting factor is noise - weak signals are not limited by the actualsignal level, but by the noise masks them out. This noise can come from a variety ofsources. It can be picked up by the antenna or it can be generated within the radioreceiver.It is found that the level of noise that is picked up externally by a receiver from theantenna falls as the frequency increases. At HF and frequencies below this thecombination of galactic, atmospheric and man-made noise is relatively high and thismeans that there is little point in making a receiver particularly sensitive. Normally radioreceivers are designed such that the internally generated noise is much lower than anyreceived noise, even for the quietest locations.At frequencies above 30 MHz the levels of noise start to reach a point where the noisegenerated within the radio receiver becomes far more important. By improving the noiseperformance of the radio receiver, it becomes possible to detect much weaker signals.Design for noise performanceIn terms of the receiver noise performance it is always the first stages or front end that ismost crucial. At the front end the signal levels are at their lowest and even very smallamounts of noise can be comparable with the incoming signal. At later stages in the setthe signal will have been amplified and will be much larger. The same levels of noise asare present at the front end will be a much smaller proportion of the signal and will nothave the same effect. Accordingly it is important that the noise performance of the frontend is optimised for its noise performance.It is for this reason that the noise performance of the first radio frequency amplifierwithin the receiver is of great importance. It is the performance of this circuit that iscrucial in determining the performance of the whole radio receiver. To achieve theoptimum performance for the first stage of the radio receiver there are a number of stepsthat can be taken. These include:• Determine the circuit topology required• Choose a low noise device• Determine the gain required• Determine the current through the device• Use low noise resistors• Optimise the matching• Ensure that power supply noise entering the circuit is removedDetermination of circuit topology The first step in any design is to decide upon the typeof circuit to be used. Whether a conventional common emitter style circuit is to be used,or even whether a common base should be employed. The decision will depend uponfactors including the matching input and output impedances, the level of gain requiredand the matching arrangements to be used.Choice of active device The type of device to be used is also important. There aregenerally two decisions, whether to use a bipolar based transistor, or whether to use afield effect device. Having made this, it is obviously necessary to decide upon a lownoise device. The noise performance of transistors and FETs is normally specified, andspecial high performance low noise devices are available for these applications.Determination of required gain While it may appear that the maximum level of gainmay be required from this stage to minimise the levels of amplification required later andin this way ensure that the noise performance is optimised, this is not always the case.There are two major reasons for this. The first is that the noise performance of the circuitmay be impaired by requiring too high a level of gain. Secondly it may lead to overloadin later stages of the radio receiver and this may degrade the overall performance. Thusthe level of gain required must be determined from the fact that it is necessary to optimise

the noise performance of this stage, and secondly to ensure that later stages of thereceiver are not overloaded.Determination of current through the active device The design of the first stage of theradio receiver must be undertaken with care. To obtain the required RF performance interms of bandwidth and gain, it may be necessary to run the device with a relatively highlevel of current. This will not always be conducive to obtaining the optimum noiseperformance. Accordingly the design must be carefully optimised to ensure the bestperformance for the whole radio receiver.Use of low noise resistors It may appear to be an obvious statement, but apart fromchoosing a low noise active device, consideration should also be given to the othercomponents in the circuit. The other chief contributors are the resistors. The metal oxidefilm resistors used these days, including most surface mount resistors normally offergood performance in this respect and can be used as required.Optimise impedance matching In order to obtain the best noise performance for thewhole radio receiver it is necessary to optimise the impedance matching. It may bethought that it is necessary to obtain a perfect impedance match. Unfortunately the bestnoise performance does not usually coincide with the optimum impedance matchAccordingly during the design of the RF amplifier it is necessary to undertake somedesign optimisation to ensure the best overall performance is achieved for the radioreceiver.Ensure that power supply noise entering the circuit is removed Power supplies cangenerate noise. In view of this it is necessary to ensure that any noise generated by theradio receiver power supply does not enter the RF stage. This can be achieved byensuring that there is adequate filtering on the supply line to the RF amplifier.SummaryReceiver sensitivity is one of the vital specifications of any radio receiver. The key factorin determining the sensitivity performance of the whole receiver is the RF amplifier. Byoptimising its performance, the figures for the whole of the receiver can be improved. Inthis way the specifications for signal to noise ratio, SINAD or noise figure can be broughtto the required level.There are a number of ways in which the noise performance, and hence the sensitivity ofa radio receiver can be measured. The most obvious method is to compare the signal andnoise levels for a known signal level, i.e. the signal to noise (S/N) ratio or SNR.Obviously the greater the difference between the signal and the unwanted noise, i.e. thegreater the S/N ratio, the better the radio receiver sensitivity performance.As with any sensitivity measurement, the performance of the overall radio receiver isdetermined by the performance of the front end RF amplifier stage. Any noise introducedby the first RF amplifier will be added to the signal and amplified by subsequentamplifiers in the receiver. As the noise introduced by the first RF amplifier will beamplified the most, this RF amplifier becomes the most critical in terms of radio receiversensitivity performance. Thus the first amplifier of any radio receiver should be a lownoise amplifier.Methods of measuring receiver sensitivityAlthough there are many ways of measuring the sensitivity performance of a radioreceiver, the S/N ratio or SNR is one of the most straightforward and it is used in avariety of applications. However it has a number of limitations, and although it is widelyused, other methods including noise figure are often used as well. Nevertheless the S/Nratio or SNR is an important specification, and it will be seen in many radio receiverspecification sheets.

Signal to noise ratio for a radio receiverThe difference is normally shown as a ratio between the signal and the noise (S/N) and itis normally expressed in decibels. As the signal input level obviously has an effect on thisratio, the input signal level must be given. This is usually expressed in microvolts.Typically a certain input level required to give a 10 dB signal to noise ratio is specified.Effect of bandwidthA number of other factors apart from the basic performance of the set can affect the SNRspecification. The first is the actual bandwidth of the receiver. As the noise spreads outover all frequencies it is found that the wider the bandwidth of the receiver, the greaterthe level of the noise. Accordingly the receiver bandwidth needs to be stated.Additionally it is found that when using AM the level of modulation has an effect. Thegreater the level of modulation, the higher the audio output from the receiver. Whenmeasuring the noise performance the audio output from the receiver is measured andaccordingly the modulation level of the AM has an effect. Usually a modulation level of30% is chosen for this measurement.Typical figuresThis method of measuring the performance is most commonly used for HFcommunications receivers. Typically one might expect to see a figure in the region of 0.5microvolts for a 10 dB S/N in a 3 kHz bandwidth for SSB or Morse. For AM a figure of1.5 microvolts for a 10 dB S/N in a 6 kHz bandwidth at 30% modulation for AM mightbe seen.Points to note when measuring SNRSNR is a very convenient method of quantifying the sensitivity of a receiver, but there aresome points to note when measuring and interpreting the figures. To investigate these it isnecessary to look at the way the measurements of SNR are made. A calibrated RF signalgenerator is used as a signal source for the receiver. It must have an accurate method ofsetting the output level down to very low signal levels. Then at the output of the receivera true RMS AC voltmeter is used to measure the output level.S/N and (S+N)/N With the generator signal switched off a 50 Ohm match isgiven to the receiver and the audio meter will detect the noise generated by thereceiver itself. This level is noted and the signal turned on. Its level is adjusteduntil the audio level meter reads a level which is 10 dB higher than just the noiseon its own. The level of the generator is that required to give the 10 dB signal tonoise ratio.The last statement was not strictly true. Whilst the first reading of the noise isquite accurate, the second reading of the signal also includes some noise as well.In view of this many manufacturers will specify a slightly different ratio: namelysignal plus noise to noise (S+N/N). In practice the difference is not particularlylarge, but the S+N/N ratio is more correct.PD and EMF Occasionally the signal generator level in the specification willmention that it is either PD or EMF. This is actually very important because thereis a factor of 2:1 between the two levels. For example 1 microvolt EMF. and 0.5microvolt PD are the same. The EMF (electro-motive force) is the open circuit

voltage, whereas the PD (potential difference) is measured when the generator isloaded. As a result of the way in which the generator level circuitry works itassumes that a correct (50 Ohm) load has been applied. If the load is not thisvalue then there will be an error. Despite this most equipment will assume valuesin PD unless otherwise stated.Navigation:: Home >> Radio receiver technology >> this pageRadio receiver SINAD measurement- an overview of the SINAD measurement used in specifying the sensitivityperformance of many radio receivers.One of the measurements that can be made to assess and specify the sensitivityperformance of a radio receiver is SINAD. While not used as widely as the signal tonoise ratio, or noise figure it is nevertheless used commonly and can be found in thespecifications of many radio receivers. SINAD is often used in conjunction with FMreceivers, but it can also be used for AM and SSB quite easily.As with any radio receiver, the design of the RF amplifier is key to its sensitivityperformance. A poorly performing RF amplifier will degrade the performance of thewhole receiver. However a high performance low noise RF amplifier will enable theoverall set to provide a high level of sensitivity. Accordingly time should be focussed inthe design of the RF amplifier in order that it should reach the required level ofperformance.What is SINAD?SINAD is a measurement that can be used for any communication device to look at thedegradation of the signal by unwanted or extraneous signals including noise anddistortion. However the SINAD measurement is most widely used for measuring andspecifying the sensitivity of a radio receiver.The actual definition of SINAD is quite straightforward. It can be summarised as the ratioof the total signal power level (Signal + Noise + Distortion) to unwanted signal power(Noise + Distortion). Accordingly, the higher the figure for SINAD, the better the qualityof the audio signal.The SINAD figure is expressed in decibels (dB) and can be determined from the simpleformula:SINAD = 10Log ( SND / ND )where:SND = combined Signal + Noise + Distortion power levelND = combined Noise + Distortion power levelIt is worth noting that SINAD is a power ratio and not a voltage ratio for this calculation.Making SINAD measurementsTo make the measurement a signal modulated with an audio tone is entered into the radioreceiver. A frequency of 1 kHz is taken as the standard as it falls in the middle of theaudio bandwidth. A measurement of the whole signal, i.e. the signal plus noise plusdistortion is made. As the frequency of the tone is known, the regenerated audio is passedthrough a notch filter to remove the tone. The remaining noise and distortion is thenmeasured.Although it is most common to measure the electrical output at the receiver audio outputterminals, another approach that is not as widely used, is to pass the signal into theloudspeaker and then use a transducer connected to SINAD meter to convert the audioback into an electrical signal. This will ensure that any distortion included by the speaker

is incorporated, and it may overcome problems with gaining access to the speakerconnections in certain circumstances where this may not be possible.Obtaining the figures for the signal plus noise plus distortion and the noise plus distortionit is then possible to calculate the value of SINAD for the radio receiver of other piece ofequipment.The set up used for making SINAD measurementsWhile the measurements for SINAD can be made using individual items of testequipment, a number of SINAD meters are made commercially. These SINAD metersincorporate all the required circuitry and can be connected directly to radio receivers tomake the measurements. Accordingly SINAD meters are a particularly convenientmethod of making these measurements.Filter for SINAD measurementsThe notch filter that is required for SINAD measurements to be made has an effect on themeasurement. In an ideal world the filter would be infinitely sharp a notch out only themodulating tone. However in the real world the filter will have a finite bandwidth. As itsbandwidth increases, so it will remove noise and distortion as well as the tone. Howeveras the distortion products will typically result from the second and third harmonics of thetone, the filter will not have an effect on this element of the reading. Nevertheless it maystill have an effect on the noise levels.In view of this problem some standards set down specifications or guidelines for the filterused in the SINAD measurement. ETSI (European Telecommunications StandardsInstitute) defines a notch filter (ETR 027). With the standard tone frequency of 1 kHz, itstates that a filter used for SINAD measurements shall be such that the output the 1000Hz tone shall be attenuated by at least 40 dB and at 2000 Hz the attenuation shall notexceed 0.6 dB. The filter characteristic shall be flat within 0.6 dB over the ranges 20 Hzto 500 Hz and 2000 Hz to 4000 Hz. In the absence of modulation the filter shall not causemore than 1 dB attenuation of the total noise power of the audio frequency output of thereceiver under test.In addition to the filter performance another critical area of a SINAD measurement is themeasurement of the output signal power levels. These have to be a true powermeasurements that accommodate the different form factors of the variety of waveforms,i.e. sine wave for the 1 kHz tone and its harmonics, but the noise will be random andhave a different form factor.Applications of SINAD measurementsSINAD measurements give an assessment of the signal quality from a receiver under anumber of conditions. As such SINAD measurements can be used for assessing a numberof elements of receiver performance.Receiver sensitivity: The most common use of the SINAD measurement is to assess thesensitivity performance of a radio receiver. To achieve this the sensitivity can be assessedby determining the RF input level at the antenna that is required to achieve a given figureof SINAD. Normally a SINAD value of 12 dB is taken as this corresponds to a distortionfactor of 25%, and a modulating tone of 1 kHz is used. It is also necessary to determineother conditions. For AM it is necessary to specify the depth of modulation and for FM

the level of deviation is required. For FM analogue systems ETSI specifies the use of adeviation level of 12.5% of the channel spacingA typical specification might be that a receiver has a sensitivity of 0.25 uV [microvolts]for a 12 dB SINAD. Obviously the lower the input voltage needed to achieve the givenlevel of SINAD, the better the receiver performance.Adjacent channel rejection: This parameter is a measure of the ability of the receiver toreject signals on a nearby channel. As the adjacent channel performance degrades, so thelevels of noise and extraneous signals will increase, thereby degrading the SINADperformance.An initial measurement of SINAD is made at a given level and this is known as thereference sensitivity. The RF input level of the signal for the SINAD measurement is thenincreased by 3 dB at the receiver antenna input. A second source or signal withmodulated with a 400 Hz tone is added with its frequency set to an adjacent channel or ata specific offset from the carrier source used for the basic SINAD measurement. It will befound that the interferer will cause the 400 Hz tone to appear in the audio of the receiveras its level is increased. This will be seen as a degradation in the SINAD as the 400 Hztone will pass through the SINAD meter notch filter.With the measurement system set up, the interferer signal level is raised until the SINADvalue is degraded to the original value obtained at the reference sensitivity. Then the ratioof the interfering level to the wanted signal is the adjacent channel rejection.Receiver blocking: SINAD can be used to form the basis of a receiver blockingmeasurement. As with other similar measurements a reference SINAD sensitivity level isfound. The level of the SINAD signal is increased by 3 dB at the antenna. An unmodulatedoff channel signal is then added and its level raised until the receiverdesensitises to an extent whereby the reference SINAD level is reached.SummarySINAD is a particularly useful measurement format that can be used to determine theperformance of a radio receiver under a variety of conditions. Although SINAD isprimarily used to specify the basic sensitivity performance of many radios, it can be usedfor other parameters as well. Additionally it is chiefly used for FM systems, but its use isequally applicable to AM and SSB. It may also be used for digital systems as well,although this is not common practice as a measurement known as bit error rate (BER) ismore widely used.The overall figure for SINAD will be chiefly dependent upon the performance of the RFamplifier in the receiver. A low noise RF amplifier will enable the set as a whole toprovide a good SINAD performance.Navigation:: Home >> Radio receiver technology >> this pageRadio receiver noise figure- an overview of noise figure used in specifying the sensitivity performanceof radio receivers and their components.Although there are a number of methods of determining the sensitivity of radio receiversand their associated elements, the noise figure is one of the most widely used methods.Not only is it widely used to assess the sensitivity performance or receivers, but it can beapplied to complete receiving systems or to elements such as RF amplifiers. Thus it ispossible to use the same notation to measure the noise performance of a whole receiver,or an RF amplifier. This makes it possible to determine whether a low noise amplifiermay be suitable for a particular system by judging their relative levels of performance.

BasicsEssentially the measurement assesses the amount of noise each part of the system or thesystem as a whole introduces. This could be the radio receiver, or an RF amplifier forexample. If the system were perfect then no noise would be added to the signal when itpassed through the system and the signal to noise ratio would be the same at the output asat the input. As we all know this is not the case and some noise is always added. Thismeans that the signal to noise ratio or SNR at the output is worse than the signal to noiseratio at the input. In fact the noise figure is simply the comparison of the SNR at the inputand the output of the circuit.A figure known as the noise factor can be derived simply by taking the SNR at the inputand dividing it by the SNR at the output. As the SNR at the output will always be worse,i.e. lower, this means that the noise factor is always greater than one.The noise factor is rarely seen in specifications. Instead the noise figure is always seen.This is simply the noise factor expressed in decibels.Noise figureIn the diagram S1 is the signal at the input, N1 is the noise at the inputand S2 is the signal at the output and N2 the noise at the outputAs an example if the signal to noise ratio at the input was 4:1, and it was 3:1 at the outputthen this would give a noise factor of 4/3 and a noise figure of 10 log (4/3) or 1.25 dB.Alternatively if the signal to noise ratios are expressed in decibels then it is quite easy tocalculate the noise figure simply by subtracting one from another because two numbersare divided by subtracting their logarithms. In other words if the signal to noise ratio was13 dB at the input and only 11 dB at the output then the circuit would have a noise figureof 13 - 11 or 2 dB.Typical examplesThe specifications of different pieces of equipment will vary quite widely. A typical HFreceiver may have a noise figure of 15 dB of more and function quite satisfactorily. Abetter level of performance is not necessary because of the high level of atmosphericnoise. However an amateur receiver used on Two metres, for example, might have anoise figure of 3 or 4 dB. RF amplifiers for this band often have a noise figure of around1 dB. However it is interesting to note that even the best professional wide-band VHFUHF receivers may only have a noise figure of around 8 dB.Navigation:: Home >> Radio receiver technology >> this pageRadio receiver noise floor- an overview of the noise floor or a receiver, what it is and how the noisefloor affects the performance of a receiver.Noise is a fact of life. Despite the best efforts of any design engineers, there is alwayssome background noise present in any radio receiver. The noise emanates from manysources, and although the design of the receiver is optimised to reduce it some willalways be present.

Accordingly a concept that is very useful in many elements of signal theory and hence inradio receiver design is that of a noise floor. The noise floor can be defined as themeasure of the signal created from the sum of all the noise sources and unwanted signalswithin a system.In order to reduce the levels of noise and thereby improve the sensitivity of the receiver,the main element of the receiver that requires its performance to be optimised is the RFamplifier. The use of a low noise amplifier at the front end of the receiver will ensure thatits performance will be maximised. Wither for use at microwaves or lower frequencies,this RF amplifier is the chief element in determining the performance of the wholereceiver. The next most important element is the first mixer.Receiver noise floorWhile noise can emanate from many sources, when looking purely at the receiver, thenoise is dependent upon a number of elements. The first is the minimum equivalent inputnoise for the receiver. This can be calculated from the following formula:P = k T BWhere:P is the power in wattsK is Boltzmann's constant (1.38 x 10^-23 J/K)B is the bandwidth in HertzUsing this formula it is possible to determine that the minimum equivalent input noise fora receiver at room temperature (290K) is -174 dBm / Hz.It is then possible to calculate the noise floor for the receiver:Noise floor = -174 + NF + 10 log BandwidthWhere NF is the noise figuredBm is the power level expressed in decibels relative to one milliwattNavigation:: Home >> Radio receiver technology >> this pageRadio receiver strong signal response- including intermodulation distortion, third order intercept point, cross modulationand blockingReceiver sensitivity is important but equally so is the way in which a receiver handlesstrong signals. Specifications including intermodulation distortion, third order interceptpoint, cross modulation and blocking can be equally vital. In any receiver design a goodbalance must be achieved between the sensitivity and the strong signal handlingcapability. Under some conditions receivers may need to contend with signals that areonly a few microvolts, but equally they need to handle the conditions when manymillivolts enter the front end.,/p>RF amplifierUnder normal conditions the RF amplifiers should remain linear with the outputremaining proportional to the input. Unfortunately even the best amplifiers have limits to

their output capability, and beyond this they start to overload. When this happens theiroutput starts to limit and the output is less than expected. At this point the amplifier issaid to be in compression.The characteristic curve for an amplifierCompression in itself is not a problem. The absolute values of a signal are of little valueand in any case the automatic gain control (AGC) used in most receivers means that thegain is reduced when strong signals are being received. However the side effects ofcompression give rise to major problems. Effects like intermodulation distortion, crossmodulation, blocking and others mean that the operation of the receiver can be seriouslyimpaired. It is these aspects which are of great importance in the receiver design.To help prevent these problems occurring, receivers have a number of methods ofreducing the signals levels. The most important is the AGC. This is standard on virtuallyevery receiver and operates on many of the amplifier stages within the set. It prevents thesignals from becoming too large, especially in the later stages of the set. However itcannot always prevent the front end stages from being overloaded. This is particularlytrue when the offending strong signal is slightly off channel. In this case it will enter theearly stages of the set but not pass through the IF filters (assuming the receiver is asuperhet). This will mean that the AGC will not be affected but the signal is still able tooverload some of the early stages.Some HF communications receivers have an attenuator on the input, although manyreceivers used in applications such as cellular telecommunications, PMR and the like willnot have these and the receiver will need to be able to handle the strong signals withoutthis assistance.In view of the importance of the various aspects of overloading, a number ofspecifications quantify the various problems caused. However to look at these it isnecessary to look at the effects and how they arise.DistortionThe problems from compression arise as a result of the distortion which occurs to thesignal when the amplifier runs into compression. The actual method which gives rise toproblems may not be obvious at first sight. It can be viewed as the combination of twoeffects. However to see how it arises it is necessary to look at some of the basic effects ofcompression.One of the forms of distortion which arises is harmonic distortion where harmonics of thewanted signal are produced. Depending upon the exact way in which the signal iscompressed the levels of even order harmonics (2f, 4f, 6f, etc) and odd order harmonics(3f, 5f, 7f, etc) will vary. As a result of the production of these harmonics it is possiblethat signals below that being received could be picked up. However the RF selectivity islikely to remove these signals before they enter the first stages of the receiver.Another effect which can be noticed is that the amplifier tends to act as a mixer. The nonlineartransfer curve means that signals will mix together or modulate one another. Thiseffect is known as intermodulation. It is unlikely that this effect on its own would giveany problems. The mix products from signals close to the wanted one fall well away from

the received signal. Alternatively, to produce a signal within the receiver pass-band,signals well away from the received one would need to be entering the r.f. amplifier.These would normally be rejected by the RF selectivity. Take the example of two signalson 50.00 and 50.01 MHz. These would mix together to give signals at 0.01 MHz and100.01 MHz. These are not likely to give rise to any problems.Problems start to arise when the two effects combine with one another. It is quite possiblefor a harmonic of one signal to mix with the fundamental or a harmonic of the other. Thethird order sum products like 2f1 + f2 are unlikely to cause a problem, but the differenceproducts like 2f1 - f2 can give significant problems. Take the example of a receiver set to50 MHz where two strong signals are present, one at 50.00 MHz and the other at 50.01MHz. The difference signals produced will be at 2 x 50.00 - 50.01 = 49.99 MHz andanother at 2 x 50.01 - 50 = 50.01 MHz. As it can be seen either of these could causeinterference on the band. Other higher order products can also cause problems: 3f1 - 2f2,4f1 - 3f2, 5f1 - 4f2, and so forth all give products which may could pass through thereceiver if it is tuned to the relevant frequency.Intermodulation products from two signalsIn this way the presence of a strong signal can produce other spurious signals which canappear in its vicinity. The signals mixing with one another in this way may be of a varietyof different types, e.g. AM, FM, digital modulation, etc, all of which may combinetogether to give what is effectively noise. This means that poor third orderintermodulation performance can have the effect of raising the noise floor under realoperating conditions.Third Order InterceptIt is found that the level of intermodulation products rise very fast. For a 1 dB increase inwanted signal levels, third order products will rise by 3 dB, and fifth order ones by 5 dB.This can be plotted to give a graph of the performance of the amplifier. Eventually theamplifier will run into saturation and the levels of all the signals will be limited. Howeverif the curve of the wanted signals and the third order products was continued, the twolines would intersect. This is known as the third order intercept point. Naturally thehigher the level of the intercept point, the better the performance of the amplifier. For agood receiver and intercept point of 25 dBm (i.e. 25 dB above 1 milliwatt or about 0.5watt) might be expected.

The third order intercept point of an amplifierBlockingWhen a very strong off channel signal appears at the input to a receiver it is often foundthat the sensitivity is reduced. The effect arises because the front end amplifiers run intocompression as a result of the off channel signal. This often arises when a receiver andtransmitter are run from the same site and the transmitter signal is exceedingly strong.When this occurs it has the effect of suppressing all the other signals trying to passthrough the amplifier, giving the effect of a reduction in gain.Blocking is generally specified as the level of the unwanted signal at a given offset(normally 20 kHz) which will give a 3 dB reduction in gain. A good receiver may be ableto withstand signals of about ten milliwatts before this happens.Cross modulationAnother effect which can be noticed when there are strong signals entering the receiver isknown as cross modulation. When this occurs the modulation from a strong signal can betransferred onto other signals being picked up. This effect is particularly obvious whenamplitude modulated signals are being received. In this case the modulation of anothersignal can be clearly heard.Cross modulation normally arises out of imperfect mixer performance in the radio,although it can easily occur in one of the RF amplifiers. As it is a third order effect, areceiver with a good third order intercept point should also exhibit good cross modulationperformance.To specify the cross modulation performance the effect of a strong AM carrier on asmaller wanted signal is noted. Generally the level of a strong carrier with 30%modulation needed to produce an output 20 dB below that produced by the wanted signal.The wanted signal level also has to be specified and 1mV or -47dBm (i.e. a signal 47 dBbelow 1 mW) is often taken as standard, together with an offset frequency of 20 kHz.Sensitivity is one of the main specifications of any radio receiver. However thesensitivity of a set is by no means the whole story. The specification for a set may show itto have an exceedingly good level of sensitivity, but when it is connected to an antennaits performance may be very disappointing because it is easily overloaded when strongsignals are present, and this may impair its ability to receive weak signals.The overall dynamic range of the receiver is very important. It is just as important for aset to be able to handle strong signals well as it is to be able to pick up weak ones. Thisbecomes very important when trying to pick up weak signals in the presence of nearbystrong ones. Under these circumstances a set with a poor dynamic range may not be ableto hear the weak stations picked up by a less sensitive set with a better dynamic range.Problems like blocking, inter-modulation distortion and the like within the receiver maymask out the weak signals, despite the set having a very good level of sensitivity.What is dynamic range?The dynamic range of a receiver is essentially the range of signal levels over which it canoperate. The low end of the range is governed by its sensitivity whilst at the high end it isgoverned by its overload or strong signal handling performance. Specifications generallyuse figures based on either the inter-modulation performance or the blockingperformance. Unfortunately it is not always possible to compare one set with anotherbecause dynamic range like many other parameters can be quoted in a number of ways.However to gain an idea of exactly what the dynamic range of a receiver means it isworth looking at the ways in which the measurements are made to determine the range ofthe receiver.SensitivityThe first specification to investigate is the sensitivity of a set. The main limiting factor inany receiver is the noise generated. For most applications either the signal to noise ratioor the noise figure is used as described in a previous issue of MT. However for dynamicrange specifications a figure called the minimum discernible signal (MDS) is often used.

This is normally taken as a signal equal in strength to the noise level. As the noise level isdependent upon the bandwidth used, this also has to be mentioned in the specification.Normally the level of the level of the MDS is given in dBm i.e. dB relative to a milliwattand typical values are around -135 dBm in a 3 kHz bandwidth.Strong signal handlingAlthough the sensitivity is important the way in which a receiver handles strong signals isalso very important. Here the overload performance governs how well the receiverperformance.In the ideal world the output of an amplifier would be proportional to the input for allsignal levels. However amplifiers only have a limited output capability and it is foundthat beyond a certain level the output falls below the required level because it cannothandle the large levels required of it. This gives a characteristic like that shown in Fig. 1.From this it can be seen that amplifiers are linear for the lower part of the characteristic,but as the output stages are unable to handle the higher power levels the signals starts tobecome compressed as seen by the curve in the characteristic.A typical amplifier characteristicThe fact that the amplifier is non-linear does not create a major problem in itself.However the side effects do. When a signal is passed through a non-linear element thereare two main effects which are noticed. The first is that harmonics are generated.Fortunately these are unlikely to cause a major problem. For a harmonic to fall near thefrequency being received, a signal at half the received frequency must enter the amplifier.The front end tuning should reduce this by a sufficient degree for it not to be a noticeableproblem under most circumstances.The other problem that can be noticed is that signals mix together to form unwantedproducts. These again are unlikely to cause a problem because any signals which couldmix together should be removed sufficiently by the front end tuning. Instead problemsoccur when harmonics of in-band signals mix together.Third order productsProblems occur when harmonics of in-band signals mix together. It is found that a combof signals can be produced as shown in Figure 2, and these may just fall on the samefrequency as a weak and intersting station, thereby masking it out so it cannot be heard.It is simple to calculate the frequencies where the spurious signals will fall. If the inputfrequencies are f 1 and f 2 , then the new frequencies produced will be at 2f 1 - f 2 , 3f 1 - 2f 2 ,4f 1 - 3f 2 and so forth. On the other side of the two main or original signals products areproduced at 2f 2 - f 1 , 3f 2 - 2f 2 , 4f 2 - 3f 1 and so forth as shown in the diagram. These areknown as odd order inter-modulation products. Two times one signal plus one timesanother makes a third order product, three times one plus two times another is a fifthorder product and so forth. It can be seen from the diagram that the signals either side ofthe main signals are first the third order product, then fifth, seventh and so forth.To take an example with some real figures. If large signals appear at frequencies of 30.0MHz and 30.01 MHz, then the inter-modulation products will appear at 30.02, 30.03,30.4 ...MHz and 29.99, 29.98, 29.97 ..... MHz.

Inter-modulation productsBlockingAnother problem that can occur when a strong signal is present is known as blocking. Asthe name implies it is possible for a strong signal to block or at least reduce the sensitivityof a receiver. The effect can be noticed when listening to a relatively weak station and anearby transmitter starts to radiate, and the wanted signal reduces in strength. The effectis caused when the front-end amplifier starts to run into compression. When this occursthe strongest signal tends to "capture" the amplifier reducing the strength of the othersignals. The effect is the same as the capture effect associated with FM signals.The amount of blocking is obviously dependent upon the level of the signal. It alsodepends on how far off channel the strong signal is. The further away, the more it will bereduced by the front end tuning and the less the effect will be. Normally blocking isquoted as the level of the unwanted signal at a given offset (normally 20 kHz) to give a 3dB reduction in gain.Dynamic range definitionWhen looking at dynamic range specifications, care must be taken when interpretingthem. The MDS at the low signal end should be viewed carefully, but the limiting factorsat the top end show a much greater variation tin the way they are specified. Whereblocking is used a reduction of 3 dB sensitivity is normally specified, but in some casesmay be 1 dB used. Where the inter-modulation products are chosen as the limiting pointthe input signal level for them to be the same as the MDS is often taken. Howeverwhatever specification is given, care should be taken to interpret the figures as they maybe subtlety different in the way they are measured from one receiver to the next.To gain a feel for the figures which may be obtained where inter-modulation is thelimiting factor figures of between 80 and 90 dB range are typical, and where blocking isthe limiting factor figures around 115 dB are generally achieved in a good receiver.Designing for optimum performanceIt is not an easy task to design a highly sensitive receiver that also has a wide dynamicrange. To achieve this performance a number of methods can be used. The front-endstage is the most critical in terms of noise performance. It should be optimised for noiseperformance rather than gain. Input impedance matching is critical for this. It isinteresting to note that the optimum match does not correspond exactly with the bestnoise performance. The amplifier should also have a relatively high output capability toensure it does not overload. The mixer is also critical to the overload performance. Toensure the mixer is not overloaded there should not be excessive gain preceding it. A highlevel mixer should also be used (i.e. one designed to accept a high-level local oscillatorsignal). In this way it can tolerate high input signals without degradation in performance.Care should be taken in the later stages of the receiver to ensure that they can tolerate thelevel of signals likely to be encountered. A good AGC system also helps preventoverloading and the generation of unwanted spurious signals.A receiver with a good dynamic range will be able to give a far better account of itselfunder exacting conditions than one designed purely for optimum sensitivity.

Frequency modulation is widely used in radio communications and broadcasting,particularly on frequencies above 30 MHz. It offers many advantages, particularly inmobile radio applications where its resistance to fading and interference is a greatadvantage. It is also widely used for broadcasting on VHF frequencies where it is able toprovide a medium for high quality audio transmissions.In view of its widespread use, a wide variety of receivers are able to demodulate thesetransmissions. Naturally there are specifications and figures that receiver manufacturersquote for the performance of their sets when receiving FM. These include sch figures asquieting, capture ratio and the like.Receiving FMIn order to be able to receive FM a receiver must be sensitive to the frequency variationsof the incoming signals which may be wide or narrow band. However the set is madeinsensitive to the amplitude variations. This is achieved by having a high gain IFamplifier. Here the signals are amplified to such a degree that the amplifier runs intolimiting. In this way any amplitude variations are removed and this improves the signal tonoise ratio after the point when the signal limits in the IF stages. However the high levelsof gain associated with the limiting process mean that when no signal is present, veryhigh levels of noise appear at the output of the FM demodulator.,/p>SquelchTo overcome the problem of the high noise levels when no signal is present a circuitknown as "squelch" is normally used. This detects when no signal is present and cuts theaudio, thereby removing the noise under these conditions. The level for this is normallypresent in domestic radios, but there is often a level adjustment for PMR or handheldtransceivers, or for scanners and professional receivers.Quieting specificationOne of the advantages of FM is its resilience to noise. This is one of the main reasonswhy it is used for high quality audio broadcasts. However when no signal is present, ahigh noise level is present at the output of the receiver. If a low level FM signal isintroduced and its level slowly increased it will be found that the noise level reduces.From this the quieting level can be deduced. It is the reduction in noise level expressed indecibels when a signal of a given strength is introduced to the input of the set. Typically abroadcast tuner should give a quieting level of 30 dB for an input level of around amicrovolt.Capture effectAnother effect that is often associated with FM is called the capture effect. This can bedemonstrated when two signals are present on the same frequency. When this occurs it isfound that only the stronger signal will heard at the output This can be compared to AMwhere a mixture of the two signals is heard, along with a heterodyne if there is afrequency difference.A capture ratio is often defined in receiver specifications. It is the ratio between thewanted and unwanted signal to give a certain reduction in level of the unwanted signal atthe output. Normally a reduction of the unwanted signal of 30 dB is used. To give anexample of this the capture ratio may be 2 dB for a typical tuner to give a reduction of 30dB in the unwanted signal. In other words if the wanted signal is only 2 dB stronger thanthe unwanted one, the audio level of the unwanted one will be suppressed by 30 dB.The phase locked loop or PLL is a particularly flexible circuit building block. The phaselocked loop, PLL can be used for a variety of radio frequency applications, andaccordingly the PLL is found in many radio receivers as well as other pieces ofequipment.The phase locked loop, PLL, was not used in early radio equipment because of thenumber of different stages required. However with the advent of radio frequencyintegrated circuits, the idea of phase locked loops, PLLs, became viable. Initiallyrelatively low frequency PLLs became available, but as RF IC technology improved, so

the frequency at which PLLs would operate rose, and high frequency versions becameavailable.Phase locked loops are used ain a large variety of applications within radio frequencytechnology. PLLs can be used as FM demodulators and they also form the basis ofindirect frequency synthesizers. In addition to this they can be used for a number ofapplications including the regeneration of chopped signals such as the colour burst signalon an analogue colour television signal, for types of variable frequency filter and a hostof other specialist applicationsConcepts - phaseThe operation of a phase locked loop, PLL, is based around the idea of comparing thephase of two signals. This information about the error in phase or the phase differencebetween the two signals is then used to control the frequency of the loop.To understand more about the concept of phase and phase difference, first visualise aradio frequency signal in the form of a familiar x-y plot of a sine wave. As timeprogresses the amplitude oscillates above and below the line, repeating itself after eachcycle. The linear plot can also be represented in the form of a circle. The beginning of thecycle can be represented as a particular point on the circle and as a time progresses thepoint on the waveform moves around the circle. Thus a complete cycle is equivalent to360 degrees. The instantaneous position on the circle represents the phase at that givenmoment relative to the beginning of the cycle.To look at the concept of phase difference, take the example of two signals. Although thetwo signals have the same frequency, the peaks and troughs do not occur in the sameplace. There is said to be a phase difference between the two signals. This phasedifference is measured as the angle between them. It can be seen that it is the anglebetween the same point on the two waveforms. In this case a zero crossing point has beentaken, but any point will suffice provided that it is the same on both.When there two signals have different frequencies it is found that the phase differencebetween the two signals is always varying. The reason for this is that the time for eachcycle is different and accordingly they are moving around the circle at different rates. Itcan be inferred from this that the definition of two signals having exactly the samefrequency is that the phase difference between them is constant. There may be a phasedifference between the two signals. This only means that they do not reach the same pointon the waveform at the same time. If the phase difference is fixed it means that one islagging behind or leading the other signal by the same amount, i.e. they are on the samefrequency.PLL basicsA phase locked loop, PLL, is basically of form of servo loop. Although a PLL performsits actions on a radio frequency signal, all the basic criteria for loop stability and otherparameters are the same.A basic phase locked loop, PLL, consists of three basic elements:• Phase comparator: As the name implies, this circuit block within the PLLcompares the phase of two signals and generates a voltage according to the phasedifference between the two signals.• Loop filter: This filter is used to filter the output from the phase comparator inthe PLL. It is used to remove any components of the signals of which the phase isbeing compared from the VCO line. It also governs many of the characteristics ofthe loop and its stability.• Voltage controlled oscillator (VCO): The voltage controlled oscillator is thecircuit block that generates the output radio frequency signal. Its frequency can becontrolled and swung over the operational frequency band for the loop.PLL operationThe concept of the operation of the PLL is relatively simple, although the mathematicalanalysis can become more complicated

The Voltage Controlled Oscillator, VCO, within the PLL produces a signal which entersthe phase detector. Here the phase of the signals from the VCO and the incomingreference signal are compared and a resulting difference or error voltage is produced.This corresponds to the phase difference between the two signals.Block diagram of a basic phase locked loop (PLL)The error signal from the phase detector in the PLL passes through a low pass filterwhich governs many of the properties of the loop and removes any high frequencyelements on the signal. Once through the filter the error signal is applied to the controlterminal of the VCO as its tuning voltage. The sense of any change in this voltage is suchthat it tries to reduce the phase difference and hence the frequency between the twosignals. Initially the loop will be out of lock, and the error voltage will pull the frequencyof the VCO towards that of the reference, until it cannot reduce the error any further andthe loop is locked.When the PLL is in lock a steady state error voltage is produced. By using an amplifierbetween the phase detector and the VCO, the actual error between the signals can bereduced to very small levels. However some voltage must always be present at thecontrol terminal of the VCO as this is what puts onto the correct frequency.The fact that a steady error voltage is present means that the phase difference between thereference signal and the VCO is not changing. As the phase between these two signals isnot changing means that the two signals are on exactly the same frequency.SummaryThe phase locked loop is one of the most versatile building blocks in radio frequencyelectronics today. Whilst it was not widely used for many years, the advent of the ICmeant that phase locked loop and synthesizer chips became widely available. This madethem cheap to use and their advantages could be exploited to the full. Nowadays most hifituners and car radios use them and a large proportion of the portable radios on themarket as well. With their interface to microprocessors so easy their use is assured formany years to come.The phase detector is the core element of a phase locked loop, PLL. Its action enables thephase differences in the loop to be detected and the resultant error voltage to be produced.There is a variety of different circuits that can be used as phase detectors, some that usewhat may be considered as analogue techniques, while others use digital circuitry.However the most important difference is whether the phase detector is sensitive to justphase or whether it is sensitive to frequency and to phase. Thus phase detectors may besplit into two categories:• Phase only sensitive detectors• Phase - frequency detectorsPhase only sensitive detectorsPhase detectors that are only sensitive to phase are the most straightforward form ofdetector. They simply produce an output that is proportional to the phase differencebetween the two signals. When the phase difference between the two incoming signals issteady, they produce a constant voltage. When there is a frequency difference betweenthe two signals, they produce a varying voltage. In fact the simplest form of phase only

sensitive detector is a mixer. From this it can be seen that the output signal will be havesum and difference signals.The difference frequency product is the one used to give the phase difference. It is quitepossible that the difference frequency signal will fall outside the pass-band of the loopfilter. If this occurs then no error voltage will be fed back to the Voltage ControlledOscillator (VCO) to bring it into lock. This means that there is a limited range over whichthe loop can be brought into lock, and this is called the capture range. Once in lock theloop can generally be pulled over a much wider frequency band.To overcome this problem the oscillator must be steered close to the reference oscillatorfrequency. This can be achieved in a number of ways. One is to reduce the tuning rangeof the oscillator so that the difference product will always fall within the pass-band of theloop filter. In other instances another tune voltage can be combined with the feedbackfrom the loop to ensure that the oscillator is in the correct region. This is approach isoften adopted in microprocessor systems where the correct voltage can be calculated forany given circumstance.Phase - frequency sensitive detectorsAnother form of detector is said to be phase-frequency sensitive. These circuits have theadvantage that whilst the phase difference is between +/- 180 a voltage proportional tothe phase difference is given. Beyond this the circuit limits at one of the extremes. In thisway no AC component is produced when the loop is out of lock and the output from thephase detector can pass through the filter to bring the phase locked loop, PLL, into lock.Navigation:: Home >> Radio receiver technology >> this pageVoltage controlled oscillator, VCO, for PLLs- an overview of the various types of voltage controlled oscillator, VCO,used in phase locked loops, PLLs and frequency synthesizersWithin a phase locked loop, PLL, or frequency synthesizer, the performance of thevoltage controlled oscillator, VCO is of paramount importance. In order that the PLL orsynthesizer can meet its full specification a well designed oscillator is essential.Designing a really high performance voltage controlled oscillator, VCO, is not alwayseasy as there are a number of requirements that need to be met. However by carefuldesign, and some experimentation a good VCO design can be found.VCO requirementsJust like any other circuit, with a VCO there are a number of design requirements thatneed to be known from the beginning of the design process. These basic requirements forthe VCO will govern many of the decisions concerning the circuit topology and otherfundamental aspects of the circuit. Some of the basic requirements are:• Tuning range• Tuning gain - tuning shift for a given tuning voltage change• Phase noise (low phase noise)These are some of the main requirements that need to be known from the outset of thedesign of the VCO. The overall tuning range and the gain are basic requirements that arepart of the basic design of any PLL into which the VCO may be incorporated. So too is

the phase noise characteristic. As phase noise is a basic parameter of any PLL orfrequency synthesizer, so too is the characteristic of the VCO, and low phase noise VCOsare often required. For example the VCO performance may govern the overall design ofthe frequency synthesizer or PLL, if a given phase noise performance is to be met.VCO circuitsLike any oscillator, a VCO may be considered as an amplifier and a feedback loop. Thegain of the amplifier may be denoted as A and the feedback as B.For the circuit to oscillate the total phase shift around the loop must be 360 degrees andthe gain must be unity. In this way signals are fed back round the loop so that they areadditive and as a result, any small disturbance in the loop is fed back and builds up. Inview of the fact that the feedback network is frequency dependent, the build up of signalwill occur on one frequency, the resonant frequency of the feedback network, and asingle frequency signal is produced.Many oscillators and hence VCOs use a common emitter circuit. This in itself produces aphase shift of 180 degrees, leaving the feedback network to provide a further 180degrees.Other oscillator or VCO circuits may use a common base circuit where there is no phaseshift between the emitter and collector signals (assuming a bipolar transistor is used) andthe phase shift network must provide either 0 degrees or 360 degrees.Colpitts and Clapp VCO circuitsTwo commonly used examples of VCO circuits are the Colpitts and Clapp oscillators. Ofthe two, the Colpitts circuit is the most widely used, but these circuits are both verysimilar in their configuration.These circuits operate as oscillators because it is found that a bipolar transistor withcapacitors placed between the base and emitter (C1) and the emitter and ground (C2)fulfils the criteria required for providing sufficient feedback in the correct phase toproduce an oscillator. For oscillation to take place the ratio C1: C2 must be greater thanone.The resonant circuit is made by including a inductive element between the base andground. In the Colpitts circuit this consists of just an inductor, whereas in the Clappcircuit an indictor and capacitor in series are used.The conditions for resonance is that:f^2 = 1 / (4 pi^2 L C )The capacitance for the overall resonant circuit is formed by the series combination of thetwo capacitors C1 and C2 in series. In the case of the Clapp oscillator, the capacitor inseries with the inductor is also included in series with C1 and C2.Thus the series capacitance is:Ctot= 1 / C1 + 1 / C2In order to make the oscillator tune it is necessary to vary the resonant point of the circuit.This is best achieved by adding a capacitor across the indictor in the case of the Colpittsoscillator. Alternatively for the Clapp oscillator, it can be the capacitor in series with theinductor.For high frequency applications a circuit where the inductive reactance is placed betweenthe base and ground is often preferred as it is less prone to spurious oscillations and otheranomalies.Choice of VCO active deviceIt is possible to use both bipolar devices and FETs within a VCO, using the same basic

circuit topologies. The bipolar transistor has a low input impedance and is current driven,while the FET has a high input impedance and is voltage driven. The high inputimpedance of the FET is able to better maintain the Q of the tuned circuit and this shouldgive a better level of performance in terms of the phase noise performance where themaintenance of the Q of the tuned circuit is a key factor in the reduction of phase noise.Another major factor is the flicker noise generated by the devices. Oscillators are highlynon-linear circuits and as a result the flicker noise is modulated onto VCO as sidebandsand this manifests itself as phase noise. In general bipolar transistors offer a lower levelof flicker noise and as a result VCOs based around them offer a superior phase noiseperformance.VCO tuningTo make a VCO, the oscillator needs to be tuned by a voltage. This can be achieved bymaking the variable capacitor from varactor diodes. The tune voltage for the VCO canthen be applied to the varactors.When varactor diodes are used, care must be taken in the design of the circuit to ensurethat the drive level in the tuned circuit is not too high. If this is the case, then the varactordiodes may be driven into forward conduction, reducing the Q and increasing the level ofspurious signals.There are two main types of varactor diode that may be used within a VCO: abrupt andhyper-abrupt diodes. The names refer to the junction within the diode. The abrupt ones donot have a sharp a transition between the two semiconductor types in the diode, and thisaffect the performance offered.Hyper-abrupt diodes have a relatively linear voltage : capacitance curve and as a resultthey offer a very linear tuning characteristic that may be required in some applications.They are also able to tune over a wide range, and may typically tune over an octave rangewith less than a 20 volt change in tuning voltage. However they do not offer aparticularly high level of Q. As this will subtract from the overall Q of the tuned circuitthis will mean that the phase noise performance is not optimum.Abrupt diodes, while not offering such a high tuning range or linear transfer characteristicare able to offer a higher Q. This results in a better phase noise (i.e. low phase noise)performance for the VCO. The other point to note is that they may need a high tuningvoltage to provide the required tuning range, as some diodes may require a tuning voltagefor the VCO to vary up to 50 volts or slightly more.SummaryThe design of a VCO can be interesting and challenging. Whether the aim is to design alow noise VCO, a low current VCO, a PLL VCO, or one that will cover a wide tuningrange there are many aspects that need to be addressed. Often when a successful designhas been obtained, it will slightly modified to enable it to cover a wide range of similarapplications.Navigation:: Home >> Radio receiver technology >> this pageOscillator design for low phase noise- an overview of the design of radio frequency, RF, oscillators for lowphase noise levels.One of the key requirements for any oscillator used in a radio receiver, radio transmitter,or many other applications is for the oscillator to perform with low levels of phase noise.Whether the oscillator is used in a frequency synthesizer, or in any other application, thebasic design principles for achieving low phase noise are the same.

Poor levels of oscillator can manifest themselves in slightly different ways. For ananalogue radio receiver a poor performance oscillator may result in poor reciprocalmixing performance. It may also raise the noise floor of the receiver. In a radio systemrelying on phase modulation, phase noise will degrade the bit error rate performance.Additionally transmitters exhibiting a poor phase noise performance will tend to transmitwide band noise, causing interference to users on other frequencies.Key points for oscillator designThere are some areas points to address when designing an oscillator to ensure that it has agood phase noise performance. By addressing these and other relevant points, theperformance of the oscillator meets its requirements.• High Q resonant circuit• Choice of oscillator device• Correct feedback level• Sufficient oscillator power output• Power line rejectionOscillator design methodologyIn order that the oscillator is able to provide the optimum phase noise performance it isnecessary to implement these elements into the design of the circuit from the outset.Some aspects may affect the basic design criteria, and therefore need to be included fromthe concept stage of the circuit:Q of the resonant circuit: One of the major factors in determining the phase noiseperformance of an oscillator is the Q of the resonant circuit. Broadly, the higher the Q ofthe oscillator tuned circuit, the better the phase noise performance. This inductors shouldbe chosen to provide the highest Q, as should the capacitors. This is particularly true ofvoltage controlled oscillators, VCOs where the varactor diodes normally employed havea lower Q than other capacitors.Typically high Q tuned circuits do not have the tuning range of lower Q circuits. Thismeans that when wide tuning ranges are required, it becomes more difficult to obtain ahigh level of Q and hence the optimum phase noise.As an illustration of the effect of having a high Q resonant circuit in an oscillator, crystaloscillators exhibit very low levels of phase noise as a result of the fact that the crystalsused in them possess very high levels of Q.Choice of oscillator active device: It is possible to use both bipolar devices and FETswithin an RF oscillator, using the same basic circuit topologies. The bipolar transistor hasa low input impedance and is current driven, while the FET has a high input impedanceand is voltage driven. The high input impedance of the FET is able to better maintain theQ of the tuned circuit and this should give a better level of performance in terms of thephase noise performance where the maintenance of the Q of the tuned circuit is a keyfactor in the reduction of phase noise.

Another major factor is the flicker noise generated by the devices. Oscillators are highlynon-linear circuits and as a result the flicker noise is modulated onto the oscillation assidebands. This manifests itself as phase noise. In general bipolar transistors offer a lowerlevel of flicker noise and as a result oscillators based around them offer a superior phasenoise performance.Oscillator feedback level: A critical feature in any oscillator design is to ensure that thecorrect level of feedback is maintained. There should be sufficient to ensure thatoscillation is maintained over the frequency range, over the envisaged temperature rangeand to accommodate the gain and parameter variations between the devices used.However if the level of feedback is too high, then the level of noise will also beincreased. Thus the circuit should be designed to provide sufficient feedback for reliableoperation and little more.Sufficient oscillator power output: It is found that the noise floor of an oscillator isreasonably constant in absolute terms despite the level of the output signal. In somedesigns there can be improvements in the overall signal to noise floor level to be made byusing a high level signal and applying this directly to the mixer or other circuit where itmay be required. Accordingly some low noise circuits may use surprisingly highoscillator power levels.Power line rejection: It is necessary to ensure that any supply line or other extraneousnoise is not presented to the oscillator. Supply line ripple, or other unwanted pickup canseriously degrade the performance of the oscillator. To overcome this, good supplysmoothing and regulation is absolutely necessary. Additionally it may be advisable toplace the oscillator within a screened environment so that it does not pick up any straynoise. It is worth remembering that the oscillator acts as a high gain amplifier, especiallyclose to the resonant frequency. Any noise picked up can be amplified and will manifestitself as phase noise.SummaryThere are many elements to ensuring that an oscillator circuit design meets itsrequirements for low phase noise. The points provider here give a start to some of thebasic decisions that are needed. Once initially realised, some refinement is likely to beneeded to ensure the optimum performance is obtained.Navigation:: Home >> Radio receiver technology >> this pagePLL loop filter- an overview of the loop filter used in a phase locked loop, PLL. This givesan overview of the requirements, and design.The design of the PLL, loop filter is crucial to the operation of the whole phase lockedloop. The choice of the circuit values here is usually a very carefully balancedcompromise between a number of conflicting requirements.The PLL filter is needed to remove any unwanted high frequency components whichmight pass out of the phase detector and appear in the VCO tune line. They would thenappear on the output of the Voltage Controlled Oscillator, VCO, as spurious signals. Toshow how this happens take the case when a mixer is used as a phase detector. When theloop is in lock the mixer will produce two signals: the sum and difference frequencies. Asthe two signals entering the phase detector have the same frequency the differencefrequency is zero and a DC voltage is produced proportional to the phase difference asexpected. The sum frequency is also produced and this will fall at a point equal to twicethe frequency of the reference. If this signal is not attenuated it will reach the controlvoltage input to the VCO and give rise to spurious signals.When other types of phase detector are used similar spurious signals can be produced andthe filter is needed to remove them.

The filter also affects the ability of the loop to change frequencies quickly. If the filterhas a very low cut-off frequency then the changes in tune voltage will only take placeslowly, and the VCO will not be able to change its frequency as fast. This is because afilter with a low cut-off frequency will only let low frequencies through and thesecorrespond to slow changes in voltage level.Conversely a filter with a higher cut-off frequency will enable the changes to happenfaster. However when using filters with high cut-off frequencies, care must be taken toensure that unwanted frequencies are not passed along the tune line with the result thatspurious signals are generated.The loop filter also governs the stability of the loop. If the filter is not designed correctlythen oscillations can build up around the loop, and large signals will appear on the tuneline. This will result in the VCO being forced to sweep over wide bands of frequencies.The proper design of the filter will ensure that this cannot happen under anycircumstances.Navigation:: Home >> Radio receiver technology >> this pagePLL Frequency synthesizer tutorial- an introduction to the indirect (phase locked loop - pll) synthesizerToday most receivers use a phase locked loop or PLL frequency synthesizer. Many ofthem advertise this fact by displaying words like "PLL", "Synthesized", or "Quartz" ontheir front panels or in the advertising literature. Whatever one thinks of the saleslanguage, PLL frequency synthesizers offer tremendous advantages to the operation of areceiver. Not only do frequency synthesizers enable receivers to have the same stabilityas the quartz reference, but they also enable many other facilities to be introducedbecause they can easily be controlled by a microprocessor. This enables facilities such asmultiple memories, keypad frequency entry, scanning and much more to be incorporatedinto the set.Phase locked loop, PLL, frequency synthesizers are widely used, but their operation isnot always well understood. One of the reasons for this is that their design can involvesome complicated math, but despite this the basic concepts are relatively easy to grasp.PLL BasicsA frequency synthesizer is based around a phase locked loop or PLL. This circuit uses theidea of phase comparison as the basis of its operation. From the block diagram of a basicloop shown in Fig. 1 it can be seen that there are three basic circuit blocks, a phasecomparator, voltage controlled oscillator, and loop filter. A reference oscillator issometimes included in the block diagram, although this is not strictly part of the loopitself even though a reference signal is required for its operation.Block diagram of a basic phase locked loop (PLL)The phase locked loop, PLL, operates by comparing the phase of two signals. The signalsfrom the voltage controlled oscillator and reference enter the phase comparator Here athird signal equal to the phase difference between the two input signals is produced.

The phase difference signal is then passed through the loop filter. This performs anumber of functions including the removal of any unwanted products that are present onthis signal. Once this has been accomplished it is applied to the control terminal of thevoltage controlled oscillator. This tune voltage or error voltage is such that it tries toreduce the error between the two signals entering the phase comparator. This means thatthe voltage controlled oscillator will be pulled towards the frequency of the reference,and when in lock there is a steady state error voltage. This is proportional to the phaseerror between the two signals, and it is constant. Only when the phase between twosignals is changing is there a frequency difference. As the phase difference remainsconstant when the loop is in lock this means that the frequency of the voltage controlledoscillator is exactly the same as the reference.SynthesisizersA phase locked loop, PLL, needs some additional circuitry if it is to be converted into afrequency synthesizer. This is done by adding a frequency divider between the voltagecontrolled oscillator and the phase comparator as shown in Fig. 2.A programmable divider added into a phase locked loop, PLL, enables thefrequency to be changed.Programmable dividers or counters are used in many areas of electronics, including manyradio frequency applications. They take in a pulse train like that shown in Fig. 3, and giveout a slower train. In a divide by two circuit only one pulse is given out for every two thatare fed in and so forth. Some are fixed, having only one division ratio. Others areprogrammable and digital or logic information can be fed into them to set the divisionratio.Operation of a programmable dividerWhen the divider is added into the circuit the phase locked loop, PLL, still tries to reducethe phase difference between the two signals entering the phase comparator. Againwhen the circuit is in lock both signals entering the comparator are exactly the same infrequency. For this to be true the voltage controlled oscillator must be running at afrequency equal to the phase comparison frequency times the division ratio.It can be seen that if the division ratio is altered by one, then the voltage controlledoscillator will have to change to the next multiple of the reference frequency. This meansthat the step frequency of the synthesizer is equal to the frequency entering thecomparator.Most synthesizers need to be able to step in much smaller increments if they are to be ofany use. This means that the comparison frequency must be reduced. This is usually

accomplished by running the reference oscillator at a frequency of a megahertz or so, andthen dividing this signal down to the required frequency using a fixed divider. In this waya low comparison frequency can be achieved.Comparison frequency reduced by adding a fixed divider after the referenceoscillatorAnalogue TechniquesPlacing a digital divider is not the only method of making a synthesizer using a phaselocked loop, PLL. It is also possible to use a mixer in the loop as shown in Fig. 5. Usingthis technique places an offset into the frequency generated by the loop.A phase locked loop, PLL, with mixerThe way in which the phase locked loop, PLL, operates with the mixer incorporated canbe analyzed in the same manner that was used for the loop with a divider. When the loopis in lock the signals entering the phase detector are at exactly the same frequencies. Themixer adds an offset equal to the frequency of the signal entering the other port of themixer. To illustrate the way this operates figures have been included. If the referenceoscillator is operating at a frequency of 10 MHz and the external signal is at 15 MHz thenthe VCO must operate at either 5 MHz or 25 MHz.. Normally the loop is set up so thatmixer changes the frequency down and if this is the case then the oscillator will beoperating at 25 MHz.It can be seen that there may be problems with the possibility of two mix products beingable to give the correct phase comparison frequency. It happens that as a result of thephasing in the loop, only one will enable it to lock. However to prevent the loop gettinginto an unwanted state the range of the VCO is limited. For phase locked loops, PLLs,that need to operate over a wide range a steering voltage is added to the main tunevoltage so that the frequency of the loop is steered into the correct region for requiredconditions. It is relatively easy to generate a steering voltage by using digital informationfrom a microprocessor and converting this into an analogue voltage using a digital to

analogue converter (DAC). The fine tune voltage required to pull the loop into lock isprovided by the loop in the normal way.Multi-loop synthesizersMany high performance synthesizers use several loops that incorporate both mixers anddigital dividers. By using these techniques it is possible to produce high performancewide range signal sources with very small step sizes. If only a single loop is used thenthere may be short falls in the level of performance.There is a large variety of ways in which multi-loop synthesizers can be made, dependentupon the requirements of the individual system. However as an illustration a two loopsystem is shown in Fig. 6. This uses one loop to give the smaller steps and the secondprovides larger steps. This principle can be expanded to give wider ranges and smallersteps.An example of a synthesizer using two loopsThe first phase locked loop, PLL, has a digital divider and operates over the range 19 to28 MHz. Having a reference frequency of 1 MHz it provides steps of 1 MHz. The signalfrom this loop is fed into the mixer of the second one. The second loop has division ratiosof 10 to 19, but as the reference frequency has been divided by 10 to 100 kHz to givesmaller steps.

The operation of the whole loop can be examined by looking at extremes of thefrequency range. With the first loop set to its lowest value the divider is set to 19 and theoutput from the loop is at 19 MHz. This feeds into the second loop. Again this is set tothe minimum value and the frequency after the mixer must be at 1.0 MHz. With the inputfrom the first loop at 19 MHz this means that the VCO must operate at 20 MHz if theloop is to remain in lock.At the other end of the range the divider of the first loop is set to 28, giving a frequencyof 28 MHz. The second phase locked loop, PLL, has the divider set to 19, giving afrequency of 1.9 MHz between the mixer and divider. In turn this means that thefrequency of the VCO must operate at 29.9 MHz. As the phase locked loops, PLLs, canbe stepped independently it means that the whole synthesizer can move in steps of 100kHz between the two extremes of frequency. As mentioned before this principle can beextended to give greater ranges and smaller steps, providing for the needs of modernreceivers.Navigation:: Home >> Radio receiver technology >> this pagePhase noise and frequency synthesizersOne of the main problems with frequency synthesisers and frequency synthesis usingphase locked loops is the fact that some designs generate high levels of phase noise.However it is possible to design some very good low phase noise synthesizers. Theproblem is often that receivers and transceivers are designed for low production costs,and this naturally means that some short cuts are needed.What is phase noise?Phase noise is present on all signals to some degree and it is caused by small phase (andhence frequency) perturbations or jitter on the signal. It manifests itself as noisespreading out either side from the main carrierPhase noise on a signalSome signal sources are better than others. Crystal oscillators are very good and havevery low levels of phase noise. Free running variable frequency oscillators normallyperform well. Unfortunately synthesizers, and especially those based around phase lockedloops, do not always fare so well unless they are well designed. If significant levels ofphase noise are present on a synthesizer used as a local oscillator in a receiver, it canadversely affect the performance of the radio in terms of reciprocal mixing.Some oscillators have phase noise levels that are quoted in their specifications. Any highquality signal generator will have the level of phase noise specified, as do many highperformance crystal oscillators used as standards. Their performance is generallyspecified in dBc/Hz and at a given offset from the carrier. The term dBc simply refers tothe level of noise relative to the carrier, i.e. -10 dBc means that the level is 10 lower thanthe carrier.

The bandwidth in which the noise is measured also has to be specified. The reason forthis is that noise spreads over the frequency spectrum. Obviously the wider bandwidththat is used, the greater the level of noise that will pass through the filter and bemeasured. To prove this, just try selecting a different bandwidth on a receiver and checkwhat happens to the noise level. It will rise for a wider bandwidth and fall when a narrowbandwidth is used. Technically the most convenient bandwidth to use a 1 Hz bandwidthand so this is used. When measuring this a wider bandwidth is usually used because it isdifficult to obtain 1 Hz bandwidth filters and a correction is made mathematically.Finally the level of noise varies as different offsets from the carrier are taken.Accordingly this must be included in a specification. A very good oscillator might have aspecification of -100 dBc/Hz at 10 kHz offset.It has already been mentioned that the level of phase noise changes as the offset from thecarrier changes and for "simple" signal sources like crystal oscillators or variablefrequency oscillators the phase noise reduces as the frequency from the main carrier isincreased. For frequency synthesizers the picture is a little more complicated as we shallsee.Phase noise in synthesizersEach of the components in a frequency synthesizer produces noise that will contribute tothe overall noise that appears at the output. The actual way in which the noise from anyone element in the loop contributes to the output will depend upon where it is produced.Noise generated by the VCO will affect the output in a different way to that generated inthe phase detector for example.To see how this happens take the example of noise generated by the voltage controlledoscillator. This will pass through the divider chain and appear at the output of the phasedetector. It will then have to pass through the loop filter. This will only allow throughthose components of the noise that are below the loop cut-off frequency. These willappear on the error voltage and have the effect of cancelling out the noise on the voltagecontrolled oscillator. As this effect will only take place within the loop bandwidth, it willreduce the level of noise within the loop bandwidth and have no effect on noise outsidethe loop bandwidth.Noise generated by the phase detector is affected in a different way. Again only thecomponents of the noise below the loop bandwidth will pass through the low pass filter.This means that there will be no components outside the loop bandwidth appearing on thetune voltage at the control terminal of the voltage controlled oscillator, and there will beno effect on the oscillator. Those components inside the loop bandwidth will appear atthe oscillator control terminal. These will affect the oscillator and appear as phase noiseon the output of the voltage controlled oscillator.Matters are made worse by the fact that the division ratio has the effect of multiplying thenoise level. This arises because the synthesizer effectively has the effect of multiplyingthe frequency of the reference. Consequently the noise level is also multiplied by a factorof 20 log N, where N is the division ratio.Noise generated by the reference undergoes exactly the same treatments as that generatedby the phase detector. It too is multiplied by the division ratio of the loop in the same waythat the phase detector noise is. This means that even though the reference oscillator mayhave a very good phase noise performance this can be degraded significantly, especiallyif division ratios are high.Dividers normally do not produce a significant noise contribution. Any noise theyproduce may be combined with that of the phase detector.The combined noise of the loop at the output generally looks like that shown in Figure 2.Here it can be seen that the noise within the loop bandwidth arises from the phasedetector and the reference. Outside the loop bandwidth it arises primarily from thevoltage controlled oscillator. From this it can be seen that optimisation of the noiseprofile is heavily dependent upon the choice of the loop bandwidth. It is also necessary to

keep the division ratio in any loop down to reasonable levels. For example a 150 MHzsynthesizer with a 12.5 kHz step size will require a division ratio of 12000. In turn thiswill degrade the phase detector and reference phase noise figures by 81 dB inside theloop bandwidth - a significant degradation by anyone's standards! Provided that divisionratios are not too high then a wide loop bandwidth can help keep the voltage controlledoscillator noise levels down as well.Noise profile of a typical synthesizerEffects of phase noisePhase noise can have a number of effects. For SSB transmitters like those used for HFcommunications for ship to shore, amateur radio and other applications the main effect isthat splatter appears either side of the main signal. This results from the phase noiseeither side of the signal will rising and falling in line with the amplitude variations of themain signal. For digital transmissions using frequency or phase modulation, the noise canintroduce errors causing the bit error rate (BER) to rise.For receivers the main problem is an effect known as reciprocal mixing. To look at howthis occurs take the case of a superhet receiver tuned to a strong signal. The signal willpass through the radio frequency stages, and then in the mixer it will be mixed with thelocal oscillator to produce a new signal at the right frequency to pass through the IFfilters. When the local oscillator is tuned away by ten kilohertz, for example the signalwill no longer be able to pass through the IF filters. However it will still be possible forthe phase noise on the local oscillator to mix with the strong incoming signal to produce asignal that will fall inside the receiver pass-band as shown in Figure 3. This could besufficiently strong to mask out a weak station.The way in which phase noise on a signal results in reciprocal mixingSpecificationsA number of different methods are used to define the level of reciprocal mixing.Generally they involve the response of the receiver to a large off channel signal. Toperform a reciprocal mixing measurement is rarely easy. The signal generator mustalways be much better than the receiver, otherwise the performance of the signalgenerator will be measured! To overcome this many people use an old valve generatorbecause their performance is often very good in this respect.

A measurement can be made by noting the level of audio with a BFO on from a smallsignal. The signal is then tuned off channel by a given amount, normally about 20 kHzand then increased until the audio level rises to the same level as a result of the phasenoise from the receiver. As the noise level is dependent upon the bandwidth of thereceiver this has to be specified as well. Generally a bandwidth useable for SSB is usedi.e. 2.7 kHz.For example a good HF communications receiver might have a figure of 95 dB at a 20kHz offset using a 2.7. kHz bandwidth. This figure will improve as the frequency offsetfrom the main channel is increased. At 100 kHz one might expect to see a figure inexcess of 105 dB or possibly more.Another way of measuring the phase noise response is to inject a large signal into thereceiver and monitor the level needed to give a 3 dB increase in background noise level.Navigation:: Home >> Radio receiver technology >> this pageDirect digital synthesis (DDS)Direct digital synthesis (DDS) is a powerful technique used in the generation of radiofrequency signals for use in a variety of applications from radio receivers to signalsgenerators and many more. The technique has become far more widespread in recentyears with the advances being made in integrated circuit technology that allow muchfaster speeds to be handled which in turn enable higher frequency DDS chips to be made.Although often used on its own, Direct Digital Synthesis is often used in conjunctionwith indirect or phase locked loop synthesizer loops. By combining both technologies itis possible to take advantage of the best aspects of each. In view of the fact that integratedcircuits are now widely available, this makes them easy to use.How it worksAs the name suggests this form of synthesis generates the waveform directly using digitaltechniques. This is different to the way in which the more familiar indirect synthesizersthat use a phase locked loop as the basis of their operation.A direct digital synthesizer operates by storing the points of a waveform in digital format,and then recalling them to generate the waveform. The rate at which the synthesizercompletes one waveform then governs the frequency. The overall block diagram isshown below, but before looking at the details operation of the synthesizer it is necessaryto look at the basic concept behind the system.The operation can be envisaged more easily by looking at the way that phase progressesover the course of one cycle of the waveform. This can be envisaged as the phaseprogressing around a circle. As the phase advances around the circle, this corresponds toadvances in the waveform.Block Diagram of a Basic Direct Digital Synthesizer (DDS).The synthesizer operates by storing various points in the waveform in digital form andthen recalling them to generate the waveform. Its operation can be explained in more

detail by considering the phase advances around a circle as shown in Figure 2. As thephase advances around the circle this corresponds to advances in the waveform, i.e. thegreater the number corresponding to the phase, the greater the point is along thewaveform. By successively advancing the number corresponding to the phase it ispossible to move further along the waveform cycle.The digital number representing the phase is held in the phase accumulator. The numberheld here corresponds to the phase and is increased at regular intervals. In this way it canbe sent hat the phase accumulator is basically a form of counter. When it is clocked itadds a preset number to the one already held. When it fills up, it resets and starts countingfrom zero again. In other words this corresponds to reaching one complete circle on thephase diagram and restarting again.Operation of the phase accumulator in a direct digital synthesizer.Once the phase has been determined it is necessary to convert this into a digitalrepresentation of the waveform. This is accomplished using a waveform map. This is amemory which stores a number corresponding to the voltage required for each value ofphase on the waveform. In the case of a synthesizer of this nature it is a sine look up tableas a sine wave is required. In most cases the memory is either a read only memory(ROM) or programmable read only memory (PROM). This contains a vast number ofpoints on the waveform, very many more than are accessed each cycle. A very largenumber of points is required so that the phase accumulator can increment by a certainnumber of points to set the required frequency.The next stage in the process is to convert the digital numbers coming from the sine lookup table into an analogue voltage. This is achieved using a digital to analogue converter(DAC). This signal is filtered to remove any unwanted signals and amplified to give therequired level as necessary.Tuning is accomplished by increasing or decreasing the size of the step or phaseincrement between different sample points. A larger increment at each update to thephase accumulator will mean that the phase reaches the full cycle value faster and thefrequency is correspondingly high. Smaller increments to the phase accumulator valuemeans that it takes longer to increase the full cycle value and a correspondingly low valueof frequency. In this way it is possible to control the frequency. It can also be seen thatfrequency changes can be made instantly by simply changing the increment value. Thereis no need to a settling time as in the case of phase locked loop based synthesizer.From this it can be seen that there is a finite difference between one frequency and thenext, and that the minimum frequency difference or frequency resolution is determinedby the total number of points available in the phase accumulator. A 24 bit phaseaccumulator provides just over 16 million points and gives a frequency resolution ofabout 0.25 Hz when used with a 5 MHz clock. This is more than adequate for mostpurposes.These synthesizers do have some disadvantages. There are a number of spurious signalswhich are generated by a direct digital synthesizer. The most important of these is onecalled an alias signal. Here images of the signal are generated on either side of the clockfrequency and its multiples. For example if the required signal had a frequency of 3 MHzand the clock was at 10 MHz then alias signals would appear at 7 MHz and 13 MHz as

well as 17 MHz and 23 MHz etc.. These can be removed by the use of a low pass filter.Also some low level spurious signals are produced close in to the required signal. Theseare normally acceptable in level, although for some applications they can cause problems.Navigation:: Home >> Radio receiver technology >> this pageGraphical method for designing a PLL frequencysynthesizer to meet a phase noise specification- a simple graphical and understandable approach to understanding wherephase noise is generated within a PLL frequency synthesizer and designingit to meet a requirementPhase noise in PLL frequency synthesizers if of great importance because it determinesmany factors about the equipment into which it is incorporated. For receivers itdetermines the reciprocal mixing performance, and in some circumstances the bit errorrate. In transmitters the phase noise performance of the frequency synthesizer determinesfeatures such as adjacent channel noise and it contributes to the bit error rate for thewhole system.Phase noise in a synthesizer loopPhase noise is generated at different points around the synthesizer loop and dependingupon where it is generated it affects the output in different ways. For example, noisegenerated by the VCO has a different effect to that generated by the phase detector. Thisillustrates that it is necessary to look at the noise performance of each circuit block in theloop when designing the synthesizer so that the best noise performance is obtained.Apart from ensuring that the noise from each part of the circuit is reduced to an absoluteminimum, it is the loop filter which has the most effect on the final performance of thecircuit because it determines the break frequencies where noise from different parts of thecircuit start to affect the output.To see how this happens take the example of noise from the VCO. Noise from theoscillator is divided by the divider chain and appears at the phase detector. Here itappears as small perturbations in the phase of the signal and emerges at the output of thephase detector. When it comes to the loop filter only those frequencies which are belowits cut-off point appear at the control terminal of the VCO to correct or eliminate thenoise. From this it can be seen that VCO noise which is within the loop bandwidth isattenuated, but that which is outside the loop bandwidth is left unchanged.The situation is slightly different for noise generated by the reference. This enters thephase detector and again passes through it to the loop filter where the components belowthe cut-off frequency are allowed through and appear on the control terminal of the VCO.Here they add noise to the output signal. So it can be seen that noise from the reference isadded to the output signal within the loop bandwidth but it is attenuated outside this.Similar arguments can be applied to all the other circuit blocks within the loop. Inpractice the only other block which normally has any major effect is the phase detectorand its noise affects the loop in exactly the same way as noise from the reference. Also ifmulti-loop synthesizers are used then the same arguments can be used again.Effects of multiplicationAs noise is generated at different points around the loop it is necessary to discover whateffect this has on the output. As a result it is necessary to relate all the effects back to theVCO. Apart from the different elements in the loop affecting the noise at the output indifferent ways, the effect of the multiplication in the loop also has an effect.The effect of multiplication is very important. It is found that the level of phase noisefrom some areas is increased in line with the multiplication factor (i.e. the ratio of the

final output frequency to the phase comparison frequency). In fact it is increased by afactor of 20 log10 N where N is the multiplication factor. The VCO is unaffected by this,but any noise from the reference and phase detector undergoes this amount ofdegradation. Even very good reference signals can be a major source of noise if themultiplication factor is high. For example a loop which has a divider set to 200 willmultiply the noise of the reference and phase detector by 46 dB.From this information it is possible to build up a picture of the performance of thesynthesizer. Generally this will look like the outline shown in Fig. 6. From this it can beseen that the noise inside the loop bandwidth is due mainly to components like the phasedetector and reference, whilst outside the loop the VCO generates the noise. A slighthump is generally seen at the point where the loop filter cuts off and the loop gain falls tounity.By predicting the performance of the loop it is possible to optimise the performance orlook at areas which can be addressed to improve the performance of the wholesynthesizer before the loop is even built. In order to analyse the loop further it isnecessary to look at each circuit block in turn.Voltage controlled oscillatorThe noise performance of the oscillator is of particular importance. This is because thenoise performance of the synthesizer outside the loop is totally governed by itsperformance. In addition to this its performance may influence decisions about otherareas of the circuit.The typical noise outline for a VCO is flat at large frequency offsets from the carrier. It isdetermined largely by factors such as the noise figure of the active device. Theperformance of this area of the oscillator operation can be optimised by ensuring thecircuit is running under the optimum noise performance conditions. Another approach isto increase the power level of the circuit so that the signal to noise ratio improves.Closer in the noise starts to rise, initially at a rate of 20 dB per decade. The point at whichthis starts to rise is determined mainly by the Q of the oscillator circuit. A high Q circuitwill ensure a good noise performance. Unfortunately VCOs have an inherently low Qbecause of the Q of the tuning varactors normally employed. Performance can beimproved by increasing the Q, but this often results in the coverage of the oscillator beingreduced.Still further in towards the carrier the noise level starts to rise even faster at a rate of 30dB per decade. This results from flicker or 1/f noise. This can be improved by increasingthe level of low frequency feedback in the oscillator circuit. In a standard bipolar circuit asmall un-bypassed resistor in the emitter circuit can give significant improvements.To be able to assess the performance of the whole loop it is necessary to assess theperformance of the oscillator once it has been designed and optimised. Whilst there are anumber of methods of achieving this the most successful is generally to place theoscillator into a loop having a narrow bandwidth and then measure its performance with aspectrum analyser. By holding the oscillator steady this can be achieved relatively easily.However the results are only valid outside the loop bandwidth. However a test loop islikely to have a much narrower bandwidth than the loop being designed the noise levelsin the area of interest will be unaltered.ReferenceThe noise performance of the reference follows the same outlines as those for the VCO,but the performance is naturally far better. The reason for this is that the Q of the crystalis many orders of magnitude higher than that of the tuned circuit in the VCO.Typically it is possible to achieve figures of -110 dBc/Hz at 10 Hz from the carrier and140 dBc/Hz at 1 kHz from a crystal oven. Figures of this order are quite satisfactory formost applications. If lower levels of reference noise are required these can be obtain, butat a cost. In instances where large multiplication factors are necessary a low noisereference may be the only option. However as a result of the cost they should be avoided

wherever possible. Plots of typical levels of phase noise are often available with crystalovens giving an accurate guide to the level of phase noise generated by the reference.Frequency dividerDivider noise appears within the loop bandwidth. Fortunately the levels of noisegenerated within the divider are normally quite low. If an analysis is required then it willbe found that noise is generated at different points within the divider each of which willbe subject to a different multiplication factor dependent upon where in the divider it isgenerated and the division ratio employed from that point.Most divider chains use several dividers and if an approximate analysis is to beperformed it may be more convenient to only consider the last device or devices in thechain as these will contribute most to the noise. However the noise is generally difficultto measure and will be seen with that generated by the phase detector.Phase detectorLike the reference signal the phase detector performance is crucial in determining thenoise performance within the loop bandwidth. There are a number of different types ofphase detector. The two main categories are analogue and digital.Mixers are used to give analogue phase detectors. If the output signal to noise ratio is tobe as good as possible then it is necessary to ensure that the input signal levels are as highas possible within the operating limits of the mixer. Typically the signal input may belimited to around -10 dBM and the local oscillator input to +10 dBm. In some instanceshigher level mixers may be used with local oscillator levels of +17 dBm or higher. Themixer should also be chosen to have a low NTR (noise temperature ratio). As the outputis DC coupled it is necessary to have a low output load resistance to prevent a backwardbias developing. This could offset the operation of the mixer and reduce its noiseperformance.It is possible to calculate the theoretical noise performance of the mixer under optimumconditions. An analogue mixer is likely to give a noise level of around -153 dBc/Hz.There are a variety of digital phase detectors which can be used. In theory these give abetter noise performance than the analogue counterpart. At best a simple OR gate typewill give figures about 10 dB better than an analogue detector and an edge triggered type(e.g. a dual D type or similar) will give a performance of around 5 dB better than theanalogue detector.These figures are the theoretical optimum and should be treated as guide although theyare sufficient for initial noise estimates. In practice other factors may mean that thefigures are different. A variety of factors including power supply noise, circuit layout etc.can degrade the performance from the ideal. If very accurate measurements are requiredthen results from the previous use of the circuit, or from a special test loop can providethe required results.Loop filterThere are a variety of parameters within the area of the loop filter which affect the noiseperformance of the loop. The break points of the filter and the unity gain point of the loopdetermined by the filter govern the noise profile.In terms of contributions to the noise in the loop the major source is likely to occur if anoperational amplifier is used. If this is the case a low noise variety must be usedotherwise the filter will give a large contribution to the loop phase noise profile. Thisnoise is often viewed as combined with that from the phase detector, appearing todegrade its performance from the ideal.Plotting PerformanceHaving investigated the noise components from each element in the loop, it is possible toconstruct a picture of how the whole loop will perform. Whilst this can performedmathematically, a simple graphical approach quickly reveals an estimate of theperformance and shows which are the major elements which contribute to the noise. In

this way some re-design can be undertaken before the design is constructed, enabling thebest option to be chosen on the drawing board. Naturally it is likely to need someoptimisation once it has been built, but this method enables the design to be made asclose as possible beforehand.First it is necessary to obtain the loop response. This is dependent upon a variety offactors including the gain around the loop and the loop filter response. For stability theloop gain must fall at a rate of 20 dB per decade (6 dB per octave) at the unity gain point.Provided this criterion is met a wide variety of filters can be used. Often it is useful tohave the loop response rise at a greater rate than this inside the loop bandwidth. By doingthis the VCO noise can be attenuated further. Outside the loop bandwidth a greater falloff rate can aid suppress the unwanted reference sidebands further. From a knowledge ofthe loop filter chosen the break points can be calculated and with a knowledge of the loopgain the total loop response can be plotted.With the response known the components from the individual blocks in the loop can beadded as they will be affected by the loop and seen at the output.First take the VCO. Outside the loop bandwidth its noise characteristic is unmodified.However once inside this point the action of the loop attenuates the noise, first at a rate of20 dB per decade, and then at a rate of 40 dB per decade. The overall affect of this is tomodify the response of the characteristic as shown in Fig. 10. It is seen that outside theloop bandwidth the noise profile is left unmodified. Far out the noise is flat, but further inthe VCO noise rises at the rate of 20 dB per decade. Inside the loop bandwidth the VCOnoise will be attenuated first at the rate of 20 dB per decade, which in this case gives aflat noise profile. Then as the loop gain increases at the filter break point, to 40 dB perdecade this gives a fall in the VCO noise profile of -20 dB per decade. However furtherin the profile of the stand alone VCO noise rises to -30dB per decade. The action of theloop gives an overall fall of -10 dB per decade.The effects of the other significant contributions can be calculated. The referenceresponse can easily be deduced from the manufacturers figures. Once obtained these musthave the effect of the loop multiplication factor added. Once this has been calculated theeffect of the loop can be added. Inside the loop there is no effect on the noisecharacteristic, however outside this frequency it will attenuate the reference noise, first ata rate of 20 dB per decade and then after the filter break point at 40 dB per decade.The other major contributor to the loop noise is the phase detector. The effect of this istreated in the same way as the reference, having the effect of the loop multiplicationadded and then being attenuated outside the loop bandwidth.Once all the individual curves have been generated they can be combined onto a singleplot to gain a full picture of the performance of the synthesizer. When doing this it shouldbe remembered that it is necessary to produce the RMS sum of the components becausethe noise sources are not correlated.Once this has been done then it is possible to optimise the performance by changingfactors like the loop bandwidth, multiplication factor and possibly the loop topology toobtain the best performance and ensure that the required specifications are met. In mostcases the loop bandwidth is chosen so that a relatively smooth transition is made betweenthe noise contributions inside and outside the loop. This normally corresponds to lowestoverall noise situation.SummaryAlthough this approach may appear to be slightly "low tech" in today's highlycomputerised engineering environment it has the advantage that a visual plot of thepredicted performance can be easily put together. In this way the problem areas can bequickly identified, and the noise performance of the whole synthesizer optimised beforethe final design is committed.Navigation:: Home >> Radio receiver technology >> this page

Radio receiver amplitude modulation (AM)demodulation- using a simple diode detector ( demodulator )One of the advantages of amplitude modulation (AM) is that it is cheap and easy to builda demodulator circuit for a radio receiver. The simplicity AM radio receivers AM is oneof the reasons why AM has remained in service for broadcasting for so long. One of thekey factors of this is the simplicity of the receiver AM demodulator.A number of methods can be used to demodulate AM, but the simplest is a diodedetector. It operates by detecting the envelope of the incoming signal. It achieves this bysimply rectifying the signal. Current is allowed to flow through the diode in only onedirection, giving either the positive or negative half of the envelope at the output. If thedetector is to be used only for detection it does not matter which half of the envelope isused, either will work equally well. Only when the detector is also used to supply theautomatic gain control (AGC) circuitry will the polarity of the diode matter.The AM detector or demodulator includes a capacitor at the output. Its purpose is toremove any radio frequency components of the signal at the output. The value is chosenso that it does not affect the audio base-band signal. There is also a leakage path to enablethe capacitor to discharge, but this may be provided by the circuit into which thedemodulator is connected.A simple diode detector (demodulator) for AM signalsThis type of detector or demodulator is called a linear envelope detector because theoutput is proportional to the input envelope. Unfortunately the diodes used can introduceappreciable levels of harmonic distortion unless modulation levels are kept low. As aresult these detectors can never provide a signal suitable for high quality applications.Additionally these detectors ( demodulators ) are susceptible to the effects of selectivefading experienced on short wave broadcast transmissions. Here the ionosphericpropagation may be such that certain small bands of the signal are removed. Undernormal circumstances signals received via the ionosphere reach the receiver via a numberof different paths. The overall signal is a combination of the signals received via eachpath and as a result they will combine with each other, sometimes constructively toincrease the overall signal level and sometimes destructively to reduce it. It is found thatwhen the path lengths are considerably different this combination process can mean thatsmall portions of the signal are reduced in strength. An AM signal consists of a carrierwith two sidebands.

Spectrum of an amplitude modulated (AM) signalIf the section of the signal that is removed falls in one of the sidebands, it will change thetone of the received signal. However if carrier is removed or even reduced in strength, thesignal will appear to be over modulated, and severe distortion will result. This is acomparatively common occurrence on the short waves, and means that diode detectorsare not suitable for high quality reception. Synchronous demodulation ( detection ) is farsuperior.Navigation:: Home >> Radio receiver technology >> this pageSynchronous demodulation / detectionToday's radio receivers offer very high levels of performance and boast many facilities.Many radio receivers incorporate memories, phase locked loops, direct digital synthesis,digital signal processing and much more. One facility that can be very useful on the shortwave bands is synchronous detection or synchronous demodulation as this can give muchimproved performance for receiving amplitude modulation (AM) transmissions.Unfortunately little is written about this form of modulation, and often it is a matter ofaccepting that it must be better than any normal options because it is included as a featurein the receiver specification.Synchronous detection is used for the detection or demodulation of amplitude modulation(AM). This form of modulation is still widely used for broadcasting on the long, mediumand short wave bands despite the fact that there are more efficient forms of modulationthat can be used today. The main reason for its use nowadays is that it is very wellestablished, and there are many millions of AM receivers around the world today.In any receiver a key element is the detector. Its purpose is to remove the modulationfrom the carrier to give the audio frequency representation of the signal. This can beamplified by the audio amplifier ready to be converted into audible sound by headphonesor a loudspeaker. Many receivers still use what is termed an envelope detector using asemiconductor diode for demodulating AM. These detectors have a number ofdisadvantages. The main one is that they are not particularly linear and distortion levelsmay be high. Additionally their noise performance is not particularly good at low signallevels.These detectors also do not perform very well when the signal undergoes selective fadingas often occurs on the short wave bands. An AM signal contains two sidebands and thecarrier. For the signal to be demodulated correctly the carrier should be present at therequired level. It can be seen that the signal covers a definite bandwidth, and the effectsof fading may result in the carrier and possibly one of the sidebands being reduced inlevel. If this occurs then the received signal appears to be over-modulated with the resultthat distortion occurs in the demodulation process.

The spectrum of an amplitude modulated signalDiode envelope detectorIn virtually every receiver a simple diode envelope detector is used. These circuits havethe advantage that they are very simple and give adequate performance in manyapplications.The circuit of a typical detector is shown in Figure 2. Here the diode first rectifies thesignal to leave only the positive or negative going side of the signal, and then a capacitorremoves any of the remaining radio frequency components to leave the demodulatedaudio signal. Unfortunately diodes are not totally linear and this is the cause of thedistortion.An envelope detector for AM signalsWhat is synchronous demodulationSignals can be demodulated using a system known as synchronous detection ordemodulation. This is far superior to diode or envelope detection, but requires morecircuitry. Here a signal on exactly the same frequency as the carrier is mixed with theincoming signal as shown in Figure 2. This has the effect of converting the frequency ofthe signal directly down to audio frequencies where the sidebands appear as the requiredaudio signals in the audio frequency band.The crucial part of the synchronous detector is in the production a local oscillator signalon exactly the same frequency as the carrier. Although it is possible to receive an AMsignal without the local oscillator frequency on exactly the same frequency as the carrierthis is the same as using the BFO in a receiver to resolve the signal. If the BFO is notexactly on the same frequency as the carrier then the resultant audio is not very good.Synchronous demodulation

Fortunately this is not too difficult to achieve and although there are a number of ways ofachieving this the most commonly used method is to pass some of the signal into a highgain limiting amplifier. The gain of the amplifier is such that it limits, and therebyremoving all the modulation. This leaves a signal consisting only of the carrier and thiscan be used as the local oscillator signal in the mixer as shown in Fig. 4. This is mostconvenient, cheapest and certainly the most elegant method of producing synchronousdemodulation.A synchronous detector using a high gain-limiting amplifier to extract the carrierAdvantages of synchronous detectionA synchronous detector is more expensive to make than an ordinary diode detector whendiscrete components are used, although with integrated circuits being found in manyreceivers today there is little or no noticeable cost associated with its use as the circuitryis often included as part of an overall receiver IC.Synchronous detectors are used because they have several advantages over ordinarydiode detectors. Firstly the level of distortion is less. This can be an advantage if a betterlevel of quality is required but for many communications receivers this might not be aproblem. Instead the main advantages lie in their ability to improve reception underadverse conditions, especially when selective fading occurs or when signal levels are low.Under conditions when the carrier level is reduced by selective fading, the receiver isable to re-insert its own signal on the carrier frequency ensuring that the effects ofselective fading are removed. As a result the effects of selective fading can be removed togreatly enhance reception.The other advantage is an improved signal to noise ratio at low signal levels. As thedemodulator is what is termed a coherent modulator it only sees the components of noisethat are in phase with the local oscillator. Consequently the noise level is reduced and thesignal to noise ratio is improved.Unfortunately synchronous detectors are only used in a limited number of receiversbecause of their increased complexity. Where they are used a noticeable improvement inreceiver performance is seen and when choosing a receiver that will be used for shortwave broadcast reception it is worth considering whether a synchronous detector is oneof the facilities that is required.Frequency modulation is widely used in radio communications and broadcasting,particularly on frequencies above 30 MHz. It offers many advantages, particularly inmobile radio applications where its resistance to fading and interference is a greatadvantage. It is also widely used for broadcasting on VHF frequencies where it is able toprovide a medium for high quality audio transmissions.In view of its widespread use receivers need to be able to demodulate thesetransmissions. There is a wide variety of different techniques and circuits that can be suedincluding the Foster-Seeley, and ratio detectors using discreet components, and whereintegrated circuits are used the phase locked loop and quadrature detectors are morewidely used.

What is FM?As the name suggests frequency modulation uses changes in frequency to carry the soundor other information that is required to be placed onto the carrier. As shown in Figure 1 itcan be seen that as the modulating or base band signal voltage varies, so the frequency ofthe signal changes in line with it. This type of modulation brings several advantages withit. The first is associated with interference reduction. Much interference appears in theform of amplitude variations and it is quite easy to make FM receivers insensitive toamplitude variations and accordingly this brings about a reduction in the levels ofinterference. In a similar way fading and other strength variations in the signal have littleeffect. This can be particularly useful for mobile applications where charges in locationas the vehicle moves can bring about significant signal strength changes. A furtheradvantage of FM is that the RF amplifiers in transmitters do not need to be linear. Whenusing amplitude modulation or its derivatives, any amplifier after the modulator must belinear otherwise distortion is introduced. For FM more efficient class C amplifiers may beused as the level of the signal remains constant and only the frequency varies.Frequency modulating a signalWide band and Narrow bandWhen a signal is frequency modulated, the carrier shifts in frequency in line with themodulation. This is called the deviation. In the same way that the modulation level can bevaried for an amplitude modulated signal, the same is true for a frequency modulated one,although there is not a maximum or 100% modulation level as in the case of AM.The level of modulation is governed by a number of factors. The bandwidth that isavailable is one. It is also found that signals with a large deviation are able to supporthigher quality transmissions although they naturally occupy a greater bandwidth. As aresult of these conflicting requirements different levels of deviation are used according tothe application that is used.Those with low levels of deviation are called narrow band frequency modulation(NBFM) and typically levels of +/- 3 kHz or more are used dependent upon thebandwidth available. Generally NBFM is used for point to point communications. Muchhigher levels of deviation are used for broadcasting. This is called wide band FM(WBFM) and for broadcasting deviation of +/- 75 kHz is used.Receiving FMIn order to be able to receive FM a receiver must be sensitive to the frequency variationsof the incoming signals. As already mentioned these may be wide or narrow band.However the set is made insensitive to the amplitude variations. This is achieved byhaving a high gain IF amplifier. Here the signals are amplified to such a degree that theamplifier runs into limiting. In this way any amplitude variations are removed.

In order to be able to convert the frequency variations into voltage variations, thedemodulator must be frequency dependent. The ideal response is a perfectly linearvoltage to frequency characteristic. Here it can be seen that the centre frequency is in themiddle of the response curve and this is where the un-modulated carrier would be locatedwhen the receiver is correctly tuned into the signal. In other words there would be nooffset DC voltage present.The ideal response is not achievable because all systems have a finite bandwidth and as aresult a response curve known as an "S" curve is obtained. Outside the badwidth of thesystem, the response falls, as would be expected. It can be seen that the frequencyvariations of the signal are converted into voltage variations which can be amplified byan audio amplifier before being passed into headphones, a loudspeaker, or passed intoother electronic circuitry for the appropriate processing.Characteristic "S" curve of an FM demodulatorTo enable the best detection to take place the signal should be centred about the middle ofthe curve. If it moves off too far then the characteristic becomes less linear and higherlevels of distortion result. Often the linear region is designed to extend well beyond thebandwidth of a signal so that this does not occur. In this way the optimum linearity isachieved. Typically the bandwidth of a circuit for receiving VHF FM broadcasts may beabout 1 MHz whereas the signal is only 200 kHz wide.DemodulatorsThere are a number of circuits that can be used to demodulate FM. Each type has its ownadvantages and disadvantages, some being used when receivers used discretecomponents, and others now that ICs are widely used.Slope detectionThe very simplest form of FM demodulation is known as slope detection ordemodulation. It simply uses a tuned circuit that is tuned to a frequency slightly offsetfrom the carrier of the signal. As the frequency of the signal varies up and down infrequency according to its modulation, so the signal moves up and down the slope of thetuned circuit. This causes the amplitude of the signal to vary in line with the frequencyvariations. In fact at this point the signal has both frequency and amplitude variations.The final stage in the process is to demodulate the amplitude modulation and this can beachieved using a simple diode circuit. One of the most obvious disadvantages of thissimple approach is the fact that both amplitude and frequency variations in the incomingsignal appear at the output. However the amplitude variations can be removed by placinga limiter before the detector. Additionally the circuit is not particularly efficient as itoperates down the slope of the tuned circuit. It is also unlikely to be particularly linear,especially if it is operated close to the resonant point to minimise the signal loss.Ratio and Foster-Seeley detectorsWhen circuits employing discrete components were more widely sued, the Ratio andFoster-Seeley detectors were widely used. Of these the ratio detector was the mostpopular as it offers a better level of amplitude modulation rejection of amplitude

modulation. This enables it to provide a greater level of noise immunity as most noise isamplitude noise, and it also enables the circuit to operate satisfactorily with lower levelsof limiting in the preceding IF stages of the receiver.The operation of the ratio detector centres around a frequency sensitive phase shiftnetwork with a transformer and the diodes that are effectively in series with one another.When a steady carrier is applied to the circuit the diodes act to produce a steady voltageacross the resistors R1 and R2, and the capacitor C3 charges up as a result.The transformer enables the circuit to detect changes in the frequency of the incomingsignal. It has three windings. The primary and secondary act in the normal way toproduce a signal at the output. The third winding is un-tuned and the coupling betweenthe primary and the third winding is very tight, and this means that the phasing betweensignals in these two windings is the same.The primary and secondary windings are tuned and lightly coupled. This means that thereis a phase difference of 90 degrees between the signals in these windings at the centrefrequency. If the signal moves away from the centre frequency the phase difference willchange. In turn the phase difference between the secondary and third windings alsovaries. When this occurs the voltage will subtract from one side of the secondary and addto the other causing an imbalance across the resistors R1 and R2. As a result this causes acurrent to flow in the third winding and the modulation to appear at the output.The capacitors C1 and C2 filter any remaining RF signal which may appear across theresistors. The capacitor C4 and R3 also act as filters ensuring no RF reaches the audiosection of the receiver.The ratio detectorThe Foster Seeley detector has many similarities to the ratio detector. The circuittopology looks very similar, having a transformer and a pair of diodes, but there is nothird winding and instead a choke is used.

The Foster-Seeley detectorLike the ratio detector, the Foster-Seeley circuit operates using a phase differencebetween signals. To obtain the different phased signals a connection is made to theprimary side of the transformer using a capacitor, and this is taken to the centre tap of thetransformer. This gives a signal that is 90 degrees out of phase.When an un-modulated carrier is applied at the centre frequency, both diodes conduct, toproduce equal and opposite voltages across their respective load resistors. These voltagescancel each one another out at the output so that no voltage is present. As the carriermoves off to one side of the centre frequency the balance condition is destroyed, and onediode conducts more than the other. This results in the voltage across one of the resistorsbeing larger than the other, and a resulting voltage at the output corresponding to themodulation on the incoming signal.The choke is required in the circuit to ensure that no RF signals appear at the output. Thecapacitors C1 and C2 provide a similar filtering function.Both the ratio and Foster-Seeley detectors are expensive to manufacture. Woundcomponents like coils are not easy to produce to the required specification and thereforethey are comparatively costly. Accordingly these circuits are rarely used in modernequipment.Quadrature detectorAnother form of FM detector or demodulator that can be these days is called thequadrature detector. It lends itself to use with integrated circuits and as a result it is inwidespread use. It has the advantage over the ratio and Foster-Seeley detectors that itonly requires a simple tuned circuit.For the quadrature detector, the signal is split into two components. One passes through anetwork that provides a basic 90 degree phase shift, plus an element of phase shiftdependent upon the deviation and into one port of a mixer. The other is passed straightinto another port of the mixer. The output from the mixer is proportional to the phasedifference between the two signals, i.e. it acts as a phase detector and produces a voltageoutput that is proportional to the phase difference and hence to the level of deviation onthe signal.The detector is able to operate with relatively low input levels, typically down to levels ofaround 100 microvolts and it is very easy to set up requiring only the phase shift networkto be tuned to the centre frequency of the expected signal. It also provides good linearityenabling very low levels of distortion to be achieved.Often the analogue multiplier is replaced by a logic AND gate. The input signal is hardlimited to produce a variable frequency pulse waveform. The operation of the circuit isfundamentally the same, but it is known as a coincidence detector. Also the output of the

AND gate has an integrator to "average" the output waveform to provide the requiredaudio output, otherwise it would consist of a series of square wave pulses.Phase locked loop (PLL)Another popular form of FM demodulator comes in the form of a phase locked loop. Likethe quadrature detector, phase locked loops do not need to use a coil, and therefore theymake a very cost effective form of demodulator.The way in which they operate is very simple. The loop consists of a phase detector intowhich the incoming signal is passed, along with the output from the voltage controlledoscillator (VCO) contained within the phase locked loop. The output from the phasedetector is passed into a loop filter and then sued as the control voltage for the VCO.Phase locked loop (PLL) FM demodulatorWith no modulation applied and the carrier in the centre position of the pass-band thevoltage on the tune line to the VCO is set to the mid position. However if the carrierdeviates in frequency, the loop will try to keep the loop in lock. For this to happen theVCO frequency must follow the incoming signal, and for this to occur the tune linevoltage must vary. Monitoring the tune line shows that the variations in voltagecorrespond to the modulation applied to the signal. By amplifying the variations involtage on the tune line it is possible to generate the demodulated signal.It is found that the linearity of this type of detector is governed by the voltage tofrequency characteristic of the VCO. As it normally only swings over a small portion ofits bandwidth, and the characteristic can be made relatively linear, the distortion levelsfrom phase locked loop demodulators are normally very low.Navigation:: Home >> Radio receiver technology >> this pageRadio receiver filter options- including LC filter, crystal filter, mechanical filter, ceramic filter, and roofingfilterThere is a wide variety of different types of filter used within superhet radios. Someradios will simply use filters made up from the tuned transformers (LC filters based oncapacitors and inductors) linking the different intermediate frequency stages within theradios or used with an IC in the radio. Other radio receivers may incorporate highlyselective crystal filters, whereas others may use mechanical filters (like those used by theCollins Radio Company some years ago) or ceramic filters. Each radio will have its ownrequirements, and the choice of filter to meet its needs of performance and cost.LC tuned circuitsThe simplest type of filter is an ordinary L-C tuned circuit. In many older radio usingdiscrete semiconductors, or older radios using vacuum tubes they take the form oftransformers to couple the individual stages in an IF amplifier chain. Often there are twoor three stages with tuned circuits. Using them it is usually possible to achieve sufficientselectivity for a medium wave AM or VHF FM broadcast radio. However for a goodquality communications receiver it is rarely possible to be able to achieve the requireddegree of selectivity using just L-C filters.

In more modern radios using integrated circuits a single tuned circuit could be used inconjunction with an integrated, as the concept of inter-stage coupling is not employed inthe same manner. Typically a ceramic filter, rather than an LC circuit is more likely to beused.If L-C filters were used in a radio using interstage transformers then it would be possibleto increase the degree of selectivity by increasing the number of tuned circuits betweeneach stage. This is not ideal for a number of reasons. In the first case it increases thedifficulty of aligning the set. In addition to this each tuned circuit will introduce a certainamount of loss. Increasing the number of tuned circuits will increase the amount of gainrequired, sometimes necessitating a further stage of gain. A further disadvantage is that itis not easy to alter the degree of selectivity by switching in additional L-C filters. If this isto be achieved then it is often preferable to switch in a further type of filter such as acrystal filter.Crystal FiltersCrystal filters provide the main selectivity in of most of today's high performance sets.They provided exceedingly high degrees of selectivity which are hard to equal in terms ofperformance and cost.The crystals in the filters are made from a substance called quartz. This is basically aform of crystalline silicon. Originally natural deposits were used to manufacture thecrystals required for the electronics industry. Now quartz crystals are grown syntheticallyunder controlled conditions to produce very high quality material.The crystals use the piezo-electric effect for their operation. This effect occurs in anumber of substances and it converts a mechanical stress into a voltage and vice versa.Many electrical transducers use the effect converting electrical impulses or signals intomechnical vibrations and vice versa.In quartz crystal resonators the piezo-electric effect is used in conjunction with themechanical resonances which occur in the substance. The electrical signals passing intothe crystal are converted into mechanical vibrations which interact with the resonances ofthe crystal. In this way the crystal uses the piezo-electric effect to enable the mechanicalresonances to tune the electrical signals. These mechanical resonances have exceedinglyhigh Q factors. Many crystals will exhibit values of several thousand. This is many ordersof magnitude higher than ordinary tuned filters made from conventional inductors andcapacitors where values of a hundred or so are considered high. Typically the Q of an LCtuned circuit may be reach values of a few hundred. For quartz crystals values of Q mayexceed 100 000.Further details about quartz, its properties and the ways in which crystals aremanufactured and used can be found on the Electronic components section of this site -see side menu for the link.The response of a single crystal is too narrow for many applications. Normally a filter isrequired to have a passband, possibly of a few hundred Hertz, or a few kilohertz, andoutside this bandwidth, other signals should be totally rejected. While it is not possible toachieve the perfect filter very high degrees of selectivity can be achieved. By addingseveral crystals together it is possible to obtain the performance that is required. Oftencrystal filters are referred to as having a certain number of poles. This terminology comesfrom the filter analysis design process, but effectively there is one crystal in the filter forevery pole.A two pole filter (i.e. one with two crystals) is not normally adequate to meet manyrequirements. The shape factor which is the ratio between the bandwidth where thestopband attenuation starts and the bandwidth of the passband) can be greatly improvedby adding further sections. Typically ultimate rejections of 70 dB and more are requiredin a receiver. As a rough guide a two pole filter will generally give a rejection of around20 dB; a four pole filter, 50 dB; a six pole filter, 70 dB; and an eight pole one 90 dB.

Monolithic filtersWith more items being integrated onto single chips these days it is hardly surprising tofind that a similar approach is being adopted for crystal filters. Instead of having severalseparate or discrete crystals in a filter, even if they are all contained in the same can, it ispossible to put a complete filter onto a single quartz crystal, hence the name monolithiccrystal filter.In essence the filter is made up by placing two sets of electrodes at opposite sides of asingle AT cut crystal. The coupling between the two electrodes acts in such a way that ahighly selective filter is produced.Monolithic filters have only been available since the 1970s. Even now a large number offilter manufacturers do not produce them, preferring to use the more traditional filtersmade from individual crystals.While it had been known for a long while that a two pole filter could be made up on asingle crystal, the idea was not developed because the way in which it worked was notunderstood. After much work, scientists at Bell Laboratories in the USA discovered itsmode of operation. Very simply it consists of two acoustically coupled resonators.A monolithic crystal filter consists of a crystal blank onto which two sets of electrodes orplates are placed at opposite ends of the blank. Each set consists of an electrode on eitherside of the blank. When the electrical signal is placed across one pair of electrodes, thepiezo-electric effect converts this into mechanical vibrations. These travel across thecrystal to the other electrodes where they are converted back into an electrical signalagain. However if the acoustic signal is to travel across the crystal then its frequencymust match the resonance of the crystal.Often these filters are manufactured for operation below about 30 MHz, because abovethese frequencies the manufacturing costs tend to rise. However manufacturingtechniques are improving all the time it is possible to use them above this. If this isrequired then the normal way of accomplishing this is to use an overtone mode. Thisconsiderably increases the maximum possible frequencies, although the performance isnot usually quite as good.Monolithic filters are used in many areas now. They offer better performance than theirdiscrete counterparts and they can be made smaller - a feature which is becomingincreasingly important in today's miniaturised electronics industry. The main drawback ofthese filters is that they require very specialised equipment for their manufacture.Ceramic filtersQuartz is not the only substance to exhibit the piezo-electric effect combined with a sharpresonance. A number of ceramics are also used successfully to perform this function.Although filters made from these ceramics are not nearly as selective as their higherquality quartz relatives, they are cheaper and offer great improvements over their L-Ccounterparts.Ceramic filters are made from a specialised family of ceramics, and the elements forfilters are normally in the form of a small disc. They operate in exactly the same way ascrystal filters, the signal being linked to the mechanical resonances by the piezo-electriceffect. Generally ceramic filters have a much wider bandwidth and a poorer shape factorthan their crystal counterparts. As a result they are rarely used in high performancecommunications receivers as the primary form of filtering, although their performancehas improved dramatically in recent years and some examples of ceramic filters offeringexceedingly good levels of performance are available. As a result they find widespreaduse in broadcast receivers for AM and VHF FM reception and some wirelessapplications.Mechanical filtersWhen high performance filters are needed there is another type which can be considered.Although not nearly as popular as crystal filters these days, mechanical filters foundwidespread use a number of years ago. The Collins Radio Company (now Rockwell

Collins) was a famous manufacturer of these devices, introducing their first designs in1952, these filters are still manufactured.In essence their operation is very similar to that of a crystal, although the variousfunctions are performed by individual components within the filter. At either end of thefilter assembly there are transducers which convert the signals from their electrical formto mechanical vibrations, and back again at the other end. These vibrations are applied toa series of discs which are mechanically resonant at the required frequency. Each of thesediscs has a Q of which can be about 5000 or more, and they are arranged close to oneanother but not touching to form a long cylinder. A number of coupling rods are attachedto run along the side of the assembly to transfer the vibrations from one section to thenext. By altering the amount of coupling between the sections and the resonance of eachdisc, the response of the overall unit can be tailored to meet the exact requirements.Operation of these mechanical filters is normally confined to frequencies between about50 and 500 kHz. Below these frequencies the discs become too large, whilst at the topend of the range they are too small to manufacture and mount in the filters with anydegree of reliability. Apart from the limited frequency range the other disadvantage isthat the resonant frequency of these filters drifts with temperature. However one of theirmain advantages is that exceedingly narrow bandwidths can be achieved relatively easily,and the low levels of intermodulation distortion they introduce. Additionally the costs ofthese devices have been reduced over the years and the number of resonators that can beused can be between 2 and 12 dependent upon the requirements.Roofing filtersIn many receivers the main filtering occurs only after there have been many stages ofamplification. This means that a strong signal which is outside the pass-band of the mainreceiver filter can cause overloading especially in the early IF stages before the filter.This occurs because the AGC does not see the signal and reduce the gain of the earlierstages to take account of it, or the operator may not be aware of the signal and reduce theRF gain if a control is available.To overcome this problem a wider bandwidth filter is placed early on in the IF stages toreduce the level of any strong off channel signals. The main filtering, however, is stillprovided late on in the receiver by the main full specification filter.Roofing filters are often found in multi-conversion superhet receivers where the mainfilter is found after two or possibly three conversion stages. The roofing filter can beplaced soon after the first mixer to reduce the effects of any strong off-channel signals.SummaryThere is a good selection of filters that can be used in radio receivers. The actual type thatis eventually decided upon a balance of performance, cost and other factors. For manyapplications where the highest levels of performance are not needed, ceramic filtersprovide the ideal solution being very cheap and easy to use while providing levels ofperformance that are quite adequate for many applications. For applications where onlythe highest levels of performance are required, crystal filters are the most commonsolution either as units made from discrete crystals or as monolithic filters. Howevermechanical filters could ebb considered for some applications. These days LC filters arenot widely used because the cost of winding coils is high, and often ceramic filters aremore convenient, cheaper, and offer a better level of performance.Navigation:: Home >> Radio receiver technology >> this pageQuartz crystal band pass filters- including the single crystal filter, half lattice filter and ladder filter for use in radioreceivers

Crystal filters are widely used in many applications including radio receivers. The veryhigh level of Q makes them ideal for use as the primary band pass filter in a radioreceiver. As a result of this there are a number of circuits that have been used to providethe required level of selectivity and performance over the years. These include the singlecrystal filter, the half lattice crystal filter and the ladder filter.Single crystal filterThe simplest crystal filter employs a single crystal. This type of filter was developed inthe 1930s and was used in early receivers dating from before the 1960s but is rarely usedtoday. Although it employs the very high Q of the crystal, its response is asymmetric andit is too narrow for most applications, having a bandwidth of a hundred Hz or less.In the circuit there is a variable capacitor that is used to compensate for the parasiticcapacitance in the crystal. This capacitor was normally included as a front panel control.Diagram of filter using a single quartz crystalHalf lattice crystal filterThis form of band pass filter provided a significant improvement in performance over thesingle. In this configuration the parasitic capacitances of each of the crystals cancel eachother out and enable the circuit to operate satisfactorily. By adopting a slightly differentfrequency for the crystals, a wider bandwidth is obtained. However the slope responseoutside the required pass band falls away quickly, enabling high levels of out of bandrejection to be obtained. Typically the parallel resonant frequency of one crystal isdesigned to be equal to the series resonant frequency of the other.Despite the fact that the half lattice crystal filter can offer a much flatter in-band responsethere is still some ripple. This results from the fact that the two crystals have differentresonant frequencies. The response has a small peak at either side of the centre frequencyand a small dip in the middle. As a rough rule of thumb it is found that the 3 dBbandwidth of the filter is about 1.5 times the frequency difference between the tworesonant frequencies. It is also found that for optimum performance the matching of thefilter is very important. To achieve this, matching resistors are often placed on the inputand output. If the filter is not properly matched then it is found that there will be more inbandripple and the ultimate rejection may not be as good.

Diagram of half lattice crystal filterA two pole filter (i.e. one with two crystals) is not normally adequate to meet manyrequirements. The shape factor can be greatly improved by adding further sections.Typically ultimate rejections of 70 dB and more are required in a receiver. As a roughguide a two pole filter will generally give a rejection of around 20 dB; a four pole filter,50 dB; a six pole filter, 70 dB; and an eight pole one 90 dB.Crystal ladder filterFor many years the half lattice filter was possibly the most popular format used forcrystal filters. More recently the ladder topology has gained considerable acceptance. Inthis form of crystal pass band filter all the resonators have the same frequency, and theinter-resonator coupling is provided by the capacitors placed between the resonators withthe other termination connected to earth.Four pole ladder crystal filterAlthough crystal filters are widely used as the high performance filters in receivers,mechanical filters are another option. Mechanical filters have been used for many yearsand are able to provide excellent service at frequencies up to just under 1 MHz. Thesefilters can offer advantages over crystal filters in some instances being small, very stable,and rugged. In fact they are not subject to deterioration due to continuous exposure toshock and vibration, a factor that can be particularly important in some applications. Afurther advantage is that they offer low levels of intermodulation distortion, a factor thatis often overlooked in many receiver designs.Principles of operationIn essence their operation is very similar to that of a crystal, although the variousfunctions are performed by individual components within the filter. At either end of thefilter assembly there are transducers which convert the signals from their electrical formto mechanical vibrations, and back again at the other end. These vibrations are applied toa series of mechanical resonators which are mechanically resonant at the requiredfrequency. The resonators are mechanically coupled, typically with coupling wires totransfer the vibrations from one section to the next. By altering the amount of couplingbetween the sections and the natural frequency of each resonator, the response of theoverall unit can be tailored to meet the exact requirements.

Types of mechanical filterThere are several types of mechanical filter. The choice of the type depends upon thefrequency in use and the application.The first type is known as the torsional filter. This type of mechanical filter uses rods thatvibrate in torsion. Electrical energy is coupled in by means of a piezoelectric ceramictransducer into torsional motion. These filters are used for frequencies in the range frombelow 100 kHz to just under 1 MHz.Seven-resonator torsional mechanical filterImage Courtesy Rockwell CollinsA second type of filter is known as the bar flexural mode mechanical filter. This is usedfor low frequency designs, typically having a centre frequency between 5 to 100 kHz andwith bandwidths of .2 to 1.5 percent.Bar flexural mode mechanical filterImage Courtesy Rockwell CollinsSummaryAlthough mechanical filters are not used as widely as crystal filters, they can neverthelessoffer an excellent solution in some instances. In view of this they are the ideal solutionfor many applications where high performance filters are required at frequencies belowabout 700 kHz to 1MHz.Navigation:: Home >> Radio receiver technology >> this pageSAW filters- an overview or tutorial about SAW filters used as RF and IF filters.Surface Acoustic Wave (SAW) technology is used in many areas of electronics toprovide resonators for oscillators, filters and transformers. One of the major uses of thesedevices is as SAW filters which find widespread use in radio applications. These SAW

filters provide good performance filtering solutions while offering a cost effectivesolution.SAW filters are widely used in cell phone applications for filtering. Here they provideconsiderable advantages in terms of cost and size, in an environment where these twoaspects are of considerable importance. Additionally their importance in the cellularindustry has meant that considerable amounts of research and development have beenundertaken on SAW filter s in the last 20 years, and their performance has improvedconsiderably in this period.Navigation:: Home >> Radio receiver technology >> this pageDSP - Digital Signal Processing tutorial- an overview or introduction to the basics of Digital Signal Processing, DSP, andhow it can be used in radio receiver technology to improve performance andflexibilityToday, Digital Signal Processing, DSP, is widely used in radio receivers as well as inmany other applications from television, radio transmission, or in fact any applicationswhere signals need to be processed. Today it is not only possible to purchase digitalsignal processor integrated circuits, but also DSP cards for use in computers. Using theseDSP cards it is possible to develop software or just use a PC platform in which to run theDSP card.DSP has many advantages over analogue processing. It is able to provide far better levelsof signal processing than is possible with analogue hardware alone. It is able to performmathematical operations that enable many of the spurious effects of the analoguecomponents to be overcome. In addition to this, it is possible to easily update a digitalsignal processor by downloading new software. Once a basic DSP card has beendeveloped, it is possible to use this hardware design to operate in several differentenvironments, performing different functions, purely by downloading different software.It is also able to provide functions that would not be possible using analogue techniques.For example a complicated signal such as Orthogonal Frequency Division Multiplex(OFDM) which is being used for many transmissions today needs DSP to become viable.Despite this DSP has limitations. It is not able to provide perfect filtering, demodulationand other functions. There are mathematical limitations. In addition to this the processingpower of the DSP card may impose some processing limitations. It is also moreexpensive than many analogue solutions, and thus it may not be cost effective in someapplications. Nevertheless it has many advantages to offer, and with the wide availabilityof cheap DSP hardware and cards, it often provides an attractive solution for many radioapplications.What is DSP?As the name suggests, digital signal processing is the processing of signals in a digitalform. DSP is based upon the fact that it is possible to build up a representation of thesignal in a digital form. This is done by sampling the voltage level at regular timeintervals and converting the voltage level at that instant into a digital number proportionalto the voltage. This process is performed by a circuit called an analogue to digitalconverter, A to D converter or ADC. In order that the ADC is presented with a steadyvoltage whilst it is taking its sample, a sample and hold circuit is used to sample thevoltage just prior to the conversion. Once complete the sample and hold circuit is ready toupdate the voltage again ready for the next conversion. In this way a succession ofsamples is made.

Sampling a waveform for DSPOnce in a digital format the real DSP is able to be undertaken. The digital signalprocessor performs complicated mathematical routines upon the representation of thesignal. However to use the signal it then usually needs to be converted back into ananalogue form where it can be amplified and passed into a loudspeaker or headphones.The circuit that performs this function is not surprisingly called a digital to analogueconverter, D to A converter or DAC.Block diagram of a Digital Signal Processor, DSP)The advantage of DSP, digital signal processing is that once the signals are convertedinto a digital format they can be manipulated mathematically. This gives the advantagethat all the signals can be treated far more exactly, and this enables better filtering,demodulation and general manipulation of the signal. Unfortunately it does not mean thatfilters can be made with infinitely steep sides because there are mathematical limitationsto what can be accomplished.Navigation:: Home >> Radio receiver technology >> this pageFPGAs for DSP Hardware- the advantages and disadvantages of using FPGAs rather than DSPprocessors in the DSP hardware.When designing the hardware system for a DSP application it is necessary to carefullyconsider the approach that will be taken. One of the fundamental decisions involveswhether to use a standard DSP processor, or whether to use an FPGA in the DSPhardware. Each has its own advantages and they need to be carefully balanced at theearliest stages of the design.DSP processorA DSP processor is a specialised processor that is designed specifically for operating

complex mathematically orientated intensive calculations very swiftly. As processingneeds to be undertaken almost in real time, the speed of the processor is one of the mainlimiting performance criteria for the performance of the system For example very steepfilters need more processing than those that are not so steep, etc..While DSP processors, despite their sophistication in terms of processing havelimitations, they also have advantages. One of these is in their cost. They may still beexpensive by some standards, but they are nevertheless cheaper than their counterparts,the FPGA.FPGAs for DSPThe other approach that many adopt is to use an FPGA as the core of the DSP hardware.These devices can be programmed and there are many set cores that can be used toprovide the routines that are required. For example if a filter is required, then it ispossible to tailor circuitry within the FPGA to undertake this. Similarly other functionscan be programmed in on top of the basic processor. In this way the FPGA is able to beprogrammed to provide a highly efficient and tailored solution.The main disadvantage of the FPGA is its cost. FPGAs are more costly that DSPprocessors and therefore performance has to be weighed against cost.SummaryFPGAs and DSP processors provide two very different approaches to the design of DSPhardware systems. Each have their own advantages. There are many high sampling rateapplications that an FPGA does easily, while the DSP could not. Equally, there are manycomplex software problems that the FPGA cannot address.