Friday, November 6, 2015

Crystal Filters and Phasing

       While looking at older receivers for SWL DX purposes, there are a few features that are particularly helpful.  After the usual  " default " requirements of at least one tuned RF amplifier stage and at least two IF amplifier stages comes some means of tightening selectivity.
       When it gets down to it, almost all boat anchor receivers that have two IF stages with a 455 kHz Intermediate Frequency  ( which covers about 95% of those built in the 50's, 60's and 70's) the selectivity of all of them are pretty much equal. About all you can count on is about 10 kHz selectivity without some kind of additional selectivity enhancement.  This is not to say that two broadcast stations 5 kHz apart could not be identified, but that if one was particularly stronger than the other, there would be considerable  "slop" between the two of them and listening would not be comfortable. CW reception on the amateur bands would be possible but digging out the real DX would be difficult because of interference from neighboring stronger signals.
       We have covered the Q multiplier in a previous post.  The next selectivity enhancing feature is the crystal filter.  This was considered a significant expense to add to receivers in that time period and would boost the price considerably. Now because of collectors looking for even the lower level, or entry level receivers, things have kind of evened out.
       The simplest form of crystal filter consists of one crystal cut for the Intermediate Frequency and at some point in the IF chain in series with the signal path. The crystal will pass frequencies on its frequency fairly well, with response falling off rapidly on either side of its resonant frequency.
       The peak on a single, simple crystal filter is fairly sharp. The center frequency is the resonant frequency with response falling off in a curve.  The selectivity figures for receivers are usually referred to in how many DB down the signal is from the resonant frequency.  The points usually used are 6  DB down and 60 DB down.  The difference between the two gives some idea of the sharpness of the filter or tuned circuits in the radio. The 6 DB down figure indicates the frequency separation at which the signal strength is half what it is on the peak. The 60 DB down figure indicates the frequency separation at which all but the strongest signal would be practically gone.
       A graph of the response can be either very "pointed' or sharp, or slope down gently.  The lower parts of the curve are referred to as the "skirts".  The closer in the skirts as the response goes down, the sharper the selectivity.
       More complex filters can be built to shape that response.  These filters make use of more than one crystal and are sometimes referred to as " crystal lattice" filters or  " multi pole" filters.   The purpose of the filter determines how the lattice is set up.  A filter designed predominantly for cw operation might be made up of crystals very close in frequency to give a very sharp, narrow filter with steep skirts. There might be a design that leaves the top a little flat to avoid ringing in the filter.
       A multi pole filter designed for Single Side Band or AM might be made quite differently.  It would be made of several crystals spaced in frequency to give a fairly flat topped response over a 2.5 to perhaps 6 or 8 kHz bandwidth to allow the sidebands of the SSB signal or AM signal to be passed without muffling the sound.  At the same time, the multiplicity of crystals will also insure very steep skirts outside the desired passband. These kinds of filters are often optional filters and are usually found in more complex, communications grade receivers.  They are also very expensive!!
      If you are lucky enough to have a receiver with positions or sockets for optional filters, such as the Icom R-75 or others,  you may have a choice of purchasing filters made by the receiver manufacturer or other after market filters made by other companies.  When looking at the information for these filters,  you will often find charts and graphs that illustrate the responses of the various models of filter.  Careful study should be given to these graphs with thought toward what kind of listening you plan to do.  Casual listening to AM signals on shortwave broadcast bands or Medium Wave might call for a slightly wider bandwidth while serious DX-in and digging for weaker signals might call for something sharper. But once again, I am wandering a bit from the central subject!
       Most common boat anchor receivers have the more simple, single crystal filters. The question sometimes is, which ones have them and which ones don't.  With Hallicrafters,  this is easy.  The receivers without filters have a model number with just the "S" and a number,  such as S-118,  S-120, S-85, S-38, S-40B etc.  The receivers with the filters have an "X" in the model number,  such as
SX-62, SX-71, SX-96,  SX-99, SX-100, etc.  With the others, its a matter of looking them up.
       When looking at a receiver,  it is often easy to tell by reading labels on controls.  If there is a selectivity switch with multiple positions,  its a pretty good clue.  If there is a control labeled crystal phasing" or just " phasing" it is a certainty! 
         Using a crystal filter is a little easier than using a Q multiplier, but there are some things to know beyond just turning it on or setting to a tighter selectivity position.  The phasing control affects its operation by in some ways bypassing it partially and in some ways shifting its peak and null frequency a little.  Some folks get the idea its like a passband tuning control,  and while it may act a little like that, it is not exactly what is going on.\
       While the crystal filter has a passband narrower than the bare IF amplifier, it also has some nulls and secondary peaks off the main frequency.  The phasing control affects them somewhat.  If you are tuning casually on AM,  I find it useful  to have it set off to one side a bit. Then when a station is encountered that you are interested in, rotate the phasing control back and forth to get the best response. You might find that one of the off peak nulls might just take out a pesky heterodyne, though it may affect the desired signal a bit.
       A good thing to do when first obtaining a receiver with a crystal filter is to conduct  a little "experiment and training" session.  If you get a manual with the receiver, there will probably be a general discussion of the operation of the various controls including the crystal filter. There may even be a graphic representation of the crystal's response.  This can be helpful if you can picture what is happening as you are tuning and trying to get the filter to help you.
       As part of the "experiment and training" session, the best thing to do is find a good, steady signal, preferably a not-too-local broadcast band station.  If the receiver has an S-meter, you can get a really good picture of what is going on. 
       First tune the station in with the crystal filter "out" or selectivity control in the broadest position. Peak the signal carefully with the receiver tuning control.  Then switch the filter in or go to the next tighter position.  You might actually see the signal strength drop a bit.  Depending on the alignment condition of the receiver and the insertion loss of the filter, it might drop a little or it could drop a lot.  All crystal filters have some insertion loss and its not unusual to see a drop of an S-unit or two even on those in the best condition.
       Turn the phasing control back and forth and see if the signal peaks up or drops down.  If it peaks, set the phasing control for maximum signal.  Then carefully rock the tuning dial back and forth and see if you can get any more out of it.  Go back and forth a few times to make sure you have it at its highest. Note the position of the phasing control for future reference.  This will be the position you will want it in while you are in the general "hunt" mode for AM or broadcast stations.
       Next watch the S-meter and tune the dial off the station, then slowly go across it to center and beyond.  Note the actions of the S-meter.  There should be the one main peak, some dips and maybe a minor peak again, probably just on one side.  These filters are not usually perfectly symmetrical,  with the symmetry somewhat affected by the position of the phasing control.
       This exercise will give you some idea of the basic shape of the passband of the will also give you the best starting point while tuning.
       After noting the best " center" position for the phasing control, center the station up well and move the phasing control back and forth, noting the actions of the s-meter.  This will give you some feel for the shape or impact the control can have when you may need it.
       Some users of these radios will come away with the idea that the crystal filter is of no use or help or has no effect.  Going through this little exercise should cure that notion.  Often those who get that impression are just turning knobs aimlessly and don't get much of an effect simply because the randomness does not take them into an area where the effect can be seen!    
       The crystal filter can really help out on CW and SSB.  To set up for CW tune to a quiet place or disconnect the antenna.  Put the selective switch in the narrowest position. Turn the phasing control off to one side or extreme range of its setting.Turn on the Beat Frequency Oscillator or set the radio in the "CW" mode. Rotate the BFO pitch control back and forth, listening for a light "swishing" sound as it moves through the receiver passband.  Find the lowest pitch spot which should indicate the center of the passband.  Note this position.  Hopefully,  if the receiver does not need a serious realignment, it will be near the center of the travel of the control. Remember this position for future use. Now offset the pitch control one way or the other just slightly from that center position so that it is toward one side of the center of the  passband. Turn off the BFO.
       Now with the antenna reconnected, tune in a steady AM carrier, preferably a medium wave station within groundwave range but one not too strong. Center it up in the passband by either watching the S-meter or listening for the slight swishing sound that sort of "bottoms out" as you get it centered. The selectivity switch if there is one should be in the narrow, or if there are multiple settings, in the narrowest position.
          Turn on the BFO.  Rotate the pitch control back and forth, finding the zero-beat position over the carrier you have tuned in.  It should be near or at the spot where you had centered it earlier if your tuning has been careful.  Now offset the pitch control slightly giving a heterodyne of a pitch that would be comfortable copying CW for you.
       A short aside here: Normally I would say to set up for copying with a low pitch.  However, with simple crystal filters anything much lower than 800-1000 Hz might put the signal on the " other side of zero beat" too close to allow what is known as single signal reception.  After you finish this set up you can experiment with what pitches give the best compromise between copying at a low pitch and selectivity.
       Now leaving the pitch control in the same spot, using the receiver tuning, tune back to zero beat and on to the other side where the pitch comes back up to about the same pitch as you had when you set the BFO.  Now adjust the crystal phasing control for minimum signal or lowest indication on the S-meter.
       Once this is done, with the receiver tuning, go back through zero beat and take a look at the signal strength of the desired signal.   It should be significantly higher than the nulled side.  With some receivers, this may take a little juggling of the pitch control, offset and phasing adjustment to get the effect you close to single signal reception of  CW as you can get.  Finding the " sweet spots" sometimes takes a bit of experimentation,  but once you have found them, you will find that you are getting the maximum effect out of your crystal filter and can most easily separate CW stations on a crowded band!  Once again, the "art" of tuning comes into play and this is another example of operator skills that can be developed to get the most out of any receiver.
       For SSB, once again turn the BFO pitch "off".  You will find that there will need to be different settings for  BFO pitch and crystal phasing for Upper Sideband ( USB) and Lower Sideband ( LSB).
In each case you will want to have the desired signal centered in the receiver passband but because of the nature of needing to have the missing carrier re-injected in different places to make the SSB signals audible you will need to have the BFO pitch control set differently for the two modes.  And to have the crystals peak and null features in the right spot, there will need to be different settings of the phasing control for the two modes as well.
       Fortunately, finding signals that are definitely known to be USB and LSB is easy.  By convention in the amateur bands, operation on 160, 80, and 40 meters is on LSB while operation on 20,17,15,12,10 and 6 meters is on USB ( there is no phone operation on 30 meters).
       Once again turn the BFO " off " or set the receiver into the AM mode.  Tune in a signal on one of the bands where LSB is used and pick out a signal that is kind of by itself to make set up easier.  You will not be able to understand the talk at this point, but as best you can, center the signal up in the passband.  Selectivity should be in the "narrow" position if there are just " wide" and "narrow" positions or in one of the mid selectivity positions if there are multiple selections.  Set the crystal phasing control off to one side or extreme.
       Turn on the BFO and rotate the pitch control until the signal becomes intelligible or sounds most natural. If the signal is really strong, you might have to back off on the RF gain or sensitivity control on some receivers.
       A this point there needs to be another aside.  If during your initial CW set up you found that the BFO hit the center of the passband well off center of its range, the receiver may be slightly out of alignment and you might not be able to get the control " far enough over " to reach the intelligibility spot.  If such is the case, its ok to "cheat" a little and move the receiver tuning slightly to hit the spot. At some point a receiver alignment would help this situation out, but most basic crystal filters are not so super selective as to make tuning in the signals with a little "cheating" impossible.
       Once the signal is intelligible and the sound is comfortable to hear, use the receiver tuning to move the signal through the passband, going toward the zero beat or in the direction that makes the voice lower pitched until you are on the " other side " of it and it is unintelligible. Make sure you are completely on the " other side" and are hearing the signal effectively " upside down".  At this point, adjust the phasing control for minimum signal.  Then take the receiver tuning control back through zero beat and to the position where your signal is intelligible. It should still be readable and not too reduced in signal strength.  As with the adjustment for CW, there might need to be a little jockeying of the positions of tuning,  BFO pitch and crystal phasing,  but with a little experimentation you can find the settings that will give maximum intelligibility and selectivity.  Note the positions of the BFO pitch and phasing controls and when you start looking for LSB signals, preset them to those positions and you will find your initial search for signals much easier.
       Now go to a band where you know USB signals are located and repeat the process.  Again, note the positions of the controls for future reference.  You will find that the BFO pitch control is at the opposite end of its range from where it ended up for LSB signals. 
       Again, every receiver is a little different,  every crystal is a little different and every receiver alignment condition is a little different.  It take patience and practice, practice, practice to get to the optimum settings.  These initial settings will get you " into the ballpark" and you may find slightly different settings as you further learn your radio. Under certain QRM conditions you might find yourself playing with the controls a bit to hear the station you want.  But always going to these initial settings will save you a lot of time and prevent simple "flailing in the dark" trying to tune a signal in and generating a lot of frustration.  I cannot over-emphasize the existence of a real art to tuning in DX!!
       One other note about tuning SSB signals on older boat anchor receivers.  Many were built before what are know as "product detectors" were routinely put in radios.  In  fact,  many were built before SSB was routinely used in amateur bands and was predominately found in commercial point-to-point operations. They were not designed for optimum SSB operation.  That does not mean they cannot do an admirable job in copying SSB signals.  It DOES mean that some additional care will be needed to make it possible.
       If there is difficulty in getting the SSB signal fully intelligible even when you hit the position where the pitches appear right,  back off on the RF gain or sensitivity control.  It may be that the signal is too strong for the BFO injection level of your receiver.  Tuning SSB signals with these receivers sometimes requires more frequent adjustment of RF gain to make it work out.
       By the way,  the description of tuning in SSB signals with a BFO pitch control outline above also applies to the newer portables that can copy SSB.  The method up to the point of adjusting the crystal phasing should be the same to avoid some wild-eyed frustration in getting the signals audible.  For those portables with pitch controls or an equivalent there-of  the method outlined will help considerably.
       One other note about crystal filters and receiver alignment.  Because crystals can "wander" somewhat with age and because quality control may not have been the best with some manufacturers, the crystal may not be centered exactly on 455 kHz ( or what ever the design Intermediate Frequency of the receiver might be).  For this reason,  when doing an alignment of these receivers it is highly important that a signal generator be fed into the IF's before the filter, meaning in some cases may need to be fed into one of the grids of the mixer tube.  It also means that instead of arbitrarily setting the signal generator at 455 that it be swept slightly back and forth two or three kHz to find the peak of the crystal ( with the phasing control off to one side or extreme) and aligning the IF to the crystal. The difference of one or two kHz will not be enough to affect the tracking and calibration of the main tuning.  The important thing is that the Intermediate Frequency amplifiers be peaked to the frequency of the crystal because that will keep the narrowest response possible.  The frequency of the IF peak can be shifted a little, but the response peak of the crystal cannot!
      One other note about alignment at this point that applies to any boat anchor regardless of whether it has a crystal filter:  Once alignment has been complete, turn on the BFO and swing it through its range with the signal generator still feeding at IF signal in.  You should always have the signal generator output level set at the lowest setting possible to allow alignment.  Rotate the BFO pitch control for zero beat against the signal generator carrier.  Note if the BFO pitch control is at the center of its range.  If it is not,  look in the service materials for the receiver and find a below chassis adjustment for the BFO frequency.  It will probably be a slug tuned coil.  Once you find it, set the BFO pitch control on the front panel to the center of its range, then bring the BFO frequency to zero beat with the other adjustment.
       Older receivers with crystal filters are not all exactly the same,  but the principal of the operation is. There are still differences in quality of components and even of the crystals themselves.  The single crystal filter may not be as tight as some of the multi pole filters that appear in the more expensive radios,  but they do tend to be more effective than some of the ceramic filters that appear in some of today's budget portables.
       Once again, I will note that the methods described here are not to be taken as " the only way" to do things,  but are the methods I personally have found useful in fifty plus years of spinning the dials.  They work well for me and have put 326 countries in the log.
       For those who enjoy tuning with the vintage receivers and capturing some of the adventure of the golden age of SWL,  a crystal filter equipped radio will provide the additional satisfaction of greatly improved selectivity. There is something special about tuning one of these "beasts" though for a beginner, there are many things to learn and some  frustrations to avoid.  They are not necessarily for the " dial in the frequency and forget it and listen " crowd. For that group, even the most expensive, digital dial and DSP equipped radio might be found to be frustrating for them.  But for others seeking a bit of adventure from the hobby the boat anchors do offer a level of performance and sound quality that is difficult for many of the inexpensive portables to match. DX-ing, like life in general, offers a bit more to those willing to put a bit more into it ( a little personal philosophy there, too!!)
       Whatever type receiver you chose to use, the important thing is to learn how to use it to the fullest of its capabilities.  Tuning shortwave is not like turning on your cable-fed television and dialing to a channel number and having a station pop up!
       Good DX.  Good Hunting!


  1. Great article Roland. I have an AR88 which has several crystal positions, and I remember the narrowest is great for CW reception. Unfortunately at the moment, that receiver is in my storage area, as the area of the lounge I have for listening would not accommodate the '88 (and the XYL would protest!) But hopefully in the future, I can dust off the venerable beast and get using it again. I can just smell it now....!
    73's, Charlie