[Hammarlund] Crystal Filter Alignment

[Richard Knoppow]

----- Original Message ----- 
From: "k2cby" <k2cby at optonline.net>
To: <Hammarlund at mailman.qth.net>
Sent: Monday, February 07, 2011 11:04 AM
Subject: [Hammarlund] Crystal Filter Alignment


>I would be grateful for any more information on aligning 
>the SP-600 crystal
> filter, starting with the procedure for orienting the 
> little arrowhead on
> the dial (which never seems to be explained in the service 
> manuals).
>
>
>
> Although I have followed the standard 455 kHz and 3.955 
> MHz procedures
> outlined in both the commercial and military service 
> manuals with apparently
> good results, I have never been able to square the results 
> with the very
> roughly described procedures for sweep oscillator 
> alignment. Is there a more
> complete description of the sweep oscillator procedure?
>
>
    All of the crystal filters used on Hammarlund receivers 
since the HQ-120-X are the same basic circuit and is one 
Hammarlund had the patent on. It is far superior to the 
original Lamb filter used in National and Hallicrafters 
receivers.
    The entire IF must be adjusted to the actual crystal 
filter frequency. While it will be close to the nominal IF 
frequency it will not be exact. The crystal is then used 
either to set the signal generator to the correct frequency 
or to establish a center marker if a sweep generator is 
used. Once the IF is set up the crystal filter is done using 
a modulated signal. There are two adjustments: one is a 
peaking adjustment, the other establishes the loading on the 
crystal to broaden it out. Both are set with the crystal 
bandwidth at maximum. The first is set using a low 
modulation frequency, say 400 hz, and peaking for maximum 
audio level with the AVC OFF and input kept fairly low. The 
second is then set using a higher modulating frequency, say 
2 khz, and again adjust for the maximum output. What you are 
doing is setting the crystal filter _wide_ bandwidth for its 
maximum, the narrower ones will follow. Once set check that 
the phasing is centered. The easiest way is by ear. Tune to 
plain noise with the BFO OFF and the crystal in its 
_narrowest_ setting. The noise will have its narrowest 
bandwidth, that is the least highs, at the center. You must 
first check to make sure the phasing capacitor is centered. 
It is a butterfly type. In the HQ-129-X the shaft has a flat 
on it so the knob is automatically in the right place but 
this is not true of many other receivers using this circuit. 
The minimum noise should occur with the phasing knob 
centered. If it does not check the trimmer capacitance in 
the filter can. In the HQ-129-X this is a short piece of 
insulated wire bent into a U shape, you may have to move it 
or bend it to center the phasing knob. In the older 
Super-Pro there is a compression trimmer in the can for the 
same purpose. Some RX using the Hammarlund circuit, for 
instance the Collins 51J series do not have adjustable 
trimmers but do have a fixed capacitor serving the same 
purpose which can be padded or changed.
    When set correctly, the null will also be reasonably 
symmetrical. To test for this set the filter for its 
narrowest bandwidth. Then set the generator for the exact 
crystal frequency. Then set the filter for its _widest_ 
bandwidth and then off set the generator to one side by a 
known amount, say 1 khz. Null it with the phasing control 
and note the position of the control. Then offset the 
generator the same amount in the other direction, again null 
the signal and note the position of the phasing control. It 
should be approximately the same distance from center in 
both directions. Note that the position of the phasing 
control does NOT follow the frequency of the null: the 
control will be at its maximum displacement for frequencies 
close to the crystal center frequency and move toward the 
center as the null moves away from the center. It is not 
completely symmetrical but will be close for frequencies 
around the middle of its range in either direction. The 
Hammarlund filter is the best of the single crystal filters, 
it was used in Hammarlund receivers and later, by contract, 
or when the patent expired, by TMC, RCA, and Collins and I 
think later by National and Hallicrafters as well. It has 
the advantage over the original Lamb circuit that the 
bandwidty control and phasing control do not vary the center 
frequency or gain of the circuit. I suspect that the reason 
the original version of the RCA AR-88 did not have a front 
panel phasing control was to avoid infringment on the 
Hammarlund patent.


--
Richard Knoppow
Los Angeles
WB6KBL
dickburk at ix.netcom.com