Re: {Collins} Crystal Filter

----- Original Message ----- From: "Clarence G. Marshall" <bonaire@xxxxxxx> To: "'Bill Carns'" <wcarns@xxxxxxxxxxxxx>; "'Richard Knoppow'" <dickburk@xxxxxxxxxxxxx>; <collins@xxxxxxxxxxxxx>
Sent: Sunday, August 25, 2013 2:26 PM
Subject: RE: {Collins} Crystal Filter


Could you give a little more color on the phasing adjustment. I have an early A-3 that seems to have a lop sided mechanical filter (FB455B3). I am not sure how to trouble shoot the phasing circuit because I am not sure what is the expected response. The filter does pass a fairly good signal but
seems weak on the high side audio wise.

Thanks in advance


This brings up a question I don't think I've seen addressed. The 51J-4 has the same crystal filter as in earlier versions of the receiver but also has three mechanical filters. The bandwidth of the narrowest mechanical filter is 1400 Hz. The crystal filter has several steps but the widest is wider than this and the narrowest only a couple of hundred Hz. Now, presumably, the crystal filter was left in the circuit in order to provide a notch to reduce the effect of interfering signals and also to provide a very narrow bandwidth for weak CW signals or where desired. The problem comes in because of the IF frequency. In the 51J the whole tuning scheme is based on the IF being at exactly 500 kHz. If the IF is set to another frequency there will be an alternating error in the calibration. While the adjustable pointer could compensate for this the J4 has an additional consideration: the mechanical filters are designed for the exact 500 kHz frequency. Now, comes the crystal; ideally the crystal resonance should also be exactly 500 kHz so that the crystal bandpass is centered on the mechanical filter bandpass. However, no matter how accurately adjusted the crystals were when new they tend to drift with age. One military handbook (for the R-388 I think) gives the tolerance of the crystal as +/- 70 Hz. In my R-388 the crystal resonance is rather farther off than this. Because my receiver does not have mechanical filters the only effect is that the setting for the BFO to zero beat a signal which is exactly centered in the crystal bandpass is offset a bit from the nominal zero setting at 500 kHz. The 500 kHz setting is necessary to calibrate the dial. This is a minor inconvenience, however, in a receiver with mechanical filters the simultaneous use of the crystal filter, especially with a narrow mechanical filter, could affect the symmetry of the passband or even the gain if the crystal was _very_ far off. I think the latter is not very likely. Of course, the ideal thing would be to obtain a crystal at exactly 500 kHz but such are not very easily available. I may be making a mountain out of a molehill. I think the R-390/A has the same arrangement but using a 455 kHz IF. Note that in conventionally designed receivers the actual IF frequency is set to the crystal resonance. As pointed out above this is not possible in the 51J without creating other effects which might be worse.
    Pardon the thinking out loud.

Richard Knoppow
Los Angeles

This archive was generated by a fusion of Pipermail (Mailman edition) and MHonArc.