[Collins] KWM2A receive alignment - USB/LSB

[Gerald]

On Wed, 2006-02-01 at 21:21 -0500, Steve Garwood wrote:
> Fern,
> 
> I have a set of the crystals on order from Inrad, since it's almost the path
> of least resistance.  I need to do some measurements of the filter itself -
> hooking up the signal generator and seeing if I can determine anything
> there.

The signal generator will confuse the measurement unless you couple it a
stage before the filter. The signal generator probably won't have the
frequency calibration of the 70E-2 PTO. Use the calibrator crystal and
the PTO to measure the response curve, you won't modify it with the
signal generator and detector.
> 
> I am looking into the filter replacement, but they don't seem to be readily
> available, the prospect of getting one worse than what I have exists, and
> it's not a plug in.

The prospect of a filter failing depends a lot on its history, one used
at home is less likely to have failed than one used mobile and having
traveled around the world in military applications.
> 
> Since I haven't been able to do anything to change the audio one way or the
> other, I'm somewhat reluctant to go into the "shotgun" mode without doing
> some more checking first.  There are some other capacitors in the circuit
> that may get replaced first since they are original and appear that they
> could at least influence the circuit if they are out of tolerance.

Two capacitors in the oscillators affect the frequency and the feedback.
There is one from the tube grid to ground and the other is from the tube
screen to ground. The 1000 pf in series with the crystal has no
significant effect, its purpose is to keep DC off the crystals.
> 
> This has proven to be very interesting - I'll most likely know a lot more
> about it before I'm done....
> 
> I'll check back in when I've done some more checks.
> 
> Steve, N0CZV 
> 
> -----Original Message-----
> From: Fern Rivard [mailto:crc at cyberlink.bc.ca] 
> Sent: Wednesday, February 01, 2006 9:12 PM
> To: Gerald; Steve Garwood
> Cc: collins at mailman.qth.net
> Subject: Re: [Collins] KWM2A receive alignment - USB/LSB
> 
>     As per your report on the frequencies of  both the USB and LSB 
> crystals, I would recommend that you replace both of them and if still 
> having some problems with the frequency response to then replace the 
> collins mechanical filter.
> 73 from Fern  VE7GZ
> 
I don't think the measured errors are great enough to cause the
complaint. I think the crystals are no at fault.
> 
> 
> ----- Original Message ----- 
> From: "Gerald" <geraldj at ispwest.com>
> To: "Steve Garwood" <n0czv at willard-oh.com>
> Cc: <collins at mailman.qth.net>
> Sent: Wednesday, February 01, 2006 11:56 AM
> Subject: RE: [Collins] KWM2A receive alignment - USB/LSB
> 
> 
> > On Wed, 2006-02-01 at 12:45 -0500, Steve Garwood wrote:
> >> I haven't checked on the center frequency of the filter yet.  I'll be
> >> setting up the signal generator - I would like to try the 10 pf 
> >> across the
> >> crystal leads with clip leads just to see what difference it makes. 
> >> Part of
> >> the problem now is that I can take the measurements but I'm not able 
> >> to
> >> affect any change.  Shifting the crystal frequency a bit would shed 
> >> some
> >> light on things.
> >>
> >> Steve, N0CZV
> >

I repeat:
> > Its not only center frequency, its tilt of the pass band, or excess
> > ripple at one part of the filter pass band. You can do a decent check
> > with the receiver and calibrator along. Take your readings from the S-
> > meter and you tune the receiver past the calibrator.
> >
> > Its unlikely the filter elements have changed frequency. Its more 
> > likely
> > that coupling between them which is by wires spot welded to the
> > resonators have come loose and affect only a part of the passband. 
> > Since
> > the response curve is inverted for LSB compared to USB a tilt in the
> > response curve gets an acoustical sound of strong highs on one 
> > sideband
> > and weak highs on the other. BFO shifts  (which typically are set 20 
> > dB
> > down the response curve) add or remove mostly bass pitches by shifting
> > the filter sideways, with little perceived change on the highs because
> > 100 Hz shift changes the low cutoff from 300 Hz to 400 or 200 Hz, but
> > the high cutoff from 2400 to 2500 or 2300 Hz which is hard to 
> > perceive.
> >
> > -- 
> > 73, Jerry, K0CQ, Technical Advisor to the CRA
> > All content copyright Dr. Gerald N. Johnson, electrical engineer
> >
-- 
73, Jerry, K0CQ, Technical Advisor to the CRA
All content copyright Dr. Gerald N. Johnson, electrical engineer