[R-390] xtal wave spikes
Dan Merz
[email protected]
Tue, 26 Nov 2002 15:34:17 -0800
Bob, the diodes in the ring detector are germanium 1N82a's so I think the
turn-on voltage is lower if I remember right compared to silicon type. I was
down measuring some more and found lower oscill. output voltages than before for
some reason and thought maybe I was reading my scope wrong before I checked
the 2nd receiver I have to see if I could get what I measured before since I had
both of them on the bench when I made the first set of measurements. I got the
lower voltages there too so I must have made a slip between my notes and the
scale factor, and the highest voltage is around 0.5 volt (not 1.7). And now I
don't see the spikes that I saw before !!! The waveforms are now very smooth.
I'm at a loss on that. I've been passing one 6BL8 tube between the two sets for
the oscill/cathode follower as I'm short on that tube type. It may be possible
that in exchanging tubes I used a different 6BL8 before for the oscillators;
the sets have 2 each and at one point I only had 2 to use between the two sets
and was exchanging both; now I found another one and have 3 and only exchange
one of them. Maybe there's something in one of those 3 tubes that would account
for the spikes. And I noticed that for one of the xtals that I thought was
bad, I could get a 0.1 volt pp wave or maybe a bit lower at the cathode
follower output but on the ring I didn't get the waveform showing up so I
interpret that to mean that the output is barely there and not strong enough,
near the threshold. But there may be another reason for this behavior, maybe
the operation is just erratic for that particular crystal. This'll take a
little more snooping on my part. This particular crystal was diagnosed as not
working several weeks ago but was working when I checked it later and then was
working again today at the particular location mentioned but was not at the
other location. By the way, I'm wondering if there is a replacement unit that
might be put here in place of the 4 diodes since these were put in back when
this type of mixer was just coming into the picture and they might not be as
good as later units. I'm also wondering if you would ever find a matched set of
1N82a diodes to replace the originals. Mini-Circuits makes a balanced mixer
but I don't know if you could stick one of these in and I don't know what kind
of diodes are in these mixers. At this point I have no reason to think I should
replace the diodes but am wondering what-if.... One of these receivers works
pretty good and is on a par with the 390a as far as detecting weak signals so I
have no reason to think the oscillator level is really too low for the good
xtals at least. I haven't noticed any problems with strong signals other than
bc stations need about 20 db atten on the front end (built in attenuator is
there for that) to behave nicely, a known characteristic of this radio. And
the other receiver also is comparable but has some other dial repairs and
intermittents that I'm trying to remedy Dan.
Bob Tetrault wrote:
> Dan,
>
> The more amplitude the better, subject to the limitations of maximum diode
> current at the higher amplitudes. It does concern me that your minimal
> amplitude approaches the turn-on voltage of a silicon diode; those bands
> will be more likely to perform poorly in the presence of strong signals, as
> well as suffering from increased conversion loss. OTOH, if it means that
> you'll have to replace all the crystals and find a hot tube and re-bias for
> increased output, etc, etc, one can reach a point of decision...
>
> Good luck
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]]On
> Behalf Of Dan Merz
> Sent: Monday, November 25, 2002 1:17 PM
> To: Bob Tetrault; R-390 List
> Subject: Re: [R-390] xtal wave spikes
>
> Bob, the wave amplitude ranges from about 1 to 1.7 volts pp output from the
> cathode follower that feeds to the single-balanced diode ring detector,
> with a
> couple of the xtals producing 0.8 volts pp. I don't think there is gross
> distortion other than what I described but I'm just looking at it with a
> scope.
> I'll order the dead crystals after trying a couple of different tubes in the
> oscillator but I don't expect a miracle at this point with tube
> substitution.
> Your comments about the square wave brought back some faint memories of
> fourier
> series representation of square waves but my memory bank need refreshing
> about
> that and what a mixer really does besides adding and subtracting the
> frequencies of waves put into it. I do remember that a square wave is
> generously
> endowed with waves of many frequencies, and hence its utility in testing.
> thanks for the comments, Dan
>
> Bob Tetrault wrote:
>
> > Dan,
> >
> > The presence of irregularities in a waveform is not immediate evidence of
> a
> > failing crystal. Either they don't work or they begin to lose output and
> > ultimately don't work. Unless the oscillator is tuned, at the output, for
> > the crystal frequency, and unless the oscillator is biased explicitly for
> > that particular crystal and mode of operation, there is very great
> > likelihood that some harmonic energy is being generated by the oscillator.
> > Also, some oscillator circuits are better at harmonic generation than
> > others. You would need a spectrum analyzer or a VHF/UHF receiver to
> > investigate this, and the fact that some of your crystals do not exhibit
> the
> > spike on a scope doesn't mean that significant energy isn't being produced
> > at *their* harmonics. The fact that this energy is being injected into a
> > mixer mitigates the presence of any harmonics, as the mixer itself is a
> > harmonic generator. They are sometimes referred to as a comb generator in
> > the older literature. In other words, don't worry. Is there sufficient
> > amplitude? Is it grossly distorted?
> >
> > This begins to get off-topic, but there is ample and long standing
> evidence
> > that a square wave with very fast edges is the best waveform to drive a
> > mixer. Such a waveform has quite high energy in the harmonics. It is the
> > relatively slow transition in the interval between either fully saturated
> ON
> > or fully cut OFF that promotes susceptibility to intermod. Thus a fast
> > square wave minimizes the susceptible time.
> >
> > Bob
> >
> > -----Original Message-----
> > From: [email protected] [mailto:[email protected]]On
> > Behalf Of Dan Merz
> > Sent: Monday, November 25, 2002 9:22 AM
> > To: R-390 List
> > Subject: [R-390] xtal wave spikes
> >
> > Hi, in looking at the output from the front end high freq. oscillator
> > (40 to 60 Mhz xtals) on the Mackay 3010, I noted double peaks on some
> > of the waveforms, though the nominal frequency was ok. My first
> > reaction was that this could be symptomatic of a failing crystal though
> > the output works ok and the right signals are received in the radio on
> > the band in use. The spike occurs on the downside of the wave, kind
> > of a minor glitch as the wave is decreasing. Does anyone know what this
> > means or what is causing it? Most of the xtals don't exhibit this
> > characteristic, but a couple are dead. I'm going to replace the dead
> > ones - MH Electronics says they will provide replacements for $15 ea if
> > I'll send them a good crystal as example of spec's. Is this a
> > reasonable price compared to other possible providers? thanks, Dan.
> >
> > _______________________________________________
> > R-390 mailing list
> > [email protected]
> > http://mailman.qth.net/mailman/listinfo/r-390
> >
> > _______________________________________________
> > R-390 mailing list
> > [email protected]
> > http://mailman.qth.net/mailman/listinfo/r-390
>
> _______________________________________________
> R-390 mailing list
> [email protected]
> http://mailman.qth.net/mailman/listinfo/r-390
>
> _______________________________________________
> R-390 mailing list
> [email protected]
> http://mailman.qth.net/mailman/listinfo/r-390