Richard - love that last sentence!
I've been reluctant to mess with the caps because the transistors and
mechanical issues are easier to do when you don't have the test fixture(s)
and procedures that were available at the factory. As a long-time
technician, I like to stay on the conservative route as far as component
replacement and also have an appreciation as to why there are not one, but
two S.A.T. (select-at-test) capacitors in parallel with the main part of the
frequency-determining circuit. Every replacement has the potential of
affecting the calibration of the entire unit. Once I had determined that
one of the previous owners had changed out the transistors with a "generic"
NPN, that was the first component replacement I wanted to try. I have to
say that the quality of the signal improved somewhat, but not entirely. It
does sound less "chirpy" but that may also be a placebo effect. I have no
measurements of the FM'ing of the PTO assembly frequency to bolster any
argument.
Once I get a chance at checking out my club's junkbox, I'll know what's
next. If any of you have a 3000 pF SM cap, I'll pay your cost plus postage.
Steve Wedge, W1ES/4
"I can't complain, but sometimes I still do."
- Joe Walsh
If the above message appears, it came from Steve's Son of Laptop!
----- Original Message -----
From: "Richard Knoppow" <[email protected]>
To: "Steve Wedge" <[email protected]>; <[email protected]>
Sent: Friday, July 29, 2011 4:00 PM
Subject: Re: [Drakelist] R-4A PTO: It wasn't the transistors
----- Original Message -----
From: "Steve Wedge" <[email protected]>
To: <[email protected]>
Sent: Friday, July 29, 2011 11:54 AM
Subject: Re: [Drakelist] R-4A PTO: It wasn't the transistors
You know, Garey, I've been thinking about the xtal oscillator. It would
have to be part that doesn't get bandswitched - which would include THAT
transistor.
When the T-4X controls the frequency, it sounds great and never jumps
frequency. In looking at the schematic, I have been aware that the two
solid-state oscillators (PTO and LO for the band) get mixed in V8 (IIRC)
and that this combination gets overridden by the output of the T-4X's
premixer.
Looks like I am going to need a counter to find out. I am loathe to dig
into the T-4X because it's working so well and the PTO is much more
difficult to remove from it due to the volume of wires in the area.
73,
Steve, W1ES
I am now a bit confused. Is the PTO in question in the TX or RX? If
the RX I can see some possibilities from the circuit diagram. These are
perhaps far fetched but, since you have checked all the obvious stuff they
may have some merit.
1, There is a connection for an FSK adaptor. Its a sort of solder tab
going through a hole in the shield can. This goes directly to the tap on
the PTO coil that couples the output to the buffer, Q3. Perhaps something
is affecting it. The output of the PTO goes to the PTO-Crystal switch. I
am not sure how much influence on frequency this point can have since it
is on the output side of the buffer but worth checking that there isn't
something flaky about the switch.
C-90 is the main frequency determining capacitor. I think this is the
"dogbone" cap which has been discussed. I am curious as to what kind of
cap this really is. Most silvered mica caps I've seen were either the old
molded bakelite case kind or the more modern type in dipped epoxy cases. I
think the migration problem, also called scintillation, is mostly in the
molded case kind. I am not sure if this is due to some moisture being able
to get into the molded cases or to a change in manufacturing technique but
it does not seem to happen, at least not as frequently in the dipped case
kind. All the "dogbone" style caps I've seen were ceramic. I don't know if
ceramic caps can develop this or a similar problem. The diagram shows two
seleced value caps in parallel with the main cap. I presume these are
temperature compensating caps, which means they are ceramic types.
Temperature compensating caps are always ceramics, they can be in at least
three varieties of casings, molded bakelite, ceramic shells sealed at the
ends, and the dogbone type. There may also be disc types.
I don't think a counter will be of much help because they respond too
slowly. To isolate the variation I think you need a separate receiver
capable of hearing the PTO frequency and also the other oscillators. You
should be able to get an isolated signal from any of them with the aid of
a one-turn loop on the end of a coax. Set the external receiver up with
its BFO adjusted for a convenient beat note and just let it go. You might
also have a signal tuned in on the Drake so you can tell if one jumps when
the other doesn't. Obviously, if the PTO is jumping both will jump but if
the Drake jumps but the external RX indicates the PTO is stable you have
to look elsewhere. The fact that the RX is stable when driven by the TX
PTO seems to eliminate other oscillators in the RX.
Now, I am thinking out loud so take that into account.
I think you said you had opened the PTO and poked a the circuitry. If
not try it. I find those big wooden toothpicks that are used to hold
sandwiches together handy for this. Some of the component positions in the
PTO are evidently critical so don't move thing much but wiggle all the
leads and also check carefully for a cracked trace on the circuit board.
They can be quite invisible and intermittant in effect. I am guessing at a
couple of things because I've had only one Drake PTO open and that was
partly curiousity and partly because I managed to slip a gear in the drive
and wanted to make sure it was back right.
I don't remember if the core itself is exposed all, I think it is. If
so poke at it. Its highly unlikely but possible for it to have developed a
crack. If that's so I am not sure what to do. I think cracked cores can be
repaired with cement but getting it out to work on it may be difficult. I
think this is _very_ unlikely, but we are looking for unlikely things.
I just looked at the Garey's Drake service manual but it does not show
the PTO so I have no memory refresher.
Also, if you have a friend who has some electronic skill get him or
her to look also. Its very easy to overlook something over and over, even
when its obvious. Sometimes another person will spot it right away.
Since so many Drake PTO's seem to work fine with no additions or
modifications I would be very leary of adding anything. This problem
exists because some component has gone wrong and not because the _design_
is wrong.
Lastly, this is just a dumb machine and its outsmarting a bunch of
fairly smart people, we should be able to beat it!
--
Richard Knoppow
Los Angeles
WB6KBL
[email protected]
_______________________________________________
Drakelist mailing list
[email protected]
http://mailman.zerobeat.net/mailman/listinfo/drakelist
