[ARC5] Oscillator Stability and Old-Time gear. (Was OT: Hally Instability)

Sun Nov 15 19:05:44 EST 2015

Hello Richard,
You make a good point here.  Several actually.  

About 6 months ago I bought a Hallicrafters S-38 with the specific
purpose of putting it 'on air'.  Yes, instability and all that.  I think
it will work on 80m.  Above that, I expect trouble, especially with
S.S.B.  The first thing I want to do is actually measure stability - but
after I make some mods.  I will rewire the heaters and replace the
'tubes' so the set will operate on 24V filament supply.  I will replace
the rectifier with a 12SL7 and use that as a "Q" multiplier.  I will
build a new (and very essential) missing back-panel board.  I will add a
solid state 25kHz "spot" marker-generator.  I will try to "tie-down" the
tank components in the local oscillator to improve the stability.  I
will run the set on a variable B+ supply and note the difference in
performance with the set running on 45 Volts and 180 Volts.  (Please
note, Richard, I wrote "Volts" not "volts" - just as I try to correctly
write MHz when I mean MegaHertz, not mHz (i.e. milliHertz)).  I hope
no-one objects when I 'hack' my set into a slightly different form.

On the point of stability, I spent many hours trying to build a stable
V.F.O.  I learned many things about oscillator stability.  List-readers
will have seen articles promoting "this" circuit or "that" circuit as
being "stable".   I plotted the stability of oscillator frequency vs
temperature many times.   I cycled the temperature inside the VFO "box".
 I noticed the graphs (frequency v temperature) made clover-leaf like
loops at the extremes - and the shape of these "loops" revealed the
sensitivity of particular components as they heated (or cooled), when
heat flowed into or out of the circuit components.

I came to understand that the stability of an oscillator depends on the
aggregate stability of components, as well as the stability of EACH
component used.  To obtain stability in the order of 50 or 100Hz the
aggregate value of tank "C" must be held within fempto-Farad limits.  I
recall reading articles declaring that the Vackar circuit so stable that
it would not drift, even sitting in an open box near  an open window.   
I doubt that.  I really, really doubt that, because EVERY component has
some temperature coefficient, and EVERY "C" drifts a few fempto-Farads
in a breeze.  Even NPO types.  Oscillators are so susceptible to the
effect of temperature that even throwing a plastic sheet (flexible
plastic) over the circuit will improve frequency stability "heaps".  

I measured the stability of the famous BC-221, because it is regarded as
the "gold" standard.   How stable was this piece of gold?  The measured
difference between my (rather old) Fluke 1912A and the BC-221 was in the
order of 160Hz per degree C.  Was the drift in my Fluke or the BC-221?  
(Note: The BC-221 was unmodified - it may perform better than this with
some new components.)  Where was the drift?  I don't know)     I just
report my observations.

I have come to respect the work done in the 30s and 40s.  Even my humble
workshop has "nicer" test gear than the pioneers from times-past had
available, yet they did so much!    Now, all the knowledge from that
time is slowly dissipating.  Soon it will be gone.  Perhaps even the
M.F. broadcast band will be obsolete.  (I hope not!)  Communications
systems will be dominated by new, more capable systems, designed and
built, not by hundreds of capable women working at a line of tables, but
by robotic "pick and place" machines.  That is the role of the present
batch of graduates.    

This gear will be more stable, more selective and work on frequencies
never-imagined by Collins, Hallicrafters or Hammarlund.  The new stuff
will be better, but I'll miss the old unstable gear.  I can make "the
old gear" work.  Modern gear is sooooo specialized, so custom, so
microprocessorized - it rarely has a role outside its intended sphere of
operation.   The old drifty stuff still has appeal!

73 de Les Smith
  vk2bcu at operamail.com

On Mon, Nov 16, 2015, at 05:36, Richard Knoppow wrote:
>     Hallicrafters strong point was good value.  Their sets mostly 
> delivered good performance at an affordable price but they didn't make 
> much world-class stuff.  There is a propaganda film called "Voice of 
> Victory" available at you tube and at www.archive.org that is mainly 
> about production of the BC-610.  What is interesting is how much change 
> the Signal Corps required to make it suitable for mobile military use. 
> Nearly all was undoing cost cutting in the original design.
>     There is also a WW-2 vintage Signal Corps communications manual(I 
> don't remember the number) which shows spurious response charts for 
> several SC receivers; the Super Pro has only a single image while the 
> SX-28 looks like a corn field. OTOH, the SX-28 was a little more than 
> half the price of the Super-Pro and had all sorts of features that would 
> make the owner feel as though he/she had the latest technology even 
> though it didn't work that well.
> 
> On 11/15/2015 10:18 AM, Todd, KA1KAQ wrote:
> > I don't think there's any big mystery involved, Dave. Hallicrafters
> > produced a lot of models, primarily for the 'entertainment'/consumer
> > market. The better the model, the better the performance. Aging of
> > components would play a role in any receiver, of course. Age of design is
> > another consideration. The early Hammarlund Super Pros  for
> > example(pre-war) were some of the finest receivers of their time, yet
> > suffered from the usual issues above 20mcs on the SX models. They also
> > suffered from drift until warmed up, but the manual makes it pretty clear
> > that they were meant to be turned on and left on. The SPs got a bad rap
> > later for drift issues related to folks not heeding this advice, along with
> > the well-used examples that hit the surplus market and got used without any
> > repair or replacement of aged components.
> >
> > You're right about the SX-101 - both that I owned drifted like crazy. They
> > added a resistor aka damp chaser to keep a couple critical components
> > 'pre-warmed' in an effort to deal with some of the drift, so later 101
> > models performed better. My SX-28A drifts, mainly because it's never been
> > gone through. Stopped using it until it gets a good re-cap and resistor
> > check. Someday....
> >
> > Halli did make a few decent sets like the SX-73 and SX-115. But they were
> > pricey. The SX-88 is probably one of the most over-rated receivers out
> > there, swooned over by collectors due to low production numbers (it was a
> > dog as far as sales go) far more than actual performance. By the 60s they
> > seem to have figured it out but their halcyon days were behind them.
> >
> > The long and short of it is, you got what you paid for. The cheaper the
> > set, the more marginal the performance. And now of course, older technology
> > and components don't have the performance of later designs. If you use your
> > Halli primarily for listening to AM, drift is less of an issue. Far more
> > noticeable on CW/SSB signals. I think it was Collins who was first to
> > conquer the drift issue with their ground breaking PTO design.
> >
> > ~ Todd,  KA1KAQ/4
> >
> > BTW, the SX-42 was the post-war upgrade of the SX-28 if I remember
> > correctly. SX-62 shared a similar design, without bandspread.
> >
> > On Sat, Nov 14, 2015 at 10:26 AM, David Stinson <arc5 at ix.netcom.com> wrote:
> >
> >> I'm not doubting anyone who says their Hallys were stable.
> >> I do know in 40+ years of hamming, I've handled dozens
> >> of them and never seen a single one from the "ham" market line that didn't
> >> show mechanical instablilty and "wibble-wobbles" above about 12 MC.
> >> Is it possible this was some sort of aging-related
> >> design flaw that a few were spared?
> >> Has anyone ever done a through investigation
> >> to find the root of it?  Coils mounted on a long
> >> bandswitch that will transmit shocks like a tuning fork always seemed a
> >> candidate to me.
> >> Most likely there are multiple causes.
> >>
> >> 73 D.S.
> >>
> >> ______________________________________________________________
> >> ARC5 mailing list
> >> Home: http://mailman.qth.net/mailman/listinfo/arc5
> >> Help: http://mailman.qth.net/mmfaq.htm
> >> Post: mailto:ARC5 at mailman.qth.net
> >>
> >> This list hosted by: http://www.qsl.net
> >> Please help support this email list: http://www.qsl.net/donate.html
> >>
> > ______________________________________________________________
> > ARC5 mailing list
> > Home: http://mailman.qth.net/mailman/listinfo/arc5
> > Help: http://mailman.qth.net/mmfaq.htm
> > Post: mailto:ARC5 at mailman.qth.net
> >
> > This list hosted by: http://www.qsl.net
> > Please help support this email list: http://www.qsl.net/donate.html
> >
> 
> -- 
> Richard Knoppow
> 1oldlens1 at ix.netcom.com
> WB6KBL
> 
> ______________________________________________________________
> ARC5 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/arc5
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:ARC5 at mailman.qth.net
> 
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html

-- 
http://www.fastmail.com - Faster than the air-speed velocity of an
                          unladen european swallow