[ARC5] ARC-5 Crystal Question

Fri Aug 13 16:08:12 EDT 2010

I have used, restored, and etched many an FT-243 or CR-1 style crystal
because I am into CW using older rigs. I've found many cases of crystals
that are very sluggish or inactive where there is nothing obviously wrong
with the holder or rock itself (e.g. broken Cu leg at pad-to-leg joint or
bad solder joint at pin or cracked crystal).  In those mystery cases the
problem is usually due to the re-distribution of loosely attached ground
quartz dust left over from the manufacturing process prevalent during WWII.
It wasn't until the latter third of WWII that the US discovered the cause of
crystals drifting the field which is the early stage manifestation of the
manufacturing artifacts.  The loose quartz moves with time and vibration and
this changes the distribution of mass across the faces of the crystal and
along the edges.  Bliley finally correctly identified the root cause and
came up with a relatively simple extra manufacturing step that killed three
birds with one stone.  They solved the drift problem, speeded up overall
manufacturing and improved frequency accuracy.  The solution was to grind
until you were within a few Kc and then use an acid dip for a pre-calculated
time in order to etch the blanck up in freq up to the correct value.  The
war dept arm-twisted Bliley to share this valuable technique with the other
makers. This technique became standard practice near the end of the war and
dominated the industry after the war.

I have experimented with many WWII surplus rocks and can attest to this
effect.  If you take an old FT-243 and first measure its freq, then open it
up and do a brief 1 min dip in a weak buffered HF solution (be careful with
this stuff!!) you will find upon scrupilous washing and subsequent
reassembly that the freq has jumped up a Kc or more and that the activity
has improved.  If you put the crystal back into the acid solution for
another minute the freq may only change 50 to 100 cycles and the next minute
even a little less.  What's going on?  The crystal had its loosely attached
quartz dust rapidly dissolved (probably within the first 30 seconds) and
then settled into a new much slower etch rate determined by the acid's
ability to break down the single crystal structure that remains.  Your are
probably thinking the same effect can be achieved by simple washing in soap
and water, but that is only a partial solution.  The really loose stuff does
wash off, but left behind is stuff that is stuck in the fine etch grooves or
is still weakly attached by a dangling atomic bond.  The acid makes short
work of anything that is not strongly attached to its cystalline neighbor.

Dennis AE6C

On Fri, Aug 13, 2010 at 9:17 AM, <WA5CAB at cs.com> wrote:

> The calibration crystals are in holders DC-8-A or ARC #7785.  The
> photograph of the O-4/ARC-5 Test Oscillator in the AN/ARC-5 Maintenance
> Manual shows
> it with two of these holders installed.  The instructions later in the
> manual say to use any ARC #77785 holder with crystal in the range of 1 to
> 10 MC.
>
>
> This has nothing to do with the Command Set calibration crystals but if,
> while testing FT-243 crystals (in a GDO or otherwise) you find a totally
> dead
> one, and it does not appear to have ever been opened since new, the odds
> approach 1:1 that the failure problem is that one or both of the soldering
> legs
> on the brass contact plates have broken.  Repair is preferably by
> replacement of the broken plate, as they almost always break at the point
> where the
> leg becomes the plate.  And a solder-joint bump at that point is not
> tolerable.  If you have a lot of other FT-243 holders on "uninteresting"
> frequencies, you can repair by salvage.  However, there are several
> different shape
> plates as almost every contractor made theirs a little differently.  And
> due to
> the varying shapes of the rest of the contacting parts (spring, rubber
> gasket, etc.) they don't all substitute successfully.  So the best solution
> (time wise) if you have one is to open another holder made by the same
> contractor as the one that you want to repair.
>
> In a message dated 8/13/2010 10:40:01 AM Central Daylight Time,
>  dennis.monticelli at gmail.com writes:
> > It's just as Henry says; You connect it to directly to the coil socket
> > with
> > short leads.  Any GDO relies upon an external inductor which is all a
> > crystal is electrically when operating in its parallel resonance mode
> > (less
> > the DC path which the GDO doesn't care about).  Of course the crystal
> > represents a lot more inductance than the coils and likewise the tuning
> > capacitance of the GDO represents more shunt capacitance than the crystal
> > is
> > likely to see in a real circuit.  But it still oscillates readily and you
> > can spin the tuning dial while watching the meter to maximize the
> reading.
> > You won't find a peak per se, but usually the oscillation gets stronger
> at
> > one end of the dial or the other.
> >
> > The post-Millen GDO's started using smaller coil sets whose pin dia and
> > spacing match that of an FT-243 so it's particularly convenient to use
> one
> > of those units to check crystals.  I have built small adaptors that adapt
> > other crystal holders to the FT-243 both for use in equipment and for use
> > with my GDO.  The smaller transistorized GDO's are particularly useful to
> > have in your bag at the hamfest, flea market, or surplus store. I picked
> > up
> > just such a unit recently very cheaply because it was missing all its
> > coils.  It won't "GDO" but it makes a dandy portable crystal checker.  In
> > fact, that's what I used to check the crystal in my ARC-5.
> >
> > BTW, if you leave the sensitivity control at one setting, you can test a
> > whole batch of crystals at one time for relative activity.  I do this to
> > identify sluggish crystals.  The more active the crystal, the higher the
> > meter reading.
> >
> > Dennis AE6C
> >
> >
>
> Robert & Susan Downs - Houston
> wa5cab dot com (Web Store)
> MVPA 9480
>  ______________________________________________________________
> 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
>