[time-nuts] Re: Crystal sweet-spot (was: Best frequency to start for GHz synth ?)

[Alan Melia]

Bob somewhere buried in my collection of interesting bits I have a rather 
battered demo sample of what the British GPO Research crystal labs refered 
to as an Essen Ring. It is indeed around 3 inches in outer diameter and 
almost an inch thick and wide. this would have been cut from natural quartz. 
I suspect this specimen is a 'failed' sample. The ring is suspended in a 
couple of silk threads.I also have what I think is a higher frequency ring 
which is mouned in a 1.5inch diameter evacuated glass holder.
I was told that if the large ring was tapped gently with a pencil it would 
'ring' for 5 minutes (that there may be some exageration  there) I believe 
Qs of 5E^6 were mentioned. More reliable info would be found in the reports 
:-))  This was around 60 years ago.

Alan
G3NYK

----- Original Message ----- 
From: "Bob kb8tq" <kb8tq at n1k.org>
To: "Discussion of precise time and frequency measurement" 
<time-nuts at lists.febo.com>
Sent: Monday, April 05, 2021 9:43 PM
Subject: [time-nuts] Re: Crystal sweet-spot (was: Best frequency to start 
for GHz synth ?)


> Hi
>
> Back it the “old days” ( so 1960’s in this case ) glass packages
> were very commonly used for precision crystals. They were available
> in large diameters ( think of transmitting sized vacuum tubes). This 
> allowed
> use of larger diameter blanks than what fit in today’s much smaller 
> packages.
>
> The result was that things like 2.5 MHz fifth overtone parts could be 
> made.
> Cute things like silk thread supports for the blank were not uncommon. 
> Yields
> simply due to the “thread tweaking” process often ran in the 10 to 15% 
> range
> (as in 8 or 9 out of ten failed …) .
>
> Since there was no way to re-do the process once the part was under
> vacuum ( and no way to test it before that) this was indeed black magic.
> Occasionally somebody would do a batch and 25% would work. They
> would then talk about that event for at least the next 20 years ….
>
> One would *not* want to go back and do it the “good old way”.
>
> Bob
>
>> On Apr 5, 2021, at 4:02 PM, Attila Kinali <attila at kinali.ch> wrote:
>>
>> On Wed, 31 Mar 2021 16:44:40 +0000
>> "Poul-Henning Kamp" <phk at phk.freebsd.dk> wrote:
>>
>>> It used to be that 5MHz was the "hot spot" for crystals on the
>>> parameters we care about as time-nuts.
>>
>> Depends on what kind of time-nut you are ;-)
>>
>> If you are going for high frequencies beyond 1GHz, then a mesa
>> type high frequency BAW is the best you can do (given you don't
>> want to use frequency comb to divide down a cryogenic silicon
>> cavity).
>>
>> As Bob wrote, for low-frequency, high stability applications,
>> the lower the frequency of the crystal the better. Or rather,
>> the thicker the crystal the better. I.e. you want to use an
>> as low frequency crystal with an as high as possible overtone.
>> Unfortunately, to make full use of the properties of low
>> frequency crystals, you need to scale the diameter with the
>> frequency. Otherwise, the energy loss due to the edges of the
>> crytal will limit the Q.
>>
>> For historical reasons, 3rd overtone 5MHz turned out to be the
>> lowest that could be done economically with the avaible tools
>> and methods and still fit the size constraints.
>>
>> Today we could probaly go lower, but the market demands for
>> large crystal units is shrinking steadily and, as Bob wrote
>> a few times in the past, nobody has the tooling to do so.
>>
>> Attila Kinali
>> -- 
>> The driving force behind research is the question: "Why?"
>> There are things we don't understand and things we always
>> wonder about. And that's why we do research.
>> -- Kobayashi Makoto
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at lists.febo.com -- To unsubscribe send 
>> an email to time-nuts-leave at lists.febo.com
>> To unsubscribe, go to and follow the instructions there.
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com -- To unsubscribe send 
> an email to time-nuts-leave at lists.febo.com
> To unsubscribe, go to and follow the instructions there.