[time-nuts] Glass Envelope Quartz Crystals

Bernd Neubig BNeubig at t-online.de
Wed Feb 3 11:51:14 EST 2016



Von: time-nuts [mailto:time-nuts-bounces at febo.com] Im Auftrag von Bob Camp
Gesendet: Mittwoch, 3. Februar 2016 13:12
 >There likely are long stories that explain just why this or that package got used in this or that application. 

The historical reasons for this type of package is more related to the production technology than to a particular application.
In the olde days, before cold-welding technology came up, glass-encapsulation was the packaging of choice to achieve high stability crystals. Because of the high temperatures associated with the melting during the sealing, all humidity, moisture and "dirt" was burned away. In addition the vacuum inside contributes for high stability and high Q
The sealing of glass tubes was a well-established technology for vacuum tubes, and could be adapted straight-forward by the crystal manufacturers.
Therefore most of these tube-style crystal packages had an "Octal" or "Noval" type socket as used for vacuum tubes.

To complete the story:
Later the all-glass enclosures HC-27 and HC-26 came up, which have the same outer dimensions and pin spacing as the old HC-6/U and HC-18/U metal enclosures. The sealing technology was as follows:
There was a ring of Kovar embedded in the upper side of the header. The glass-cover was positioned on top of this Kovar ring.   The output power ( around 1 kW) of a RF (e.g. 13.56 MHz) transmitter was inductively coupled through a coil which was arranged around the crystal header, so the Kovar ring was forming the secondary winding of this RF transformer. Through RF induction the Kovar ring was melting and with it the edge of the glass cover and of the header. The whole thing happened under a vacuum bell, and the sealing process was accompanied by water cooling of the pins and parts of the header to avoid overheating of the crystal during the melting process (its temperature must be maintained well below the Curie temperature (573°C). The degree of overheating could be seen afterwards by you measuring the resonance spectrum:
At about 1.5 times the nominal frequency you could see resonances from the BT-cut mode, caused by partial twinning of the crystal structure ...

With this I will close the crystal history book for today.

Best regards

Bernd
DK1AG



More information about the time-nuts mailing list