Re: [time-nuts] Super stable BVA Quartz resonators... BVA??
BNeubig@t-online.de said:
The key points which yield the BVA's improved aging are, that - the
whole resonator package is made from quartz, consisting of the
resonationg quartz plate in the middle section and the two mounting &
sealing plaates on top and bottem - you may call it a Hamburger style
;-) - the resonating plate is held through quartz bridges rather than
metal suspensions, thus reducing mounting stress.
Neat. Thanks.
Why is the top domed? I assume flat would be easier to manufacture.
Why is it not symmetrical? If the top is domed, why not the bottom too?
--
These are my opinions, not necessarily my employer's. I hate spam.
Hal Murray wrote:
Why is the top domed? I assume flat would be easier to manufacture.
Why is it not symmetrical? If the top is domed, why not the bottom too?
For a plano-parallel qaurtz resonator the diameter must be at least 60
times larger than the thickness, otherwise the vibration amplitude would
be too high at the edges, and the suspension would damp teh vibration
hence the Q would be lowered. For a 10 MHz 3rd overtone SC-cut you would
need a crystal of 40 to 50 mm diameter!
By making a convex contour, the vibration is trapped in the center of
the plate, and therefore the Q keeps high (and other parameters improve
also).
Bi-convex contours are more difficult to manufacture, as it is required
that the symmetry axis of the upper and lower contour must coincide.
Also other parameters become worse. For the BVA the manufacturability
would also be much worse as it already is.
Regards
Bernd
DK1AG
Hal Murray wrote:
BNeubig@t-online.de said:
The key points which yield the BVA's improved aging are, that - the
whole resonator package is made from quartz, consisting of the
resonationg quartz plate in the middle section and the two mounting &
sealing plaates on top and bottem - you may call it a Hamburger style
;-) - the resonating plate is held through quartz bridges rather than
metal suspensions, thus reducing mounting stress.
Neat. Thanks.
Why is the top domed? I assume flat would be easier to manufacture.
Why is it not symmetrical? If the top is domed, why not the bottom too?
One surface is convex to facilitate suppression of unwanted harmonics.
Its actually quicker and easier to make a planoconvex "lens" than a
biconvex one.
Bruce
Bernd T-Online wrote:
Hal Murray wrote:
Why is the top domed? I assume flat would be easier to manufacture.
Why is it not symmetrical? If the top is domed, why not the bottom too?
For a plano-parallel qaurtz resonator the diameter must be at least 60
times larger than the thickness, otherwise the vibration amplitude would
be too high at the edges, and the suspension would damp teh vibration
hence the Q would be lowered. For a 10 MHz 3rd overtone SC-cut you would
need a crystal of 40 to 50 mm diameter!
By making a convex contour, the vibration is trapped in the center of
the plate, and therefore the Q keeps high (and other parameters improve
also).
Bi-convex contours are more difficult to manufacture, as it is required
that the symmetry axis of the upper and lower contour must coincide.
That problem was solved over a century ago in optical lens manufacture.
If the two surfaces are spherical, then such decentering is equivalent
to adding a wedge, which is easily removed by optical centering and
edging techniques.
Also other parameters become worse. For the BVA the manufacturability
would also be much worse as it already is.
Regards
Bernd
DK1AG
Bruce
From: Bernd T-Online BNeubig@t-online.de
Subject: Re: [time-nuts] Super stable BVA Quartz resonators... BVA??
Date: Sat, 08 Dec 2007 09:27:28 +0100
Message-ID: 475A5570.2010005@t-online.de
Hal Murray wrote:
Why is the top domed? I assume flat would be easier to manufacture.
Why is it not symmetrical? If the top is domed, why not the bottom too?
For a plano-parallel qaurtz resonator the diameter must be at least 60
times larger than the thickness, otherwise the vibration amplitude would
be too high at the edges, and the suspension would damp teh vibration
hence the Q would be lowered. For a 10 MHz 3rd overtone SC-cut you would
need a crystal of 40 to 50 mm diameter!
By making a convex contour, the vibration is trapped in the center of
the plate, and therefore the Q keeps high (and other parameters improve
also).
Bi-convex contours are more difficult to manufacture, as it is required
that the symmetry axis of the upper and lower contour must coincide.
Also other parameters become worse. For the BVA the manufacturability
would also be much worse as it already is.
While I higly enjoy reading this thread and learning things as I read on, I
notice that several fellow time-nuts have designer skills at advanced SC-level
OCXOs. For those of you, what reading material is there to get better into
depths with these issues? The only sufficiently indepth book I have is Gerber,
Ballato "Precision Frequency Control, Volume 1; Acoustic Resonators and
Filters" (I also got Volume II).
Cheers,
Magnus
Bernd T-Online wrote:
Hal Murray wrote:
Why is the top domed? I assume flat would be easier to manufacture.
Why is it not symmetrical? If the top is domed, why not the bottom too?
For a plano-parallel qaurtz resonator the diameter must be at least 60
times larger than the thickness, otherwise the vibration amplitude would
be too high at the edges, and the suspension would damp teh vibration
hence the Q would be lowered. For a 10 MHz 3rd overtone SC-cut you would
need a crystal of 40 to 50 mm diameter!
By making a convex contour, the vibration is trapped in the center of
the plate, and therefore the Q keeps high (and other parameters improve
also).
Bi-convex contours are more difficult to manufacture, as it is required
that the symmetry axis of the upper and lower contour must coincide.
Also other parameters become worse. For the BVA the manufacturability
would also be much worse as it already is.
Regards
Bernd
DK1AG
Bernd
Whilst traditional optical edging techniques can easily remove any wedge
when both surfaces are convex, maintaining alignment of the "lens" axis
with respect to the crystal axes is another matter, at least when using
traditional surfacing techniques, with computer controlled surfacing
techniques even this can be done.
With a planoconvex "lens" bias polishing or an equivalent technique can
be used to adjust the inclination of the plano surface with respect to
the crystal axes and this alignment is maintained during edging leaving
only axial thickness adjustments to be made. The etching process used to
remove cracks and defects after mechanical polishing is anisotropic
which may introduce further complications in maintaining alignment and
shape.
Bruce
Bruce Griffiths wrote:
Bernd T-Online wrote:
Bi-convex contours are more difficult to manufacture, as it is required
that the symmetry axis of the upper and lower contour must coincide.
Also other parameters become worse. For the BVA the manufacturability
would also be much worse as it already is.
Regards
Bernd
DK1AG
Bernd
Whilst traditional optical edging techniques can easily remove any wedge
when both surfaces are convex, maintaining alignment of the "lens" axis
with respect to the crystal axes is another matter, at least when using
traditional surfacing techniques, with computer controlled surfacing
techniques even this can be done.
With a planoconvex "lens" bias polishing or an equivalent technique can
be used to adjust the inclination of the plano surface with respect to
the crystal axes and this alignment is maintained during edging leaving
only axial thickness adjustments to be made. The etching process used to
remove cracks and defects after mechanical polishing is anisotropic
which may introduce further complications in maintaining alignment and
shape.
Bruce
So Bruce, what source did you read that led you to change your original
answer from --it's easy--, to --it's difficult--? (reference your quoted
text below:
Bi-convex contours are more difficult to manufacture, as it is required
that the symmetry axis of the upper and lower contour must coincide.
That problem was solved over a century ago in optical lens manufacture.
If the two surfaces are spherical, then such decentering is equivalent
to adding a wedge, which is easily removed by optical centering and
edging techniques.
Also other parameters become worse. For the BVA the manufacturability
would also be much worse as it already is.
-Chuck Harris
Chuck Harris wrote:
Bruce Griffiths wrote:
Bernd T-Online wrote:
Bi-convex contours are more difficult to manufacture, as it is required
that the symmetry axis of the upper and lower contour must coincide.
Also other parameters become worse. For the BVA the manufacturability
would also be much worse as it already is.
Regards
Bernd
DK1AG
Bernd
Whilst traditional optical edging techniques can easily remove any wedge
when both surfaces are convex, maintaining alignment of the "lens" axis
with respect to the crystal axes is another matter, at least when using
traditional surfacing techniques, with computer controlled surfacing
techniques even this can be done.
With a planoconvex "lens" bias polishing or an equivalent technique can
be used to adjust the inclination of the plano surface with respect to
the crystal axes and this alignment is maintained during edging leaving
only axial thickness adjustments to be made. The etching process used to
remove cracks and defects after mechanical polishing is anisotropic
which may introduce further complications in maintaining alignment and
shape.
Bruce
So Bruce, what source did you read that led you to change your original
answer from --it's easy--, to --it's difficult--? (reference your quoted
text below:
No reference (unless you want references to optical manufacturing
techniques), its obvious when you stop to think about it, the various
crystal cuts (AT, BT, SC etc) require a specific crystallographic
orientation, changing the orientation changes the characteristics.
When I realised that Bernd hadn't explicitly stated the requirement to
maintain crystallographic alignment when contouring the 2 surfaces the
real cause of the difficulty of manufacture became obvious.
Introducing wedge when generating the spherical (other contours are much
more difficult to produce unless one uses computer controlled equipment
together with optical shape measurement) surfaces is geometrically
equivalent to changing the crystallographic orientation of the blank.
This doesnt arise when imparting a spherical contour to only one
surface, as the plane surface defines the crystallographic orientation
before and after removing any wedge using optical centering/edging
techniques.
The original statement merely indicated that well known techniques
provide a solution to his actual statement of the problem of centering
the 2 surfaces.
Bi-convex contours are more difficult to manufacture, as it is required
that the symmetry axis of the upper and lower contour must coincide.
That problem was solved over a century ago in optical lens manufacture.
If the two surfaces are spherical, then such decentering is equivalent
to adding a wedge, which is easily removed by optical centering and
edging techniques.
Also other parameters become worse. For the BVA the manufacturability
would also be much worse as it already is.
-Chuck Harris