Quartz Crystal Specification Confusion.

[Jim Flanagan]

Hi..
I would appreciate collaboration from some of you who are familiar
with specifying quartz crystals. My question revolves around
the series versus parallel mode of operation.

I understand the difference between series and parallel (or so I think).
The following article makes some statements that seem to contradict what
I understand- www.crystek.com/appnotes/VCXOarticle.pdf

In the article the following statement is made:

"This is tricky and is misstated in many papers. The capacitive load is
always placed effectively in series with the crystal and never in
parallel. In other words, a so-called parallel crystal does not mean you
place a capacitor in parallel with the crystal, but in series."

Am I reading this properly? My understanding is that when when
specifying a xtal for use in the parallel mode, the crystal is
calibrated against a parallel cap. A crystal calibrated as 'series'
uses no capacitor in either a series or parallel with the xtal.
When a crystal is specified as 'series' the intent is that the crystal
frequency desired is the fs frequency, whereas a crystal specified with
a load is used in the anti-resonance mode.

The article talks about using a crystal in a VCXO circuit, where a
varactor is used in series to slightly 'pull' the motional C (Cm) to
effectively change fs.

Anyway, I am a bit confused by this article. Would anyone please share
their expertise with me? What I am doing is trying to determine how
to specify some crystals for use with in a crystal filter. I would like
to use some generic 10 MHz microprocessor crystals and am not sure
whether to buy some series or 20-30pf load caps.


I am not sure whether I have properly expressed my confusion, but in a
nutshell the article seems to indicate that the calibration is done
with a series cap. This would 'pull' the fs into the 'parallel '
frequency range. My current understanding was that the calibration
was done with a parallel cap, looking for a parallel frequency
calibration. Your help would be appreciated.

Thanks..

-jim WB5KYE

 

[Tim Wescott]

Hi..
I would appreciate collaboration from some of you who are familiar
with specifying quartz crystals. My question revolves around
the series versus parallel mode of operation.

I understand the difference between series and parallel (or so I think).
The following article makes some statements that seem to contradict what
I understand- www.crystek.com/appnotes/VCXOarticle.pdf

In the article the following statement is made:

"This is tricky and is misstated in many papers. The capacitive load is
always placed effectively in series with the crystal and never in
parallel. In other words, a so-called parallel crystal does not mean you
place a capacitor in parallel with the crystal, but in series."

Am I reading this properly? My understanding is that when when
specifying a xtal for use in the parallel mode, the crystal is
calibrated against a parallel cap. A crystal calibrated as 'series'
uses no capacitor in either a series or parallel with the xtal.
When a crystal is specified as 'series' the intent is that the crystal
frequency desired is the fs frequency, whereas a crystal specified with
a load is used in the anti-resonance mode.

The article talks about using a crystal in a VCXO circuit, where a
varactor is used in series to slightly 'pull' the motional C (Cm) to
effectively change fs.

Anyway, I am a bit confused by this article. Would anyone please share
their expertise with me? What I am doing is trying to determine how
to specify some crystals for use with in a crystal filter. I would like
to use some generic 10 MHz microprocessor crystals and am not sure
whether to buy some series or 20-30pf load caps.


I am not sure whether I have properly expressed my confusion, but in a
nutshell the article seems to indicate that the calibration is done
with a series cap. This would 'pull' the fs into the 'parallel '
frequency range. My current understanding was that the calibration
was done with a parallel cap, looking for a parallel frequency
calibration. Your help would be appreciated.

Thanks..

-jim WB5KYE

A crystal looks like a series LC in parallel with a capacitor. The
series LC characteristics come from the chuck of quartz, while the
parallel capacitance comes from the metalization and case.

When it's operated in series resonant mode the only real determinant is
the series LC part -- so in theory things are nice and stable. When
it's operated in the parallel resonant mode the series LC part must
supply some inductive reactance to counter the capacitive reactance of
the external crystals. As this external capacitance (and the case
capacitance) changes, the resonant frequency changes also.

So at resonance the crystal just looks like an LC circuit, with the
crystal acting as the inductor and the external capacitance acting as
the capacitor.

This was all a very long way to lead up to the following: Consider two
circuits. Circuit A is a series-resonant LC circuit, working into a
very low impedance -- what is it's resonant frequency? Circuit B is a
parallel-resonant LC circuit, working into a very high impedance -- what
is _it's_ resonant frequency?

.-------o-----o
| |
--- |
C --- |
| .-. R =
| | | really low .-------o--------o------o
| | | | | |
| '-' | | .-.
C| | C| --- | | R =
L C| | L C| C --- | | really high
C| | C| | '-'
| | | |
| | | | |
=== === === === ===
GND GND GND GND GND


Circuit A Circuit B
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

Next question: Does it matter whether the author thinks the circuit is
"really" in series or parallel?

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google? See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html

 

[Genome]

Hi..
I would appreciate collaboration from some of you who are familiar
with specifying quartz crystals. My question revolves around
the series versus parallel mode of operation.


Thanks..

-jim WB5KYE

I am probably wrong but, just to confuse the issue, one thing you might want
to think about is wether you are dealing with a series or parallel 'cut'
crystal. That refers to the way the crystal is dealt with mechanically with
hugely massive whirling blades bathed in blood for cooling that splashes all
over the place as they grind grindingly through the very body of the pure
crystal of life.

All horribly horror show.

Anyway, I'm not sure about this, but one way after you metalllise the ends,
after treatment with the acidic, in boiling vats of viciously hot metal that
will scar ten thouasand fanatics of the ancient days such that they will bow
down to the commands as given by those who are truly worthy...

Cough

One way, the houses are lined up so you excite the streets and the other way
they are lined up so you jump up and down on the roofs. Yea verily such is
the way that the pure can be selected or we can produce overtones. Such it
was moted, let it moted be.

Let us pray.

DNA

 

[colin]

Hi..
I would appreciate collaboration from some of you who are familiar
with specifying quartz crystals. My question revolves around
the series versus parallel mode of operation.

I understand the difference between series and parallel (or so I think).
The following article makes some statements that seem to contradict what
I understand- www.crystek.com/appnotes/VCXOarticle.pdf

In the article the following statement is made:

"This is tricky and is misstated in many papers. The capacitive load is
always placed effectively in series with the crystal and never in
parallel. In other words, a so-called parallel crystal does not mean you
place a capacitor in parallel with the crystal, but in series."

Am I reading this properly? My understanding is that when when
specifying a xtal for use in the parallel mode, the crystal is
calibrated against a parallel cap. A crystal calibrated as 'series'
uses no capacitor in either a series or parallel with the xtal.
When a crystal is specified as 'series' the intent is that the crystal
frequency desired is the fs frequency, whereas a crystal specified with
a load is used in the anti-resonance mode.

The article talks about using a crystal in a VCXO circuit, where a
varactor is used in series to slightly 'pull' the motional C (Cm) to
effectively change fs.

Anyway, I am a bit confused by this article. Would anyone please share
their expertise with me? What I am doing is trying to determine how
to specify some crystals for use with in a crystal filter. I would like
to use some generic 10 MHz microprocessor crystals and am not sure
whether to buy some series or 20-30pf load caps.


I am not sure whether I have properly expressed my confusion, but in a
nutshell the article seems to indicate that the calibration is done
with a series cap. This would 'pull' the fs into the 'parallel '
frequency range. My current understanding was that the calibration
was done with a parallel cap, looking for a parallel frequency
calibration. Your help would be appreciated.

The crystal looks like a series LC resonator, therefore the external
capacitance is always in series with the internal motional capacitance (wich
is extremly small), even though it maybe in parallel with the crystal
itself.

Colin =^.^=

 

[John Popelish]

www.crystek.com/appnotes/VCXOarticle.pdf

In the article the following statement is made:

"This is tricky and is misstated in many papers. The capacitive load is
always placed effectively in series with the crystal and never in
parallel. In other words, a so-called parallel crystal does not mean you
place a capacitor in parallel with the crystal, but in series."

Picture any two terminal device. How do you connect another two
terminals device to it, without the result being seen as either a pair
of devices in parallel, or a pair of devices in series? The terms
series and parallel, in this case, refer to the resonance mode.
Series resonance hits minimum and real impedance at resonance, while
parallel resonance hits real and maximum impedance, when in resonance.
A crystal is a series resonator, all by itself. You must connect
capacitance, effectively across it to produce parallel resonance
across the crystal.

Am I reading this properly? My understanding is that when when
specifying a xtal for use in the parallel mode, the crystal is
calibrated against a parallel cap.

Yes, and with a specific value of capacitance, to achieve the
specified resonant frequency.

A crystal calibrated as 'series'
uses no capacitor in either a series or parallel with the xtal.

Well, none small enough to affect the resonant frequency, which is
based on a mechanical capacitance it the femto farad range.

When a crystal is specified as 'series' the intent is that the crystal
frequency desired is the fs frequency,

Yes. It is intended to be operated so that it achieves minimum real
impedance.

whereas a crystal specified with
a load is used in the anti-resonance mode.

Yes. Intended to operate as a parallel resonance, that achieves
maximum and real impedance at resonance.

The article talks about using a crystal in a VCXO circuit, where a
varactor is used in series to slightly 'pull' the motional C (Cm) to
effectively change fs.

The varactor acts as a variable parallel capacitor, as far as the
resonance goes. It may be in series with some larger capacitance to
close the loop around the crystal.

Anyway, I am a bit confused by this article. Would anyone please share
their expertise with me? What I am doing is trying to determine how
to specify some crystals for use with in a crystal filter. I would like
to use some generic 10 MHz microprocessor crystals and am not sure
whether to buy some series or 20-30pf load caps.

If you want the filter to have a tunable resonance near the crystal's
specified resonant frequency, then you need one specified for parallel
resonance, so that you can trim the exact resonance on either side of
that frequency, with a variable capacitance that varies through the
specified load capacitance.

 

[Jim Flanagan]

The crystal looks like a series LC resonator, therefore the external
capacitance is always in series with the internal motional capacitance (wich
is extremly small), even though it maybe in parallel with the crystal
itself.

Colin =^.^=

Yes... of course!!

'If it had been a snake, it would have bitten me..'

Thanks. Sometimes the obvious is hard to see for some of us.
Jim WB5KYE

 

[Ancient_Hacker]

There's a heck of a lot of confusion about this.

A crystal is a crystal is a crystal. It has no intrinsic "series" or
"parallel" flavor.

If you hook a crystal up between a signal generator and a voltmeter,
and slowly and carefully tune the generator across the band, you'll
notice a peak, that's where the crystal is acting a a series-tuned
circuit, presenting a low impedance at a very narrow frequency band
(often just cycles wide), then a little ways off, the voltmeter goes
to darn near zero over another narrow band-- there the crystal is
acting as a very high impedance, a parallel resonance.

To get the crystal to osciallte in its series mode, you have to put it
between the output and the input of an amplifier-- feeding back
CURRENT, which will preferentially happen at it's low impedance point,
its series resonance. If you see a crystal with both ends above
ground, that's usually a series-resonance oscillator circuit.


To get the crystal to oscillate in its parallel mode, you have to put
it between the output and the input of an amplifier-- feeding back
VOLTAGE , which will preferentially happen at it's HIGH impedance
point, its parallel resonance. If you see a circuit where the crystal
has one side at ground, that's usually a parallel oscillation circuit.