[SI-LIST] Re: follow up: Re: Query about Oscillator problem
- From: Richard Jungert <r_jungert@xxxxxxxxxxx>
- To: ray anderson <ray.anderson@xxxxxxxxxx>, <r_jungert@xxxxxxxxxxx>, steve weir <weirsi@xxxxxxxxxx>, <tlx6f@xxxxxxx>, <chundis@xxxxxxxxx>, si list freelist <si-list@xxxxxxxxxxxxx>
- Date: Mon, 1 Feb 2010 15:39:07 -0800
Its in this app note. What I am trying to say is that the bias type in the
oscillator determines if it must run with parallel or series resonance. It
would not make sense if the crystal is switching between resonance modes.
http://www.chrontel.com/pdf/an06.pdf
Some oscillator circuits are designed for series resonance and the oscillation
frequency shall equal the
specified series resonance value. These series mode oscillators, however, are
more sensitive to temperature
and component variations. In fact, most crystals oscillators in today's ICs are
of the parallel resonance type.
The oscillation frequency of a parallel mode oscillator is always higher than
fseries. The actual oscillation
frequency of a parallel mode oscillator is dependent on the equivalent
capacitance seen by the crystal.
here is another good point.
At parallel resonance, the crystal behaves inductively and resonates with
capacitance shunting the crystal
terminals. Depending on the application, especially in microprocessors where
Pierce oscillators are used
predominantly, a crystal manufacturer may specify parallel resonance frequency
instead of series resonance
frequency. Since fparallel is a function of the load capacitance Ceq, it should
also be specified along with
fparallel.
For PC CPU clock and VGA clock applications, the frequency accuracy required is
usually not very stringent
and can easily be satisfied with a 14.318 MHz crystal that has been specified
for operation in either series or
parallel resonance modes.
REFERENCES
Intel Application Note AP-155, Oscillators for Microcontrollers, 1983.
> Subject: [SI-LIST] Re: follow up: Re: Query about Oscillator problem
> Date: Mon, 1 Feb 2010 15:09:33 -0800
> From: ray.anderson@xxxxxxxxxx
> To: r_jungert@xxxxxxxxxxx; weirsi@xxxxxxxxxx; tlx6f@xxxxxxx;
> chundis@xxxxxxxxx; si-list@xxxxxxxxxxxxx
> CC: raya@xxxxxxxxxx
>
> Richard-
>
>
> That might be true if there were physical differences between "series"
> and "parallel" resonant crystals, but there aren't. From the Maxim app
> note you referred to:
>
>
> Resonance Mode
>
>
> Crystals have two modes of resonance, parallel and series, and all
> crystals exhibit both resonance modes. The oscillator circuit is
> calibrated for one mode or the other, but not both. For applications
> requiring no tighter than 100ppm frequency accuracy, resonance mode is
> usually not an issue. If you are attempting to control frequency (or
> time) to within 100ppm, however, the resonance-mode specification
> becomes important. The crystal vendor's principal concern is in which
> mode the crystal is calibrated during manufacturing. Knowing that
> information, the crystal vendor then sets up an oscillator circuit with
> the crystal in a customer-specified series resonance or parallel
> resonance, and calibrates the crystal. Figure 2 shows crystal impedance
> behavior versus frequency, and the relative location of each resonance
> mode.
>
>
>
>
>
> There are various "cut's" that are used in crystal manufacturing.
> Depending on the angle of the cut with respect to the lattice structure
> the vendor can set the inflection point of the tempco curve such that
> crystal frequency change wrt to temp is relatively small (or not). There
> are probably other factors that effect the temperature coefficient of a
> quartz resonator, but the cut angle is a first order effect. As far as I
> know the calibration frequency has little or no effect. (haven't seen
> the Intel app note you referred to. Which one is it?)
>
>
>
> -Ray
>
>
>
> ________________________________
>
> From: Richard Jungert [mailto:r_jungert@xxxxxxxxxxx]
> Sent: Monday, February 01, 2010 2:50 PM
> To: Ray Anderson; steve weir; tlx6f@xxxxxxx; chundis@xxxxxxxxx; si list
> freelist
> Subject: RE: [SI-LIST] Re: follow up: Re: Query about Oscillator problem
>
>
>
> Here is a great app note on how to specify a crystal.
>
> http://www.maxim-ic.com/app-notes/index.mvp/id/726
>
> According to Intel app note I just looked at the series resonant ones
> are more sensitive to temp variation than parallel types.
>
>
> Richard
>
>
> > Subject: RE: [SI-LIST] Re: follow up: Re: Query about Oscillator
> problem
> > Date: Mon, 1 Feb 2010 14:19:47 -0800
> > From: ray.anderson@xxxxxxxxxx
> > To: r_jungert@xxxxxxxxxxx; weirsi@xxxxxxxxxx; tlx6f@xxxxxxx;
> chundis@xxxxxxxxx; si-list@xxxxxxxxxxxxx
> > CC: raya@xxxxxxxxxx
> >
> > The terminology of "series" or "parallel" resonance when speaking of
> > quartz crystal resonators refers to the resonance mode where
> calibration
> > was established.
> >
> > Physically there is no difference between a series and parallel
> resonant
> > crystal. The type of resonance utilized in the oscillator has to do
> with
> > the oscillator topology. Some oscillators require the crystal to
> exhibit
> > a low impedance at resonance (the series mode), whereas other
> oscillator
> > types require a high impedance (parallel mode) to sustain oscillation.
> >
> > Using a "parallel" resonant crystal in a "series" resonant oscillator
> > will result in a slightly different frequency of oscillation than the
> > crystal vendor intended, but it shouldn't materially affect the
> > oscillator functionality.
> >
> > -Ray Anderson
> > Xilinx Inc.
> >
> > > -----Original Message-----
> > > From: si-list-bounce@xxxxxxxxxxxxx
> > [mailto:si-list-bounce@xxxxxxxxxxxxx]
> > > On Behalf Of Richard Jungert
> > > Sent: Monday, February 01, 2010 1:49 PM
> > > To: steve weir; tlx6f@xxxxxxx; chundis@xxxxxxxxx; si list freelist
> > > Subject: [SI-LIST] Re: follow up: Re: Query about Oscillator problem
> > >
> > > Guys
> > >
> > > It may very well be that the series resonant quartz won't work and
> > that
> > > you may need to get a parallel resonant XTAL. Series resonant can
> get
> > you
> > > into trouble. I would go with the crystal type/mfg that Motorola
> > > recommends.
> > >
> > > I would bet dollars to doughnuts that the resonant type is the issue
> > here.
> > >
> > > Richard Jungert
> > >
> > >
> > >
> > >
> > >
> > > > Date: Sun, 31 Jan 2010 15:20:46 -0800
> > > > From: weirsi@xxxxxxxxxx
> > > > To: tlx6f@xxxxxxx
> > > > CC: chundis@xxxxxxxxx; si-list@xxxxxxxxxxxxx
> > > > Subject: [SI-LIST] Re: follow up: Re: Query about Oscillator
> problem
> > > >
> > > > Li, that is not correct. NPO capacitors are high stability and
> high
> > Q.
> > > > They are great for the crystal tuning caps where both
> > characteristics
> > > > are a plus. But the high Q can actually create unwanted resonance
> > > > problems in the PLL power filter.
> > > >
> > > > Steve.
> > > > Li, Tianqi . (S&T-Student) wrote:
> > > > > Sree,
> > > > >
> > > > > I forget sth. You need to check the PLL filtering capacitors of
> > > MCF5212.
> > > > > In most cases CPU will leave out 2 or 3 power pins for filtering
> > PLL
> > > > > noise. Capacitors on these pins should also use NPO for a better
> > > > > performance.
> > > > >
> > > > > And you also need to find at what value of Vpp of XTAL droped to
> > the
> > > PLL
> > > > > start to lose locking. If the value is out of MCF5212 spec, the
> > root
> > > > > cause is surely on XTAL tank circuit; If not, it might be XTAL's
> > > glitch
> > > > > aroused. You can use scope to test the eye diagram and find the
> > > glitch.
> > > > >
> > > > > Tianqi Li
> > > > >
> >
> > This email and any attachments are intended for the sole use of the
> named recipient(s) and contain(s) confidential information that may be
> proprietary, privileged or copyrighted under applicable law. If you are
> not the intended recipient, do not read, copy, or forward this email
> message or any attachments. Delete this email message and any
> attachments immediately.
> >
> >
>
> ________________________________
>
> Hotmail: Free, trusted and rich email service. Get it now.
> <http://clk.atdmt.com/GBL/go/201469228/direct/01/>
>
>
>
> This email and any attachments are intended for the sole use of the named
> recipient(s) and contain(s) confidential information that may be proprietary,
> privileged or copyrighted under applicable law. If you are not the intended
> recipient, do not read, copy, or forward this email message or any
> attachments. Delete this email message and any attachments immediately.
>
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>
> List technical documents are available at:
> http://www.si-list.net
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
>
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
_________________________________________________________________
Hotmail: Free, trusted and rich email service.
http://clk.atdmt.com/GBL/go/201469228/direct/01/
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List technical documents are available at:
http://www.si-list.net
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
