Searching \ for '[EE]: Oscillator Testing' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/timers.htm?key=oscillator
Search entire site for: 'Oscillator Testing'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Oscillator Testing'
2007\01\17@163334 by Mike Hord

picon face
Does anyone know of a company which will do a
comprehensive set of tests on crystal oscillator
units to diagnose possible failure causes?

Mike H.

2007\01\21@092721 by Dave Lag

picon face
Mike Hord wrote:
> Does anyone know of a company which will do a
> comprehensive set of tests on crystal oscillator
> units to diagnose possible failure causes?
>
> Mike H.

I would ask these guys?

http://www.wenzel.com/

2007\01\21@093908 by Vasile Surducan

face picon face
On 1/17/07, Mike Hord <spam_OUTmike.hordTakeThisOuTspamgmail.com> wrote:
> Does anyone know of a company which will do a
> comprehensive set of tests on crystal oscillator
> units to diagnose possible failure causes?

These has manufactured some custom osc for us:
http://www.suntsuinc.com/

Vasile

2007\01\21@130206 by Brooke Clarke

flavicon
face
Hi Mike:

What do  you mean by "possible failure causes"?
For example are you trying to determine future reliability, like for a
space mission, or . .
do you have units that have failed and you want to know why, or . .

Have Fun,

Brooke Clarke

--
w/Java http://www.PRC68.com
w/o Java www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com

2007\01\21@214431 by Bob Axtell

face picon face
Brooke Clarke wrote:
> Hi Mike:
>
> What do  you mean by "possible failure causes"?
> For example are you trying to determine future reliability, like for a
> space mission, or . .
> do you have units that have failed and you want to know why, or . .
>
> Have Fun,
>
> Brooke Clarke
>
>  
I'll bet Mike has a legitimate reason for asking. I have gone through a
spate of bad crystals in the
past, and the experience was NOT pleasant.

You will notice that I no longer design-in crystals, I use ceramic
resonators. And I have never regretted
the change.

--Bob

2007\01\21@220218 by Derward

picon face

----- Original Message -----
From: "Bob Axtell" <.....engineerKILLspamspam@spam@neomailbox.com>
To: "Microcontroller discussion list - Public." <piclistspamKILLspammit.edu>
Sent: Sunday, January 21, 2007 8:42 PM
Subject: Re: [EE]: Oscillator Testing


{Quote hidden}

> --

2007\01\22@004226 by Vasile Surducan

face picon face
On 1/22/07, Bob Axtell <.....engineerKILLspamspam.....neomailbox.com> wrote:
{Quote hidden}

You are not clocking then some PLL of RF transcievers to get 5Ghz, or
some FPGA  PLL having 1GHz main clock... because ceramic resonators
are good only for low speed microcontrollers running at 0-30C (and not
allways).

I had very bad experiences with resonators and PIC16F676 and
PIC16F877A, caused by an temperature/amplitude variation at the oscout
pin, below the internal trigger limit.

I had bad experiences with some VTCXO too...I'll say it depends by your luck.

greetings,
Vasile

2007\01\22@025409 by Bob Axtell

face picon face
Vasile Surducan wrote:
{Quote hidden}

Of course, ceramic resonators do not operate above 16Mhz in my
experience, and RF requires
a crystal.

> I had very bad experiences with resonators and PIC16F676 and
> PIC16F877A, caused by an temperature/amplitude variation at the oscout
> pin, below the internal trigger limit.
>  
But ceramic resonators can handle incredible amounts of shock. For
crystals, shock is a risk,
because the crystal chip can sustain a fracture and cause strange
effects. Most of my clients are
military or truck-mounted industrial, and once I showed 'em how
sensitive crystals were to
shock, they never looked back.

If I _MUST_ use a crystal, I mount it manually (on a daub of silicone
gel) and allow the leads
to accept an 'S' curve before they are soldered. I NEVER use SMT
versions,  they  break
easily.

Con: ceramic resonators are also hydroscopic (attract water), so some
need a better seal: ECS
units need to be dipped or sprayed to obtain a good seal.
> I had bad experiences with some VTCXO too...I'll say it depends by your luck.
>
> greetings,
> Vasile
>  

2007\01\22@043514 by Alan B. Pearce

face picon face
>Most of my clients are military or truck-mounted industrial

So why not use a Mil-spec item? We use /883 or better parts in instruments
that get shot into space, and they have to survive 9G wideband vibration for
a minute - a couple of times before launch, plus the launch itself, and
still work.

2007\01\22@070724 by Bob Axtell

face picon face
Alan B. Pearce wrote:
>> Most of my clients are military or truck-mounted industrial
>>    
>
> So why not use a Mil-spec item? We use /883 or better parts in instruments
> that get shot into space, and they have to survive 9G wideband vibration for
> a minute - a couple of times before launch, plus the launch itself, and
> still work.
>
>  
actually I have, and still do. When I have to.

--Bob

2007\01\22@101931 by Mike Hord

picon face
> > What do  you mean by "possible failure causes"?
> > For example are you trying to determine future reliability, like for a
> > space mission, or . .
> > do you have units that have failed and you want to know why, or . .

I'm seeing odd failure modes- suddenly, other frequencies appear in the
spectrum.  It only happens at a particular temp, which varies from one
osc to the next.  Above or below that temp, the effect vanishes.  Some
of them, the temp is above the operating range.  Others, it's right in
the room-temp ambient breadbasket (which is how we found it, of
course).

No one seems to know WHY this particular failure mode is occuring.
No one seems to know what the implication of this failure mode is
(i.e., if some of these slip through, is that a Portent of Doom for that
product, or will that oscillator last the effective lifetime of the device?).

> I'll bet Mike has a legitimate reason for asking. I have gone through a
> spate of bad crystals in the
> past, and the experience was NOT pleasant.

No, it isn't pleasant.  We can't tell if it's a process problem (handling,
baking, etc.) or a manufacturer defect in the parts.  Since nobody
knows why it's failing, we don't know where the flaw was introduced.

> You will notice that I no longer design-in crystals, I use ceramic
> resonators. And I have never regretted
> the change.

That would be nice, but I don't think 100MHz resonators are available.
Plus, we need more than .5% accuracy of frequency.

Mike H.

2007\01\22@105209 by David VanHorn

picon face
>
> I'm seeing odd failure modes- suddenly, other frequencies appear in the
> spectrum.  It only happens at a particular temp, which varies from one
> osc to the next.  Above or below that temp, the effect vanishes.  Some
> of them, the temp is above the operating range.  Others, it's right in
> the room-temp ambient breadbasket (which is how we found it, of
> course).


How carefully was the crystal desgined into the oscillator?
I can think of a few reasons why this might happen.

No one seems to know WHY this particular failure mode is occuring.
> No one seems to know what the implication of this failure mode is
> (i.e., if some of these slip through, is that a Portent of Doom for that
> product, or will that oscillator last the effective lifetime of the
> device?).


Been there, done that.. :(  It wasn't pretty.


> No, it isn't pleasant.  We can't tell if it's a process problem (handling,
> baking, etc.) or a manufacturer defect in the parts.  Since nobody
> knows why it's failing, we don't know where the flaw was introduced.


Has anyone pulled the suspect crystals, and checked them against the specs?

Failure analysis isn't something that I routinely do for other companies,
but I would be interested in taking it on as a project.


--
> Feel the power of the dark side!  Atmel AVR

2007\01\22@110411 by Alan B. Pearce

face picon face
>No, it isn't pleasant.  We can't tell if it's a process problem
>(handling, baking, etc.) or a manufacturer defect in the parts.
>Since nobody knows why it's failing, we don't know where the
>flaw was introduced.

1. Did it used to work (i.e. is this a new batch of a previously produced
product)?

2. If you take a unit that has the problem at room temp, and a unit that has
the problem at a different temp, and swap crystals, which unit now has the
problem at room temp?

My suspicion would be a subtle change in the way the crystals are
processed - slight change in the angle of cut, that may be within the
manufacturers tolerance, but causes your particular oscillator circuit
problems. maybe you don't have quite enough gain at the desired overtone and
too much at others.

With a 100MHz crystal you are gong to be operating at 5th or 7th overtone I
would expect. Is the ferrite in the inductors from the same manufacturer as
previously? A change in the loss curve here may be making the crystals shift
between overtone modes due to ferrite loss dependence due to temperature.

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...