[SI-LIST] Re: PLL Loop filter
- From: "Girish Bangalore" <bvg@xxxxxxxxxxx>
- To: <reanderson@xxxxxxxxxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Wed, 8 Sep 2004 10:50:36 -0400
Gentlemen & Ladies:
Thank you very much for your time and insightful thoughts on this topic.
Ray, sorry for the lack of details in my original post. Here is the exact
scenario with as much detail as possible.. please let me know if you need
more clarification with specifics,
1. The ADC & PLL in question was a ADI AD9884 used to digitize RGB VGA
signal from a normal PC (application is LCD controller)
2. The PLL generates a clock and is synced to the Horizontal Sync input
2. The loop filter caps were SMD X7R/0603 of 39nF and 3.9nF values
3. Any kind of vibration or even tapping on the board with the back of a
screw driver would make the digitized displayed (on an LCD) image to
momentarily shake, more like a ghosting effect with one image static and a
ghost image moving away from the static image and then coming back to merge
with it
We have a similar product but with a Philips ADC/PLL combo, TDA8752, which
did not exhibit this problem.
I hope I have given all the details as I know it.. I would be happy to
provide some more info. if you could let me know.
Also, we did some experiments as follows, we lifted both the SMD caps off
the board and connected them through short wires. We found that the problem
was more or less gone (no jumping).
I am now looking to replace the X7R caps with Film capsand found that
Panasonic has an ECHU series which they specifically recommend for PLL
filter applications and also talk about "no shock noise"/no piezo-electric
effect. Are you aware of other similar caps from other manufacturers?
Panasonic ECHU 39nF comes in 1206 size but I would prefer 0603 as the board
is already laid out for 0603.
Again, all your inputs have been tremendous. I have learnt a lot.
Girish
----- Original Message -----
From: "Ray Anderson" <reanderson@xxxxxxxxxxxxx>
To: <si-list@xxxxxxxxxxxxx>
Cc: <billw@xxxxxxxxxxx>
Sent: Wednesday, September 08, 2004 10:16 AM
Subject: [SI-LIST] Re: PLL Loop filter
> Bill Wurst wrote:
>
>>Steve, Ray,
>>Certainly if the dielectric is piezo-electric, such as ceramic, that
>>could be the source of the problem. A problem could also occur if the
>>plate separation is changing with time, regardless of the dielectric
>>material. One way to analyze such a time-varying capacitor is to start
>>with the charge on the loop filter capacitor, i.e., q = CV. To find the
>>current in the circuit, one takes the time derivative of each side of
>>the equation. Normally, the capacitance is considered time invariant
>>and we write i = C*dV/dt. However, if C is not constant, in this case
>>due to mechanical vibration, then the current must be expressed as the
>>sum of the partial derivatives. Makes for an interesting analysis. Way
>>back in the early days of electronic ignition, I designed a capacitive
>>pickup to take advantage of this phenomena. It worked to a degree but
>>harmonic suppression proved difficult as I recall. So proper material
>>selection as well as mechanical damping may be required.
>> -Bill
>>==============
>>
>>
> Bill-
>
> Thanks for sharing your thoughts and experience on the topic. You
> provide a good description of the mechanism that allows a time varying
> capacitance to generate a changing voltage. (This effect is utilized in
> capacitance microphones where the microphone's diaphram is one plate of
> a capacitor. A DC bias is applied to the microphone's 'capacitor'. When
> sound waves impinge on the diaphram and cause it to move an audio
> voltage results as you've just described)
>
> The original poster described a problem where the timing of a video
> signal varied in response to mechanical vibration. _If_ there were some
> way for the plates in a ceramic capacitor to vary their physical spacing
> in response to mechanical vibration I could see the time varying
> capacitance causing a noise voltage. However, if you consider the
> construction of a typical ceramic chip capacitor it is rather difficult
> to visualize the plate to plate spacing changing due to vibration. The
> whole part is rather monolithic in construction. A stack of metallic
> plates that are insulated from each other by thin layers of ceramic
> material and then encased in ceramic material doesn't seem to provide
> much opportunity for mechanical movement.
>
> Stress and strain coupled from the PCB to the part can excite a
> piezeoelectric effect, but that doesn't cause the plate to plate
> separation to vary. Perhaps if the part were stressed to the breaking
> point there could be some internal movement in the part as it was
> mechanically breaking, but that doesn't seem to be the case here.
>
> I guess the this whole discussion is predicated on several assumptions
> of what was going on in the construction of the circuit that was alluded
> to by the original posted as precious few details were given.
> If anyone can relate further details on how a ceramic chip capacitor can
> be induced to exhibit time varying capacitance due to mechanical
> vibration coupled from a PCB I would be interested in learning. I agree
> that if relative movement of the plates were to occur then a capacitance
> change would occur which could induce a noise voltage, but it just seems
> highly unlikely to me that the plate spacing would change due to
> vibration.
>
> -Ray Anderson
> ------------------------------------------------------------------
> 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 FAQ wiki page is located at:
> http://si-list.org/wiki/wiki.pl?Si-List_FAQ
>
> List technical documents are available at:
> http://www.si-list.org
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
> or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
------------------------------------------------------------------
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 FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl?Si-List_FAQ
List technical documents are available at:
http://www.si-list.org
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
