Microchip ref design

Was looking at one of Microchip's reference designs for a Wheatstone bridge 
sensor.
http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1406&dDocName=en556317

Was curious about how they justify using 20K resistors feeding the ADC input 
that calls for a 2.5K maximum source impedance. Granted that the ADC input 
hold  capacitor is 25pF and the output of the instrumentation amp has 1uF 
diff and common mode caps. I would guess that this is valid for only low 
sampling rates. High sampling rates would make for a larger load on the ADC 
input and cause error.
  Am I correct or missing something?

Aside from the schematic and component data sheets, aren't many details.
From the schematic, there are two instrumentation amps for comparison. They 
both are fed from the came Wheatstone bridge which has some common mode 
'noise' injected.
The Hexfile that is available is likely programmed into the PIC, no source 
included. The Thermal Management Utility appears to run on the PC and 
accesses the demo board. 

On Wed, 23 Nov 2011 20:44:23 -0500, the renowned "Oppie"
<Oppie@127.0.0.1> wrote:

>Was looking at one of Microchip's reference designs for a Wheatstone bridge 
>sensor.
>http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1406&dDocName=en556317
>
>Was curious about how they justify using 20K resistors feeding the ADC input 
>that calls for a 2.5K maximum source impedance. Granted that the ADC input 
>hold  capacitor is 25pF and the output of the instrumentation amp has 1uF 
>diff and common mode caps. I would guess that this is valid for only low 
>sampling rates. High sampling rates would make for a larger load on the ADC 
>input and cause error.
>  Am I correct or missing something?

Looks like a pretty miserable circuit to me. The ADC input error may
not be the biggest problem. Rushed out to meet the introduction date
of the Inamp?  

>Aside from the schematic and component data sheets, aren't many details.
>From the schematic, there are two instrumentation amps for comparison. They 
>both are fed from the came Wheatstone bridge which has some common mode 
>'noise' injected.

IOW, to demo the CMRR. "Reference design"? 

>The Hexfile that is available is likely programmed into the PIC, no source 
>included. The Thermal Management Utility appears to run on the PC and 
>accesses the demo board. 




Best regards, 
Spehro Pefhany
-- 
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

"Oppie" <Oppie@127.0.0.1> wrote:

>Was looking at one of Microchip's reference designs for a Wheatstone bridge 
>sensor.
>http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1406&dDocName=en556317
>
>Was curious about how they justify using 20K resistors feeding the ADC input 
>that calls for a 2.5K maximum source impedance. Granted that the ADC input 
>hold  capacitor is 25pF and the output of the instrumentation amp has 1uF 
>diff and common mode caps. I would guess that this is valid for only low 
>sampling rates.

What would be the point of sampling the output of a low pass RC filter with
20ms time constant at high rates? 

My rough calculations indicate at 100Hz sampling rate the sampling
capacitor charge current will give about 1/5th of an ADC bit error from 20k
source impedance.