AVR-Chat

What Tim is talking about is the equivalent of an RC time constant.  The input of the ADC is a capacitor based sample/hold type.  Which means, as Tim points out, the longer the signal is at that pin, at a fixed level, the less current will be required for the sample cap to charge up.  The 10K equivalent number in the datasheet allows you to run the ADC at more or less full speed. 

I believe the RC time should be linear.  So if the samples are running at 1/10 max (around 25Khz?) then a 100khz equivilant input would be ok.  And so on.  

Regards/Jeff


From: tim gilbert 
Sent: Friday, February 05, 2010 2:41 PM
To: AVR-Chat@yahoogroups.com 
Subject: Re: [AVR-Chat] ADC source impedance


  
Steve,
How fast is the input signal varying and how often are you sampling? If it's a slow signal you'll probably get away with it.

Tim Gilbert
JEM Innovation Inc.
303-926-9053 (office)
303-437-4342 (cell)
720-890-8582 (fax)
www.jeminnovation.com
www.pdksolutions.com

----- Original Message ----- 
From: Steve Hodge 
To: AVR-Chat@yahoogroups.com 
Sent: Friday, February 05, 2010 1:35 PM
Subject: RE: [AVR-Chat] ADC source impedance

I didn't give enough information. The particular application does not need
the full 10 bits of resolution. 9 bits or even 8 bits would be ok. I'm
trying to keep the board to minimal size so I'd prefer not to have to cram
in an opamp. In addition, the application will allow things to be
calibrated, so a correction could be applied in software. Anyone have a gut
feeling given all that? That's all I'm looking for. Steve

From: AVR-Chat@yahoogroups.com [mailto:AVR-Chat@yahoogroups.com] On Behalf
Of David VanHorn
Sent: Friday, February 05, 2010 12:22 PM
To: AVR-Chat@yahoogroups.com
Subject: Re: [AVR-Chat] ADC source impedance

On Fri, Feb 5, 2010 at 12:37 PM, Steve Hodge <steve@terrafirma.us
<mailto:steve%40terrafirma.us> > wrote:
> The Atmel data sheets, at least the ones I'm dealing with, say the "ADC is
> optimized for analog signals with an output impedance of approximately 10
K
> or less". Does anyone have any experience about how much you can exceed
> this 10 K?

Depends on how much precision you want to loose, and there are no hard
answers.
A buffer amp will let you have >1Gohm, and there are some flea-powered
opamps out there.

__________ Information from ESET Smart Security, version of virus signature
database 4839 (20100205) __________

The message was checked by ESET Smart Security.

http://www.eset.com

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]





[Non-text portions of this message have been removed]