AVR-Chat

Even if you have a source of 100K, put a small cap, 1uF, at the input to the
ADC. It now becomes very low impedance to the ADC.  (even 100n will be fine)

Time constant is 0.1s, so that's not aproblem. The biggest loss is the
charge/discharge of the cap if the signal varies a lot.

Roland



On 6 February 2010 00:32, Steve Hodge <steve@terrafirma.us> wrote:

>
>
> Thanks a lot for everyone's help. Sounds like I'm good to go. Steve
>
>
> From: AVR-Chat@yahoogroups.com <AVR-Chat%40yahoogroups.com> [mailto:
> AVR-Chat@yahoogroups.com <AVR-Chat%40yahoogroups.com>] On Behalf
> Of Jeff Blaine AC0C
> Sent: Friday, February 05, 2010 2:12 PM
>
> To: AVR-Chat@yahoogroups.com <AVR-Chat%40yahoogroups.com>
> Subject: Re: [AVR-Chat] ADC source impedance
>
> Steve,
>
> Yes, sorry for the error. 100k. There will be some finite upper range to
> that due to leakage but it's got to me miniscule. But that's the general
> idea. And if you are talking about sampling 1x/sec then the 20K mentioned
> in
> the other message is not a problem at all. Good luck.
>
> Regards/Jeff
>
> From: Steve Hodge
> Sent: Friday, February 05, 2010 2:53 PM
> To: AVR-Chat@yahoogroups.com <AVR-Chat%40yahoogroups.com> <mailto:
> AVR-Chat%40yahoogroups.com <AVR-Chat%2540yahoogroups.com>>
> Subject: RE: [AVR-Chat] ADC source impedance
>
> Ah, got it. So it sounds like I'm just fine merely doubling it to 20 Kohms
> and only sampling once a second. I assume you intended to say "100k
> equivalent input"? Steve
>
> [Non-text portions of this message have been removed]
>
>  
>


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