John, you hit the nail on the head. Here is a good article on the
subject:
http://images.rfdesign.com/files/4/0599ANDERSON56.pdf
Clearly specifying #bits isn't enough. Noise and non-linearities
are a major consideration. The article does hint that very high-
dynamic range digital receivers may be possible with certain
types of converters, but I'd say its still too soon to throw away
the analog design books.
73 de Mike, W4EF.....................................
----- Original Message -----
From: "John Kaufmann" <john.kaufmann@verizon.net>
To: <topband@contesting.com>
Sent: Friday, July 02, 2004 3:12 PM
Subject: Re: Re: Topband: Analog to Digital converter for 1.8MHz
> CHARLES HUTTON wrote:
>
> > By the way, there are some 18 bit ADC's with spurious free dynamic range
in
> > excess of 103 dB and sampling rates of over 1 MSPS. For 455 kHz IF's,
they
> > would provide a fine A/D conversion that should meet even our most
demanding
> > needs.
>
> Spurious-free dynamic range has to be referenced to a measurement noise
bandwidth to have any meaning because dynamic range is measured against a
noise floor. Furthermore SFDR varies nonlinearly with noise bandwidth
because we are dealing with nonlinearities. If the 103 dB SFDR is
referenced to 1 Hz (which is common practice for many systems and
components), then the SFDR would degrade to 63 dB when measured in a 100 Hz
noise bandwidth if the dominant nonlinearity is third-order. That 63 dB now
seems less appealing as a dynamic range figure.
>
> 73, John W1FV
>
> _______________________________________________
> Topband mailing list
> Topband@contesting.com
> http://lists.contesting.com/mailman/listinfo/topband
>
_______________________________________________
Topband mailing list
Topband@contesting.com
http://lists.contesting.com/mailman/listinfo/topband
|