Re: DAC Reconstruction Filter
- From: Tim Wescott <tim@xxxxxxxxxxxxxxxx>
- Date: Thu, 08 Mar 2007 21:02:39 -0800
rrodgers wrote:
rrodgers wrote:
How do I specify the requirements of the DAC reconstruction filter
(AD9777).
fDATA = 160MSPS
fout = fDATA/3 = 53.33MHz
SNR is specified at 75dB for:
fDATA = 160 MSPS, fOUT = 5 MHz; 0 dBFS
Well, that's an open-ended question!
I'm not clear on exactly what you're trying to do here, other than it's pretty clear you're sampling data out at 160 * 10^6 samples/second.
You seem to be saying that you're going to drive a 53.33MHz sine wave out of your DAC. This is reasonable, maybe. This will generate aliases
at 106.67MHz (160MHSP - 53.33MHz) and 213.33MHz, along with inevitable glitches at 160MHz.
So you need to specify your filter to pass what you want (53.33MHz) and attenuate what you don't want (anything above that). If you're going to
make marked changes in frequency you need to specify the pass band of the filter, and you need to specify how much the alias frequencies need to be attenuated.
What did I miss?
Tim,
Since SNR is specified over the range dc to fc/2 (160/2=80MHz), I was
under the impression that the LP filter was specified as a brickwall
filter with a 3db cutoff at fc/2. I'm not sure if I need to specify the
filter such that I have at least -80db at fc/2?
Thanks for your reply.
They may have done that calculation using a brick wall filter, or they may have integrated the area under the curve of a spectrum analyzer sweep -- if you're lucky the data *** will say. Remember that the IC vendor doesn't have to build a test circuit that would ever be practical in the real world. They can use $100 inductors and $20 capacitors, or they can build up an answer in several bits from related measurements, or they can build a filter that only works at room temperature on a day with no wind.
The question isn't what Analog Devices did to measure it, however. The question is what system are _you_ building, and what does it need to do for you?
Hence my confusion. There's no one best reconstruction filter -- only the best one for the problem at hand. I still don't know enough about your problem at hand to be much help.
--
Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
Posting from Google? See http://cfaj.freeshell.org/google/
"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html
.
- References:
- DAC Reconstruction Filter
- From: rrodgers
- Re: DAC Reconstruction Filter
- From: Tim Wescott
- Re: DAC Reconstruction Filter
- From: rrodgers
- DAC Reconstruction Filter
- Prev by Date: Reed solomon-error correction
- Next by Date: Re: DSD 1796
- Previous by thread: Re: DAC Reconstruction Filter
- Next by thread: query on DSP 2810
- Index(es):