AD/DA discussion

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
G

Guest

Guest
I have been following this...

http://www.soundonsound.com/forum/showflat.php?Cat=&Number=453703&page=0&view=collapsed&sb=5&o=&fpart=1#Post453956453703
 
Nothing new, as far as I can tell. All sigma-delta converters' appnotes recommend a capacitor across the input lines (or to ground/Vref, for single-ended designs). This capacitor is part of the input anti-aliasing filter. Furthermore most S/D ADCs have unbuffered inputs, and tend to have current pulses flowing into/out of them at the modulator frequency. Without this cap you either need a very fast external buffer amp or your dynamic range goes down the drain. You really need this cap, but AFAICT there is no seekrit insider knowledge involved.

Or I could produce converters with a dynamic range of more than the theoretical maximum of 144dB (yes, it's possible, but it is protected by patents, and I don't know if the patent owners will ever actualy produce it).
Riiiiiight.... patented by the same guy who invented the 1 million hour light bulb, I guess.

JDB.
 
If it's for getting rid of the (fixed) sample frequency coming out backwards and interfereing, why don't we use a L/C notch for getting rid of this in stead?

Ya know, like the bias trap of the rec amp in the good 'ole tape-days..?

Jakob E.
 
[quote author="gyraf"]If it's for getting rid of the (fixed) sample frequency coming out backwards and interfereing, why don't we use a L/C notch for getting rid of this in stead? [/quote]
It's not a fixed frequency. The S/D architecture feeds a sample of the modulator output back to its input. While the modulator frequency puts an upper limit on the current pulse rate, the polarity of the pulses depends on the input waveform.

JDB.
 
Back
Top