cd4517b --- CMOS Dual 64-Stage Static Shift Register

Thread Starter

13hm13

Joined Jun 1, 2023
48
FYI, this is where the cd417b resides in a 1995 schematic (Cambridge Audio, DACMAGIC 2). Two cd4517's are used. The ad hoc purpose is to "split" a DATA (i2S) signal into separate L and R DATA signals. This allows the a stereo dac IC to function as a single-channel mono unit.



There are other other "dirtier" ways to do the same thing, but much more glue logic is involved, as is in the case of this Rotel cd player:

 

Papabravo

Joined Feb 24, 2006
20,383
When I worked at Wadia 15 years ago all of the processing of the audio chain was done in FPGAs. There was nary an MSI part to be found on any of the boards except for the RS-485 differential transceivers.
 

Poyntat

Joined May 24, 2022
60
FPGA would be the way forward for this task if the cost of development is justified.
There are a few lower cost development systems available these days which you could investigate, it really depends on volumes expected for this unit.
 

Thread Starter

13hm13

Joined Jun 1, 2023
48
I don't know how to program FPGA. But, yes, that would be an option. The question is, are FPGA's as fast and clean as stand alone chips and ASICs. In audio, jitter and other issues can creep in out of no where.
 

Papabravo

Joined Feb 24, 2006
20,383
I don't know how to program FPGA. But, yes, that would be an option. The question is, are FPGA's as fast and clean as stand alone chips and ASICs. In audio, jitter and other issues can creep in out of no where.
FPGAs are faster, better and more flexible than standalone chips. The development software is free in most cases. An FPGA usually programs itself from a large serial non-volatile memory co-located on the board with the FPGA. This allows field upgrades to the FPGA if required. The serial EEPROM is programmed from the development software.
 
Top