AL1201G based DAC - looking for input

[Russ White]

I am designing a balanced output AL1201G based DAC from wavefront.

The DAC will take S/PDIF in, and uses the CS8414 receiver. The output is provided by 4 single opamps, and there is an auto reset function provided by a TLC7705 and SN74HC04. When there is no signal lock the voltage supervisor resets the receiver repeatedly until the signal is locked.

The digial and analog GND planes are seperate and there is a leaded ferrite bead between them.

Does it look sane? This is my very first attempt at a DAC so any advice is welcome.

I am especially looking for input around the CS8414 and the AL1201G.

[Russ White]

The board may end up looking something like this (it is an early layout subject to change)

[hayenc]

I would be curious to know reasons for selecting the AL1201G over other choices, such as the Wolfson 8740. I could not find much in any threads other than the Burr Brown chips.

[Russ White]

Originally posted by hayenc
I would be curious to know reasons for selecting the AL1201G over other choices, such as the Wolfson 8740. I could not find much in any threads other than the Burr Brown chips.

I am not familiar with that part, but my part choices here are based partly on input from a couple "recording biz" engineers I know here in Nashville who really like the AL1201. Also I chose it because it is SOIC which is as small as I want to go, no SSOP for me. (well not yet anyway)

The AL1201 is also very easy to use. And since this is my first run at a DAC I want something extremely hard to muck up.

Cheers!
Russ

[hayenc]

I have the eval kit for the 8740, plus a Benchmark DAC1 and Peter's nos 1543 DAC's, so I look forward to this project and doing some comparisons.

[metalman]

Ah! Russ! You are beating to me to the punch on producing a balanced output DAC, a project I have been slowly tinkering with for some time without getting far enough to have anything worth posting. Your approach looks very appealing, I was getting led astray by overthinking and overdesigning the power feeds and clocking arrangements.

Last question: When do you find time to sleep?

Cheers, Terry

[b-square]

are you planning to include the obligatory high performance clock input?

[Russ White]

I don't feel obliged to.

[ble0t]

Only advice I would have is to change some of the routing (it's obvious from even the EAGLE3D pic). The FILT pin is very sensitive and should have very short traces to the filter caps/resistors (i.e. use SMD parts in this case). Also, the corresponding VA+ and AGND pins should also have decoupling as close as possible i.e. use SMD if possible.

[Russ White]

Wow, so I guess I didn't screw it up too bad? I am honestly surprised as this was my first shot at it. I will try changing the caps as you suggest.

Cheers!
Russ