Thoughts on a bunch of DACs (and why delta-sigma kinda sucks, just to get you to think about stuff)

Nov 3, 2014 at 5:25 PM Post #1,733 of 6,500
  I'm guessing...
http://www.analog.com/en/digital-to-analog-converters/da-converters/ad5791/products/product.html
 
Not an audio DAC, but is designed for military/aerospace.
 

This looks like a decent guess to me.  How common are voltage output dac chips?  Mofat seems to have hinted strongly that he's using a 'volts', which I take to mean no i-v conversion stage and high voltage (which this chip does up to 33 V).  20 bit is 'good enough' for what we want.
 
This write-up appears to be relevant:   The 20-Bit DAC Is the Easiest Part of a 1-ppm-Accurate Precision Voltage Source:
Its design, shown in Figure 2, features precision voltage-mode R-2R
architecture, exploits state-of-the-art thin-film resistor-matching
techniques, and employs on-chip calibration routines to achieve
1-ppm accuracy levels. Because the device is factory calibrated and,
therefore, requires no run-time calibration routines, its latency is
no greater than 100 ns, so the AD5791 can be used in waveformgeneration
applications and fast control loops.

 
Nov 3, 2014 at 6:44 PM Post #1,734 of 6,500
The only problem with that one is it doesn't take standard digital i2s or even anything resembling audio in digital format. Look at the write mode timing diagram. Doesn't exactly take 44.1khz straight up and need to run through a few loops to load up all the bits in word.
 
Nov 3, 2014 at 7:04 PM Post #1,736 of 6,500
   
The picture shows the R2R ladder arrangements inside of a Totaldac d1. The laser trimmed Vishay VAR Series "Naked" Z-Foil Resistors are a dead giveaway. I own a Totaldac d3-single, so I should know. It's not vintage but an actual production DAC and it definitely belongs in the Class E tier.

 
Yeah, I didn't cruise the totaldac website in too much depth, basically looking for the exact foto source purrin used, but you'd think that I could have mentally rotated the top down interior board shot to an isomeric view, wouldn't you?
 
The d1 does have its enthusiasts, agreed.  Have to sell it to purrin, tho, it's his category.  ;)
 
You know you might be right though: that analog implementation looks pretty serious, even sans tubes.

 
Who are you addressing here, me or estreeter?
 
Originally Posted by purrin /img/forum/go_quote.gif
 
...other DACs based on industrial R2R chips with bad looking INL plots...

 
Do you include the DAC8811 chip in this statement?  Have you looked at its datasheet?  INL ≈ ±0.7 @ 25 °C work for you?
 
Does this DAC circuit look capable of sustaining the output VOLTS that Baldr alluded to?  I suppose that its not requiring an output buffer as he also included is a mark against it.  Not a whole lot of other candidates in active production that I could divine...
 
Nov 3, 2014 at 8:28 PM Post #1,737 of 6,500
  Besides the fact that it's single bit, you mean?

It is 'single 20-bit' --- what does that mean?  My reading, not knowing the terrain here, was "single '20-bit'".  
 
Nov 3, 2014 at 8:32 PM Post #1,738 of 6,500
  The only problem with that one is it doesn't take standard digital i2s or even anything resembling audio in digital format. Look at the write mode timing diagram. Doesn't exactly take 44.1khz straight up and need to run through a few loops to load up all the bits in word.

Doesn't sound impossible with some pre-processing...but not at all my field of expertise.   
 
Nov 3, 2014 at 8:40 PM Post #1,739 of 6,500
Plus an incredibly complex streaming data (audio signal) is orders of magnitude more demanding from the range hi to low than even the most stringent requirements of distributive process control analogs measurements, let alone gas chromatography. Does not look like to me that circuitry would be able to process in real time a complex word structure representing streaming audio bandwidth. But I am a chromotography specialist not an engineer so take that for whatever it's worth...
 
Nov 3, 2014 at 8:42 PM Post #1,740 of 6,500
I'm not ashamed to admit that the technical details are way over my head, but that's fine - I just want to see Yggy in beta. purrin and jacal01 can pull theirs apart and post photos of all 7 pcbs - its time someone got some serious listening time with this dac. As purrin said, the techno babble means nothing without results. 
 

Users who are viewing this thread

Back
Top