Build thread for Diyinhk ES9018 DAC on Ebay - Page 33 - diyAudio
Go Back   Home > Forums > Source & Line > Digital Line Level

Digital Line Level DACs, Digital Crossovers, Equalizers, etc.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 17th February 2013, 10:25 PM   #321
glt is offline glt  United States
diyAudio Member
 
Join Date: Oct 2004
Maciej,

Thank you for sharing your measurements. I like your results. Certainly gives credibility for a "budget and good sounding" solution.

Have you tried adding a large capacitor after the ADP-151? or increasing the three 10 uF near the chip?

According to the ESS sabre white paper, the difference between voltage mode and current mode is -108db vs 120 db THD...
__________________
www.hifiduino.wordpress.com

Last edited by glt; 17th February 2013 at 10:31 PM.
  Reply With Quote
Old 18th February 2013, 01:06 AM   #322
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
Quote:
According to the ESS sabre white paper, the difference between voltage mode and current mode is -108db vs 120 db THD...
under ideal circumstances for the voltage mode connection, not connecting directly through a cap with a 20V RMS (1763) regulator powering AVCC.

Last edited by qusp; 18th February 2013 at 01:18 AM.
  Reply With Quote
Old 18th February 2013, 01:14 AM   #323
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
if 0.01% is OK with you, just use an old ipod.

measurements were with an AP2, but not this dac board, not this IV and of course they were using the AP2 as a signal source...

better yet, just use the EMU...

characterizing an es9018 dac with the EMU0202 ADC
Quote:
Dynamic Range (A-weighted, 1kHz, min gain): 113dB
Signal-to-Noise Ratio (A-weighted, min gain): 113dB
THD+N (1kHz at - 1dBFS, min gain): -103dB (.0007%)
I would hope any ESS implementation would make the EMU the limiting factor, if not youve done something very very wrong.

still much better than 0.01% though

Last edited by qusp; 18th February 2013 at 01:30 AM.
  Reply With Quote
Old 18th February 2013, 04:54 AM   #324
glt is offline glt  United States
diyAudio Member
 
Join Date: Oct 2004
Quote:
Originally Posted by qusp View Post
under ideal circumstances for the voltage mode connection, not connecting directly through a cap with a 20V RMS (1763) regulator powering AVCC.
I am quoting that the difference between V mode and I mode is 12 db....
__________________
www.hifiduino.wordpress.com
  Reply With Quote
Old 18th February 2013, 05:10 AM   #325
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
yes, people using voltage mode quote this all the time, but they do not give a 'voltage mode' circuit in the datasheet, just connecting directly through a cap will not give this performance. then of course we have the issue that the power supply and IV are not suitable for getting -120db THD+N either

the main point was that the ADC being used doesnt have the performance needed to get proper and ideal 'voltage mode' measures, let alone judge a comparison with a proper IV, as the difference would be lost in the noise, both leveled to the performance of the ADC

Last edited by qusp; 18th February 2013 at 05:13 AM.
  Reply With Quote
Old 18th February 2013, 07:10 AM   #326
glt is offline glt  United States
diyAudio Member
 
Join Date: Oct 2004
Why would you need a "circuit" if the DAC can be used as a voltage out DAC?.
__________________
www.hifiduino.wordpress.com
  Reply With Quote
Old 18th February 2013, 07:54 AM   #327
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
yes, because without some sort of buffer and a very short connection the performance will vary depending on the load the dac sees. the DNR will probably stay roughly the same unless you do something silly, but the THD will vary quite a bit, as will the frequency response. it will depend on the load impedance, capacitance and inductance. it will also vary depending on the voltage compliance/drive capability of the output, offhand I cant remember what that is for the ESS

Last edited by qusp; 18th February 2013 at 07:56 AM.
  Reply With Quote
Old 18th February 2013, 12:49 PM   #328
diyAudio Member
 
Join Date: Feb 2011
Quote:
Originally Posted by glt View Post
Have you tried adding a large capacitor after the ADP-151? or increasing the three 10 uF near the chip?
Yes, I've tried this, and it caused an increase of distortion (again measured with my not pro equipment). Now I have 3.3V VD and VA bypassed with 3300uF low-esr each. I also tried non-low-esr caps, but the results were worse.

Quote:
Originally Posted by qusp View Post
under ideal circumstances for the voltage mode connection, not connecting directly through a cap with a 20V RMS (1763) regulator powering AVCC.
I'm sorry: V, not nV - my mistake.

Quote:
Originally Posted by qusp View Post
if 0.01% is OK with you, just use an old ipod.
I still use my old compact cassette player
According to you, I am unable to make any adequate measurements of the diyinhk ES9018 board because I do not posess or have access to pro analysers. I think it applies to most of the diyers. Does that mean I should simply copy solutions of others (mostly untested), populate the board with parts as described in schematics and believe, that it will work well? So diy implementations of ES9018 are pointless, because it's parameters are beyond any amateur home-lab measuring equipment?
I would really love to see someone's diyinhk board tuning process with measurements done by proper pro-tools. I am waiting.
  Reply With Quote
Old 18th February 2013, 01:38 PM   #329
vuki is offline vuki  Croatia
diyAudio Member
 
vuki's Avatar
 
Join Date: Jun 2002
Location: zagreb, croatia
Thanks for the measurements Maciej! Your measurements are much more useful then someone else's armchair engineering.
  Reply With Quote
Old 18th February 2013, 02:20 PM   #330
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
yep, 20nV vs 20V RMS noise power is ~30dB difference, not insignificant and a further ~10dB for the LM317
Quote:
Thanks for the measurements Maciej! Your measurements are much more useful then someone else's armchair engineering. Click the image to open in full size.
whatever you say vuki, if you can measure a difference with the EMU, basically if youve got either measure poking above the noise floor, youve got some serious issues. i've been working with the ES9018 and 9012 for ~3 years now, have built 5 of them (3 I still own) and have built numerous IV stages for them, myriad power supply configurations and gone through significant objective and subjective characterization. this is along with using direct out and transformers for a while too.

For sure i'm still learning, but this is pretty basic good practice, the measurement device needs to be able to capture the measurement if you hope to gain useful data... you disagree?

Quote:
I still use my old compact cassette player
According to you, I am unable to make any adequate measurements of the diyinhk ES9018 board because I do not posess or have access to pro analysers. I think it applies to most of the diyers. Does that mean I should simply copy solutions of others (mostly untested), populate the board with parts as described in schematics and believe, that it will work well? So diy implementations of ES9018 are pointless, because it's parameters are beyond any amateur home-lab measuring equipment?
I would really love to see someone's diyinhk board tuning process with measurements done by proper pro-tools. I am waiting.
I still use old audio gear and a crappy radio in the car too, but I wouldnt bother upgrading to an ESS dac if 0.01% was good enough, particularly if I already had an EMU.

This is not some opinion i'm trying to convince you of, its simply scientific method. feel free to make measurements while ignoring the scientific method, but why bother?

if you are trying to differentiate between 2 levels of output performance, the sampling device needs to have enough bandwidth to be able to at the very least measure a little bit better than the expected lower performing device/arrangement. If you do not have this minimum bandwidth, you will just be bouncing against the performance of the ADC and its analogue input stage, rather than giving any useful information to separate the 2 DUT.

you said you needed to adjust the levels for the ESS direct output vs IV because the IV was too hot and clipping I guess, did you do this with any method? how much attenuation and what with? the EMU digital attenuator? this will directly lower DNR and THD performance. how and by how much will depend.

i'm afraid its a fact of life, if you are going to measure the ESS dac at its best (as well as the effect of different low noise supplies), thats pretty ambitious IMO, its arguably the highest performance audio dac on the planet. while its performing correctly (even performing sub-optimally it outstrips the EMU by a fair margin), then you do need some pretty serious kit, you just do, thats no comment on you personally, nothing elitist about it, just a fact of life. ESS's own ADC, the ES9102 does not really have good enough performance, or is at least very borderline without long FFT

the ESS dac performing well makes it difficult for even some top grade measurement gear. the EMU is going to do fine for some measures, like some (but not all) power amps, definitely speakers etc, but its ADC is hardly top notch and neither is its preamp stage, it was designed to a budget and not with measuring top flight line level input in mind.

Last edited by qusp; 18th February 2013 at 02:37 PM.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
BUILD THREAD - A New Take on the Classic Pass Labs D1 with an ESS Dac opc Digital Line Level 667 27th August 2014 10:15 AM
High Performance WM8741 Upsampling DAC New Version build thread (show 'n tell too) merlin2069er Digital Source 354 7th April 2013 08:51 PM
Pass D1V3 DAC - build thread spzzzzkt Pass Labs 224 8th December 2012 03:40 AM
Gamma-1 DAC build thread mattmcl Digital Line Level 8 20th January 2012 11:27 PM


New To Site? Need Help?

All times are GMT. The time now is 12:50 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2