Go Back   Home > Forums > >

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

ES9038Q2M Board
ES9038Q2M Board
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
Old 23rd April 2018, 08:39 PM   #791
JensH is offline JensH  Denmark
diyAudio Member
 
Join Date: Jul 2009
ES9038Q2M Board
I measured -74.5dB or 0.02% as documented here.
  Reply With Quote
Old 23rd April 2018, 08:41 PM   #792
Spartacus is offline Spartacus  United Kingdom
diyAudio Member
 
Join Date: Jan 2002
Location: Gloucestershire
Quote:
Originally Posted by Markw4 View Post
Without serious mods, I forget exactly what it is for voltage mode operation with this board but distortion might be around -70dB or so. Not very good. Not even good enough for good 16-bit playback if one considers that 16-bits goes all they way down to -96dB.

All in IMHO, of course.
I'd love to see a source for -70dB. The older ESS chips are specified at -108dB in voltage mode, although I would accept this is the absolute best case.
  Reply With Quote
Old 23rd April 2018, 09:48 PM   #793
madds1 is offline madds1
diyAudio Member
 
Join Date: Oct 2010
I had posted this earlier but posting it again

Measurements Of Generic ES9038Q2M DAC Board | Audio Science Review (ASR) Forum

These measurements correlate with JensH measurements

Even with the change to the output stage (Current mode) the distortions persist albeit at lower levels (See JensH second set of tests)

I tested these boards too recently and found that my unbalanced I/V measurements correlated with JensH

However the question remains. What could be causing these distortions. Random jitter will raise the noise floor. Correlated jitter will usually create 2nd harmonics and the jitter test in ASR seems to negate any jitter

I plan to measure the raw output sometime soon

Cheers
  Reply With Quote
Old 23rd April 2018, 09:59 PM   #794
Markw4 is offline Markw4  United States
diyAudio Member
 
Join Date: Jul 2016
Location: California
Actually, percent THD is a fairly meaningless number in the sense that it does not correlate with human perception almost at all. For harmonic distortion it really depends on which harmonics are present and in what amount. Very low order harmonics, primarily the 2nd and 3rd are much less objectionable than higher orders. 2nd harmonic can be hard to hear (for an untrained listener) sometimes at a few or several percent. On the other hand, higher order harmonics can be quite objectionable at .01%, IIRC. Earl Geddes did quite a bit of research in this area and he says THD at a performance metric is completely useless.

For DACs there are other types of linear distortion (most objectionably, group delay) and types of distortion and ugly sounding non-random noise such as can result from timing jitter. A comprehensive set of measurments is really needed to fully evaluate a DAC from the technical perspective.

The reference DAC here that I am using is a Benchmark DAC-3 which was reviewed by Stereophile. Their measurements can be seen here as one example: Benchmark DAC3 HGC D/A preamplifier-headphone amplifier Measurements | Stereophile.com The DAC3-HGC they reviewed retails for about $2.2k.

The Chinese DAC we are working on here probably costs a bit over $200 to fully mod (includng the cost of the DAC board itself), that we know of so far (and we are very near the end of the project). That is a budget price compared to what you would have to pay for that level of sound quality commercially. Probably up around $1k or maybe more. Thing is as price goes up, manufacturers tend to add bells and whistles rather than put all the money into sound quality. (Benchmark is an exception. They make a plain box with a lot of very well engineered circuitry inside primarily aimed at sound quality above all else. But they mostly sell to professional recording and mastering studios, movie soundscore editing, etc., and somewhat to audiophiles. So, they have a different business model than most.)

The low-cost Chineses DAC manufacturers also have a business model. They try to figure out what will sell in the rock bottom price range aimed at people looking for a bargain and thinking they can get something cheap, self-customized, good, etc., based on use of a few brand name parts and the rest mostly cheap parts made to look similar to good parts, and using the cheapest low-cost circuits imagniable. So what? Unfortunately, it just so happens that with DACs that the brand name of the chip is much less important than the implementation, the quality of the circuit that the DAC chip is put into, and including the quality of all the other parts in the circuit. That's why we have to mod to get pretty good sound quality if we start with the DAC we have.

Last edited by Markw4; 23rd April 2018 at 10:07 PM.
  Reply With Quote
Old 23rd April 2018, 10:21 PM   #795
Markw4 is offline Markw4  United States
diyAudio Member
 
Join Date: Jul 2016
Location: California
Quote:
Originally Posted by madds1 View Post
ASR seems to negate any jitter
DACs are very sensitive to clock quality. cdgames who makes the Allo DACs has said the same things I do, BTW.

The clock that comes with this DAC is what is known is the trade as a standard crystal clock. Such clocks are specified for jitter performance down to about 12kHz.

DACs are highly sensitive to jitter far below 12kHz. Good DACs use clocks known in the trade as ultra-low jitter. They are specified for jitter down at 10Hz or maybe even 1Hz. There are a variety of clocks that could be used for this DAC but to play back at the highest sample rates it supports requires 100MHz. There are some Japanese clocks that are very low jitter but they are only available up to 50MHz. In reels of 3,000 they only cost maybe $2 each. One or two at a time, if you can find them, probably go for $15 or so depending on the seller, shipping costs, etc. Good 100Mhz clocks are available for under $30. No surprise they don't come included with $39 DACs where the DAC chip itself costs $15.

Last edited by Markw4; 23rd April 2018 at 10:24 PM.
  Reply With Quote
Old 23rd April 2018, 10:25 PM   #796
madds1 is offline madds1
diyAudio Member
 
Join Date: Oct 2010
Oh.. I totally agree with you regarding the harmonics present. I have said this too several times in the past. In fact in phase second harmonics great for my magnepans. I deliberately introduce them in my preamp using a low biased jfet. For me thus the source has to be clean or I have to bypass my preamp

The question however is what is causing these harmonics to be present in such high numbers. The ASR tests are pretty decent and it gives me no clue as to what it would be. I understand the clock has to be clean but a bad clock would show up in the J test. Maybe the ASR test is flawed?

I also agree with you that after a certain point mods are not worth it.

I ll wait for the other board that I ordered with the sync SA receiver. Hopefully that one is worth it out of the box
  Reply With Quote
Old 23rd April 2018, 10:29 PM   #797
Markw4 is offline Markw4  United States
diyAudio Member
 
Join Date: Jul 2016
Location: California
I put a Chrystek 575 clock in my Chinese DAC and sure enough, it made as big a difference as any of the other big differences, such as going from voltage to current modes, regulating AVCC as ESS recommends, getting the analog circuits off the digital power, etc.

Actually, there are not a lot of mods it is just a lot of work to do them.
They are:
Use IV output stages with differential summing
Clean up all power issues, including AVCC as ESS recommends
Use ultra-low jitter clock
Use minimum-phase slow-transition reconstruction filter to minimise audible group delay
Upsample to move reconstruction filter transition band out of the audio band

Only 5 mods, they all make an audible difference, and they all make sense that they should help.

I am working on getting ready to test one last possible mod which is 'master mode I2S,' for want of a better name. The DAC chip will be I2S clock master which means it should be possible to turn off ASRC in the DAC without penalty. In fact, cdgames says it sounds better and the Allo Katana DAC will use that mode. I will see if I can reproduce his results with my DAC and if so I will try to make available a mod anyone can do, and do so in a way that is compliant with the ESS non-disclosure agreement.

Last edited by Markw4; 23rd April 2018 at 10:46 PM.
  Reply With Quote
Old 23rd April 2018, 10:51 PM   #798
DPH is offline DPH  United States
diyAudio Member
 
Join Date: Dec 2008
Location: Sunny San Diego, CA
For all these assertions about clock sensitivity, it doesn't align with the JTest result on a stock board shown earlier (measured with a QA400). It's midway down: Measurements Of Generic ES9038Q2M DAC Board | Audio Science Review (ASR) Forum

His claimed mods were to use a bench PSU for the +/- 15 V and an XMOS usb to i2s board. Fast roll off linear filter.

The sidelobes were -120 dB intermodulations with 60 Hz. No other spray. Sure, one can do another 20 dB better with careful effort of the layout and better power isolation, but we're not showing a defective clock here. There's no spray of harmonics anywhere. Clean clean clean, which is good because ESS advertises this very fact!

Which certainly leaves me scratching my head about this mod.
__________________
Happy DIYing, Daniel

Last edited by DPH; 23rd April 2018 at 10:54 PM.
  Reply With Quote
Old 23rd April 2018, 11:06 PM   #799
Markw4 is offline Markw4  United States
diyAudio Member
 
Join Date: Jul 2016
Location: California
Clock jitter rises rapidly as frequency goes down, much like 1/f noise. It can look pretty good at 100Hz and much worse at 10Hz or 1Hz.

It is known that human attention tends to be attracted to very LF sound variations. People have speculated as to why, and why humans might not like it in DACs. But, who knows?

For whatever the reason, clocks specified for very low jitter at very low frequencies sound better in DACs. Try it you will like it is all I can say.

EDIT: Regarding ESS and clock jitter, they have a white paper on their downloads page. They say so long as the clock used locally by the DAC is good, that the clock, cabling, etc., sending audio to the DAC doesn't need to have such a clean clock. The DAC will use its own clock and ASRC to fix the incoming data. The clock replaced in the mod was the DAC's own clock. (Just so we are clear we are all talking the same thing.)

Don't know about any measurements made with the clocks currently shipping. If measurements showed the original clock to be pretty good, it would be very interesting to replace the clock with an ultra-low jitter and measure again. It is always nice if sense can be made out of all the data.

Last edited by Markw4; 23rd April 2018 at 11:21 PM.
  Reply With Quote
Old 24th April 2018, 02:17 AM   #800
keilau is offline keilau  United States
diyAudio Member
 
Join Date: Mar 2013
Location: Windy City, US
Mark, your insight on high end gear is much appreciated. But we are talking different animal here. My reference point is the Topping D50 DAC which currently sells at $250 ($199 in DROP earlier). The review and measurement looked very good.

Review and Measurements of Topping D50 DAC | Audio Science Review (ASR) Forum

It is not the $2200 DAC that you referred to. My question at this DIY forum is whether I can build an improved ES9038Q2M DAC for less than the Topping. The -106dB D+N of the Topping D50 is much better than the -70dB you quoted. I do understand that THD measurement is not very meaningful to audible results. I just use the parameter that you mentioned first.

There are positive report on the Chinese ES9038Q2M boards in other threads on this forum. It is true that the quality may vary, but this is one of the good one.

ES9038 Q2M DAC DSD Decoder Support IIS DSD 384KHz Coaxial Fiber DOP USA | eBay

Quote:
Originally Posted by Markw4 View Post
Actually, percent THD is a fairly meaningless number in the sense that it does not correlate with human perception almost at all. For harmonic distortion it really depends on which harmonics are present and in what amount. Very low order harmonics, primarily the 2nd and 3rd are much less objectionable than higher orders. 2nd harmonic can be hard to hear (for an untrained listener) sometimes at a few or several percent. On the other hand, higher order harmonics can be quite objectionable at .01%, IIRC. Earl Geddes did quite a bit of research in this area and he says THD at a performance metric is completely useless.

For DACs there are other types of linear distortion (most objectionably, group delay) and types of distortion and ugly sounding non-random noise such as can result from timing jitter. A comprehensive set of measurments is really needed to fully evaluate a DAC from the technical perspective.

The reference DAC here that I am using is a Benchmark DAC-3 which was reviewed by Stereophile. Their measurements can be seen here as one example: Benchmark DAC3 HGC D/A preamplifier-headphone amplifier Measurements | Stereophile.com The DAC3-HGC they reviewed retails for about $2.2k.

The Chinese DAC we are working on here probably costs a bit over $200 to fully mod (includng the cost of the DAC board itself), that we know of so far (and we are very near the end of the project). That is a budget price compared to what you would have to pay for that level of sound quality commercially. Probably up around $1k or maybe more. Thing is as price goes up, manufacturers tend to add bells and whistles rather than put all the money into sound quality. (Benchmark is an exception. They make a plain box with a lot of very well engineered circuitry inside primarily aimed at sound quality above all else. But they mostly sell to professional recording and mastering studios, movie soundscore editing, etc., and somewhat to audiophiles. So, they have a different business model than most.)

The low-cost Chineses DAC manufacturers also have a business model. They try to figure out what will sell in the rock bottom price range aimed at people looking for a bargain and thinking they can get something cheap, self-customized, good, etc., based on use of a few brand name parts and the rest mostly cheap parts made to look similar to good parts, and using the cheapest low-cost circuits imagniable. So what? Unfortunately, it just so happens that with DACs that the brand name of the chip is much less important than the implementation, the quality of the circuit that the DAC chip is put into, and including the quality of all the other parts in the circuit. That's why we have to mod to get pretty good sound quality if we start with the DAC we have.

Last edited by keilau; 24th April 2018 at 02:41 AM.
  Reply With Quote

Reply


ES9038Q2M BoardHide this!Advertise here!
Thread Tools

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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
My Cambridge 500SE, needs an output board (to replace internal DAC Board) slbender Digital Source 1 11th July 2018 02:43 AM
JLSounds XMOS USB board + AK4396 board + Direct DSD PCB ravid Swap Meet 1 29th December 2016 11:36 AM
Unknown Board DC12V TPA3116 amplifier board DIY L163-9 AUS Ben Class D 0 8th December 2016 04:05 AM
FS: Ian FIFO i2s-to-PCM board, Single XO clock board trung224 Swap Meet 3 26th November 2016 02:18 PM
Lost 1 Zv4 ckt board..looking to buy 1 circuit board Blues Pass Labs 1 22nd March 2004 10:09 PM


New To Site? Need Help?

All times are GMT. The time now is 12:45 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Copyright ©1999-2021 diyAudio
Wiki