Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

PC Based Computer music servers, crossovers, and equalization

ES9018K2M, ES9028Q2M, 9038Q2M DSD/I2S DAC HATs for Raspberry Pi
ES9018K2M, ES9028Q2M, 9038Q2M DSD/I2S DAC HATs for Raspberry Pi
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 8th December 2017, 02:14 AM   #71
abraxalito is offline abraxalito  United Kingdom
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Send a message via Yahoo to abraxalito
Quote:
Originally Posted by markusA View Post
Itís all quite an enigma. Op amps measure great, but for some reason they seem to lack some magic ingredient? Going by measurements alone, they would be hard to beat, but once you stop comparing technical data, your ears often tell a very different tale. Itís both frustrating and what makes hifi so special, all at once.
Agreed - I have a hunch that its in part to do with RF, opamps don't like it and DACs are full of it. Its why nowadays I put a passive filter right after my DAC chips, so the the RF doesn't get to interfere with the active circuits. When there's RF present, passive is the way to go. Which is one reason transformers sound so good - they also provide some filtering to the downstream stages in addition to common-mode noise reduction.
__________________
I know you think you understand what you thought I said but I'm not sure you realize that what you heard is not what I meant - Alan Greenspan
  Reply With Quote
Old 8th December 2017, 01:07 PM   #72
iancanada is offline iancanada  Canada
diyAudio Member
 
iancanada's Avatar
 
Join Date: Dec 2009
Location: Toronto
ES9018K2M, ES9028Q2M, 9038Q2M DSD/I2S DAC HATs for Raspberry Pi
Quote:
Originally Posted by abraxalito View Post
OPA861 would certainly do the job and its got a good reputation for I/V duty as far as I'm aware. I'd be a bit concerned though about its 3rd harmonic distortion figure, given that one of the things attracting customers is ESS's very low THD+N numbers, these might be thrown out the window by attaching an OPA861. The DS only shows 2nd/3rd harmonic at 5MHz so its anyone's guess how much better it'll be at audio freqs.

Also is OPA861's 10ohm input impedance low enough to be considered 'current mode' from the point of view of the DAC chip?

My own reservations with the NTD1 are that its such a power hog - I've an I/V stage where I'm running much lower quiescent currents and much lower supplies but its operation is very similar to NTD1's. There are more recently released MOSFETs which look to provide very high gm at quite reasonable output currents. SSM3K339R is one such but being sot23 it cannot dissipate much power.
Hi Abraxalito,

I don't want to make a oven. So, I'm more interested in your approach. Do you have more information about your IV "with much lower quiescent currents and much lower supplies" ?

There are thousands of different kinds of MOSFETs, but most of them are targeted switching mode, even the one you mentioned SSM3K339R.

My question is that, according to what specification you chose MOSFETs for this kind of audio applications? High gm? Linearity of Vgs-Id curve? Or, something else?

Regards,
Ian
  Reply With Quote
Old 8th December 2017, 01:31 PM   #73
abraxalito is offline abraxalito  United Kingdom
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Send a message via Yahoo to abraxalito
Hi Ian

I've attached a schematic showing the output stage I'm using in my latest DAC design. This illustrates the idea - the I/V transistor (BC807-40) needs to run at a substantial bias to present a low enough impedance to the passive filter (around 2ohm or so in this case). It could just as easily be a MOSFET as a bipolar, I'm using a BC807 for cheapness. However no MOSFET has as high a gm as a bipolar, the K339R comes about the closest I've seen, about half as much transconductance. It therefore needs higher bias currents from the CCSs.

The reason this circuit doesn't dissipate as much power as the NTD1 does is that the whole of the bias current isn't run through the I/V resistor. If it were, the supply voltages would have to be much higher. Instead the current is 'bled off' via a current source across the resistor.

As for the choice of MOSFET - its about gm primarily, coupled with not having too high capacitances. MOSFET capacitances are quite non-linear so will contribute harmonic distortion which rises with frequency.
Attached Images
File Type: png DACoutputstage.png (16.2 KB, 787 views)
__________________
I know you think you understand what you thought I said but I'm not sure you realize that what you heard is not what I meant - Alan Greenspan
  Reply With Quote
Old 8th December 2017, 04:51 PM   #74
iancanada is offline iancanada  Canada
diyAudio Member
 
iancanada's Avatar
 
Join Date: Dec 2009
Location: Toronto
ES9018K2M, ES9028Q2M, 9038Q2M DSD/I2S DAC HATs for Raspberry Pi
Quote:
Originally Posted by abraxalito View Post
Hi Ian

I've attached a schematic showing the output stage I'm using in my latest DAC design. This illustrates the idea - the I/V transistor (BC807-40) needs to run at a substantial bias to present a low enough impedance to the passive filter (around 2ohm or so in this case). It could just as easily be a MOSFET as a bipolar, I'm using a BC807 for cheapness. However no MOSFET has as high a gm as a bipolar, the K339R comes about the closest I've seen, about half as much transconductance. It therefore needs higher bias currents from the CCSs.

The reason this circuit doesn't dissipate as much power as the NTD1 does is that the whole of the bias current isn't run through the I/V resistor. If it were, the supply voltages would have to be much higher. Instead the current is 'bled off' via a current source across the resistor.

As for the choice of MOSFET - its about gm primarily, coupled with not having too high capacitances. MOSFET capacitances are quite non-linear so will contribute harmonic distortion which rises with frequency.
Thanks Abraxalito for the schematic, it's a very smart design. Looks fantastic.

Does it use single supply? Do you have simulation result? Thant will be much help for me to understand the principle.

Thanks again.
Ian
  Reply With Quote
Old 9th December 2017, 10:23 AM   #75
abraxalito is offline abraxalito  United Kingdom
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Send a message via Yahoo to abraxalito
Quote:
Originally Posted by iancanada View Post
Thanks Abraxalito for the schematic, it's a very smart design. Looks fantastic.
Thank you!

Quote:
Does it use single supply? Do you have simulation result? Thant will be much help for me to understand the principle.
Yes it does use a single supply though that entails using output capacitors as the output has a half-rail offset. Or with a negative rail generator (say ICL7660 or similar) the offset can be taken down to a much lower level.

Simulations I've done for the frequency response and I've built a few and measured frequency response and THD for a sinewave input. THD performance isn't stellar as you'd expect for such a simple circuit. It reaches around 0.2% THD (2nd and 3rd harmonics predominantly) at high frequencies (5kHz say). There are ways to improve on this - a cascode arrangement will help.
__________________
I know you think you understand what you thought I said but I'm not sure you realize that what you heard is not what I meant - Alan Greenspan
  Reply With Quote
Old 9th December 2017, 05:05 PM   #76
markusA is offline markusA  Sweden
diyAudio Member
 
markusA's Avatar
 
Join Date: Dec 2009
Location: Gothenburg
Default Design DSD/I2S DAC HATs for Raspberry Pi, start from ES9018K2M

If weíre exploring different options, it would be really cool with a tube buffer. My wishlist would contain 1. Balanced topology to match my amps + I like xlr connectors. 2. Using 6n23p (6dj8) tubes since I have a box full of them.

Having said that, I have some annoying questions I canít shake. Can you make it better than transformers? Or if transformers are used in the circuit, do you really need the tubes? And just because Iím curious, could you make something interesting with tubes and Autoformers for volume controls?
And why would one even bother since the digital volume control is already built in? This is the kind of stuff that keep me awake at night, lol.
For some perverse reason, I keep coming back to circlotrons?

P.S Sonic pros & cons aside, who doesnít love the glow from tubes when you turn the light down while listening to some favorite piece of music. ES9018K2M, ES9028Q2M, 9038Q2M DSD/I2S DAC HATs for Raspberry Pi
__________________
Current build to be found in my blog. Not so current anymore, everything is put on hold until march.

Last edited by markusA; 9th December 2017 at 05:07 PM.
  Reply With Quote
Old 9th December 2017, 05:48 PM   #77
iancanada is offline iancanada  Canada
diyAudio Member
 
iancanada's Avatar
 
Join Date: Dec 2009
Location: Toronto
ES9018K2M, ES9028Q2M, 9038Q2M DSD/I2S DAC HATs for Raspberry Pi
Quote:
Originally Posted by markusA View Post
Itís all quite an enigma. Op amps measure great, but for some reason they seem to lack some magic ingredient? Going by measurements alone, they would be hard to beat, but once you stop comparing technical data, your ears often tell a very different tale. Itís both frustrating and what makes hifi so special, all at once.
I have a very interesting tube circuit somewhere on the hdd but it would be sweet with a good discrete solution as well. Either way, I think buffer and fully balanced outputs is a good goal. Gain isnít really needed unless you have a very low amplifying amp.
My tube circuit was fully balanced with a Russian version of the ecc88, thatís the name of the tube, right? Not schoolbook balanced but sort of a sneaky dirty way, lol.
A box of ECC88? What brand?

Ian
  Reply With Quote
Old 9th December 2017, 06:05 PM   #78
markusA is offline markusA  Sweden
diyAudio Member
 
markusA's Avatar
 
Join Date: Dec 2009
Location: Gothenburg
Oh, I wouldnít know. Itís the Russian 6n23p tube, a 6dj8 equivalent I believe.
It could quite possibly be a grab bag of different brands?
__________________
Current build to be found in my blog. Not so current anymore, everything is put on hold until march.
  Reply With Quote
Old 9th December 2017, 06:07 PM   #79
iancanada is offline iancanada  Canada
diyAudio Member
 
iancanada's Avatar
 
Join Date: Dec 2009
Location: Toronto
ES9018K2M, ES9028Q2M, 9038Q2M DSD/I2S DAC HATs for Raspberry Pi
Quote:
Originally Posted by markusA View Post
Oh, I wouldnít know. Itís the Russian 6n23p tube, a 6dj8 equivalent I believe.
It could quite possibly be a grab bag of different brands?
Good luck

Ian
  Reply With Quote
Old 9th December 2017, 06:18 PM   #80
markusA is offline markusA  Sweden
diyAudio Member
 
markusA's Avatar
 
Join Date: Dec 2009
Location: Gothenburg
Thanks, but first I need a DAC.
__________________
Current build to be found in my blog. Not so current anymore, everything is put on hold until march.
  Reply With Quote

Reply


ES9018K2M, ES9028Q2M, 9038Q2M DSD/I2S DAC HATs for Raspberry PiHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
I/V stage for ES9028Q2M gandolf Digital Line Level 25 9th February 2018 06:04 AM
ES9028Q2M balanced output evert1 Digital Line Level 0 26th December 2017 01:30 PM
ES9018K2M: best (Raspberry)-OS/384 khz + dtoverlay sfb2 PC Based 2 30th May 2017 09:43 PM
Hats off to Mr Pass, encouragement to new members Russellc Pass Labs 10 17th May 2015 08:23 AM
Hats make better bass? xstephanx Subwoofers 11 9th February 2005 10:59 AM


New To Site? Need Help?

All times are GMT. The time now is 02:32 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki