Build thread for ES9023 + JG Buffer boards (betatest) - Page 5 - 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 30th March 2013, 08:07 AM   #41
grufti is offline grufti  United States
wombat
diyAudio Member
 
grufti's Avatar
 
Join Date: Jun 2004
Location: Hillside
Nice!

I'm still listening to music on the test setup, having fun with it.
  Reply With Quote
Old 30th March 2013, 08:09 AM   #42
grufti is offline grufti  United States
wombat
diyAudio Member
 
grufti's Avatar
 
Join Date: Jun 2004
Location: Hillside
What about the charge pump cap[s]? I like the new XO placement a lot. I also like the now missing optional C103/C104 as you can tell from my beta build.

Last edited by grufti; 30th March 2013 at 08:16 AM. Reason: more
  Reply With Quote
Old 30th March 2013, 08:26 PM   #43
diyAudio Member
 
Join Date: Feb 2011
Location: Madison, Wisconsin USA
Default What Charge Pump? [kidding]

Quote:
Originally Posted by grufti View Post
What about the charge pump cap[s]? I like the new XO placement a lot. I also like the now missing optional C103/C104 as you can tell from my beta build.
I did it! I have -3.4V on the ES9023 NEG pin, no charge-pump cap. I do not have anything but a VOM handy so I hesitate to tweak the -V any more than I have. The SOUND? As good or better than before I started, and that is with no filtering on the POT wiper {yet}

9V battery [8.something] --> 79L05 --> 100R multi-turn POT, wiper set under load to -3.4Vdc [my prior NEG with +3.6 source and 100K to GND on ES9026 pin #6]

From this experience I would say it is worth determining what is the best negative voltage to place there. Now, mind you, my UAE23 device under test has 100K ohms on the Vreg pin ES9023 pin #6 to ground -- I could remove THAT however that is something I pretty much like, not having tried ESS suggestion of 130K or ODAC value of 118K there [100K is what I had when I did it].

SO that is another test yet to consider, a temporary POT {wired as variable R} and fixed-R on ES9023 pin #6 to ground so could have say 75K--200K test range. Test with a very "hot" recording like a female vocalist up close on the microphone example Musica Nuda.

I am currently playing The Bridge by Jens KrÜger and it sounds superb. The FET tester is not getting it's 9V back anytime soon
  Reply With Quote
Old 30th March 2013, 09:47 PM   #44
diyAudio Member
 
Join Date: Aug 2010
Quote:
Originally Posted by grufti View Post
What about the charge pump cap[s]? I like the new XO placement a lot. I also like the now missing optional C103/C104 as you can tell from my beta build.
Since it turned out () that a suitable pad layout for the ES9023 was not available I made a new one. Some pads are a little bit longer in order to put Cs directly at the pins For the charge pump 1-2.2µF X7R (0402 or 0603) is perfectly suited.

Have a look at the layout for the XO decoupling and grounding. The small cap C22 for local decoupling (10nF COG/NPO) is as close a possible to VDD. GND is routed with a low impedance trace to the Cap. Note that only the bigger cap C2 (e.g. 2.2µF) is connected to the groundplane. The XO will draw current during rising or trailing edge. This will be delivered only from C22 (C2 is too slow and too far away) and thus the current will flow from C22 through the XO and back to C22. All the noise will stay there and not affect the main ground! I used the same layout for the flipflops

I omitted the TH foil caps (103, 104) since I am convinced that the SMD PPS caps are really good here and this gave some more room for improvements on the lower board area

BTW: I will clean the part numbering for the final version (e.g. order part numbers according to placement within schematic). Still a lot of work...

Have fun!
Attached Images
File Type: jpg ES9023 new pad layout.jpg (46.6 KB, 667 views)
File Type: jpg XO new layout.jpg (67.1 KB, 509 views)
  Reply With Quote
Old 30th March 2013, 10:54 PM   #45
diyAudio Member
 
Join Date: Feb 2011
Location: Madison, Wisconsin USA
As to the charge-pump and noise, as I told Daniel in another thread, I'll try and look for the prior noise he saw on the NEG pin around 1.5MHz...{amplitude unknown} Daniel expects there could be CP-related noise yet, with CP cap missing and NEG fed with -3.4Vdc. Only one way to tell, I have to look. I tried to Google charge-pump circuits and found "too many". I can only guess what is inside ES9023. I had another odd thought, not knowing the circuit inside....could I use a cap to ground on one or both of the CP pins I depopulated for the external NEG test? SWAMP it? 100R to GND? It's only a test and a crude one. I have a stack of SMD caps on the NEG pin to ground, ok, some mm away, I COULD make a very short path NEG --||--GND top-side on the UAE23 [another model of ES9023-based DAC sold on eBAy]. Maybe 0.01uF? .1? I can haul a scope up to the living room but maybe not on a Saturday evening, I've come too far not to get an answer as to the nature of the NEG pin and noise.
  Reply With Quote
Old 30th March 2013, 11:20 PM   #46
grufti is offline grufti  United States
wombat
diyAudio Member
 
grufti's Avatar
 
Join Date: Jun 2004
Location: Hillside
Good reasons to drop C103, C104. I had noticed the close proximity of C2, C22. That sounds like a good way to go, possibly even overkill, but that would not hurt. The longer pads around the ES9023 are a neat idea. I can't wait to try that version.

One additional question: how about moving the u.fl for MCLK closer to the ES9023 into the left part of the open space above the three other u.fl connectors? They don't really need to be in one line and that way the distance to the XO / ES9023 would be even shorter.
  Reply With Quote
Old 30th March 2013, 11:52 PM   #47
diyAudio Member
 
Join Date: Feb 2011
Location: Madison, Wisconsin USA
Ferrite bead in series to V+ to OSC? Should I2s lines have same length if MCK is used? {USB receiver}. I LOVE the new custom ES9023 pads!! Would ES9023 benefit in any way from a ground plate on top of it as shown in ES9018 thread? A swatch of copper foil and some GND pads to tack it to topside?
  Reply With Quote
Old 31st March 2013, 08:27 AM   #48
diyAudio Member
 
Join Date: Aug 2010
Quote:
Originally Posted by grufti View Post
Good reasons to drop C103, C104. I had noticed the close proximity of C2, C22. That sounds like a good way to go, possibly even overkill, but that would not hurt. The longer pads around the ES9023 are a neat idea. I can't wait to try that version.

One additional question: how about moving the u.fl for MCLK closer to the ES9023 into the left part of the open space above the three other u.fl connectors? They don't really need to be in one line and that way the distance to the XO / ES9023 would be even shorter.
Don't think that C2 + C22 are overkill. You'll need a fast cap (e.g. <100nF) to serve the current need of the XO (3ns rise time -> 333MHz + harmonics up to GHz!) and with this small cap the reg will not be stable (at least MIC needs some ESR as you also mentioned -> Tantal). Also reg is much too slow to reload the small cap between the transitions.
Moving the u.fl is an option that i will check. Please note that I designed the traces for the I2S lines to have 50Ohm impedance (trace thickness, gap between trace and GND, board thickness,...) so the lenght of the trace will only add some (minor) damping. BTW: if you don't need the external MCLK you should cut the trace close to the ES9023 to avoid reflections

Quote:
Ferrite bead in series to V+ to OSC? Should I2s lines have same length if MCK is used? {USB receiver}. I LOVE the new custom ES9023 pads!! Would ES9023 benefit in any way from a ground plate on top of it as shown in ES9018 thread? A swatch of copper foil and some GND pads to tack it to topside?
Ferrite beads together with close Cs are used to clean power just before the reg and to prevent dirt e.g. coming from the XO to spread to other parts (the reg is "open" for HF). Frequencies on I2S lines are still quite low (<< GHz) so the same length should not be necessary here. Impedance is more important. GND isn't routed yet and will be also on top layer under ES9023. Though directly on board and not using a copper foil Thats a point I am just thinking over anyway: the routing of VDD to mute pin. This might be reworked...

Last edited by curryman; 31st March 2013 at 08:29 AM.
  Reply With Quote
Old 31st March 2013, 05:57 PM   #49
grufti is offline grufti  United States
wombat
diyAudio Member
 
grufti's Avatar
 
Join Date: Jun 2004
Location: Hillside
It's always difficult to communicate glitch free on a forum like this. I didn't mean to suggest overkill. I really like the C2 , C22 arrangement as proposed. As far as I can tell it could handle frequencies many times higher than the ones we're dealing with here. That's where the idea came in.
  Reply With Quote
Old 31st March 2013, 06:08 PM   #50
diyAudio Member
 
Join Date: Aug 2010
Everythings OK, just wanted to explain a bit what may be obvious to most of us here but perhaps not to all

Last edited by curryman; 31st March 2013 at 06:09 PM. Reason: wording
  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
Fastron Inductor Group Buy for JG Filter Buffer pchw Group Buys 176 8th February 2013 09:19 PM
ES9023 DAC build heartwinter Vendor's Bazaar 8 7th July 2012 09:27 AM
Preamp / buffer suggestions for TA2020 / TA2024 boards lossfound Class D 11 29th January 2012 10:35 AM
ES9023 and buffer before 10K pot. Neutrality Digital Line Level 1 12th December 2011 07:00 PM
DIY Op Amp Buffer Boards aziltz Swap Meet 20 18th October 2011 12:35 AM


New To Site? Need Help?

All times are GMT. The time now is 03:57 AM.


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