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

Equipment & Tools From test equipment to hand tools

Behringer UMC 202HD for measurements
Behringer UMC 202HD for measurements
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 23rd August 2019, 10:03 PM   #11
RNMarsh is offline RNMarsh  United States
diyAudio Member
 
RNMarsh's Avatar
 
Join Date: May 2012
Location: 2457 Cascade Trail; Cool, CA. 95614
Quote:
Originally Posted by MagicBus View Post
Bingo! I had the feeling that it had to do with Vcom. The ADC inputs need DC bias at Vcom level but apparently not the actual Vcom, not even buffered. There is a buffer for this on board. It's the half of a RC4580 dual op amp. The other half is also a DC buffer but referenced to a resistance voltage divider from power rail. I couldn't trace where this is going but I simply lifted the other from Vcom and fed it from the voltage divider too. Seems to work! The high frequency peak is still there but it doesn't dominate the spectrum any more.


While working on this, I tried the ADC without DC bias at all. Distortion rises dramatically and it is not about clipping. It seems that the DC bias at half the rail voltage keeps the balance internally in the digital domain. At least this is how I understand it, maybe it's something else. Anyway, all this gave me the idea to replace the fixed voltage divider with a pot to allow trimming "on the fly". I'll check this and also optimize the input capacitors of the SE to BAL converter because this has to do with 50Hz peak.






-RNM
  Reply With Quote
Old 24th August 2019, 04:55 AM   #12
1audio is online now 1audio  United States
diyAudio Member
 
Join Date: Mar 2004
Location: SF Bay Area
Behringer UMC 202HD for measurements
There is lots on the UMC204HD here Budget DAC Review: behringer UMC204HD | Audio Science Review (ASR) Forum which probably apply to the UH202 as well.

Sometimes the odd spurs come from the host PC Vbuss or it not being quite strong enough. Check that its at 4.75 or more at the UH202 end of the cable. Or the peak may be crosstalk from the USB communications which are at around an 8 KHz rate but the data modulate the rate so it moves.

It is basically great performance for not much money. For 6X more you get a 10% improvement. For 20X a 15% improvement and for 250X a 15.5% improvement depending on how you value the improvement.

Now the hard part- learning what to make of what you see with this tool.
__________________
Demian Martin
Product Design Services
  Reply With Quote
Old 24th August 2019, 09:17 AM   #13
MagicBus is offline MagicBus  Greece
diyAudio Member
 
MagicBus's Avatar
 
Join Date: Aug 2016
Location: Athens, Greece
Thank you all!



@1audio: I came across this thread, it was not all very clear to me but I checked the power supply. From USB Vbus it comes exactly 4,75V but after on board filtration, 4,5V is what all the chips see. I hacked a USB cable and fed the soundcard with external psu. I forced it up to 5,5V real Vcc at the chips but I didn't see any difference.


As I said in the first post, my needs are bellow the average. Even if it was not for the money, the more advanced equipment would simply left unused on the shelf.

Last edited by MagicBus; 24th August 2019 at 09:26 AM.
  Reply With Quote
Old 25th August 2019, 01:23 PM   #14
MagicBus is offline MagicBus  Greece
diyAudio Member
 
MagicBus's Avatar
 
Join Date: Aug 2016
Location: Athens, Greece
Quote:
Originally Posted by MagicBus View Post
...Anyway, all this gave me the idea to replace the fixed voltage divider with a pot to allow trimming "on the fly"...

I tried it. Yes, it does allow direct control on THD but I didn't manage to get any better figures than with the fixed voltage divider. So, an unnecessary complexity. The project is pretty much completed. I'll take some good photos to demonstrate the tweaks. It turned out to be fairly easy!
Attached Images
File Type: jpg untitled.JPG (139.0 KB, 183 views)
  Reply With Quote
Old 25th August 2019, 02:23 PM   #15
MagicBus is offline MagicBus  Greece
diyAudio Member
 
MagicBus's Avatar
 
Join Date: Aug 2016
Location: Athens, Greece
Default How to do it!

Remove the screws that secure the input jacks in the front panel and the knobs. Remove the rear panel too -another three screws. Then the pcb slides out of the chassis. The photo shows what I have. I'm not aware if there have been any revisions so please compare with yours. The areas of interested are marked. The audio codec (CS4272), the ADC input buffer (AD8694), its coupling capacitors and the Vcom buffer (behind the electrolytic cap). Right in front of that cap is the copper trace that needs to be cut. Three steps to improve performance, each one independent from the others.
Attached Images
File Type: jpg Modifications.JPG (629.6 KB, 171 views)
  Reply With Quote
Old 25th August 2019, 02:54 PM   #16
MagicBus is offline MagicBus  Greece
diyAudio Member
 
MagicBus's Avatar
 
Join Date: Aug 2016
Location: Athens, Greece
Default Step #1: Access to ADC input buffer.

Remove the coupling electrolytics and solder wires in their pads. I used cat5 solid core. In the picture, the colored wires are what comes from the mic preamps and the white are what goes to input buffer.
Behringer UMC 202HD for measurements-coupling-caps-coloured-preamps-jpg


I wired them as shown in the schematic.
Behringer UMC 202HD for measurements-wiring-jpg


This way when nothing is inserted, the circuit is restored and all front panel inputs and controls are fully functional. When the jack goes in, the mic preamps are bypassed. Please resist the temptation to use audiophile caps. We need at least 10μF in compact size. I used the original 47μF but with polarity reversed. Be sure you do that otherwise distortion and noise go up. The 100k resistors are good to add. The question is what to do with the ground. I think it should be left floating. It makes sense for loop measurements. However, it doesn't seem to have any impact in performance if attached to ground. Finally, I drilled a hole in the rear panel to hold the jack. That's the only sign of the modifications.
Attached Images
File Type: jpg Coupling Caps-Coloured Preamps out.jpg (628.3 KB, 365 views)
File Type: jpg Direct Input.jpg (580.7 KB, 50 views)
File Type: jpg Wiring.JPG (32.3 KB, 280 views)
File Type: jpg Rear.JPG (50.7 KB, 48 views)
  Reply With Quote
Old 25th August 2019, 03:10 PM   #17
MagicBus is offline MagicBus  Greece
diyAudio Member
 
MagicBus's Avatar
 
Join Date: Aug 2016
Location: Athens, Greece
Default Step #2: Improve ADC input buffer performance.

SMD job! You can do it! I did it with an ordinary soldering iron. And I reviewed it many times the following night in my dreams...
The picture is a close up of the input buffer. The colored frames is what needs to be removed. In blue are the resistors that are in parallel with the ADC inputs -see first post- and they will stay open. Easy! In green are the the op amp's output resistors. We need something there but not the original 1k. The lower the better and I've seen as low as 22ohm in some schematics. I used 91ohm salvaged from another pcb. There won't be a close photo of my smd rework and that's final!
Attached Images
File Type: jpg Green Buffer Out.JPG (644.0 KB, 81 views)
  Reply With Quote
Old 25th August 2019, 03:22 PM   #18
MagicBus is offline MagicBus  Greece
diyAudio Member
 
MagicBus's Avatar
 
Join Date: Aug 2016
Location: Athens, Greece
Default Step #3: Kill the high frequency peak.

You need a sharp cutter for this. Cut the trace that goes to the Vcom buffer close to pin #5. I cut it at the other side too, close to the ADC. It may not be of any importance though. If you do that make sure you leave intact the decoupling caps at pin #15. Then on the Vcom buffer just bridge pin#5 with pin#3. You should do it with the shortest wire that leans on the op amp, otherwise... noise.


That's all!


PS. Be gentle with the cutter. Although I haven't verified, this is likely a multilayer pcb. Don't go too deep!
Attached Images
File Type: jpg Vcom Buffer.jpg (644.9 KB, 80 views)

Last edited by MagicBus; 25th August 2019 at 03:24 PM.
  Reply With Quote
Old 25th August 2019, 03:37 PM   #19
phofman is offline phofman  Czech Republic
diyAudio Member
 
Join Date: Apr 2005
Location: Pilsen
Very interesting, thanks a lot for sharing! Do I understand correctly there was no active component replacement, only removing/changing resistors and capacitors?

The lowered output resistors - another option would be keeping the existing resistor and soldering a new one on top.
  Reply With Quote
Old 25th August 2019, 03:49 PM   #20
MagicBus is offline MagicBus  Greece
diyAudio Member
 
MagicBus's Avatar
 
Join Date: Aug 2016
Location: Athens, Greece
Yes, only 6 smd resistors and 2 electrolytic capacitors. And a few wires. You are right about paralleling the output resistors. Another thing I did in my preliminary tests was to use through hole resistors directly from the op amp -solered at pins 1,7,8,14- to the ADC - right on the antialiasing capacitors. No noise problems. Unfortunately, I haven't kept a photo of that.
  Reply With Quote

Reply


Behringer UMC 202HD for measurementsHide 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
Behringer iNuke NU3000 Measurements bolserst Class D 126 31st October 2019 03:05 PM
Usefulness of In-Room LF Measurements vs. No Measurements (ref FH) sayrum Full Range 66 19th February 2015 09:44 PM
FS: Behringer B2031A Pair (Some say they beat Orion!) and Behringer DCX2496 bmwparts Swap Meet 10 17th December 2014 02:58 PM
Behringer USB mixer for speaker measurements? tcpip Multi-Way 22 13th April 2014 04:30 AM
SW measurements Grandma´s_SUB Everything Else 9 10th January 2005 02:34 PM


New To Site? Need Help?

All times are GMT. The time now is 02:01 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