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

Headphone Systems Everything to do with Headphones

An inverting version of the NwAvGuy O2 headphone amp vs. the original: THD+N
An inverting version of the NwAvGuy O2 headphone amp vs. the original: THD+N
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 6th September 2015, 04:14 AM   #1
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
An inverting version of the NwAvGuy O2 headphone amp vs. the original: THD+N
Default An inverting version of the NwAvGuy O2 headphone amp vs. the original: THD+N

EDIT: 5/19/2017

I just realized that I never posted the Google Drive link to all the Inverting O2 Project materials! Here it is:

https://drive.google.com/drive/folde...00?usp=sharing

then the V3.61 folder. That contains the schematic, layout, bill of materials, build instructions, etc.

The original first post is below.

************************************************** ***********************************

Have you ever wondered what the result would have been if NwAvGuy had used inverting amplifier stages for the gain and output sections of his O2 Headphone Amplifier rather than non-inverting? Well wonder no more.

I've put together an inverting O2 version over the last few months and finally had a chance to run some THD+N tests on it vs. the original NwAvGuy O2. What I've expected would happen is cancellation of the higher-order even harmonics by the inverting stages at a cost of slightly higher background noise. The additional noise is due to the higher value of gain resistors (10K, 24.9K) required by an inverting amplifier (Johhnson noise and current noise). The input impedance is still 10K, but with an inverting amplifier that 10K is in series with the input. With a non-inverting amplifier it is from the input to ground, as in the orignal O2, with just 274 ohms or so in series for the RF filter.

Well, sure enough, that is exactly what happens! The plots below show some AC powerline noise leaking in via the fluorescent room lighting, which makes a good input test signal. The 4th harmonic at 240Hz is gone, as is the 6th at 360Hz, etc. The overall background noise has increased by about 10dB it appears. The gain chip in this one is a FET-input LME49880 which helps mitigate some current noise, as does the output stage which is a FET-input OPA2140 wrapped around BUF634s, set up as an inverting amp. I've also used the Vishay RN50C series 1/8W resistors throughout to lower current noise a bit.

Although this is the general format of the O2 headamp with (now inverting) gain stage feeding a pot in the middle (5K here for some reduced noise) and the (now inverting) output stage, it looks like that same overall design was patented at one point in 1984 using inverting stages for exactly the same reason, even harmonic cancellation:

https://www.google.com/patents/US4628278

Lol, I *think* that patent has expired, it has been more than 20 years, but I'll check further. The patent was from John Fluke Manufacturing, the DMM people.

* The first plot below shows a 10KHz test signal into the inverting version of the O2. The even harmonics at 240Hz (4th), 360Hz (6th), and 480Hz (8th) are gone but the noise floor is about 10dB higher than the stock O2 below.
* The second plot shows a 10KHz test signal into a standard O2 headamp. This one is built up exactly from NwAvGuy's BOM, no modifications or part substitutions of any kind. It is running on batteries for this test with no AC adaptor plugged in.
* The first photo shows the test setup. Signal generator output (one of Victor's low THD oscillators) BNC T'ed into a QA400 tester and then into one channel the DUT's input, either an O2 or the inverting O2 here.
* The second photos shows the two O2s. The inverting version is about an inch wider than the standard O2. Fits into the blue B5-080 case. The black one is the standard O2 in a B2-080 case.
Attached Images
File Type: png 10KHz Victor osc into inverting O2 red_is_out 3.png (50.6 KB, 1756 views)
File Type: png 10KHz Victor osc into standard O2 red_is_out 3.png (50.1 KB, 1718 views)
File Type: jpg IMG_3586.JPG (82.2 KB, 1717 views)
File Type: jpg IMG_3587.JPG (65.1 KB, 1717 views)
File Type: jpg inverting O2 schematic.jpg (459.5 KB, 1770 views)
File Type: jpg inverting O2 layout.jpg (992.8 KB, 822 views)

Last edited by agdr; 19th May 2017 at 09:47 PM.
  Reply With Quote
Old 6th September 2015, 04:31 AM   #2
Nrik is offline Nrik  Denmark
diyAudio Member
 
Join Date: Feb 2005
Location: Copenhagen
...Great job agdr.

I think I read a thread some time back here on diyaudio that inverting amps has a proven better performance, but I can't find anything.
Since I am also doing some experimenting with inverted vs noninverted amplifier circuits, I want to investigate the theory about this a little deeper...can you point me in a direction?

...and now the important question:How does it sound compared to the original? :-)
__________________
Regards
Henrik
  Reply With Quote
Old 6th September 2015, 04:46 AM   #3
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
An inverting version of the NwAvGuy O2 headphone amp vs. the original: THD+N
Nrik - thanks! I really haven't found a lot out there via Google searching on inverting amps.

One of the big benefits is elimination of common-mode distortion since both op-amp inputs stay at zero volts with respect to ground. sgrossklass did an excellent post about that in a thread months ago, or maybe even was last year, but I probably couldn't find it now.

And it sounds great! I'm still trying to determine if I can actually hear any audible difference vs. the standard O2 though. Goes back to NwAvGuy's "sighted bias" comments. To do it right I really should set up a blind A/B test sometime.
  Reply With Quote
Old 6th September 2015, 04:48 AM   #4
jcx is offline jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
what's the connection of the current circuit with NwAvGuy/Rocket Scientist? - I really don't see it



patents expire in 20 years now, formerly US patents ran 17 years from issue date - anything from 1984 is thoroughly public domain - no added research needed
  Reply With Quote
Old 6th September 2015, 03:27 PM   #5
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
An inverting version of the NwAvGuy O2 headphone amp vs. the original: THD+N
jcx - thanks for the confirm on the patent lifespan.

Yeah just the overall topology is simlar to the O2. The gain stage followed by the pot in the middle followed by unity gain (-1 in this case) current buffer output stage (I spotted a typo on the print, that should be a 24.9K feedback resistor in the output stage for unity gain), two NiMH rechargeable batteries with a power management circuit to cut then off when low, a voltage doubler power supply simiar to the O2 (I've used heat sinked LDOs here though), and fits into a similar case.

But it is different enough that it probably wouldn't be considered a "derivative" as per the O2's license. I've come to learn that for the "derivative" test to hold a circuit apparently has to be extemely similar. Plus he disappeared years ago and is non-contactable. I'm told that in several countries that alone is enough for a license to revert to the public domain (non contactable for more than 1 year).

I'll post the Gerber files once it is done and tested, probably a couple of months from now, if anyone wants to build one. The photos are V2.0 but the schematic is a V3.0 that is still being fiddled with. George's ODAC fits in the battery area in place of the batteries too, same as the O2. V3.0 has an input select switch the ODAC can be sent through.

Although the FET-input chips help with current noise they have higher voltage noise than bipolar, of course. I want to try building one up with (bipolar) OPA1612 chips instead and see where the noise floor lands.

Last edited by agdr; 6th September 2015 at 03:39 PM.
  Reply With Quote
Old 7th September 2015, 05:51 PM   #6
CHiroshi is offline CHiroshi  Sweden
diyAudio Member
 
CHiroshi's Avatar
 
Join Date: Nov 2014
Location: Stockholm
How exciting agdr! I saw the article on your web a while ago, then it disappeared! Happy to see you kept working on that.

I guess we'll all have to practice surface mount soldering!
Opamp rolling will be a bit trickier too... but WTH?!

Seriously, do you have any estimates about price, bare board or with just the surface mount components already soldered?
  Reply With Quote
Old 7th September 2015, 05:55 PM   #7
Turbon is offline Turbon  Sweden
diyAudio Member
 
Turbon's Avatar
 
Join Date: Aug 2011
Location: EU-Southern part of Sweden, Europe.
Hopefully its just about to shovel the ic's around a bit

... Ok, I'll have another go

Regards

Last edited by Turbon; 7th September 2015 at 06:02 PM.
  Reply With Quote
Old 8th September 2015, 01:58 AM   #8
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
An inverting version of the NwAvGuy O2 headphone amp vs. the original: THD+N
Hi guys!

Yep if anyone should be interested in this amp once I get the final version tested I'll sell the PC board at-cost to forum members, plus I'll post the Gerbers if anyone wants to fab their own board. I mainly did this one for fun, just out of curiousity wanting to hear what two inverting stages would sound like.

Way back when, a month or so after the O2 was released, I asked NwAvGuy via PM if he had considered inverting stages for the O2 headamp. Since the O2 has two stages it works out especially well, you wind up with two inversions = non-inverting output. I can't remember his his exact reply now, but he had thought about it.

The total parts price so far on the BOM looks like about $70 at Mouser, not including the transformer and case. The parts for the low-battery cutoff power management circuit can be left off on this one if not using batteries, which saves some money. The new per-battery PM circuit uses a couple of new-generation optical solid state relay chips (just 0.015 ohms "on", 1/20th of the O2's mosfets!) that are $6 each at Mouser, so that is $12 of the $70 right there.

I've debated a bit whether to even post this amp, given all the controversy over NwAvGuy's license. Lol, adydula has been good enough to listen to some of my rantings on the issue via email over the last couple of months. I was glad to see jcx' posting. If folks can't even really see a correlation to the O2 that is great news in terms of the amp not being a "derivative"!

The whole amp is largely through-hole, except the power management circuit. On this amp the PM circuit is *only* for the batteries, to cut them off when low. The AC power doesn't go through it at all. This amp has a separate headphone relay, like my O2 Booster Board, to take care of thumps. NwAvGuy tried to make his PM circuit do both low battery cutoff and thump elimination. In my view it didn't quite succeed well at either.

So if you are not using batteries then all those power management surface mount parts (the stuff under the PCB under the batteries on the layout) can be left off. The other surface mount parts are the two op-amps (all the op-amps "better" than the NJM2068 are surface mount, it seems), two transistors under the board for current sources for the on/off LEDs, and 2.2uF caps across the .22uF film caps. Those latter 2.2's could also be done as through hole parts under the board though.

I "corrected" a bunch of things about the O2 that had been bugging me ever since it was released. All the good stuff from the O2 Booster Board is rolled into the amp too. I used BUF634's as the output current buffers rather than LME49600's because they come in TO-220 and I can (and do) heat sink them.

Last edited by agdr; 8th September 2015 at 02:09 AM.
  Reply With Quote
Old 8th September 2015, 10:47 AM   #9
Turbon is offline Turbon  Sweden
diyAudio Member
 
Turbon's Avatar
 
Join Date: Aug 2011
Location: EU-Southern part of Sweden, Europe.
Yep, go for it agdr

I remember testing things regarding the low battery problem. Never got it solved though. The obvious solution came from you agdr - oda with more parts .


Regarding the non-inverting vs inverting versions I expect not to hear any difference unless you decide to change the sonic signature. The performance has gone beyond what is hearable... Which reminds me of control over the gain at different frequensies :P... Still at the top of life at 55 I suppose it is showing that I take planning for the future seriously ...

Regards
  Reply With Quote
Old 8th September 2015, 01:58 PM   #10
adydula is offline adydula  United States
diyAudio Member
 
Join Date: Aug 2010
Absolutely great news!!!

I am glad you published this.... in the spirit of " I wonder if...." LOL!!

This is one amp that I would build just to listen to it in comparison to the original O2 and ODA.

Also would like to listen to it in comparison to my fairly new BH Crack, which is a tube amp full of THD!

Alex
  Reply With Quote

Reply


An inverting version of the NwAvGuy O2 headphone amp vs. the original: THD+NHide 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
O2 headamp output booster PCB agdr Headphone Systems 409 16th October 2018 04:49 PM
NwAvGuy O2 and ODAC: measurements versus listening Expansion2 Digital Line Level 9 29th November 2017 03:38 AM
O2 headamp PM latch PC board to stop O2 low battery buzzing agdr Headphone Systems 0 24th April 2014 01:20 AM
FS - Fully populated & working O2 headamp ThuyTN Swap Meet 0 16th January 2013 03:30 AM
WTT O2 headamp PCB for CCS PCBs regal Swap Meet 0 13th October 2011 04:42 AM


New To Site? Need Help?

All times are GMT. The time now is 03:36 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki